Evidence Based Standards For Measuring Nurse Staffing and Performance

Evidence-based Standards for Measuring Nurse Staffing and Performance Project # RC1 0621 06 Report for the Canadian
an Health Services Research Foundation December, 2003 Revised and Resubmitted, September, 2004
Prepared by Linda OBrien-Pallas, RN, PhD Donna Thomson, RN, MBA Linda McGillis Hall, RN, PhD George Pink, PhD Mickey Kerr, PhD Sping Wang, PhD Xiaoqiang Li, PhD Raquel Meyer, RN, PhD Student
Acknowledgements
The investigators wish to thank the Canadian Health Services Research Foundation, the Ontario Hospital Association Change Foundation, the nursing effectiveness, utilization, and outcomes research unit of the faculty of nursing at the University of Toronto, and the contributing hospitals for the financial support that made this research project possible. The advisory committee members are acknowledged for their guidance in the development of the data collection tools and for their assistance in interpreting the results and their input on the feasibility of collecting significant data elements on an ongoing basis. Dr. Judith Shamian Health Canada Kathleen MacMillan Health Canada Jill Strachan Canadian Institute for Health Information Barbara McGill and Nancy Savage Atlantic Health Sciences Corporation Jane Moser University Health Network David McNeil Sudbury Regional Hospital Margaret Keatings Hamilton Health Sciences Heather Sherrard Ottawa Heart Institute Carol Wong London Health Sciences Centre Lucille Auffrey Canadian Nurses Association Sue Williams Ontario Joint Provincial Nursing Committee Beverly Tedford New Brunswick Department of Health and Wellness Sue Matthews Ontario Ministry of Health and Long-Term Care Hospital and site co-ordinators and data collectors are recognized for their efforts to collect comprehensive and accurate data about their organization, patients, and nurses in order to support this project. Staff and patients at participating hospitals are thanked for their willingness to participate in this study by completing surveys. Health records departments are thanked for providing patient-specific diagnoses and outcomes. Hospitals and Site Co-ordinators: Sudbury Regional Hospital: Claire Gignac London Health Sciences Centre: Nancy Hilborn University Health Network: Elke Ruthig Hamilton Health Sciences: Bernice King Atlantic Health Sciences Corporation: Trevor Fotheringham Ottawa Heart Institute: Judith Sellick A special thank you is given to project co-ordinators Shirliana Bruce and Min Zhang and research assistant Irene Cheung.
Evidence-based Staffing
Key Implications for Decision Makers

Variations in nursing productivity/utilization and staffing patterns are frequently observed between, as well as within, hospitals. Decision makers are challenged to maximize productivity/utilization and minimize staffing costs, while ensuring the quality of care. Recommendations from this study inform decision-making on these important issues within hospital cardiac and cardiovascular units. Nursing unit productivity/utilization levels should target 85 percent, plus or minus five percent. Levels higher than this lead to higher costs, poorer patient care, and poorer nurse outcomes. Maximum productivity/utilization is 93 percent (because seven percent of the shift is made up of paid, mandatory breaks). Units where nurses frequently work at or beyond maximum productivity/utilization must urgently reduce productivity/utilization and implement acceptable standards. Productivity/Utilization targets can be met by enhancing nurse autonomy, reducing emotional exhaustion, and having enough staff to cope with rapidly changing patient conditions. Overall costs are reduced when experienced nurses are retained. Retention is more likely when there is job security, when nurses can work to their full scope of practice, and when productivity/utilization levels are below 83 percent. Retention strategies must address the physical and mental health of nurses, balancing the efforts and rewards associated with work, nurse autonomy, full scope of practice, managerial relationships, innovative work schedules, hiring more nurses into full-time permanent positions, and reasonable nurse-to-patient ratios based on targeted productivity/utilization standards. These will minimize the effect of persistently high job demands and reduce absenteeism and the use of overtime. Investment is needed for infrastructure to collect data that will monitor and improve care delivery processes and measurement of performance outcomes. Data that should be routinely captured, but are not yet, include valid workload measurement; environmental complexity; patient nursing diagnoses and OMAHA ratings of knowledge, behaviour, and status; nurse and patient SF-12 health status; nurse to patient ratios; and productivity/utilization.
ii
Executive Summary
Policy makers and hospital administrators are seeking evidence to support nursing staffing decisions that includes both the volume and mix of nurses required to provide efficient and effective care. The principal objective of this study was to examine the interrelationships between variables thought to influence patient, nurse, and system outcomes. The results provide quality, evidence-based standards for adjusted ranges of nursing productivity/utilization and for staffing levels for patients receiving cardiac and cardiovascular nursing care. Although hospitals have little control over patient severity and complexity, organizations can manage nurse characteristics, system characteristics and behaviours, and environmental factors that influence patient, nurse, and system outcomes. Numerous findings provide important evidence to guide policy and management decisions related to the deployment and use of nursing personnel. These findings suggest that organizations can implement many strategies to improve the cost and quality of care. In the past, actions to minimize expenses have focused on reducing the cost of inputs, the number of nurses, and the skill level. The findings of this study suggest that to actually reduce the cost and improve the quality of patient care, organizations will benefit from 1) hiring experienced, full-time, baccalaureate-prepared nurses; 2) staffing enough nurses to meet workload demands; and 3) creating work environments that foster nurses mental and physical health, safety, security, and satisfaction. The evidence supports the need for a significant change in the way organizations view costs and suggests that the emphasis on cost of inputs should shift to the cost of outputs and the quality of care. The study found nursing productivity/utilization should be kept at 85 percent, plus or minus five percent. When rates rise above 80%, costs increase and quality of care decreases. Patient health is more likely to be improved at discharge if productivity/utilization levels are below 80 percent and
iii
if patients are cared for by nurses who work less overtime. When productivity/utilization levels are kept below 80 percent, nurses are more likely to be satisfied with their jobs and absenteeism is reduced, and nurses are less likely to want to leave their jobs when productivity/utilization is less than 83 percent. Costs are lower when hospitals maintain productivity/utilization levels below 90% and implement strategies to improve nurse health and incentives to retain experienced nurses. Autonomy can be enhanced by balancing the number of patients assigned to each nurse and each nursing unit, and emotional exhaustion is less likely when nurses are satisfied, mentally and physically healthy, and feel that they receive appropriate rewards for their efforts. Nurses are more likely to be physically healthy when there are good relationships with the physicians on the unit, and these relationships tend to improve when nurses autonomy and decision-making abilities are respected. Aggression- and violence-free workplaces are key to enabling nurses to do their nursing interventions on time. There also needs to be enough nursing staff to deal with the rapidly changing conditions in hospitalized patients, so that nurses have enough time to complete patient care. Patient care is improved when units are staffed with degree-prepared nurses and when nurses can work to their full scope of practice. This not only improves job satisfaction, but nurses are also less likely to leave their jobs. Patients health behaviour improves when nurses have a satisfying work environment, secure employment, and when unit productivity/utilization does not exceed 88 percent. Enhanced nurse autonomy, full-time employment, and fewer shift changes are shown to improve patients knowledge about their conditions when they are discharged.
iv
Table of Contents
Acknowledgements.......................................................................................................................... i Key Implications for Decision Makers ........................................................................................... ii Executive Summary ....................................................................................................................... iii I. Context......................................................................................................................................... 7 II. Implications................................................................................................................................ 8 System Implications.................................................................................................................... 9 Patient Implications .................................................................................................................. 13 Nursing Implications................................................................................................................. 13 III. Approach................................................................................................................................. 15 IV. Results..................................................................................................................................... 18 Descriptives............................................................................................................................... 18 Research Question 1. ................................................................................................................ 21 Intermediate System Outputs..........................................................................................21 Patient Outcomes............................................................................................................22 Nurse Outcomes..............................................................................................................24 System Outcomes............................................................................................................27 Research Question 2. ................................................................................................................ 30 Research Question 3. ................................................................................................................ 30 Research Question 4. ................................................................................................................ 31 V. Additional Resources ............................................................................................................... 32 VI. Further Research..................................................................................................................... 32 VII. References ............................................................................................................................. 32
Appendices
A. B. C. D. Annotated Bibliography...............................................................................................35 Patient Care Delivery Model.......................................................................................84 Tables.............................................................................................................................85 Instruments, Psychometric Properties, and Variables at Individual and Unit Levels............................................................................................................................117 E. F. G. Data Collection Forms................................................................................................124 Methods........................................................................................................................162 Descriptive Analyses...................................................................................................168
vi
I. Context Nurse staffing is closely linked to patient outcomes and system effectiveness. A greater understanding of the causes and outcomes of hospital nurse staffing is essential to meet increasing demands for both cost and quality accountability in healthcare. Recent Canadian reports highlight the urgent need to identify methods for valid measurement of nursing workload and productivity/utilization, and to understand their relationship with patient, nurse, and system outcomes,1,2,3,4 a need further underscored by the current and predicted nursing workforce shortages.2,5 Policy makers and hospital administrators are seeking evidence to support nursing staffing decisions that includes both the volume and mix of nurses required to provide efficient and effective care. Prior studies have provided insight into some of the factors contributing to the need for nurses and the effect of different staffing approaches on patients, providers, and systems (Appendix A). Recent evidence suggests that adding one patient to each nurses caseload in acute-care hospitals is associated with increases in 30-day mortality (seven percent), failure-torescue (seven percent), nurse burnout (23 percent), and job dissatisfaction (15 percent).6 Another study demonstrated that an increase of one hour of overtime per week increases the odds of a work-related injury by 70 percent.7 Part-time and casual employment can also negatively impact continuity of care and the nurses ability to influence clinical and work related decisions.8 A review of relevant studies is presented in Appendix A. The principal objective of this study was to examine the interrelationships between variables thought to influence patient, nurse, and system outcomes, in order to provide quality evidencebased standards for adjusted ranges of nursing productivity/utilization and for staffing levels for patients receiving cardiac and cardiovascular nursing care. This evidence will help policy makers
develop mechanisms and policies to measure the need for nursing service in light of appropriate staffing and productivity/utilization standards. By examining specific cardiac and cardiology diagnoses, as well as nurse and nursing work indicators within hospital cardiac and cardiovascular unit settings, this research study examined four questions: 1. To what extent do patient, nurse, and system characteristics and behaviours, and environmental complexity measures, explain variation in nursing worked hours and patient, nurse, and system outcomes, such as length of stay? 2. To what extent is there agreement between the estimates generated by a gold standard for measuring nursing resource needs (PRN workload methodology) and the worked hours of care per patient, and how does variance affect patient and nurse outcomes? 3. At what nurse-patient ratio and with what proportion of registered nurse worked hours are productivity/utilization and patient and nurse outcomes improved, after controlling for the influence of patient, nurse, organizational, and environmental factors? 4. Which data elements, in addition to those routinely collected within administrative databases, are critical for routine data collection in Canada? To what extent do policy and administrative decision makers support the feasibility of routine data collection? II. Implications Numerous findings provide important evidence to guide policy and management decisions related to the deployment and use of nursing personnel. Although hospitals have little control over patient severity and complexity, organizations can manage nurse characteristics, system characteristics and behaviours, and environmental factors that influence patient, nurse, and system outcomes. The implications of this study are directed at those latter factors, which are amenable to policy and management intervention.
System Implications 1. Results of this study suggest a target of 85 percent (plus or minus five percent) unit productivity/utilization on a daily basis. Sustained productivity/utilization outside this range will result in higher costs and poorer quality of care. Rationale: Different levels of unit productivity/utilization are associated with different outcomes as summarized in Table 1. Although the goal is to maximize nurse activity, at productivity/utilization levels above 80 percent, negative outcomes emerge because there arent enough nurses to meet demands. The maximum work capacity of any employee is 93 percent, because seven percent is allocated to paid breaks during which time no work is contractually expected. At 93 percent, nurses are working flat out with no flexibility to meet unanticipated demands or rapidly changing patient acuity. This study demonstrates that significant benefits, both fiscal and human, can be achieved by moderating productivity/utilization levels within a range of 85 percent, plus or minus five percent. It must be noted however, the suggested range may not be applicable to specialty units with variable patient flow demands, such as emergency and labour and delivery departments. Depending on performance goals, organizations may wish to target specific unit productivity/utilization values in Table 1. These values are cumulative in nature, such that, if a unit works at a 92 percent productivity/utilization level, not only will lengths of stay be longer, but all of the other negative outcomes that occur with productivity/utilization values below 92 percent will apply.
Table 1. Productivity/Utilization Levels and Associated Outcomes Productivity/Utilization Levels (%) > 91 > 90 > 88 > 85 > 83 > 80 Outcomes Longer length of stay Higher costs per resource intensity weight Less improvement in patient behaviour scores at discharge Higher nurse autonomy Deteriorated nurse relationships with physicians Higher intention to leave among nurses More nurse absenteeism Less improvement in patient physical health at discharge Less nurse job satisfaction
Although the Canadian Institute for Health Information defines productivity/utilization as workload over worked hours,9 this neither accounts for the quality and outcomes of care delivered, nor the impact of length of stay on total cost. This definition is not a measure of productivity/utilization as an output, but rather a measure of use as a process. Workload over worked hours actually measures use of nursing resources and thus evaluates an organizations ability to operate to meet patient care standards and needs. 2. Unit productivity/utilization levels below 90 percent, strategies to address nurse health, and incentives to retain experienced nurses who are expert in their field should lower resource intensity weight costs (the cost of providing services to groups of people with different characteristics). Rationale: Lower costs per resource intensity weight are associated with higher physical health scores for nurses, expert clinical practice, reduced length of stay, and unit productivity/utilization levels below 90 percent. 3. Attendance at pre-operative clinics as a routine process for surgical patients, adequate staffing to prevent medical problems, and unit productivity/utilization levels below 91 percent are recommended. Rationale: Shorter-than-expected length of stay is 185 percent
10
more likely when patients attend pre-operative clinics and 57 percent less likely when patients suffer medical problems as a consequence of their treatment. 4. Maintaining unit productivity/utilization levels below 90 percent and recognizing the effect of complex and numerous nursing diagnoses will optimize the actual worked hours per patient. Rationale: Increases in actual worked hours per patient are associated with increases in nursing worked hours and with higher numbers of nursing diagnoses. Actual patient care hours decline as unit productivity/utilization exceeds 90 percent and with increases in the proportion of both full-time nurses and average clinical expertise on the unit. 5. Efforts should be made to prevent adverse events to reduce overall costs. Rationale: Patients who suffer medical consequences are 319 percent more likely to be referred to homecare, and for each additional hour of care given, the patient is 13 percent more likely to suffer a medical consequence. 6. Staffing should be sufficient to account for the rapidly changing conditions in hospitalized patients so that all key nursing interventions can be done. Rationale: Patient interventions are more likely to be left undone when there are more unanticipated changes in patient acuity or when nurses experience violence. The likelihood of patient interventions not being completed increases by 260 percent for nurses at risk of feeling their efforts are not properly rewarded. 7. Providing innovative programs to create aggression-free work environments will enable nurses to complete key nursing interventions on time. Rationale: Delays in interventions are more likely when nurses on the unit experience violence, but they are 27 percent less likely for every 10 percent increase in degree-prepared nurses on the unit. 8. Efforts to improve the job satisfaction of nurses will lead to better ratings of quality of nursing care. Rationale: Nurse ratings of good/excellent quality of nursing care are 606
11
percent more likely when nurses rate the quality of patient care over the past year as improved and 159 percent more likely when nurses are satisfied. 9. Staffing units with degree-prepared nurses and ensuring that nurses can provide the quality nursing care that they deem appropriate will improve nurse perceptions of patient care quality over the last year. Rationale: Ratings of improved quality of patient care over the past year are 915 percent more likely when nurses report good/excellent quality of nursing care and are 40 percent more likely for every 10 percent increase in degree-prepared nurses on the unit. 10. Unit productivity/utilization levels should be kept below 80 percent, and work environments should be assessed to determine why there is higher absenteeism among full-time nurses. Rationale: Absenteeism is reduced when unit productivity/utilization remains below 80 percent. Full-time nurses are 152 percent more likely to be absent than those who work parttime or casually. Nurses who are physically healthy are five percent less likely to be absent. 11. Job security and allowing nurses with degrees to work to their full scope of practice will prevent nurses from leaving. Rationale: Intent to leave is 197 percent more likely among nurses who are concerned about job security and 101 percent more likely among degreeprepared nurses. As unit productivity/utilization exceeds 83 percent, intent to leave increases. However, intent to leave is 97 percent less likely for every 10 percent increase in proportion of nurse ratings of improved quality of nursing care on unit, 58 percent less likely when nurses are satisfied, and 51 percent less likely when nurses work full-time.
12
Patient Implications 12. Reducing overtime hours and unit productivity/utilization levels below 80 percent will improve patients physical status at discharge. Rationale: Improvements in patient SF-12 physical scores at discharge are 45 percent less likely when productivity/utilization exceeds 80 percent and seven percent less likely for each additional hour of nurse overtime. 13. Creating satisfying work environments, offering secure employment, and ensuring unit productivity/utilization does not exceed 88 percent enhances changes in patient behaviours related to nursing diagnoses. Rationale: Patient behaviour scores are more likely to decrease when unit productivity/utilization exceeds 88 percent. Improvements in patient behaviour scores at discharge are 176 percent more likely when nurses are satisfied but 53 percent less likely when nurses were forced to change units within the past year or anticipate forced changes in units in the next year. 14. Employing more nurses in full-time positions, facilitating autonomy, and reducing the frequency of shift changes improves patients knowledge about their conditions at discharge. Rationale: Improved patient knowledge scores at discharge are 74 percent more likely for every 10 percent increase in nurses worked hours on the unit and 24 percent more likely for every 10 percent increase in full-time nurses on the unit. Patient knowledge scores are 44 percent less likely to improve for every 10 percent increase in nurses on the unit with more than one shift change during the past two weeks. Nursing Implications 15. Ensuring sufficient numbers of nurses who are physically healthy and continuity of care providers, as well as facilitating autonomy and decision-making will improve nursephysician relationships. Rationale: Improved nurse-physician relationships are associated
13
with higher proportions of physically healthy nurses and increases in nurses hours worked on the unit. Deterioration in nurse-physician relationships is associated with unit productivity/utilization beyond 85 percent. 16. Finding balance between the number of patients assigned to a nurse, the rate of occupancy on the unit, and unit productivity/utilization is recommended to enhance autonomy. Rationale: Lower nurse autonomy is associated with higher unit occupancy rates, nurses experiencing effort and reward imbalance, more degree-prepared nurses, and greater nurse clinical expertise. Higher nurse autonomy is associated with unit productivity/utilization greater than 85 percent, nurse satisfaction, and higher nurse-patient ratios. 17. Hiring degree-prepared nurses, increasing average hours per patient, promoting autonomy, ensuring good quality nursing care, and maintaining unit productivity/utilization levels below 80 percent are recommended to improve nurse job satisfaction. Rationale: Higher nurse job satisfaction is 301 percent more likely when nurses rate the quality of nursing care as good or better, and 10 percent more likely for every hour increase in the average worked hours on the unit. Improved job satisfaction is also 56 percent more likely for every 10 percent increase of nurses with degree preparation and 24 percent more likely for each one point increase in ratings of nurse autonomy. Higher job satisfaction is 57 percent less likely when unit productivity/utilization levels exceed 80 percent. 18. Environmental scanning for factors that cause full-time nurses to be more emotionally exhausted is recommended. Rationale: Emotional exhaustion is 242 percent more likely when nurses experience effort and reward imbalance and 179 percent more likely when nurses work full-time. However, emotional exhaustion is 66 percent less likely when nurses are satisfied, 10 percent less likely with every one point increase in mental health scores, and
14
four percent less likely with every one point increase in physical health scores. For every 10 percent increase in satisfied nurses on the unit, nurses are 32 percent less likely to suffer from emotional exhaustion. 19. Improving nurse-physician relationships at the unit level, balancing the demands placed on nurses and the rewards they receive for their work, and enhancing job satisfaction will improve nurses physical health. Rationale: Nurses are 49 percent less likely to be physically healthy when they experience an effort and reward imbalance and 41 percent less likely to be physically healthy when they are emotionally exhausted. However, as relationships between nurses and physicians improve, nurses are more likely to be physically healthy. III. Approach This study, which comprised cross-sectional and longitudinal components, included the cardiac and cardiovascular care units of six hospitals in Ontario and New Brunswick. The Patient Care Delivery System Model10 was adapted for this study (Appendix B). This model emphasizes that characteristics of patients, nurses, and the system, as well as system behaviours, interact with communication and co-ordination, environmental complexity, and care delivery activities to produce system outputs (intermediate outputs include unit productivity/utilization and daily hours of care per patient; overall outputs include patient, nurse, and system outcomes) and provide feedback for the entire system. Ethical approval was received from the University of Toronto and from hospital sites. Patient and nurse consent was obtained on site. Eight hospitals met the inclusion criteria (high volumes of patients in the cardiac case mix groups of interest). The first six hospitals approached agreed to participate. Each hospitals chief nursing officer or designate joined the studys advisory
15
committee and became a local investigator to oversee hospital ethics approval, hiring of project staff, and data quality at the site. On participating units, data for study patients, all nurses, and the unit itself were collected on each patient for each day of stay. Data were collected from patients and nurses directly as well as from administrative sources. The key variables and data sources are summarized in Table 1 (Appendix C). A detailed summary of each measure and its related psychometric properties appears in Appendix D, and data collection forms are presented in Appendix E. In addition to this unit-level data, nurses completed a survey package questionnaire that addressed issues like burnout, the balance between work efforts and rewards, nurse-physician relationships, autonomy, and health. Nurses provided input into the PRN workload measurements, identification of nursing diagnoses, and ratings of patient knowledge, behaviour, and status. Data were collected between February and December 2002. Data collection periods averaged six months at each site to maximize the number of patients assessed, but the target of 145 patients for all specified case mix groups was not achieved. Inter-rater reliability on the application of all measures remained at 90 percent during orientation and throughout the study. Of 1,107 surveys provided to nurses at all six sites, 727 were returned (66 percent response rate). In total, 1,230 patients housed in 24 nursing units from the six hospitals were included in the full study, accounting for 8,113 patient days of data. Decision makers were involved in developing the proposal and reviewed all data collection forms and methods prior to implementation. They also reviewed drafts of the descriptive data for the studys final report. They made recommendations on additional data elements that should be routinely collected and assisted in the overall interpretation of the studys findings.
16
The findings will be published in peer reviewed and trade journals to target different audiences. The report, fact sheets, and a video will be sent to hospital executives, nongovernment bodies which influence health policy, and each ministry of health in Canada. Analysis Techniques: Data were analysed using SPSS version 11 and MLwin beta version 2.0. Initially, the distribution and transformation of variables was conducted. Descriptive statistics were compiled, and subscale scores and alpha reliabilities for the various research tools used were generated. Basic comparisons between hospitals or units were made using analysis of variance (ANOVA). Where applicable, the Pearson Product Moment Correlation was used to explore interrelationships between variables. Hierarchical linear modeling is useful for understanding relationships in multilevel structures. Since data in this study were collected at both the hospital unit level and at the individual nurse and patient level, a multilevel approach to the analysis was proposed as a way to better account for the possible clustering of effects within hospitals. That is, questionnaire responses from nurses within hospitals were likely to be affected by things that are fixed for all employees in that organization, such as the size and type of the organization. The advantage of hierarchical linear modeling methods is that they can account for this clustering or grouping of variation in scores on questionnaire measures within a given organization. Without accounting for the possible clustering of effects within hospitals, the conclusions of the study could be invalid, since other statistical measures assume that no such clustering occurs. For multilevel modeling, most variables were dichotomized and hierarchical logistic regressions were completed. Only unit productivity/utilization, worked hours per patient, cost per resource intensity weight, nurse-physician relationship, violence, and autonomy were kept as numeric variables. Worked hours per patient and cost per resource intensity weight were
17
logarithm transformed due to their highly skewed distributions. The order of entry of variables into the statistical modeling process was consistent with the theoretical framework at two levels. The first level included individual nurse and patient variables, while the second unit level included system characteristics and behaviours and throughput factors. Some of the nurse questionnaire measures were also aggregated to the unit level as a measurement of unit atmosphere or morale. Multicollinearity among independent variables was examined, but none of the variables was very strongly associated with any other. To determine whether or not variables were associated with outcomes, individual variables were sequentially added to statistical models and the properties of each newly expanded model were compared to the previous one to see if the new variable was of any importance (see Appendix F). IV. Results Descriptives Descriptive results pertaining directly to the implications outlined above are presented here. More detailed results and tables are presented in Appendix G. Patient Characteristics: For 1,230 patients in the study, the mean age was 63.5 years, and 66.7 percent were male. The most common cardiac case mix group was percutaneous transluminal coronary angioplast. Of the surgical patients, one-third (33 percent) attended a preoperative clinic and more than half (57.5 percent) received post-operative education. About one in 10 (10.9 percent) was referred to homecare. On a scale of 1 to 5, OMAHA knowledge, behaviour, and status scores regarding nursing diagnoses averaged 3.4, 4, and 3.3 respectively, upon admission or identification of new nursing diagnoses. At admission, 87 percent and 49.2 percent of patients scored below the standardized American norms for physical and mental health, respectively.
18
Nurse Characteristic: Of 727 nurses who completed the survey, most (93.9 percent) were female, registered nurses (96.6 percent), with a mean age of 40.6 years. More than 42 percent of nurses held a bachelor or higher degree. On average, 59.8 percent of nurses were employed fulltime, with 97.8 percent indicating permanent employment. Almost 40 percent of nurses rated their approach to care delivery as expert, rather than novice. System Characteristics and Behaviour: On an average day, nurses on each nursing unit admitted 6.1 and discharged 6.1 patients per 24 hour period. Overall, 64.3 percent of nurses reported significant increases in employer expectations for overtime in the last year and actual increases in overtime worked per week: zero to one hour (45.1 percent), two to four hours (32.2 percent), and greater than four hours (22.7 percent). Of the overtime reported, 26.7 percent was unpaid and 22.8 percent was involuntary. Eight percent of nurses experienced a forced change in their work unit in the previous year, and 15.1 percent anticipated such a change in the upcoming year. Nurses continue to perform tasks that could be delegated to non-nursing personnel, including ancillary services (83.5 percent), venipunctures (64.8 percent), housekeeping (55.1 percent), delivering trays (55.1 percent), and starting intravenous sites (51 percent). Intermediate System Output: Unit productivity/utilization was determined by dividing unit workload by total worked hours on the unit. The maximum capacity of any employee is 93 percent, because seven percent is allocated to paid breaks when no work is contractually expected. At 93 percent, nurses are working flat out with no flexibility to meet unanticipated demands or rapidly changing patient acuity. On 46 percent of the days, units worked beyond the ceiling value of 93 percent, and on 61.5 percent of the days units worked beyond 85 percent. Patient Outcomes: Few medical consequences were reported, although variation existed among hospitals. Medical consequences included falls with injury (0.7 percent), medication
19
errors with consequences (1.6 percent), death (0.4 percent), and complications such as urinary tract infections (1.5 percent), pneumonia (1.3 percent), wound infections (1.4 percent), bed sores (0.4 percent), and thrombosis (0.2 percent). Between admission and discharge, patients scores for SF-12 physical health status improved (41.1 percent) nearly as often as they declined (44.9 percent). A similar pattern was noted for patients mental health status (42.3 percent improving and 44.9 percent deteriorating). For physical and mental health status scores, 12.8 percent of patients showed no change. Overall, general improvement of patients was evidenced through mean changes in OMAHA knowledge (0.43), behaviour (0.25), and status (0.79) scores between admission and discharge (or appearance and resolution of new nursing diagnoses). Nurse Outcomes: On average, nurses scored 22.7 for emotional exhaustion, six for depersonalization, and 12.2 for personal accomplishment using Maslachs Burnout Inventory. Almost 30 percent of nurses were at risk for emotional burnout. Additionally, 18 percent of nurses said their work efforts exceeded work rewards. On average, 17.7 percent of nurses were dissatisfied with work, primarily due to inadequate opportunities to interact with management (45.5 percent). Of the nurse survey respondents, 34.8 percent and 49.2 percent scored below the standardized American norms for physical and mental health, respectively. During the two weeks preceding the survey, 32.4 percent of nurses changed their shift time more than once. During the week preceding the survey, nurses experienced emotional abuse (24.9 percent), threat of assault (13.6 percent), and physical assault (10.2 percent) while at work. The main sources of this workplace abuse were patients (31.1 percent), other nurses (21.5 percent), physicians (15.8 percent), and families (10.7 percent).
20
System Outcomes: Nurse ratings of quality of care and omission or delay of patient interventions comprised the measures of quality of care. Of 714 responses, 13.4 percent of nurses rated the nursing care quality on the last shift as fair/poor, while 41.9 percent said patient care quality had deteriorated over the last year. When faced with insufficient time, nurses generally omitted nursing (as opposed to physician-dependent) interventions. The most frequently omitted interventions included care planning (48.2 percent), comforting/talking (38.6 percent), back/skin care (31.4 percent), oral hygiene (28.7 percent), patient/family teaching (23.3 percent), and documentation (22.6 percent). Delayed interventions included vital signs/medications/dressings (37.3 percent), mobilization/turns (30.5 percent), call bell response (25.9 percent), and PRN pain medications (16.6 percent). In total, nurses reported missing 1,768 work episodes in the last year, with each episode averaging 2.42 shifts. Although 16.4 percent of nurses were never absent, frequency of missed episodes ranged from one to two (42.9 percent), three to four (25.2 percent), and greater than four (15.5 percent). Reasons for absenteeism were reported as physical health (71.4 percent), mental health (5.4 percent), injury (4.8 percent), and other (18.4 percent).Almost five percent of nurses planned to leave their job in the next year. Only 5.6 percent of nurses expected to have difficulty in securing a new job if they wanted one. Research Question 1. To what extent do patient, nurse, system characteristics and behaviours, and environmental complexity measures explain variation in nursing worked hours and patient, nurse, and system outcomes, such as length of stay? Intermediate System Outputs Unit productivity/utilization: As indicated earlier, at 93 percent productivity/utilization, nurses are working at maximum capacity, and high rates of productivity/utilization on the unit Evidence-based Staffing 21
directly influence patient outcomes. This analysis identifies the variables associated with higher and lower productivity/utilization at the unit level. Higher productivity/utilization levels were more likely when there were more nursing worked hours on the unit, higher nurse-to-patient ratios, higher nurse autonomy, and when nurses required more time to complete the work as specified by the patient care plan. Productivity/Utilization was more likely to be lower when units were specialized (such as units that only service patients with cardiology conditions) and where a higher proportion of nurses on the unit were emotionally exhausted or mentally healthy. When nurses are emotionally exhausted they may not be able to work at the same level of productivity/utilization than when they are not. Nurses who are mentally healthy may be inclined to say no to unrealistic work expectations. Actual Worked Hours per Patient: The actual worked hours per patient were likely to increase with a higher proportion of nursing worked hours on the unit and when patients had more nursing diagnoses. Increases in worked hours per patient were associated with increases in unit productivity/utilization up to the cut-off point of 90 percent. Units with more clinical expertise or with a higher proportion of full-time nurses were more likely to provide fewer hours of patient care. Patient Outcomes Tables 2 to 19 (Appendix C) display the variables modeled in relation to patient health and safety outcomes. Medical Consequences: Since there were so few medical consequences of any one type, all types of consequences were summed into one category. In this analysis, the factors associated with the presence or absence of any medical consequences during a patients stay were examined. As patients experienced greater numbers of nursing diagnoses, reflecting more
22
complex nursing needs, they were more likely to suffer medical consequences. Medical consequences were 53 percent more likely for each additional nursing diagnosis. In contrast, patients with better mental health at admission were less likely to have medical consequences. Patients who experienced medical consequences were more likely to require greater actual worked hours of nursing care during their stay and 319 percent more likely to be referred to homecare for follow-up after discharge, resulting in additional expense to the health system. OMAHA Knowledge, Behaviour, and Status at Discharge. Helping patients understand the cause and course of their conditions is seen to improve the overall health of patients. A ceiling effect was observed among the OMAHA knowledge, behaviour, and status scores, in that patients with higher scores at admission were less likely to demonstrate improvements in these scores at discharge (because there was less room for improvement). Improved patient knowledge scores at discharge were 74 percent more likely for every 10 percent increase in nursing worked hours on the unit and 24 percent more likely for every 10 percent increase in full-time nurses on the unit. When patients were cared for by nurses who reported higher autonomy in their jobs, they were more likely to show increases in knowledge about their condition at discharge. However, patient knowledge was 44 percent less likely to improve for every 10 percent increase in the proportion of nurses who had at least one shift change in the last two weeks. Helping patients understand which behaviours they need to change in order to improve their health status is another important role function of the nurse. When cared for by nurses who were very satisfied with their work, patients were 176 percent more likely to demonstrate improvements in their behaviour scores at discharge. Conversely, patients cared for by nurses with concerns about job security were 53 percent less likely to demonstrate improved behaviour
23
scores at discharge. Productivity/Utilization levels below 88.2 percent were associated with increased possibility of improvements in patients behaviour scores at discharge. SF-12 Health Status at Discharge: As with the OMAHA scores, patients with higher physical and mental health scores at admission were less likely to see improvements in these scores at discharge. Improvement in patients physical health status at discharge was less likely for patients with higher resource intensity weights and for patients with more nursing diagnosis. These two factors reflect the medical acuity and nursing complexity of patients needs for nursing care. Patient physical health scores were 45 percent less likely to improve when unit productivity/utilization exceeded 80 percent and were seven percent less likely to improve for each additional hour of nurse overtime. However, patients who scored higher in physical health status at admission were more likely to have improvements in mental health status at discharge. Patients who stayed longer in hospital were less likely to show improvements in mental health status scores at discharge. More hours of care were likely to be used if patient mental health was not improved at discharge.
Nurse Outcomes Although improving patient outcomes and reducing the risk of medical consequences are goals of healthcare, achievement of these goals may sometimes occur at the expense of nurse health and safety. In order to retain and recruit nurses senior and experienced nurses in particular understanding which factors influence nurse outcomes is pivotal. Ten nurse outcome variables derived from the literature were subsequently used in this analysis. Tables 10 to 19 (Appendix C) display the variables modeled in relation to nurse outcomes. Emotional Exhaustion: Physically and mentally healthy nurses were less likely to experience emotional exhaustion (burnout). The likelihood of emotional exhaustion increased by 242 Evidence-based Staffing 24
percent when nurses were at risk of an effort and reward imbalance and by 179 percent when nurses worked full-time. Nurses were 32 percent less likely to suffer high emotional exhaustion for every 10 percent increase in the proportion of satisfied nurses on units. Autonomy: Nurses reported higher autonomy in practice when they reported stronger relationships with physicians, were more satisfied with their job, or said the quality of patient care improved over the last year. Autonomy was also higher when patients had attended a preoperative clinic and when the nurse-patient ratio was high. As unit productivity/utilization exceeded 85 percent, nurses reported more autonomy, possibly since nurses have to make decisions on their own under such circumstances. However, lower autonomy scores were reported by degree-prepared nurses and by nurses who rated themselves as expert clinicians, perhaps due to organizational constraints imposed on their practice. When occupancy is high on the unit or when nurses were at risk of an effort and reward imbalance, autonomy was likely to be lower. Job Satisfaction: Nurses who were at risk for emotional exhaustion were 71 percent less likely to be satisfied with their jobs, and when unit productivity/utilization levels were higher than 80 percent, nursing staff were 57 percent less likely to be satisfied. Nurse satisfaction was 301 percent more likely when nurses rated the nursing care given on the last shift as good/excellent and 56 percent more likely among degree-prepared nurses. As the average hours available for care on the unit increased and when nurses autonomy increased, so did nurses satisfaction. Nurse-Physician Relationships: On units with higher proportions of physically healthy nurses and of nursing worked hours, nurses were more likely to have better relationships with physicians. Nurses who perceived their practice to be more autonomous and those who rated the
25
quality of nursing care on the last shift as good/excellent were also more likely to have better relationships with physicians. However, nurse-physician relationships tended to deteriorate when there was a higher proportion of nurses with frequent shift changes on the unit and as nurses took on more patients in their daily assignment or care for patients with more nursing diagnoses. Deteriorated relationships were also more likely as unit productivity/utilization levels exceeded 85 percent. SF-12 Health Status: Higher physical health status scores were 59 percent less likely for female nurses; 49 percent less likely when nurses were at risk for an effort and reward imbalance; and 41 percent less likely for nurses at risk of emotional exhaustion. In contrast, nurses were more likely to be physically healthy when stronger nurse-physician relationships were reported on the unit and as the average worked hours available for care on the unit decreased. The likelihood of being physically healthy increased by 58 percent when nurses were satisfied with their job, and decreased by 28 percent for every 10 percent increase in nursing worked hours probably because increased nursing hours came from the same nurses worked on the unit rather than from new hired nurses. Female nurses were 52 percent less likely to be mentally healthy than male nurses, and older nurses reported better mental health. Nurses with one point increases in their physical health scores were four percent less likely to be mentally healthy. Nurses were less likely to be mentally healthy when they were at risk of emotional exhaustion and as the average worked hours on the unit increased. The likelihood of being mentally healthy increased by 74 percent when nurses were satisfied with their current job and decreased by 79 percent when nurses were at risk of emotional exhaustion.
26
System Outcomes Tables 20 to 29 (Appendix C) display the variables modeled in relation to system outcomes. Length of Stay: Patients in units where the productivity/utilization of the unit exceeded 91 percent were more likely to have longer-than-expected lengths of stay. Patients with more nursing diagnoses and with higher resource intensity weights, reflecting greater medical acuity, were also more likely to have longer lengths of stay. Shorter-than-expected lengths of stay were two percent and 185 percent more likely for patients whose physical health status scores were one point higher at admission and for those who attended a pre-operative clinic, respectively. Shorter-than-expected length of stay was 57 percent less likely when patients experienced medical consequences and 13 percent less likely for each additional nursing diagnosis. Interventions Not Done or Delayed: Older, experienced nurses were less likely to have interventions not completed at the end of their shift. The likelihood of interventions not being completed increased by 260 percent when nurses were at risk for an effort-reward imbalance. The more often patients had unanticipated changes in acuity, the more often interventions were left undone. The more frequently violence was experienced by individual nurses and the higher the medical complexity (as indicated by the resource intensity weight), the more likely interventions were not completed. The greater the number of nursing diagnoses, the less likely interventions were not completed. The likelihood of interventions being left undone was reduced as units hired nurses with more clinical expertise and reduced for units that increased average overtime. Interventions not completed were 12 percent less likely with every one point increase in the ratings of nurse autonomy. The more nurses re-sequenced their activities in response to demands from others, the less often interventions were left undone.
27
Delayed interventions were 74 percent more likely when nurses worked full-time, 87 percent more likely when nurses had concerns about job security, and 123 percent more likely when nurses were at risk of an effort and reward imbalance. Interventions were 27% less likely to be delayed for every 10% increase in the proportion of degree-prepared nurses on the unit. More complex patients with increasing numbers of nursing diagnoses were less likely to experience delays in receiving interventions. However, when individual nurses experienced violence or where the average level of violence was high on a unit, interventions were more likely to be delayed. Interventions were 71 percent more likely to be delayed for every 10 percent increase in absenteeism at the unit level. Quality of Patient Care Over the Past Year: When nurses rated themselves as expert clinicians, they were less likely to rate the quality of patient care on the unit as improved. Likewise, when interventions were delayed, nurses were 46 percent less likely to report improvements in the quality of patient care. The likelihood of improved nurse ratings of patient care increased by 915 percent when nurses rated the quality of nursing care given on the unit as good/excellent (as opposed to fair/poor) and when nurse autonomy was higher. Improved quality of patient care was 41 percent less likely with every 10 percent increase in nursing worked hours on the unit but 40 percent more likely with every 10 percent increase in degree-prepared nurses on the unit. Quality of Nursing Care on the Last Shift: Good or excellent ratings by nurses of the quality of nursing care on the last shift were 606 percent more likely when individual nurses rated the quality of patient care as improved over the last year; 159 percent more likely when nurses were satisfied; and more likely when nurses rated themselves as clinical experts. Nurses reports of strong nurse-physician relationships were also associated with good/excellent ratings of nursing
28
care on the last shift. However, nurses who changed shifts at least once during the past two weeks were 50 percent less likely to rate the quality of nursing care as good/excellent. Likewise when a 10 percent increase in the proportion of ratings of quality of nursing care at the unit level were good/excellent, individual nurses on the unit were 93 percent more likely to rate individual scores of quality of nursing care as good/excellent. However, for units with higher ratings of nurse-physician relationships on average, individual nurses were less likely to rate nursing care as good/excellent. Absenteeism: Full-time nurses were 152 percent more likely than part-time and casual nurses to miss work. Nurses who scored one point higher in physical health status scores were five percent less likely to miss work. When unit productivity/utilization was below 79.7 percent, nurses tended to have fewer days absent. Intent to Leave: Degree-prepared nurses were 101 percent more likely to leave as compared to diploma-prepared nurses. Nurses who reported job instability were 197 percent more likely to report intentions of leaving than those who did not. Satisfied nurses were 58 percent less likely to intend to leave. Full-time nurses were 51 percent less likely to leave than part-time or casual nurses. When productivity/utilization was below 82.8 percent on the unit, nurses were less likely to leave. Cost Per Resource Intensity Weight: Patients who were admitted with higher mental health status scores and with a higher number of nursing diagnoses were more likely to have higher costs per resource intensity weight, as were patients who attended pre-operative and postoperative education. Lower costs per resource intensity weight were more likely when care was provided in part in step-down units, when nurses rated themselves as clinical experts, and with emergency admissions, higher nurse-patient ratios, and higher physical health status scores
29
among nurses. As length of stay increased and as unit productivity/utilization exceeded 90 percent, so did the cost per resource intensity weight. Research Question 2 To what extent is there agreement between the estimates generated by a gold standard for measuring nursing resource needs (PRN workload methodology) and the worked hours per patient, and how does variance affect the patient and nurse outcomes? Table 30 (Appendix C) reveals that only two significant variables were found when examining the PRN estimates and actual worked hours. When actual worked hours were less than PRN predicted hours, nurses were more likely to leave in the next year and productivity/utilization was more likely to be high. Research Question 3 At what nurse-patient ratio and with what proportion of nursing worked hours are productivity/utilization and patient and nurse outcomes improved, after controlling for the influence of patient, nurse, organizational, and environmental factors? As shown in Table 31 (Appendix C), when a nurse was assigned more patients, the relationship with physicians deteriorated and autonomy increased. When more patients were assigned to a nurse, unit productivity/utilization increased and cost per resource intensity weight decreased. For every additional worked hour per patient, the odds of medical consequences increased by 13 percent, and the odds of improvement in patient mental health at discharge decreased by six percent. For every additional hour increase in the average worked hours on the unit, the likelihood of nurses being satisfied with the current job increased by 10 percent, but their odds of being physically and mentally health declined by 10 percent and seven percent respectively. For every 10 percent increase in the proportion of nursing worked hours the odds of
30
patients having improved knowledge scores increased by 74 percent, but nurses were 28 percent less likely to be physically healthy, were 41 percent less likely to rate the quality of patient care as improved. Given that the maximum productivity/utilization for any unit should not exceed 93 percent, productivity/utilization levels range from 79.7 percent for absenteeism to 91.4 percent for shorter-than-expected length of stay. These findings highlight the difficulties nurses face in this study, where almost 50 percent of the nursing units worked over productivity/utilization levels of 93 percent. Research Question 4 Which data elements, in addition to those routinely collected within administrative databases, are critical for routine data collection in Canada? To what extent do policy and administrative decision makers support the feasibility of routine data collection? Discussion with our policy and practice decision-making partners identified that: 1. nurse SF-12 physical and mental health status, emotional exhaustion, autonomy, effort and reward imbalance, and quality of nurse-physician relationships should be monitored annually in the new National Nursing Health Survey; 2. unit workload data should be checked for reliability and validity at least annually, and these data, in combination with worked hours, should be tracked regularly by nursing unit managers to determine if actual values exceed those recommended in this study. The Environmental Complexity Scale should be completed on each shift by nurses. Productivity/Utilization and environmental complexity should become quality indicators used by the Canadian Council of Health Services Accreditation to monitor healthy workplaces; and
31
3. nursing diagnoses and OMAHA tool ratings should be used daily in practice. Automated care planning systems that are easy to access and use are recommended. These are important indicators of patient goal achievement. V. Additional Resources The reader is referred to the works of Aiken et al, OBrien-Pallas et al, and Shamian and OBrien-Pallas et al as referenced in Appendix A. VI. Further Research 1. Develop and validate a shorter version of the effort and reward imbalance scale. 2. Conduct studies to examine the influencing factors and nature of short- and long-term illnesses among nurses. Evaluate strategies (such as access to fitness centers, improved hot meals in the hospital, and mandatory breaks) that may enhance the health of nurses. 3. Replicate this study on other patient populations to determine if the productivity/utilization cut-off points hold. 4. Explore experienced nurses perceptions of quality and develop measures of quality that can be evaluated yearly at the nursing unit level.
VII. References 1. Canadian Nursing Advisory Committee. (2002).Our health, our future: Creating quality workplaces for Canadian nurses. Toronto, ON: Author. 2. OBrien-Pallas, L. L., Thomson, D., Alksnis, C., Luba, M., Pagniello, A., Ray, K. et al (2003). Stepping to success and sustainability: An analysis of Ontarios nursing workforce. Toronto, ON: Nursing Effectiveness, Utilization, and Outcomes Research Unit. 3. Canadian Council for Health Service Accreditation (2002). Recognition guidelines for 2003: Specific issues and related criteria. Ottawa, ON: Author. 4. Baumann, A., O'Brien-Pallas, L., Armstrong-Stassen, M., Blythe, J., Bourbonnais, R. Cameron, S. et al. (2001). Commitment and care: The benefits of a healthy workplace for
32
nurse, their patients and the system a policy synthesis. Ottawa, ON: Canadian Health Service Research Foundation. 5. OBrien-Pallas, L. L., Alksnis, C., Wang, S., Birch, S., & Tomblin Murphy, G. (2003). Bring the future into focus: Projecting RN retirement in Canada. Toronto, ON: Canadian Institute for Health Information. 6. Aiken, L., Clarke, S., Sloane, D., Sochalski, J., & Silber, J. (2002). Hospital nurse staffing and patient mortality, nurse burnout, and job satisfaction. JAMA: The Journal of American Medical Association, 288(16), 1987-1993. 7. Shamian, J., OBrien-Pallas, L., Kerr, M., Koehoorn, M., Thomson, D., & Alksnis, C. (2001). Effects of job strain, hospital organizational factors and individual characteristics on work-related disability among nurses. Toronto, ON: Ontario Workplace Safety and Insurance Board. 8. Grinspun, D. (2003). Part-time and casual nursing work: The perils of healthcare restructuring. International Journal of Sociology and Social Policy, 23(8/9), 54-70. 9. Canadian Institute for Health Information. (1999). MIS guidelines for Canadian healthcare facilities. Ottawa, ON: Author. 10. OBrien-Pallas, L., Giovannetti, P., Peereboom, E., & Marton, C. (1995). Case costing and nursing workload: Past, present and future [Working Paper 95-1]. Hamilton, ON: Quality of Nursing Worklife Research Unit.
33
Appendix A. Annotated Bibliography

Table of Contents
1. 2. 3. 4. 5. 6. 7. 8. 9.
Patient Characteristics Nurse Characteristics System Characteristics and Behaviours Throughputs Patient Outcomes Nurse Outcomes System Outcomes Glossary References
35 38 40 60 62 65 74 76 76
34
1.
Inputs Occupation, medical diagnosis. Throughputs Outputs
Patient Characteristics
Authors/Year 1. Alterman, Shekelle, Vernon, Burau (1994) Patients perceived health status, costs of care
Focus Decision latitude, psychological demand, job strain, & coronary heart disease.
2. Bull, Hansen, Gross (2000)
Discharge planning model for elders hospitalized with heart failure.
Sample Annual examine of 1,683 men employed at Hawthorne Works for 25 years. 158 elder/ caregiver dyads, before-and-after non-equivalent control group.
Findings Contrary to the hypothesis, those with highest decision latitude had lowest coronary heart disease death rates (risk of 6.8, with average risk being 7.8). No association between coronary heart disease & psychological demand. Elders who received new model of discharge planning felt more prepared to manage care (t=4.30), felt in better health (t=2.0) & spent fewer days in hospital when readmitted. Facilitated elder & caregiver participation in planning. Predictive model of risk of major complications can be used to analyze cost of care, resource utilization, & outcomes. Resource utilization increases as the probability of risk for cardiac complications increases (ex: highest risk group had 75% higher costs than the lowest risk group).
3. Calvin, Klein, Vanden Berg, , Meyer, RamirezMorgen, Parrillo (1998) Patient demographics, occupation Patient learning needs, demographics, occupation
Predicting resource utilization in patients with unstable angina.
Patient demographics, education, continuity of care, patient teaching, medical diagnosis. Patient demographics, medical diagnosis. Length of stay, costs of care, complications
4. Crilley, Farrer (2001)
Impact of a first myocardial infarction on self-perceived health status.
SF-12 patient health status
5. Czar, Engler (1997)
Perceived learning needs of patients with coronary artery disease.
465 patients admitted for unstable angina to a tertiary care university-based medical centre, prospective evaluation. 165 patients were surveyed 2 years after a first myocardial infarction Convenience sample of 28 men admitted with angina or myocardial
Patients having their first myocardial infarction have significantly lower levels of self-perceived health status after 2 years than control subjects. Health status is associated with persistent cardiac symptoms & lack of employment. No significant difference in learning needs between hospitalization & subsequent clinic visits. Most important learning needs were symptom recognition, cardiac anatomy/physiology, & medications.
35
Authors/Year
Focus
Sample infarction to a California universityaffiliated medical centre. Social supports, occupation, medical diagnosis.
Inputs
Throughputs
Outputs
6. Hemingway, Marmot (1999)
Psychosocial factors in the development & prognosis of coronary heart disease.
7. Johnson, Stewart, Hall, Fredlund, Theorell (1996)
Impact of work organization (psychological demand, work control, & social support) on cardiovascular disease mortality.
Patient occupation, social support, patient demographics education.
Patient mortality
Findings No correlation between learning needs & age, occupation, smoking or marital status. Most important learning needs are those that affect survival. A self-administered questionnaire can be used to determine patients perceived learning needs so education can focus on areas most important to the patient. Strong correlations between depression/ anxiety & development of coronary heart disease (11/11 studies). Traits such as type A/ hostility (6/14 studies), work organization (6/10 studies) & social support (5/8 studies) also have moderate correlations with coronary heart disease. Workers with low work control had a higher risk for cardiovascular mortality (after 5 year exposure, relative risk of mortality is 1.46 for low control vs. 1 .00 for high control). No significant associations between physical job demand, work social support, job hazards, & cardiovascular mortality.
8. Marchette, Holloman (1986)
Variable affecting length of stay.
Medical diagnosis, patient demographics, patient education given.
Length of stay
For every area of discharge planning the nurse carried out (nutrition, medication, etc.), there was a decrease of 0.8 days of hospitalization (2 days for CVA patients). Strong relationship between timing of planning & length of stay (for every day that the planning was postponed, there was an increase of 0.8 days of hospitalization).
9. Shi (1996)
Relationship between
12,517 Swedish men 25-74 currently or previously employed. Random sample from entire Swedish population, 80% response rate. 500 patients discharged from an acute care hospital. Stratified random sampling of 100 patients with 5 most common diagnostic categories. 274,311 patient Patient
Length of stay
Direct relationships between longer length of
36
Authors/Year
Focus patient & hospital characteristics on length of stay. Admission to or discharge from intensive care unit., post-op complications.
10. Shih, Chu, Yu, Hu, Huang, (1997)
Turning points of recovery from cardiac surgery in an intensive care unit.
Sample records & 484 hospitals, random sampling of hospitals & discharges. Convenience sample of 30 adults who had undergone cardiac surgery in 1of 3 general hospitals in northern Taiwan. Medical diagnosis Recruited 416 middle-aged blue-collar men from steel & metal plants in West Germany for prospective study, followed over 6.5 years. 73,174 patient admissions to 137 hospitals. Data from national surveys. Patient demographics, occupation Effort-reward imbalance, health status.
Inputs demographics, hospital size, medical diagnosis, social support. Patient demographics.
Throughputs
Outputs
Findings stay & older age, non-married status, being female, being African American, & having insurance. Fewer hospital beds correlated with shorter length of stay tables omitted).
11. Siegrist (1996)
12. Siegrist, Peter, Junger, Cremer, Seidel (1990)
Impact of high effort, low reward conditions in the workplace on cardiovascular health. Impact of stressful work on ischemic heart disease.
13. Silber, Rosenbaum, Ross (1995)
Predictors of hospital outcomes.
Adverse occurrences.
Turning points included events, nursing actions, & time. Components of turning points were preceding conditions, markers, & consequences. Turning point experiences included none (7%), both (57%) or one of positive (33%) & negative (3%) outcomes. May sensitize nurses to detect more quickly patients turning point experiences. Nurses may educate patients before surgery based on turning point knowledge. Variables indicating high effort & low reward (money, esteem, status) predict cardiovascular events. These conditions at work must be considered a risk for cardiovascular health. Status inconsistency (regression coefficient=1.48), job insecurity (1.23), work pressure (1.24) & need for control (1.51) predicted ischemic heart disease occurrence. Reducing burden of high workload & increasing reward & security could reduce ischemic heart disease risk. Individual prevention could be directed at coping with work demands & strengthening regenerative potential. Nearly all of predictable variation in outcomes is from differing patient characteristics, not hospital or staffing ones. Effort & reward Those with fewer social supports were at
14. Steptoe
Psychosocial factors in
Staffing ratios, medical diagnosis, patient demographics, hospital size, Occupation,
37
Authors/Year (1999)
Focus the cause of coronary heart disease.
Sample
Inputs social support, medical diagnosis
Throughputs
Outputs imbalance
Findings increased risk for cardiovascular death. Other risk factors: stressful work conditions, cynically hostile attitude.
2. Sample 121 nurses from 10 acute care hospitals with high & low nurse-injury compensation claim rates. Interviews with 5 chief executive officers, 10 chief nursing officers, &9 Occupational Health & Security Officers Nurse absenteeism Work environment Nurse job satisfaction, violence, nurses health. Inputs Nurse demographics, workload, staffing. Throughputs Organizational work environment. Outputs Absenteeism, nurse injury.
Nurse Characteristics Findings Nurses in both high-claim & low-claim hospitals identified physical work environment, claims process, & staffing as factors related to different injury claim rates among hospitals. Workload is a contributing factor to high-injury rates among nurses.
Authors, Year 15. Bruce, Sale, Shamian, O'Brien-Pallas, Thomson (2002).
Focus Describe nurses' health status, examine trends in injury compensation claims, & determine factors contributing to high-injury claim rates.
16. Coutts (2001)
Highlights issues from Canadian Health Services Research Foundations report on healthy workplaces for nurses.
When nurses are more independent & have more say in patient treatment, they find jobs more satisfying. There is a direct relationship between hours of overtime & sick time. Fatigue, too much to do & temporary staffing lead to nurses getting hurt. Good working conditions for nurses should be a standard for hospital accreditation.
38
Authors, Year 17. Josephson, Lagerstrom, Hagberg, Hjelm (1997)
Focus Musculoskeletal symptoms & job strain in nurses.
Inputs
Throughputs
Outputs Nurses health status.
Findings Job strain is a risk factor for musculoskeletal symptoms & the risk is higher when combined with perceived physical exertion (RR=1.5-2.1). Increased job strain may be associated with staff cuts, reorganization, & new requirements.
18. Josephson, Vingard, MUSICNorrtalje Study Group (1998)
Comparison of lowback pain in female nurses & other employed women.
Nurse demographics
Conflicting demands, job strain.
Low-back problems
Female nursing personnel had no increased risk of low-back pain compared with other employed women. Combinations of physical & psychosocial factors are associated with high risk. Physical load is more significant in nursing than psychosocial factors in relation to low-back pain Strong associations between poor psychosocial conditions (especially low support & high demands) & musculoskeletal disorders (especially central body region). Most prevalence ratios were above 1.0 meaning mostly positive associations between unfavourable conditions & pain.
19. Toomingas, Theorell, Michelsen, Nordemar (1997)
Associations between self-rated psychosocial conditions & characteristics of musculoskeletal symptoms, signs, & syndromes.
Sample 285 nurses at a county hospital. Repeated crosssectional surveys given to all personnel on wards with patients requiring daily care (e.g. transfers). Random sample of 333 women with back pain & 733 women in control group (81 and 188 respectively were employed in nursing). 358 men & women from various occupations (83 male furniture movers, 89 female medical secretaries; 96 men & 90 women of working population). Psychosocial work conditions (demands, social support, decision latitude) Work environment. Symptoms, signs, & syndromes of musculoskeletal origin. Nursing ratios, professional status, proportion of nurse worked Nurses health, patient outcomes, violence.
20. Wunderlich, Sloan, Davis (1996)
Summary report from Institute of Medicine about adequacy of nursing in hospitals
Increase in acuity in hospitals means that RNs may need more education. Ancillary nursing personnel should have documented evidence of competency. Nurses have high rates of work-
39
Authors, Year
Focus & nursing homes.
Sample
Inputs hours, staffing levels.
Throughputs
Outputs
Findings related injury & back injuries were related to staffing issues. Violence towards healthcare workers is increasing.
3. Sample 39 magnet hospitals matched with 195 control hospitals in U.S. Unit of analysis is hospital Inputs Hospital size & organization, staffing ratios. Throughputs Outputs Mortality rates
System Characteristics and Behaviours
Authors, Year 21. Aiken, Smith, Lake (1994)
Focus Mortality rates in hospitals with higher proportion of RN staff to total staff
22. Aiken, Clarke, Sloane, (2002).
Examine effects of nurse staffing & organizational support for nursing care on nurses' dissatisfaction with their jobs, nurse burnout, & nurse reports of quality patient care in
10,319 nurses working in medical & surgical units in 303 international hospitals
Workload, proportion of nurse worked hours, organizational support.
Nurse burnout, job satisfaction, nurses perceived quality of care.
Findings Observed mortality rates for magnet hospitals are 7.7% lower (9 fewer deaths per 1,000 Medicare discharges; p=0.011). After adjusting for predicted mortality, magnet hospital rates were 4.6% lower (p=0.026, CI 95%; 0.9 to 0.4 fewer deaths per 1,000). Magnet hospitals had significantly higher RN: total nursing personnel ratios & slightly higher nurse: patient ratios Skill mix & nurse: patient ratios do not explain the mortality effect or the variability in effects across hospitals. Authors propose that mortality effect derives from greater status, autonomy & control afforded nurses in magnet hospitals; not simply an issue of credentials & number of nurses. Organizational/managerial support for nursing had a pronounced effect on nurse dissatisfaction & burnout. Organizational support for nursing & nurse staffing were directly related to nurseassessed quality of care. Nurse reports of low quality care were three times as likely in hospitals with low staffing & support for nurses, compared to hospitals with high staffing & support.
40
Authors, Year 210 adult general hospitals; 10,184 staff nurses randomly surveyed; 232,342 patients discharged. Workload measurement. Staffing ratios, length of stay. Nurse demographics, professional status, medical diagnosis, patient demographics, hospital size. Nurse job satisfaction, burnout, patient mortality, costs, and complications.
Sample
Inputs
Throughputs
Outputs
Findings In hospitals with high patient-to-nurse ratios, surgical patients experience higher riskadjusted 30-day mortality & failure-to-rescue rates (odds ratio 1.07), & nurses are more likely to experience burnout & job dissatisfaction (odds ratio for dissatisfaction is 1.15).
23. Aiken, Clarke, Sloane, Sochalski, Silber (2002).
24. Arthur, James (1994)
Focus hospitals. Determine association between patient-to-nurse ratio & patient mortality, failure among surgical patients, & factors related to nurse retention. Literature review of various methods of nurse staffing level measurement.
25. Baker, Messmer, Gyurko, Domagala, Franklin, Eads, Harshman, Layne (2000) 39 units in 11 hospitals. Staffing mix, type of unit.
Hospital ownership, performance, & outcomes.
6,097 hospitals in the U.S.
Patient demographics, nursing interventions, patient dependency, and proportion of nurse worked hours. Hospital ownership type (public, private-for profit, private non-profit), performance Organizational work environment.
Various approaches to nurse demand methods: consensus (intuitive, consultative method), topdown management (staffing norms, staffing formulae), & bottom-up management (nursing interventions, patient dependency). Debates surrounding these methods include: comparability vs. local suitability, control, efficiency, effectiveness, & philosophy of care. Hospital ownership has an impact on hospital performance in relation to system operations: costs, prices, & financial management practices; & personal issues. Association between hospital ownership & adverse events consistently supported.
26. Blegen, Vaughn (1998)
Nurse staffing & patient occurrences.
27. Blegen, Goode, Reed, (1998).
Relationship among incidence rates of 6 adverse pt. outcomes, the hours of care
42 inpatient units in an 880-bed university hospital.
Patient acuity: (unique Patient Classification System; levels
Adverse events, morbidity, mortality, patient satisfaction, nurse satisfaction, costs, productivity. Patient complications (med errors, falls), cardiopulmonary arrests. Patient outcomes: medication errors per 10,000
Higher proportion of RNs resulted in fewer patient complications (med errors/dose R2=0.576, falls R2=-0.456) but the relationship is not linear. Units with RN proportion >85% had higher med errors (may be heightened vigilance, sicker patients with more meds). Controlling for average patient acuity adjusted at unit level, RN proportion was inversely related to the unit rates of medication errors, decubiti, & patient complaints. Total hours of
41
Authors, Year
Focus provided by all nursing personnel & the proportion of those care hours given by RNs. 21,783 discharges & 198,962 patient days of care provided by 1074 FTE nursing staff members, 832 of those FTEs were RNs. Fiscal year 1993.
Sample
Inputs range between 1 and 7 most acute/most care)
Throughputs
Findings care from all nursing personnel were associated directly with complaints, decubiti, & mortality; however total hours of care were highly correlated with acuity. A larger proportion of care delivery by RNs was associated with a decrease in adverse pt outcomes up to a level of 87.5% of RN staffing. When the proportion of RNs in the staff mix was greater than 87.5%, adverse outcomes also increased.
Outputs doses (nurse selfreport), falls, decubiti, urinary & respiratory infections, patient complaints per 1,000 patient days & mortality rates per 1,000 patient days.
28. Brown (2001)
Synopsis of ANA study on relationship between staffing & patient outcomes.
Nearly 13 million patients in 1500 hospitals from 9 states
Nurse staffing: All Hours (monthly hours of care patient per day by RNs, LPNs, NAs/patient days on unit); RN Hours (hours of direct RN patient care/patient days); RN proportion (RN hours patient per day/All Hours) Proportion of nurse worked hours, medical diagnosis. Staffing ratios, patient complications & outcomes, length of stay Costs of care, clinical outcomes
29. Buerhaus (1997)
Mandatory minimum nurse staffing levels in hospitals
Proportion of nurse worked hours, nurse-to-patient ratio
Several patient outcomes were significantly related to staffing (length of stay, pneumonia, post-op infections, ulcers, urinary tact infections). Cost savings in reductions of staff & mix may not be real savings when complications & increased length of stay are considered. Staffing regulations (if imposed) would force employers to ignore dynamic interactions of economic, technology, capital & labour supply variables. There would be significant costs associated with enforcing the regulation which would outweigh the benefits.
42
Authors, Year 30. Burke (2003)
Sample Self-report survey of 744 hospital-based nursing survivors. 1st wave Nov. 1996. 2nd wave Nov. 1999. Hospital: size Medical diagnosis Nurse: experience, employment status, education, demographics
Focus Relationship between changes in patientnurse ratios resulting from hospital restructuring & nursing staff satisfaction, psychological health, & perceptions of hospital functioning.
Inputs Patient-nurse ratio: current & changes since restructuring began
Throughputs
31. California Nurses Association. (2001) 32. Callaghan, Cartwright, ORourke, Davies (2003) 692 very low birth weight infants in an Australian intensive care unit Jan. 1996 Dec. 1999 Infant characteristics: dependency (infant:nurse ratios: intensive 1:1, high 1:2, medium 1:3, & recovery 1:5), birth history, admission & physiological data Survival to hospital discharge, adjusted for initial risk (using Clinical Risk Index for Babies) & for unit workload (infant dependency scores)
Reasons for the CNA backed nurse-topatient ratios.
Outputs Work outcomes, work experiences: extent of restructuring, perceived workload, job security, psychological health, hospital effectiveness: Staffing ratios
Findings 53% of nurses reported an increased patientnurse ratio. Increased ratios associated with less job satisfaction & job security, greater intention to quit & more restructuring initiatives, poorer psychological (but not physical) health, & less effective hospital functioning.
Strong ratios will help reduce the nursing shortage. They must be determined in accordance to individual patient care needs. Overall hospital mortality rate of 12% (80 out of 692 infants). Odds of mortality, adjusted for initial risk & unit workload improved by 82% when an infant:staff ratio of greater than 1.71 occurred.
Relationship between infant to staff ratios in first three days of life on the survival to hospital discharge
33. Campbell, Taylor, Callaghan, Shuldham
Using case mix group to predict workload.
798 patients & 30 nurses from one ward. Retrospective
Staffing: number of nurses working per shift, maximum number of infants per shift Patient demographics, admission type, case mix group.
Workload
Costs of care, length of stay.
Regression analysis showed that there was no relationship between resources used & nursing clinical hours; however, there was a good relationship between the number of patients &
43
Authors, Year (1997)
Focus
Sample data from hospital pt. administration system.
Inputs
Throughputs
Outputs
34. Canadian Labour & Business Centre (2002). Data from 799 hospitals in 11 states. Proportion of nurse worked hours. Patient complications & outcomes, length of stay, staffing ratios.
Cost associated with absenteeism, overtime, & involuntary part-time employment.
201 bed hospital in Ottawa.
Nurse demographics.
Costs of care, absenteeism, use of agency nurses, turnover.
35. Clark (2002)
Effect of nurse staffing levels on adverse events.
Findings resources used. Average lengths of stay for case mix groups were greater than predicted. For cystic fibrosis patients (representative of specialist nursing), case mix group accounted for only 18% of variation in nursing time required. Case mix group has shown to be a poor predictor of nursing requirement. Overtime hours have increased dramatically over past 3 years. Agency costs represent 34 % of overtime costs. There are estimates that 2530% of absenteeism is related to stress & injury. Additional full-time equivalents would reduce costs. Higher proportion of RNs resulted in better care. No association between lower rates of outcomes & number of hours of care by LPNs or nurses aides.
44
Authors, Year 36. Clarke, Rockett, Sloane, Aiken (2002)
Focus Organizational climate, staffing, & safety equipment
Sample 1998 survey data for 2287 medical surgical unit nurses in 22 US hospitals (20 were magnet hospitals). 1998 survey of management, infection control, purchasing officials re: equipment selection & procurement
37. Cockerill, OBrien-Pallas, Bolley, Pink (1993) 1116 nurses from 19 urban teaching hospitals. Stratified random sample, 97% response rate. Staffing ratios, proportion of nurse worked hours, patient demographics, case mix groups, nurse education, experience. Staffing mix, ratios
Measuring nursing costs using nursing workload.
256 patient records from 4 units in a large teaching hospital.
Inputs Self-report compliance with universal precautions & perceived risk Nurse characteristics. Protective Equipment. Nurse staffing: patients cared for on last shift worked; hospitallevel measure averaging patient loads, organizational climate (R-NWI) Case mix group Workload (GRASP, NISS, Medicus, PRN) Costs of care.
Throughputs
Outputs - Needlestick injuries & nearmisses: selfreport occurrence, frequency in past month & past year, circumstances, reporting
Findings - Average day shift workload ranged from 3.6 to 8.7 patients per nurse (n = 22) - In n=5, average day shift workload of more than 6 patients (heaviest workload) - nurses with heaviest workload were 50% more likely to report an injury & 40% more likely to report a near-miss in the preceding month
38. Doran, McGillis Hall, Sidani, OBrien-Pallas, Donner, Baker, Pink (2001)
Effect of nurse staffing issues on nurse communication & patient outcomes.
Patient health status.
There were significant differences among systems in estimates of care for patients. Nursing costs per case differ significantly depending solely on workload measurement system used (as much as 30% difference in cost for same patient). Higher mix of RNs improved communication & led to beneficial patient outcomes. Nursepatient ratio (r=0.138) & proportion of regulated staff (r=0.155) positively affected nurse communication. Effective communication among nurses led to better patient health outcomes (r=0.175). Work environment Costs of care, quality of care Environment, organization, staffing, & culture influence quality of care via their effect on the healthy workplace. Adequate number of staff,
39.Eisenberg, Bowman, Foster (2001)
Impact of workplace health on quality of care.
45
Authors, Year
Focus
Sample
Inputs
Throughputs
Outputs
40. Gauci Borda, Norman (1997)
Factors that influence turnover & absence of nurses, the relationship between absence & turnover.
17 studies on relationship between job satisfaction, intent to stay/leave & actual turnover.
Nurse job satisfaction, system organizational factors.
Nurse absenteeism, turnover.
41. Gaudine (2000)
Nurses views of workload & work overload.
31 staff nurses from 9 different units of a hospital in central Canada, volunteer sampling.
Nurse demographics, experience, and patient acuity.
Simultaneous demands, unanticipated events, interruptions, noise level.
Nurses feelings of workload & overload.
Findings appropriate blend of skills & proper equipment enable work. Heavy workloads inhibit staff from participating in research. Enhancing workers health & satisfaction may improve patient outcomes. Job satisfaction influences absence & intent to stay. Intent to stay in current employment influences turnover. Intent to stay is most strongly associated with job satisfaction. Pay & opportunity for alternative employment also influence intent to stay, which is supported by two studies. Absence is positively related to turnover (absence increased before turnover) & negatively related to intent to stay. Kinship responsibility is directly related to intent to stay. Meanings that nurses attributed to workload include volume, simultaneous demands, demands on self, qualitative overload, anticipation, responsibility, interdependence, non-work roles & exhaustion. The meanings of work overload include simultaneous demands, qualitative work overload, heavy load, & responsibility. These meanings include more dimensions than current measures of workload.
46
Authors, Year 42. Greenglass, Burke (2001)
Focus Impact of hospital restructuring on nurses.
Inputs Nurse education, demographics, hospital size.
Throughputs
Outputs Perceived quality of care, job satisfaction, working conditions.
43. Grillo-Peck, Risner (1995)
Implementation of a nursing partnership model.
Sample 1363 nurses employed in hospitals undergoing restructuring. Random selection from union membership, 35% response rate. 156 patients from a neuroscience unit in an 800bed not-forprofit hospital in Ohio. Proportion of nurse worked hours, care delivery system, medical diagnosis, patient demographics, and continuity of care. Nursing diagnoses, medical diagnoses (DRG), patient demographics. Costs of care, nursing workload. Patient length of stay, complications (infections, falls), medication errors.
44. Halloran (1985)
Effects of variables on nursing workload.
Findings Restructuring had a negative effect on staff (97.9% of respondents agree) & working conditions (94%). It has compromised the quality of care & reduced nurses ability to provide services for patients. During hospital restructuring workload was the most significant & consistent predictor of distress in nurses, as manifested in lower job satisfaction, professional efficacy, & job security. Greater workload also contributed to depression, cynicism, & anxiety. A nursing partnership model which included a decrease in RNs & a primary nursing model was implemented. RNs were partnered with a patient care technician & assisted by service associates. This allowed for continuity of care. Patient complications showed a downward trend after the implementation & length of stay decreased. RNs were able to spend less time in non-professional tasks. Variations in nursing workload were better explained by nursing condition than by medical condition or patient demographics (75% of the sum of the squared regression coefficient is associated with nursing diagnosis & 25 % is associated with medical diagnosis). There was a strong positive relationship between workload & length of stay in hospital (correlation coefficient=0.774). Of demographic characteristics, only age is associated with variations in workload (r=0.198). Patient mortality. Higher mortality rates were associated with forprofit (121/1000 patients vs. average of 116/1000) & public hospitals (120/1000) &
45. Hartz, Krakauer, Kuhn, Young,
Factors that affect patient mortality rates.
2560 patient records & 141 nursing staff members from a 279-bed acute care, community hospital. This included all patients admitted & discharged over a 4 month period. 3100 hospitals in the United States. Data from Patient demographics, medical diagnosis,
47
Authors, Year Jacobsen, Gay, Muenz, Katzoff, Bailey, Rimm (1989)
Focus
Inputs hospital size, proportion of nurse worked hours, nurse ratios.
Throughputs
Outputs
Findings osteopathic hospitals (129/1000). The characteristics most closely linked with mortality are related to training of medical staff (e.g. higher percentage of RNs = lower mortality). Higher occupancy rate was associated with lower mortality rate.
46. Hendrix, Foreman (2001)
Optimal nurse staffing levels in long term care.
Sample the Healthcare Financing Administration & the American Hospital Association annual survey. Over 12,000 federally certified skilled & intermediate nursing homes in the United States (data from 1994). Proportion of nurse worked hours, nursing ratios. Cost of care, costs of injury, patient outcomes (decubitus ulcers), public burden. Workload. Staffing mix, ratios.
47. Kenney. (2001)
Pilot project for implementing LPNs to maintain quality during staffing shortages.
There is an optimal level of nurse staffing that minimizes decubitus ulcers in nursing homes. The presence of RNs (superior skills; =-1.659) & nurses aides (lower wage; =-7.334)) reduces costs associated with ulcer care while the presence of LPNs (=4.544) increases the cost of ulcer care. Nursing homes should increase the number of RNs & NAs. There were no changes in quality of care (number of treatment/procedure errors & falls remained stable) or staff satisfaction with the addition of LPNs. LPNs were delegated to tasks by RNs who were still responsible for several patients. Nurses health, costs of care, patient satisfaction, complications (falls, med errors), nurse satisfaction Patient complications & outcomes. Degree of coordination of intensive care positively influences its effectiveness. Interaction & communication between physicians & nurses affect patient mortality.
48. Knaus, Draper, Wagner, Zimmerman (1986)
Influencing factors on mortality in intensive care units.
5,030 patients in intensive care units in 13 tertiary care hospitals. Hospitals were self selected & patients were convenience sampled. Proportion of nurse worked hours.
Patient demographics, admission type, medical diagnosis.
49. Kobs. (1997)
Adequacy of nurse staffing.
Patient complications,
There is a positive correlation between shorter length of stay & higher staffing levels. As RN
48
Authors, Year
Focus
Sample
Inputs
Throughputs
50. Kovner (2001)
Impact of staffing & work organization on patient outcomes & healthcare workers. Nurse staffing (FTE RNs working patient per day). Patient mortality, medication error rates, postoperative infections.
Nurses education, staffing ratios.
Outputs length of stay, staffing ratios. Nurses health, patient complications.
51. Kovner, Gergen (1998)
Relationship between nurse staffing & adverse events.
Findings staffing increased, patient complications decreased. Inverse relationship between mortality & number of RNs. Inverse relationship between complications & number of RNs. There are high rates of illness & injury among healthcare personnel. Inverse relationships between FTE RNs per adjusted inpatient day & urinary tact infections (-636.96, p<.001), pneumonia (-159.41, p<.001), thrombosis (-33.22, p<.01), pulmonary compromise (-59.69, p<.05) after major surgery. Hospitals with higher proportion of RNs had lower adjusted mortality rates (difference of 2.1-3.6%).
52. Krakauer, Bailey, Skellan, Stewart, Hartz, Kuhn, Rimm (1992)
Evaluation of a model for analyzing hospital mortality rates.
Proportion of nurse worked hours, hospital size.
Patient complications.
53. Kramer, Schmalenberg. (1988) Part 1.
Characteristics of magnet hospitals.
589 acute-care hospitals in 10 states, data from a 20% stratified probability sample to approximate US hospitals. 42,773 patients from 84 hospitals. Random sampling of discharges & hospitals from strata. 16 magnet hospitals. Hospital size, staffing ratios, nurse education, experience. Nurse turnover.
54. Kramer, Schmalenberg (1988) Part 2.
Characteristics of magnet hospitals.
16 magnet hospitals.
Hospital size, staffing ratios, nurse education, experience.
Nurse turnover.
Magnet hospitals are successful in recruiting & retaining nurses during periods of shortage. Fluidity & informality allows for communication & exchange of information. Staff nurses were allowed time for research, publication & special projects. There is support for continuing education & encouragement of autonomy & entrepreneurship. Magnet hospitals have characteristics similar to well run companies. They deal with nursing shortage by altering organizational conditions to eliminate internal shortage. Leaders are
49
Authors, Year
Focus
Sample
Inputs
Throughputs
Outputs
Organization of intensive care units & the influence on patient outcomes. Work environment.
55. Kutsogiannis, Hague, Triska, Johnston, Noseworthy (2001) 56. Manitoba Nursing Strategy (2003).
Care delivery system.
Work environment & organization.
Costs of care, patient complications, length of stay.
Findings highly visible & accessible. Magnet hospitals value quality, autonomy, informal communication, innovation, education, respect, excellence, & bringing out the best in each individual. Important factors in organization include communication, leadership, & interdisciplinary politics. Better standardized practices & coordination were related to better outcomes (lower mortality & morbidity). MNS includes: 1. Increase the supply of nurses 2. Improve access to staff development for nurses. 3. 3. Improve the utilization of nurses. 4. Improve working conditions. 5. Increase nurses' opportunities to provide input into decision-making.
57. Maxwell (2002).
The Manitoba Nursing Strategy (MNS) is a report released by the Manitoba government to address the concerns raised by nurses & other stakeholders within the healthcare system. Factors needed to create high-quality care environments. System organization, workload, nurse job control. Nurse health, job satisfaction, patient outcomes.
Job design, job rewards, organizational change, & job security can have major employee health implications. Role stressors & job insecurity influence the work environment. Workload, work pace, & work scheduling are important work environment issues facing health-care workers. Positive health outcomes for nurses are associated with high job control & a balance of job demands with sufficient resources. Hospitals with positive work environments have better staff recruitment & retention, & patient outcomes.
50
Authors, Year 58. McGillis Hall, Irvine Doran, Baker, Pink, Sidani, O'BrienPallas, Donner (2002). Costs of care, nurse job satisfaction, & burnout, patient complications, length of stay.
Focus Reports on staff mix & work status of nurses in adult medical, surgical, & obstetrical units in Ontario's teaching hospitals.
Throughputs
Outputs
59. Mitchell, Armstrong, Simpson, Lentz (1989)
The Demonstration Critical Care Unit: organizational & clinical outcomes.
Inputs Nurse demographics, experience, employment status, care delivery system, proportion of nurse worked hours. Hospital size & type, nurse demographics, admission type, patient demographics, medical diagnosis.
Findings There is a need for developing appropriate levels of knowledge & skill for complex inpatients cared for in medical/surgical & obstetrical units. The proportions of RNs within the individual unit staffing models remained relatively high (60-89%). More than one third of the nursing staff were employed on a parttime or casual basis. Positive organizational & clinical outcomes exist with valued aspects of organizational environment (high nurse-physician collaboration, highly rated nursing performance, positive organizational climate) as compared with historical comparison samples.
60. Mitchell, Shortell, (1997)
E7ffects of organizational variables in care delivery systems on adverse outcomes.
Sample 19 teaching hospitals, 2,046 patients, 1,116 nurses, 74 unit managers. Random sampling was used. 42 nurses, 68 physicians, 192 patient admissions. Patients representative of units population except for drug overdose or short stay. 81 research papers. Organizational variables in care delivery systems. Patient morbidity, mortality, & adverse effects.
Some support that nursing surveillance, quality of working environment & interactions with other professionals are related to lower mortality & complications. Patient variables have a greater impact than organizational variables. Adverse events are more closely related to organizational factors than is mortality.
51
Authors, Year 61. Needleman, Buerhaus, Mattke, Stewart, Zelevinsky (2002)
Focus Examine the relation between the amount of care provided by nurses at the hospital & patients' outcomes.
Sample 1997 admin data for 799 hospital in 11 US states (discharges: 5,075,969 medical & 1,104,659 surgical patients; accounted for 26% of 1997 discharges from non-federal US hospitals). Unit of analysis was hospital.
Inputs Inpatient staffing levels patient per day adjusted for inpatient vs. outpatient bias; differences between hospitals level of nursing care per DRG; risk adjustment for patient characteristics; & hospital characteristics (number of hospital beds, teaching status, state, & metropolitan/nonmetropolitan).
Throughputs
Outputs Length of stay, post-operative complications, adverse events, mortality. Failure to rescue defined as: death from pneumonia, shock or cardiac arrest, upper GI bleeding, sepsis, or deep venous thrombosis p.1715
62. OBrienPallas, Cockerill, Leatt (1992)
Determine equivalence of workload estimates of 5 patient classification methods (NISS, GRASP, Medicus, PRN 76 & PRN 80) Nursing
Case mix group classification
Program cost forecasting
Findings - Mean hours of nursing care per patient-day was 11.4; of which, 7.8, 1.2 & 2.4 provided by RNs, LPNs, & nursing aides respectively. Mean proportion of total hours of RN care was 68% & of nursing aides care was 21%. - Among medical patients, a higher proportion of RN hours of care patient per day & greater absolute number of RN hours of care per day associated with shorter length of stay (p=0.01 & p<0.001), lower rates of urinary tact infection ( p<0.001 & p=0.003) & lower rates of upper GI bleeding (p=0.03 & p=0.007). Higher proportion of RN hours associated with lower rates of pneumonia (p=0.001), shock or cardiac arrest (p=0.007), & failure to rescue (p=0.05). - Among surgical patients, higher proportion of RN care associated with lower rates of urinary tact infection (p=0.04). Greater number of hours of RN care patient per day associated with lower rates of failure to rescue (p=0.008). - No association found between RN staffing levels & rate of in-hospital mortality. No association found between increased staffing by LPN or nursing aides & rate of adverse outcomes. Clinically significant differences in hours of care estimates found by each system but a high correlation between the systems suggests that calibration could be used to compare data (alphas < 0.0001).
63. OBrien-
Presents a meta-
206 patients from a large urban teaching hospital, purposive sampling in selected units (critical care unit, intensive care unit, etc.). 14 nursing units
Presence of new
Amount of direct
Multifactorial intensity & complexity of care
52
Authors, Year Pallas, Irvine, Peereboom, Murray (1997)
Focus paradigm for examining nursing work & resource use.
Sample in a 489 bed pediatric tertiary care, urban, universityaffiliated hospital, crosssectional sample. Client: demographics, nursing & medical diagnoses, OMAHA scores, SF-36 health status, time on program Nurse: education, experience, professional status. Agency: geographic location, visit type, caseload, proportion of nurse worked hours & continuity of care. Nurse: perceived adequacy of care time. Agency: total visits. 38 RNs, 11 RPNs; 751 clients receiving home healthcare (6,840 visits or 7% of agency caseload during study period); convenience sample. Clients were unit of analysis. Environmental Complexity: competing demands/nurse safety, unanticipated case complexity, formal information exchange, voice mail, travel, unanticipated admissions. Agency: visit time Patient health status, OMAHA scores (knowledge, behaviour, and status).
Inputs complexity (NANDA), medical complexity case mix groups (CMGsTM), medical severity (length of stay).
Throughputs staff, relief staff, unanticipated events.
Outputs care required (PRN 80), costs of care.
64. OBrienPallas, Doran, Murray, Cockerill, Sidani, LaurieShaw, LochhassGerlach (2001)
Variables that affect nursing utilization in a home visiting nursing service.
Findings model examines costs, nursing work & variability in resource use across patients & environments. Relationships observed between 4 key variables & workload: patients nursing condition (positive linear relationship with # of diagnoses), medical condition (most have coefficients of variation > 0.5), caregiver characteristics & the environment. Overall, Client Care Delivery Model explained 47% (R2 = .46) of the variation in average visit time. Medical & nursing diagnoses explained 14.7% of variation in average visit length. Specifically, mental health diagnoses contributed to longer but not necessarily more visits. Unanticipated case complexity & unanticipated admissions were positively associated with greater average visit time, explaining 20.5% of the variation.
65. OBrienPallas, Thomson, Alksnis, Bruce
Economic impact of staffing decisions
Ontario acute care hospitals
Hospital Characteristics: earned (paid) hours patient per day (for
Overall, Client Care Delivery Model explained 35.6% (R2 = .33) of the variation in number of visits. Visits performed by degree-prepared nurses resulted in fewer total visits & improved RN perceptions of visit adequacy. Greater time per visit, higher scores for formal information exchange & continuity of care by the primary nurse were associated with fewer visits. A medical diagnosis of neoplasm, greater number of nurses visiting client, palliative & long duration visits types, & increasing use of voice mail were associated with increased number of visits. Between 1994 and 1998, inpatient cases dropped by 184,766 while outpatient cases increased by 144, 603. Data suggest that an increase in the overall resources used by
53
Focus
Sample
Inputs RN, RPN, & UCP combined), RN survey of absenteeism & overtime, staffing, compensation, workload data & productivity Patient Characteristics: relative intensity weights, complexity of inpatient hospital cases (1994/95 1998/99);
Throughputs
Outputs
Findings hospitalized patients in recent years even though the number of hospitalized cases & the average length of stay have decreased. Complexity levels have tended to increase for all age groups in each year between 1994 and 1998, while the overall number of nurses working in hospital settings ahs decreased. In 1998/99 alone, $171 million spent on overtime hours (approx. 2,250 FTEs); of which, $57 million on overtime pay premiums. $19 million spent on nursing agency personnel & $39 million spent on sick time (approx. 765 FTEs).
66. Pinkerton, Rivers (2001)
Factors that affect staffing needs.
Nurse education, experience, use of relief staff, workload, proportion of nurse worked hours
Frequency & complexity of changes.
Burnout.
Overtime costs almost perfectly correlated (r=.928, p>.01) with sick time costs. Hours patient per day were related to both overtime costs (r=.439, p<.01) & sick time costs (r= .488, p<.01) suggesting that as hours of care patient per day increased, so did the overtime that nurses were asked to work & the incidence of missed shifts due to illness. Important variables include unit cohesiveness, chaos factors, communication, organizational skills of nurses, support staff availability, & number of float staff.
54
Authors, Year 67. Potter, Barr, McSweeney, Sledge (2003)
Focus Relationship between RN staffing levels & patient outcomes.
Sample All acute inpatient care units (n=32) of one hospital. 2000.022001.01. Unit of analysis was inpatient unit. Adjusted for float percentage & acuity.
Inputs 8 hour day shift nurse staffing converted to direct nursing care daily hours per patient for all nursing personnel per month; average: number of hours of nursing care per patient daily on day shift; percentage of RN & UAP direct care hours; float percentage; total patient care hours.
Throughputs
Outputs Patient outcomes: falls per 1000 patient days, medication errors per 1,000 patient days, self-reported symptom management (VAS), self-care & health status (National Center for Health Statistics Health Interview Survey) & postdischarge patient satisfaction.
Findings Percentage of RN hours negatively correlated with patients perception of pain & positively correlated with patients perceptions of selfcare ability & health status, as well as satisfaction post-discharge.
68. Prescott (1993)
Impact of nurse staffing levels & skill mix on patient outcomes 1044 staff nurses working on 90 patient care units in 15 hospitals. Data collected in 1981 & 1982 33 predictor variables. Staff-patient ratios Unit vacancy rates, voluntary turnover, stability.
Patient characteristics: acuity (vendorbased patient classification tool). Proportion of nurse worked hours, staffing levels.
Patient mortality, quality, and costs of care.
69. Prescott (1986)
Whether organizational, administrative, & practice factors differentiate among hospitals & patient care units as to registered nurse
High percentage of RNs is associated with lower than expected mortality rates (13 studies), length of stay, costs, complications. Salary savings of declining skill mix may be offset by productivity declines. Model explained 52% of variability in vacancy rates, 56% of variability in stability rates, & 42% of variability in relative turnover. High vacancy rates associated with 7 variables including high staff-patient ratios on evening shift & perceived inadequacy of working conditions.
55
Authors, Year
Focus vacancy, stability, & turnover rates.
Sample
Inputs at organizational level.
Throughputs
Outputs
Findings High stability associated with 7 variables including increased staff-patient ratios & perceived inadequacy of working conditions.
70. Robertson, Dowd, Hassan (1997)
Staffing intensity & costs of care.
196 acute care hospitals in California from AHA annual survey.
Proportion of Nurse worked hours
Costs of care, Staffing ratios.
71. Sainfort, Karsh, Booske, Smith (2001) All acute-care hospitals in United States, secondary data from national survey. Staffing levels (shortages), patient demographics, nurse education, care delivery system.
Review of literature on characteristics & impact of healthy work organizations
Hospital size, care delivery system, staffing
Work environment
Nurse health status, patient complications, job satisfaction, burnout
72. Seago, Ash, Spetz, Coffman, Grumbach (2001)
Characteristics of acute care hospitals that report RN shortages when widespread shortage exists & when widespread shortage is no longer evident. 647 females RNs Career stage, work-
73. Seybolt (1986)
Understanding the
Nurse turnover
High turnover associated with 7 variables including low staff-patient ratios on night shift & perceived inadequacy of working conditions. Negative association between some staffing levels & costs of care (r=-0.04 to r=-0.36). Higher respiratory care technician staffing intensity is related to lower costs per episode of care (r=-0.36). Positive relationship between RN staffing intensity & costs of care (may be due to use of high skill mix strategy) r=0.08. There are limits of skill-specific staffing intensity below which costs of care are actually increased. Understaffing increases costs & reduces quality. Creates a model & research agenda for healthcare quality improvement & patient safety. Organizational factors & working conditions affect employee health & productivity. Location in South, high percentage of nonwhite county residents, high percentage of patients with Medicare as payer, higher patient acuity & use of team nursing care delivery predicted hospitals reporting shortages both when there was & when there was not a widespread shortage. Wage is not a significant predictor of shortages. RN workforce policy needs to place emphasis on distribution relative to overall supply. Turnover intentions of employees at different
56
Authors, Year
Focus causes of premature nurse turnover in order to retain nurses. Adverse patient outcomes, mortality
Sample at a large West Coast hospital
Throughputs
Outputs intentions
Findings career stages are affected by differing work-role design factors. Adverse occurrences associated primarily with patient characteristics. Failure to rescue associated more with hospital characteristics than patient characteristics (higher relative risk). Understanding reasons behind variations in mortality should be used to upgrade quality of care.
74. Silber, Williams, Krakauer, Schwartz (1992)
Hospital & patient characteristics that predict mortality after surgery.
Inputs role design (job, interactions, organizational policies) Patient characteristics, number of hospital beds, staff ratios
75. Sochalski (2001)
Quality of care, nurse staffing & patient outcomes.
Nurse demographics, education, proportion of nurse worked hours
Work environment, workload
Job satisfaction, burnout, patient complications
Medical/surgical RNs had lowest scores on quality of care (compared to other types of units), had a higher number of tasks left undone at the end of shift & are experiencing significant levels of burnout. Higher workloads were associated with lower quality (r=-0.24). 41% of nurses were moderately or very dissatisfied with their job. Outcome: Patient outcomes (annual fall rates, nosocomial pressure ulcers, urinary tact infections, patient satisfaction with various surveys) - RNs fewer in number with an increase in unlicensed assistive personnel. - increased RN hours worked patient per day associated with lower fall rates (F=11.73, p=0.002) & higher patient satisfaction with pain management (F=15.05, p=0.0007) - increased worked hours patient per day by all staff associated with lower rates of urinary tract infections.
76. Sovie, Jawad (2001)
Impact of hospital restructuring on patient outcomes
2831 patients undergoing cholecystectomy & 3141 patients undergoing transurethral prostatectomy. Random selection, from 7 states. 13,200 medicalsurgical RNs from acute care hospitals in Pennsylvania. Random sample from state board database, 52% response rate. 1997 & 1998 fiscal year data from 29 U.S. university teaching hospitals (with > 300 acute operating beds) from 8 of 9 U.S. census regions; 1 inpatient acute adult medical Structure: MECON-PEERx Operations Benchmarking Database Reports (FTE for each type of nursing personnel; proportion of nurse worked hours; hours worked per patient daily for Process: Management Practices & Organizational Processes Questionnaire & Quality of Employment Survey subscale on autonomy & decision-making
57
Authors, Year
Focus
Sample unit & surgical unit per hospital; Units of analysis: hospital nursing dept (incl. intensive care units); medical units; surgical units 9000-bed academic medical centre in Texas. Costs of care, productivity, staffing ratios
77. Strickland, Neely (1995)
Implementation of a Standard Staffing Index to allocate nursing staff.
Inputs RN, unlicensed staff, LPN, clerks, managers; labour costs per discharge; restructuring assessment tool; & interview) - nurse demographics & satisfaction (Individual Nurse Questionnaire) Staffing mix, patient acuity
Throughputs
Outputs
Findings
78. TarnowMordi, Hau, Warden, Shearer (2000)
The relationship of nursing requirements & workload measures & hospital mortality in the intensive care unit.
Mortality rates
New system determined nursing productivity quickly & efficiently & specific patient need were determined through regular & thorough evaluations. One person was responsible for staffing on each unit. The SSI allowed for an efficient & accurate utilization of nursing personnel. Unadjusted mortality was greater for patients exposed to high versus moderate overall intensive care unit workload (odds ratio 4-0 [26-6.2] (p. 187). Patients exposed to high intensive care unit workload were more likely to die (odds ratio 31 [1.9-5.0]) than those exposed to lower workload, both before & after adjustment for risk by the APACHE II equation. Two measures of intensive care unit workload most strongly associated with adjusted mortality (excluding total unit nursing care requirements) were peak occupancy & the ratio of occupied to appropriately staffed beds.
One adult intensive care unit in the UK. All admissions (n=1050) between 1992 & 1995 that met criteria for adjustment of mortality risk by the APACHE II equation (Acute Physiology & Chronic Health Evaluation)
- Patient predicted risk of mortality (APACHE II equation which uses information from the 1st 24h after admission). intensive care unit workload: occupancy (highest number of beds occupied each shift & peak occupancy as the highest occupancy per shift during patient stay), total
58
Authors, Year
Focus
Sample
Throughputs
Outputs
Findings Use of aggregate data (total) intensive care unit nursing care requirements) may explain the association between high intensive care unit nursing requirement & mortality, potentially because more seriously ill pts are more likely to die.
79. Wai, Bame, Robinson (1998)
Literature review of nursing turnover.
Inputs Intensive care unit nursing requirement-UK Intensive Care Society recommendation, ratio of occupied to appropriately staffed beds Nurse demographics, education, professional status, hospital size. Workload Costs of care, job satisfaction. Nurse demographics. Job satisfaction
80. Weisman, Alexander, Chase (1981)
Reasons for nursing turnover.
Older staff, minorities, those with higher income, emotional support, or longer tenure had lower turnover. Factors leading to turnover include job satisfaction & tension, organizational commitment, job possibilities & supervisor behaviour. Majority resigned due to job dissatisfaction (57.1% and 72.5%). Personal factors have little effect on turnover process. Reasons for resigning include work pressures due to understaffing & number or scheduling of work hours. Low pay was least frequently cited as the reason (less than 2%). Medical errors, fall rates, infections, ulcer rates. Significant inverse relationships present between staffing & falls in cardiac intensive care (r=-0.53), medication errors in cardiac & non-cardiac intensive care unit (r=-0.55 and 0.65 respectively) & restraint rates in medicalsurgical units (r=-0.48). No statistically significant relationships were found between the outcomes of central line infection rates & pressure ulcer rates & WHPPD across specialty units. An inverse relationship between WHPPD & falls was present in cardiac intermediate care (r=-0.53). The impact of staffing on outcomes
81. Whitman, Yookyang, Davidson, Wolf, Wang (2002)
Determine the relationships between nursing staffing & specific nursesensitive outcomes.
1259 full-time RNs in two large universityaffiliated hospitals. Entire population was targeted, 98% response rate. Observational data from 95 patient care units across 10 acute care hospitals in eastern US. Nurse professional status, staffing hours, patient days per unit, worked hours patient per day (WHPPD) Structural hospital or unit variations.
59
Authors, Year
Focus
Sample
Inputs
Throughputs
Outputs
Findings is highly variable across specialty units, but when present, the relationships are inversely related with lower staffing levels resulting in higher rates of all outcomes.
4. Sample 113 RNs from a tertiary care medical centre in south-eastern United States. Stratified random sampling, 66% response rate. Nurse ratios, patient teaching, and medical diagnosis. Length of stay in intensive care unit & step-down units, costs of care, patient complications Inputs Nurses experience. Throughputs Work environment, unanticipated events, environmental uncertainty. Outputs
Throughputs Findings No difference between nurses work status or experience & response patterns after chi-square analysis. Increase in environments complexity (r=0.49), changeability (r=0.34) & unpredictability (r=0.56) lead to increased environmental uncertainty.
Authors, Year 82. Allred, Michel, Arford, Carter, Veitch, Dring, Beason, Hiott, Finch (1994)
Focus Study of environmental uncertainty.
83. Brown (2000)
One-Stop Recovery: a fast-track program for cardiac surgical patients.
84. Cady, Mattes, Burton (1995)
Implementation of a step-down unit to decrease intensive care unit length of
Teaching hospital with 220 adult beds, 27% are intensive care
Patient demographics, medical diagnosis, patient
Nursing workload
Length of stay in intensive care unit. , readmissions to
Fast-track programs improve patient comfort, enhance quality of care, & reduce costs. Barriers include physician reluctance, limited resources, lack of communication, & patient & family anxiety. One-stop Recovery offers continuity of staff, limits multiple inpatient transfers, increases flexibility in human & institutional resources, reduces intensive care unit readmissions, & enhances patient comfort & family support. Patient care is provided in one location by consistent staff throughout the recovery. Time spent in intensive care unit decreased significantly for those admitted to step down unit 1.99 days instead of 3.35). Length of stay did not differ for those in a step down unit.
60
Authors, Year
Focus
Throughputs
stay.
Sample unit. Comparable institutions provided information for comparison. Multidisciplinary healthcare team at Brigham & Womens hospital. Medical diagnosis.
Inputs acuity, proportion of nurse worked hours, nursing ratios.
85. Cohn, Rosborough, Fernandez (1997)
Reducing costs & length of stay & improving efficiency & quality of care in cardiac surgery.
Outputs intensive care unit. , nurse job satisfaction, patient complications, costs of care. Length of stay, costs of care
Findings Quality of care remained excellent & costs were decreased for patient in the step down unit.
86. Drenkard (2001)
Strategic planning methodology for nursing care.
Care delivery system.
87. Duffy, Lemieux (1995)
Service-line concept & patient-centered care in a cardiac setting. Eight million patient days from Medicus National Database in US. 16 patients who had worked with Continuity of care. Hospital size, workload, proportion of nurse worked hours.
Continuity of care.
Costs of care, productivity.
88. Helt, Jelinek (1988)
Nursing productivity & quality in the wake of cost cutting.
Perceived quality of care, length of stay, costs of care.
89. Lamb, Stempel (1994)
Nurse case management from the
Patient complications,
Multidisciplinary group met weekly to discuss problems with cardiac surgical services. Care Coordination Team monitors clinical pathways & recommends ways to improve services. Results include higher volume of surgery, decreased length of stay (by about 15%), decreased costs, & increased patient satisfaction (to 95%). The methodology uses a transformational leadership assessment tool, quality planning methods & large group interventions to engage nurses in implementation of strategies. Six driving strategies for nursing were determined: leadership, practice, culture, learning, and research & role clarity. A horizontal, multidisciplinary environment with a performance-based model is described. Employee roles are broadened to focus on the entire process of care with all disciplines working together. Even in face of staffing reductions, productivity (decrease in ratio of nursing hours to workload) & quality (based on specific objectives) have improved. While acuity increased, length of stay had decreased. Increased percentage of RNs leads to increased productivity, only half of which is lost due to higher costs. The study explored patients perspective of working with a nurse case manager. Patients
61
Authors, Year
Focus patients perspective: the insider-expert.
Sample a nurse case manager during hospitalization, ranging in age from 66-100.
Inputs
Throughputs
Outputs health status.
Findings described the process of nurses becoming their insider-expert. This consists of three phases: bonding, working & changing. Relationship with a nurse insider-expert enabled patients to improve health outcomes, have fewer hospitalizations, & better quality of life.
5. Sample Inputs Proportion of nurse worked hours. Nurse education & experience, agency size. Throughputs Outputs Patient complications & outcomes, staffing ratios, costs of care. Patient satisfaction
Patient Outcomes
Authors, Year 90. Brooten, Naylor (1995)
Focus Nurses effects on patient outcomes.
91. Dansky, Brannon, Wangsness, (1994)
Staffing characteristics & patient satisfaction in home healthcare.
92. Fortinsky, Madigan (1997).
Relationship between measures of home care resource consumption & patient outcome measures.
13 not-for-profit home health agencies in Pennsylvania & Ohio. N = 201 adult medical/surgical home care patients who began new episodes of home care from 10 medicarecertified home care agencies in Ohio. Patient characteristics: the 29 item OASIS assessment (demographics, clinical & functional health status, illness & rehab prognosis & amt. of family & other informal support recd by
Patient outcomes: discharge status; change in clinical & functional health status measures between admission & discharge.
Findings Most important issues include types of patient outcomes that should be measured & the amount & type of nursing needed in a given environment, for specific patient groups & in order to affect outcomes. Higher numbers of full-time staff or of BNSprepared RNs predicted higher patient satisfaction. Size of agency had no impact on satisfaction. Agencies with medium benefits had the highest patient satisfaction. Patients whose episodes ended with discharge at home vs. hospitalization had similar total visit numbers & costs but those discharged to hospitals utilized the home care resources over less time. No significant differences were noted in the use of home care resources between patients who improved & those whose health status declined during the episode of home care. Patients who were admitted to home care from hospitals & who received home care for longer
62
Authors, Year Resource consumption: total # of home visits by all disciplines; disciplinespecific cost information; length of stay; service intensity (# of visits per day) Patient complications, staffing ratios. 38 RNs, 11 RPNs, 751 clients receiving home healthcare, convenience sample. Client: demographics, nursing & medical diagnosis, OMAHA scores, SF-36 health status, time on program Nurse: demographics, education, experience, professional status Agency: Patient health status, OMAHA (knowledge, behaviour, status)
Focus
Sample
Inputs pts at home)
Throughputs
Outputs
Findings than the 62-day study period used the greatest number of resources.
93. Lancaster (1997)
Effects of staffing on infection rates.
94. OBrienPallas, Doran, Murray, Cockerill, Sidani, LaurieShaw, LochhassGerlach (2002)
Variables that affect client outcomes with a home visiting nursing service.
Increased patient-to-nurse ratios place severe time constraints on nurses. This has been shown to result in increased infection rates with central venous catheters. Clients with degree-prepared nurses had better OMAHA scores (1.8 times better odds of improved knowledge and 2.2 times better odds of improved behaviour). Clinical, provider, organizational & environmental factors affect outcomes. Medical & nursing diagnoses explain much of variation in outcomes. Environmental complexity was negatively associated with client outcomes.
63
Authors, Year
Focus
Sample
95. Proctor, Yarcheski, Oriscello (1996)
Relationships between hospital variables & patient outcomes. 872 nursing home residents & 10 nursing homes. Organizational design included 3 structural measuresjob assignment, hierarchy, closeness of supervision & 2 nature-of-task measurespace of operations & workload. Resident outcomes: The 7item physical function scale included bladder incontinence, bowel incontinence, bathing, eating, mobility (walking or wheeling), dressing, & transferring.
68 patients diagnosed with MI from a large urban medical centre.
Inputs geographic location, visit type, caseload, proportion of nurse worked hours & continuity of care Patient demographics Patient judgment of quality of care.
Throughputs
Outputs
Findings
96. Rohrer, Momany, Chang (1993).
The relationship between nature-ofthe-task aspects of organization design, structural aspects of organization design & organizational effectiveness (operationalized as outcome resident functional ability).
Significant correlations between nursing care & patient outcome post-MI (0.40) as well as hospital environment & pt. outcome post-MI (0.28). Nursing care accounted for 16 % of variance in pt. outcomes. Pt. level of education was related to outcome (0.41). Only job assignment & hierarchy (structural variables) were related to improved physical function. In general, assuming a stable pace of operations & workload, non-specific job assignment & less hierarchy were related to better physical function. A consistent workload effect was demonstrated in that fewer heavy care residents resulted in better resident physical functioning.
64
Authors, Year 97. Roseman, Booker (1995)
Focus Workload & environmental factors in medication errors.
Sample All med errors in a 5 year period from a 140 bed hospital in Alaska.
Inputs
Throughputs Work environment, workload.
Outputs Med errors.
98. Silber, Rosenbaum, Schwartz, Ross, Williams (1995)
Complication rate as a measure of quality of care in coronary artery bypass graft surgery.
16,673 patients who underwent coronary artery bypass graft surgery (CABG) at 57 hospitals in 1991, data from American Hospital Association annual survey.
Patient demographics, hospital size, nursing ratios, medical diagnosis.
Patient mortality, complication, & failure to rescue rates.
Findings Errors were positively associated with number of shifts worked by temporary staff & with patient days but negatively associated with overtime shifts (use of experienced nurses). A seasonal pattern of errors emerged: errors corresponded with the level of darkness that occurred 2 months earlier (i.e. Winter darkness correlated with increased med errors in early spring). Many hospitals with higher quality of care had higher complication rates but lower mortality rates (ex: facilities with an MRI had 38% increase in complications). Hospital rankings based on complication rates give different information than those based on mortality rates. Complication rates should not be used to judge hospital quality of care until more is known about the difference.
6. Sample 834 nurses from England. Data collected via postal survey, hospitals chosen were stratified to include all health regions, response rate 57%. Inputs Nurse demographics.
Nurse Outcomes Throughputs Organizational characteristics. Outputs Job satisfaction. Findings There were no correlations with job satisfaction & nurses age or level of education. Some correlations were found between job satisfaction & cohesion of nursing team (0.51), staff organization including staffing & workload (0.46), level of professional practice (0.46) & collaboration with medical staff (0.41). Most important factors in job satisfaction were social & professional
Authors, Year 99. Adams, Bond (2000)
Focus Effects of individual & organizational characteristics on job satisfaction.
65
Authors, Year 820 RNs & LPNs on AIDS units. All nurses on selected units were surveyed with an 86% response rate. 1998 survey of more than 2,000 nurses in 22 hospitals; 1996 surveys from chief executive officers at 646 hospitals. 204 female nurses from a university hospital in Germany. Burnout, effort & reward imbalance, Proportion of nurse worked hours, workload, continuity of care, nurse work index. Organizational restructuring. Mortality, patient satisfaction. Care delivery system, nurse education & experience Work environment & organization. Burnout, job satisfaction, nurse health, patient complications.
Focus
Sample
Inputs
Throughputs
Outputs
100. Aiken, Sloane (1997)
Burnout in AIDS care nurses.
101. Aiken, Clarke, Sloane (2000)
Findings relationships with nursing & other colleagues. Nurses in dedicated AIDS units were less emotionally exhausted than in scattered bed units. The organizational attributes associated with lower burnout are also related to safer work environment, greater satisfaction with care & lower mortality. Organizational support accounts for 5% variance in emotional exhaustion. Nurse well-being is enhanced by autonomy & control over work. Nurse control over the practice environment explains variations in patient satisfaction. Better organizational support was associated with lower emotional exhaustion. The higher the staffing level, the lower the death rate (r=-0.49).
102. Baker, Kilmer, Synergist, Shuffle, (2000)
Describe how nurse staffing changed relative to hospital restructuring (case mix of patients receiving care), & examine changes in nursing practice environments. Investigates associations between experiences of work stress among nurses.
103. Baumann, Giovannetti, OBrien-Pallas, Mallette, Deber, Blythe, Hibberd, DiCenso (2001)
Nurses perceptions affected by job change experiences.
1662 nurses from two large teaching hospitals. The entire population was surveyed, 50.7% response rate.
Nurse demographics.
Work environment, workload.
Perceived quality of care.
Effort & reward imbalance was predictive of emotional exhaustion (F=35.33) & depersonalization (F=8.97), both dimensions of burnout. Nurses feelings of personal accomplishment were lowest among those with a mismatch between demands and rewards. Burnout is observed more often among younger nurses. All nurses were negatively affected by hospital restructuring. Nurses reported that work environments had deteriorated in the previous year with heavier workloads, greater patient acuity & more workplace injuries. Nurses perceived a decline in quality of care. Those who experienced job change (involuntary reallocation) were more dissatisfied, less confident, more concerned about patient
66
Authors, Year Meta-analysis of 48 studies & 15,048 subjects. Nurse demographics, experience. Work environment. Nurse job satisfaction, burnout.
Focus
Sample
Inputs
Throughputs
Outputs
104. Blegen (1993)
Variables that affect nurses job satisfaction.
105. Bourbonnais, Comeau, Vezina, Dion (1998)
The psychological effects of nurses work environments.
Nurse demographics, experience
Nurse burnout, psychological distress, job strain, social support at work.
Findings welfare, & less committed. Job satisfaction most strongly associated with stress & organizational commitment. Other factors included communication with supervisor, autonomy, recognition, routinization, communication with peers, fairness, & control. Low correlations with age, yrs of experience, education, & professionalism. High psychological demands & low decision latitude is associated with psychological distress (adjusted odds ratio of 2.34) & emotional exhaustion (OR=5.77). Social support at work altered mental health but not job strain.
106. Buchan (1999)
A follow-up of magnet hospitals 15 years after their designation to see how restructuring has affected the status.
1891 nurses from six acute care hospitals in Quebec, voluntary recruitment. Mostly bedside nurses working full time. 10 magnet hospitals & 5 ANCC hospitals Cost of care
Due to reorganization, some hospitals no longer exhibit characteristics of magnetism. It is not the magnet hospital label that is important, but the concepts of quality care, effective staff deployment & job satisfaction. There is a need for monitoring & re-accreditation to maintain a live register of magnet hospitals.
107. Burke, Greenglass (2000)
Effects of hospital restructuring on nurses.
1362 nurses in Ontario. Random selection from a nurses' union, 35% response rate.
Administration, professional practice & development factors (staffing, care delivery models, proportion of nurse worked hours) Nurse demographics, experience, education, & professional status. Workload, work environment.
Nurses emotional & physical health, burnout.
Full time & part time nurses experienced hospital restructuring & downsizing in similar ways. FT nurses had poorer health, were more emotionally exhausted (mean=3.6/6 vs. PT mean=3.0/6) & were more likely to be absent (mean=3.2/4 vs. PT mean=2.5/4). Restructuring was associated with less work satisfaction & poorer well-being.
67
Authors, Year 108. Cameron, Horsburgh, ArmstrongStassen (1994)
Focus Job satisfaction, propensity to leave & burnout in RNs & RN assistants.
Sample 623 RNs & 231 RN assistants from 3 community hospitals.
Inputs Nurse demographics
Throughputs Work environment
Outputs Nurses job satisfaction, burnout.
109. Carey, Campbell (1994)
Strategies for nurse retention: preceptors, mentors, & sponsors.
Nurse education, experience.
Job satisfaction
Findings Nurses were only moderately satisfied with their jobs (mean scores are lower than other employees). RNs with more experience had higher job satisfaction & less burnout. RNs in psychiatric settings were least satisfied. Greater satisfaction & less burnout when a fit was demonstrated between person & environment (tables omitted) No causal relationship between mentors & job satisfaction (R2 for factors = 0.01-0.08). Nurses leave b/c of dissatisfaction rather than needs for recognition, accomplishment, or self-worth. Environments where management supports interpersonal relationships have higher levels of satisfaction & less turnover.
110. Clarke, Laschinger, Giovannetti, Shamian, Thomson, Tourangeau, (2001)
Effects of workplace attributes on nurses satisfaction & quality of care.
Nurse demographics, experience.
Work environment.
Nurse burnout, job satisfaction, perceived quality of care, patient adverse events (falls, med errors).
Strongest predictors of nurses emotional exhaustion & satisfaction with jobs are having control over work environment, having sufficient resources & effective nursing leadership. Nurse-assessed quality was significantly correlated with occurrence of adverse events (r=-0.145for falls to -0.454 for wrong med). Hospitals with good physiciannurse collaboration & strong nursing leadership have less burnout & lower turnover intentions. Leadership & nurses length of experience on the unit were predictive of intent to leave current job.
111. Davison,
Effects of healthcare
143 staff nurses from two large teaching hospitals in Atlanta. Random sample selection with 44% and 36% response rate from respective hospitals. 17,965 RNs in 392 hospitals from Alberta, Ontario & British Columbia. Representative samples were drawn from Ontario & British Columbia while all RNs in Alberta were sampled. Response rate was 49-57%. Longitudinal Nurse Hinshaw &
Perceived high workload (Price & Mueller
68
Authors, Year Folcarelli, Crawford, Duprat, Clifford (1997)
Focus reforms on job satisfaction & voluntary turnover among hospital nurses
Sample survey of 685 nurses from one hospital between 1993 and 1994
Inputs demographics, wage, clinical advancement level, termination status, work conditions Nursing demographics, experience. Workload.
Throughputs
112. Demerouti, Bakker, Nachreiner, Schaufeli (2000)
Factors contributing to burnout & life satisfaction in nurses.
109 nurses from one hospital & two nursing homes in Germany, response rate 59%. Ethical climate shared perception of how issues should be addressed & what is ethically correct Proportion of nurse worked hours, care delivery models (primary nursing, team nursing, case management), continuity of Job satisfaction of nurses (with pay, promotion, coworkers, supervisors, work itself)
Outputs Atwoods Nurse Job Satisfaction Scale; Price & Muellers Model of Turnover; Perlin & Schoolers Personal Mastery Scale Burnout.
Findings overload subscale) was an important determinant of low job satisfaction. Insufficient time to complete the job predicted turnover.
113. Joseph, Deshpande (1997)
Hospitals can have various types of ethical climates. Managers may be able to enhance nurses satisfaction by altering this climate.
114. Kangas, Kee, McKee-Waddle (1999)
Patient & nurse satisfaction within different care delivery models & organizational structures.
144 nurses from large non-profit hospital. Average subject was 40 year old married female who has worked at hospital for 9 years. 50% response rate. 102 nurses & 102 patients from 3 different hospitals (2 traditional, 1 shared governance). Systematic random sampling
Nurse job satisfaction
Age & occupational tenure showed a positive relationship with exhaustion. Age was significantly, negatively related to life satisfaction. Job demands have a strong positive effect on exhaustion while job resources have a strong negative effect on disengagement. Job demands & job resources correlate negatively with each other (-0.61). Professional, instrumental (protect own interests), independence (decide for oneself whats right) climates had no impact on job satisfaction. Caring (whats best for everyone) climate influence pay, supervisor, & overall satisfaction. Rules (strict with policies) climate had positive impact on overall satisfaction. Efficiency (must control costs) climate had negative impact on satisfaction with supervisors. No difference in nurse job satisfaction between types of care delivery models. Somewhat higher patient satisfaction for those within the primary care delivery model (not significant). Supportive environment (=-0.709) & working in a specialized unit (=-0.305) increase nurse job satisfaction.
69
Authors, Year
Focus
Inputs care, patient demographics Nurse satisfaction
Throughputs
Outputs
Findings
115. Kovner, C., Hendrickson, G., Knickman, J., Finkler, S. (1994)
Nurse recruitment & retention.
Pilot project to improve recruitment & retention. All of the innovations enhanced satisfaction. Nurses ranked pay as most important factor, followed by autonomy & professional status. Education initiatives, reorganization, & new technology enhanced satisfaction. With each change, there was an initial dissatisfaction.
116. Kramer & Schmalenberg (1990)
Job satisfaction & retention of nurses in magnet & nonmagnet hospitals.
Sample used. Inclusion criteria included 6mos experience & type of unit. 37 hospitals, 858 RNs on 68 pilot units & 335 RNs on comparison units. Pilot units were selfselected; comparison units were chosen by evaluators to be similar to pilot units. 1800 nurses in magnet & nonmagnet hospitals across United States. Image & valuation of nurses how they see themselves & how others see them;
Magnet hospitals have higher degree of satisfaction & better staffing situations than non-magnet. Positive correlation between hospital image of nursing & adequacy of staffing.
117. Kramer, Hafner (1989)
Impact of values on nurses satisfaction & perceived productivity.
2336 staff nurses in 24 hospitals. A 1/3 sample, proportionate by regions of the country, of the magnet hospitals
Work environment.
Job satisfaction, including organizational structure, professional practice, management style, quality of leadership, professional development; also overall job satisfaction. Nurses job satisfaction, perceived quality of care.
Inverse correlation between value congruence & nurse job satisfaction, quality of care (for staff nurse-top manager dyad, correlation between value congruence & satisfaction ranged from 0.074-0.377). Staff nurses reported fewer factors as important to satisfaction & quality of care than did other members of
70
Authors, Year
Focus
Inputs
Throughputs
Outputs
118. Kutzscher, Sabiston, Laschinger, Nish (1997).
Effects of teamwork on staff perception of empowerment & job satisfaction.
Work environment
Job satisfaction.
Findings nursing. Important factors include role clarity, role evolvement, role distance, responsiveness from management, autonomy. Perceptions of work empowerment were higher for staff who were on teams (t=5.04). The staff on teams was slightly more satisfied but the difference was not significant.
119. Laschinger, Finegan, Shamian (2001)
Impact of workplace empowerment & organizational trust on nurses work satisfaction.
Nurse demographics, professional status.
Work environment
Nurses job satisfaction
Staff nurse empowerment impacts on their trust in management & their job satisfaction. Fostering environments that enhance empowerment will have positive effects on members & effectiveness. Access to information (correlation=0.49) & support (0.46) are strongly related to trust in management. Feedback & guidance are also related to trust. Nurse burnout, meaningfulness of work, patient outcomes, & satisfaction. Patients perceptions of quality were correlated with nurses relationships with work (more meaning, less exhaustion=higher perception of quality). Patients were more satisfied on units where nurses found the work meaningful & were less satisfied on units where nurses were exhausted or rated high on cynicism (Spearman rank order correlations). No correlations between professional efficacy & patient satisfaction.
120. Leiter, Harvie, Frizzell (1998)
Impact of nurse burnout on patient satisfaction.
Sample was drawn, random sampling of nurses. 210 staff who participated on multidisciplinary teams & a random sample of 185 staff (response rate 52%) who did not. 412 staff nurses from Ontario. Random selection from professional registry list, equal sampling of males & females. 711 nurses & 605 patients from sixteen hospital units. Volunteers completed the nurse surveys & patients were randomly sampled. Nurse & patient demographics.
71
Authors, Year 121. McGillis Hall, Doran, Baker, Pink, Sidani, OBrien-Pallas, Donner (2001)
Focus Impact of nursing staff mix models & organizational change strategies.
Inputs Proportion of nurse worked hours, medical diagnosis (case mix group)
Throughputs
Outputs Med errors, infections, nurse job satisfaction, perceived quality of care.
Findings Nursing leadership has positive influence on nurses job satisfaction (t=4.88). Lower complexity of patients corresponds with high job satisfaction (t=-3.17). Units with lower proportion of RNs to RPNs had more med errors & wound infections. Highest rated value was good supervisory relations. If managers support nurses in attaining values in work setting, retention may be improved. Correlations between supervisory relations & security (r=0.59) & security & achievement (r=0.60) were noted. Negative correlation with economic values & job satisfaction (r=-0.14). By using appropriate management styles, staff nurse job satisfaction may be improved. Job satisfaction improved as the style approached participative management.
122. McNeeseSmith, Crook (2003)
Variables influencing nurses values.
Nurse demographics, education
Nurse job satisfaction
123. Moss, Rowles (1997)
The effect of nurse managers management styles on staff nurse job satisfaction.
Sample 2046 patients, 1116 nurses, 63 unit managers, 50 senior executives from 19 teaching hospitals across Ontario. 412 RNs from 3 Los Angeles hospitals. Hospitals were selected for convenience, nurses were randomly sampled. 623 nurses in 3 Midwestern hospitals Job satisfaction
124. Munro (1983)
Job satisfaction among recent graduates.
Management styles (exploitive/ authoritative, benevolent/ authoritative, consultative, participative). Nurse education
Job satisfaction
Education background did not affect job satisfaction. Achievement, responsibility (33% of variance), advancement, growth (2.2% of variance), & work itself (5.5% of variance) are related to satisfaction. Administrators need to appeal to nurses needs for challenges & opportunities to grow.
125. Nakata, Saylor
Management style &
329 recent nursing graduates. Design was a stratified, twostage probability sample of high school grads in US (2% in nursing). 102 RNs &
Work
Job satisfaction.
Positive correlation between perceived
72
Focus nurse satisfaction.
Inputs
Throughputs environment.
Outputs
Findings management style & staff nurse job satisfaction (r=0.48). The closer the management style to participative group management, the higher the satisfaction. Nurses would like to be more involved in decision-making & setting of unit goals. Autonomy & authority are sources of job satisfaction while poor communication leads to dissatisfaction.
126. RNAO (2002)
To determine the extent to which RNs have seen specific changes in their work environments since the task force recommendations were released.
Sample LPNs from an acute care hospital in California. All staff in selected units was surveyed with a 43% response rate. Surveys given to RNAO members & non-member RNs, RNs age 23-67 yrs. Convenience sample, 549 responses. Nurse demographics, employment status, sector of employment, position, workload, proportion of nurse worked hours Nurses education. Work environment. Nurse job satisfaction. Work environment. Nurse job satisfaction.
127. Roedel, Nystrom (1998)
Factors that affect nurses job satisfaction.
135 RNs from a 200-bed community hospital, selfselection, all female respondents. Proportion of nurse worked hours Workload
Respondents are not experiencing a high degree of control through flexibility of the work schedules. Professional satisfaction was rated highly. RNs indicate no change in opportunities to participate in decision-making that influences patient care. When nurses have satisfactory workloads & continuity of patient assignment, their overall job satisfaction improves. Consistency in patient assignment is linked to nurses' perceptions of improved organizational commitment to nursing. Nurses ranked co-worker satisfaction as the highest satisfaction score. Less task identity, autonomy or feedback is related to lower job satisfaction. Skill variety & task significance tend to be unrelated to most facets of job satisfaction. Costs of care, nurse satisfaction. Nurse satisfaction has positive relationship with self-scheduling. Optimal skill mix of 85% RNs. Nurses remunerative value is more than what can be measured by workload.
128. Shullanberger (2000)
Literature review of cost-effective nurse staffing
73
Authors, Year 129. Stordeur, Dhoore, Vandenberghe (2001)
Focus Impact of leadership behaviours on nurses emotional exhaustion.
Inputs Work stressors (physical, psychological, social environments) & leadership behaviours Continuity of care, nurse demographics, experience. Nurses job satisfaction, burnout.
Throughputs Work environment
Outputs Emotional exhaustion component of burnout
Findings Work stressors explained 22% of emotional exhaustion whereas leadership dimensions explained 9%. Stress from physical (=0.28) & social (=0.17) environment, role ambiguity (=0.17), active management-by-exception leadership (=0.13) significantly associated with emotional exhaustion. Aspects of job important to satisfaction include continuity of care, autonomy, individual accountability, & performance feedback.
130. Tonges, Rothstein, Carter (1998)
Variables that affect nurses job satisfaction.
131. Tzeng, Ketefian (2002)
Relationship between nurse job satisfaction & inpatient satisfaction.
Sample 625 ward nurses from a university hospital. 39.2% response rate but demographics of sample similar to nursing population. 222 staff nurses in acute care hospitals. All nurses meeting the selection criteria were surveyed. 59 patients & 103 nurses from six units in a Taiwan teaching hospital. Cluster sampling technique. Patient teaching, continuity of care, patient demographics, nurse demographics, experience Work environment Nurse job satisfaction, length of stay Nurse health, burnout.
Nurses job satisfaction is correlated with inpatient satisfaction factors: explanation of care (r=0.765), pain management (r=0.866) as well as the nurses general happiness (r=0.891). Nurses general happiness positively contributed to patient satisfaction. Nurses who work overtime are also under the stress of competing job & family responsibilities, their own health, & their patients' safety.
132. Vernarec (2000).
Overtime & what nurses can do when faced with mandatory or faced overtime.
7. Sample Inputs
System Outcomes Throughputs Outputs Findings
Authors, Year
Focus
74
Authors, Year 152 patients in a large teaching hospital in Tennessee, 81% male, 19% female. Nurse demographics, experience Patient demographics, medical diagnosis.
Focus
Sample
Inputs
Throughputs
Outputs
Findings Intensive care unit length of stay was shorter when ambulation was initiated sooner (t=-2.68). Shorter hospitalization for patients who stayed in intensive care unit 1 day than those staying 2 days (t=-1.46; not significant).
133. Anderson, Higgins, Rozmus (1999)
Length of stay in intensive care unit after coronary artery bypass graft.
134. Bourbonnais, Mondor Myrto (2001)
The association between nurses job strain & sick leave.
135. Liu, Subramanian, Cromwell (2001)
Implementing global bundled payments on Hospital costs of coronary artery bypass grafting. Care delivery system, staff mix, size of hospital, nurses education, use of agency personnel.
1793 nurses from six acute care hospitals in Quebec, voluntary recruitment. Patients undergoing bypass surgery at three hospitals (in Atlanta, Ann Arbor and Boston). Patient demographics, admission type, medical diagnosis.
136. Pierce (1997)
Literature review of outcomes related to nursing.
Length of stay in intensive care unit & step-down units, patient mortality, costs of care, post-op complications. Nurse burnout, job strain, social support at work, short term & certified sick leaves. Costs of care (direct variable costs e.g. nurses wages), postoperative complications, length of stay. Nurse satisfaction, costs of care, patient complications.
Short term sick leaves were associated with job strain (incidence-density ratio = 1.20) & low social support (IDR=1.26). Certified sick leaves were significantly associated with low social support (IDR=1.27 for all diagnoses & IDR=1.78 for mental health diagnoses). All hospitals had significant reductions in total direct variable costs, many coming from reduced nursing costs (e.g. decreased length of stay). Payment method aligns physician & hospital incentives. Improvements in efficiencies were made without diminishing quality of care. Clinical nursing studies and studies of care delivery systems should be integrated. Increased nursing care, better prepared staff, more consistency among caregivers and leadership correlate with lower mortality.
75
8. 9.
Glossary BNS Bachelor of Nursing Science LPN Licensed Practical Nurse RPN Registered Practical Nurse RN Registered Nurse WHPPD worked hours per patient day References
Patient Characteristics Alterman, T., Shekelle, R., Vernon, S. & Burau, K. (1994). Decision latitude, psychological demand, job strain and coronary heart disease in the western electric study. American Journal of Epidemiology, 139(6), 620-627. 2. Bull, M. J., Hansen, H. E., & Gross, C. R. (2000). A professional-patient partnership model of discharge planning with elders hospitalized with heart failure. Applied Nursing Research, 13(1), 19-28. 3. Calvin, J. E., Klein, L. W., VandenBerg, B. J., Meyer, P., Ramirez-Morgen, L. M., & Parrillo, J. E. (1998). Clinical predictors easily obtained at presentation predict resource utilization in unstable angina. American Heart Journal, 136, 373-381. 4. Crilley, J. G., & Farrer, M. (2001). Impact of first myocardial infarction on self-perceived health status. QJM: Monthly Journal of the Association of Physicians, 94(1), 13-18. 5. Czar, M.L., & Engler, M. M. (1997). Perceived learning needs of patients with coronary artery disease using a questionnaire assessment tool. Heart & Lung, 26(2), 109-117. 6. Hemingway, H., & Marmot, M. (1999). Evidence based cardiology: Psychosocial factors in the aetiology and prognosis of coronary heart disease. Systematic review of prospective cohort studies. British Medical Journal, 318(7196), 1460-7. 7. Johnson, J., Stewart, W., Hall, E., Fredlund, P., & Theorell, T. (1996). Long-term psychosocial work environment and cardiovascular mortality among Swedish men. American Journal of Public Health, 86(3), 324-31. 8. Marchette, R., & Holloron, F. (1986). Length of stay: Significant variable. Journal of Nursing Administration, 16(3), 12-20. 9. Shi, L. (1996). Patient and hospital characteristics associated with average length of stay. Healthcare Management Review, 21(2), 46-61. 10. Shih, F. J., Chu, S. H., Yu, P. J., Hu, W. Y., & Huang, G. S. (1997). Turning points of recover from cardiac surgery during the intensive care unit transition. Heart & Lung, 26(2), 9-108. 11. Siegrist, J., Peter, R., Junge, A., Cremer, P., & Seidel, D. (1990). Low status control, high effort work & ischemic heart disease: Prospective evidence from blue-collar men. Social Science & Medicine, 31(10), 1127-1134. 1.
76
12.
Siegrist, J. (1996). Adverse health effects of high-effort/low-reward conditions. Journal of Occupational Health, 1(1), 27-41. 13. Silber, J., Rosenbaum, P., & Ross, R. (1995). Comparing the contributions of groups of predictors: Which outcomes vary with hospital rather than patient characteristics? Journal of the American Statistical Association, 90(429), 7-18. 14. Steptoe, A. (1999). Psychosocial factors in the aetiology of coronary heart disease. Heart 82(3), 258-9.
Nurse Characteristics 15. 16. 17. 18. 19. Bruce, S., Sale, J., Shamian, J., OBrien-Pallas, L., & Thomson, D. (2002). Musculoskeletal injuries, stress and absenteeism. Canadian Nurse, 98(9), 12-17. Coutts, J. (2001). Health workplaces mean more satisfied nurses. Hospital Quarterly (Summer), 57-58. Josephson, M., Lagerstrom, M., Hagberg, M., & Wigaeus, H. E. (1997). Musculo-skeletal symptoms and job strain among nursing personnel: A study over a three year period. Occupational and Environmental Medicine, 54, 681-685. Josephson, M., Vingard, E., & MUSIC-Norrtalje Study Group (1998). Workplace factors and care seeking for low-back pain among female nursing personnel. Scandinavian Journal of Work, Environment and Health, 24(6), 465-472. Toomingas, A., Theorell, T., Michelsen, H., Nordemar, R., & Stockholm MUSIC I Study Group (1997). Associations between self-rated psychosocial work conditions & musculoskeletal symptoms & signs. Scandinavian Journal of Work, Environment and Health, 23(2), 130-139. Wunderlich, G., Sloan, F., & Davis, C. (1996). Nursing staff in hospitals & nursing homes: is it adequate? In G. S. Wunderlich, F A. Sloan, & C. K. Davis (Eds.), Washington, D.C.: National Academy Press
20.
System Characteristics 21. 22. 23. 24. 25. 26. 27. Aiken, L., Clarke, S., & Sloane, D. (2002). Hospital staffing, organization, and quality of care: Cross-national findings. Nursing Outlook, 50(5), 187-194. Aiken, L., Clarke, S., Sloane, D., Sochalski, J., & Silber, J. (2002). Hospital nurse staffing and patient mortality, nurse burnout, and job satisfaction. JAMA: The Journal of the American Medical Association, 288(16), 1987-1993. Aiken, L., Smith, H., & Lake, E. (1994). Lower Medicare mortality among a set of hospitals known for good nursing care. Medical Care, 32(8), 771-787. Arthur, T., & James, N. (1994). Determining nurse staffing levels: A critical review of the literature. Journal of Advanced Nursing, 19, 558-565. Baker, C., Messmer, P., Gyurko, C., Domagala, S., Conly, F., Eads, T., et al. (2000). Hospital ownership, performance, and outcomes: Assessing the state-of-the-science. Journal of Nursing Administration, 30(5), 227-240. Blegen, M., & Vaughn, T. (1998). A multi-site study of nurse staffing and patient occurrences. Nursing Economics, 16(4), 196-198. Blegen, M., Goode, C., Reed, L. (1998). Nurse staffing and patient outcomes. Nursing Research, 47(1), 43-50.
77
28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47.
Brown, S. (2001) Research evidence linking staffing and patient outcomes. Orthopaedic Nursing, 20(1), 67-68. Buerhaus, P. (1997). What is the harm in imposing mandatory nurse staffing? Nursing Economics, 15(2), 66-72. Burke, R. J. (2003). Hospital restructuring, workload, and nursing staff satisfaction and work experiences. Healthcare Manager, 22(2), 99-107. California Nurses Association (2001). CNA On-Line, 12 reasons for the CNA backed nurse-to-patient ratios. Retrieved October 5, 2001, from http://www.calnurse.org/cna/staff3/12reasons.html. Callaghan, L. A., Cartwright, D. W., ORourke, P., & Davies, M. W. (2003). Infant to staff ratios and risk of mortality in very low birthweight infants. Archives of Disease in Childhood Fetal and Neonatal Edition, 88, F94-F97. Campbell, T., Taylor, S., Callaghan, S. Shuldham, C. (1997). Case mix type as a predictor of nursing workload. Journal of Nursing Management, 5, 237-240. Canadian Labour and Business Centre (2002). Full-time equivalents and financial costs associated with absenteeism, overtime, and involuntary part-time employment in the nursing profession (pp. 1-10). Ottawa, ON. Clark, A. (2002). Nurse staffing levels and prevention of adverse events. Clinical Nurse Specialist, 16(5), 237-238. Clarke, S. P., Rockett, J. L., Sloane, D. M., & Aiken, L. H. (2002). Organizational climate, staffing, and safety equipment as predictors of needlestick injuries and near-misses in hospital nurses. American Journal of Infection Control, 30(4), 207-216. Cockerill, R., OBrien-Pallas, L., Bolley, H., & Pink, G. (1993). Measuring nursing workload for case costing. Nursing Economics, 11(6), 342-349. Doran, D., McGillis Hall, L., Sidani, S., OBrien-Pallas, L., Donner, G., Baker, G. R., et al. (2001). Nursing staff mix and patient outcome achievement: The mediating role of nurse communication. International Nursing Perspective, 1(2-3), 74-83. Eisenberg, J., Bowman, C., & Foster, N. (2001). Does a healthy healthcare workplace produce higher-quality care? Journal on Quality Improvement, 27(9), 444-457. Gauci Borda, R., & Norman, I. (1997). Factors influencing turnover and absence of nurses: A research review. International Journal of Nursing Studies, 34(6), 385-394. Gaudine, A. P. (2000). What do nurses mean by workload and by work overload? Journal of Nursing Leadership, 13(2), 22-27. Greenglass, R., & Burke, R. (2001). Impact of restructuring scale: an instrument to measure effects of hospital restructuring. Healthcare Management Forum, 14(3), 24-28. Grillo-Peck, A. M., & Risner, P. B. (1995). The effect of a partnership model on quality and length of stay. Nursing Economics, 13(6), 367-374. Halloran, E. (1985). Nursing workload, medical diagnosis related groups and nursing diagnosis. Research in Nursing and Health, 8, 421-433. Hartz, A., Krakauer, H., Kuhn, E., Young, M., Jacobsen, S., Gay, G., et al. (1989). Hospital characteristics and mortality rates. The New England Journal of Medicine, 321(25), 1720-1725. Hendrix, T. J., Foreman, S. E. (2001). Optimal long-term care nurse-staffing levels Nursing Economics$, 10(4), 164-175. Kenney, P. (2001). Maintaining quality care during a nursing shortage using licensed practical nurses in acute care. Journal of Nursing Care Quality, 15(4), 60-68.
78
48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66.
Knaus, W., Draper, E., Wagner, D., & Zimmerman, J. (1986). An evaluation of outcome from intensive care in major medical centers. Annals of Internal Medicine, 104(3), 410418. Kobs, A. (1997). The adequacy of nurse staffing. Nursing Management, 28(11), 16-20. Kovner, C. (2001). The impact of staffing and the organization of work on patient outcomes and healthcare workers in healthcare organizations. Journal on Quality Improvement, 27(9), 458-468. Kovner, C., & Gergen, P. J. (1998) Nurse Staffing Levels & Adverse Events Following Surgery in US Hospitals. Journal of Nursing Scholarship, 30(4), 315-321. Krakauer, H., Bailey, R. C., Skellan, K., Stewart, J., Hartz, A., Kuhn, et al. (1992). Evaluation of the HCFA Model for the analysis of mortality following hospitalization. Health Services Research, 27(3), 317-335. Kramer, M., &Schmalenberg, C. (1988). Magnet hospitals: Part I, Institutions of excellence. Journal of Nursing Administration, 18(1), 13-24. Kramer, M., & Schmalenberg, C. (1988). Magnet hospitals: Part II, Institutions of excellence. Journal of Nursing Administration, 18(2), 11-19. Kutsogiannis, C. H., Triska, O. H., Johnston, R. G., & Noseworthy, T. (2001). The organizational structure of intensive care units and its influence on patient outcomes. Healthcare Management Forum, 14(4), 28-34. Manitoba Nursing Strategy (2003). Manitoba nursing strategy: Three year progress report. Winnipeg. Manitoba Government. Maxwell, J. (2002) Creating high-quality health-care workplaces. Canadian Healthcare Management, 102(11), 101-110. McGillis Hall, L., Doran, D., Baker, G., Pink, G., Sidani, S., OBrien-Pallas, L., et al. (2002). Nurse staffing and work status in medical, surgical and obstetrical units in Ontario teaching hospitals. Hospital Quarterly, 5(4), 64-69. Mitchell, P. H., Armstrong, S., Forshee Simpson, T., & Lentz, M. (1989). American Association of Critical-Care Nurses Demonstration Project: Profile of excellence in critical care nursing. Heart & Lung 18(3), 219-237. Mitchell, P. H., Shortell, S. M. (1997). Adverse outcomes & variations in organization of care delivery. Medical Care, 35(11), NS19-NS32. Needleman, J., Buerhaus, P., Mattke, S., Stewart, M., & Zelevinsky, K. (2002). Nursestaffing levels and the quality of care in hospitals. New England Journal of Medicine, 346(22), 1715-1722. OBrien-Pallas L, Irvine D, Peereboom E, & Murray M (1997). Measuring nursing workload: Understanding the variability. Nursing Economics$, 15(4), 171-182. OBrien-Pallas, L., Irvine Doran, D., Murray, M., Cockerill, R., Sidani, S., Laurie-Shaw, B., et al. (2001). Evaluation of a client care delivery model part 1: Variability in nursing utilization in community home nursing. Nursing Economic$, 19(6), 267-276. OBrien-Pallas, L. L., Cockerill, R., & Leatt, P. (1992). Different systemsdifferent costs. Journal of Nursing Administration, 22(12), 17-22. O'Brien-Pallas, L. L., Thomson, D., Alksnis, C., & Bruce, S. (2001). The economic impact of nurse staffing decisions: Time to turn down another road. Hospital Quarterly, 4(3), 42-50. Pinkerton, S., & Rivers, R. (2001). Factors influencing staffing needs. Nursing Economic$, 19(5), 236-238.
79
67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81.
Potter, P., Barr, N., McSweeney, M., & Sledge, J. (2003). Identifying nurse staffing and patient outcome relationships: A guide for change in care delivery. Nursing Economics, 21(4), 158-166. Prescott, P. A. (1986). DRG prospective reimbursement: The nursing intensity factor. Nursing Management, 17(1), 43-48. Prescott, P. A. (1993). Nursing: An important component of hospital survival under a reformed healthcare system. Nursing Economics, 11(4), 192-199. Robertson, R., Dowd, S., & Hassen, M. (1997). Skill-specific staffing intensity and the cost of hospital care. Healthcare Management Review, 22(4), 61-71. Sainfort, F., Karsh, B., Booske, B., & Smith, M. (2001). Applying quality improvement principles to achieve healthy work organizations. Journal of Quality Improvement, 27(9), 469-483. Seago, J., Ash, M., Spetz, J., Coffman, J., & Grumbach, K. (2001). Hospital registered nurse shortages: environmental, patient, and institutional predictors. Health Services Research, 36(5), 831-852. Seybolt, J. W. (1986). Dealing with premature employee turnover. Journal of Nursing Administration, 16, 26-32. Silber, J. H., Williams, S. V., Krakauer, H., & Schwartz, J. S. (1992). Hospital & patient characteristics associated with death after surgery. Medical Care, 30(7), 615-629. Sochalski, J. (2001). Quality of care, nurse staffing, and patient outcomes. Policy, Politics & Nursing Practice, 2(1) 9-18. Sovie, M. D., & Jawad, A. F. (2001). Hospital restructuring and its impact on outcomes: Nursing staff regulations are premature. Journal of Nursing Administration, 31(12), 588-600. Strickland, B., & Neely, S. (1995). Using a standard staffing index to allocate nursing staff. Journal of Nursing Administration, 25(3), 13-21. Tarnow-Mordi, W. O., Hau, C., Warden, A., & Shearer, A. J. (2000). Hospital mortality in relation to staff workload: A 4-year study in an adult intensive-care unit. The Lancet, 356, 185-189. Wai, T., Bame, S., & Robinson, C. (1998). Review of nursing turnover research. Social Science Medicine, 47(12), 1905-1924. Weisman, C. S., Alexander, C. S., & Chase, G. A. (1981). Evaluating reasons for nursing turnover-comparison of exit interview and panel data. Evaluation and the Health Professions, 4(2), 107-127. Whitman, G., Kim, Y., Davidson, L., Wolf, G., & Wang, S. (2002). The impact of staffing on patient outcomes across specialty units. Journal of Nursing Administration. 32(12), 633-639.
Throughputs 82. 83. Allred, C. A., Michel, Y., Arford P. H., Carter, V., Veitch J. S., Dring, R., et al. (1994). Environmental uncertainty: Implications for practice model redesign. Nursing Economics, 12(6), 318-325. Brown, M. M. (2000). Implementation Strategy: One-stop recovery for cardiac surgical patients. AACN Clinical Issues: Advanced Practice in Acute Critical Care, 11(3), 412423
80
84. 85. 86. 87. 88. 89.
Cady, N., Mattes, M., & Burton, S. (1995). Reducing intensive care unit length of stay: A step-down unit for first-day heart surgery patients. Journal of Nursing Administration, 25(12), 29-35. Cohn, L., Rosborough, D., & Fernandez, J. (1997). Reducing costs and length of stay and improving efficiency and quality of care in cardiac surgery. The Annals of Thoracic Surgery, 64, S58-60. Drenkard, K. (2001). Creating a future worth experiencing. Journal of Nursing Administration, 31(7/8), 364-376. Duffy, J., & Lemieux, K. (1995). A cardiac service line approach to patient-centred care. Nursing Administration Quarterly, 20 (1), 12-23. Helt, E. & Jelinek, R. (1988). In the wake of cost cutting, nursing productivity and quality improve. Nursing Management, 19(6), 36-48. Lamb, G. S., & Stempel, J. E. (1994). Nurse case management from the client's view: Growing as insider-expert. Nursing Outlook, 42(1), 7-13.
Patient Outcomes 90. 91. 92. 93. 94. Brooten, D., & Naylor, M. (1995). Nurses effect on changing patient outcomes. Image: Journal of Nursing Scholarship, 27(2), 95-99. Dansky, K. H., Brannoon, D., & Wangsness, S. (1994). Human resources management practices and patient satisfaction in home healthcare. Home Health Services Quarterly, 15(1), 43-56. Fortinsky, R. H., & Madigan, E. A. (1997). Home care resource consumption and patient outcomes: What are the relationships. Home Healthcare Services Quarterly, 16(3), 5573. Lancaster, A. (1997). Understaffing can increase infection rates. RN, 60(10), 79. OBrien-Pallas, L., Doran, D., Murray, M., Cockerill, R., Sidani, S., Laurie-Shaw, B., et al. (2002). Evaluation of a client care delivery model, part 2: variability in client outcomes in community home nursing. Nursing Economic$, 20(1), 13-21, 36. Proctor, T., Yarcheski, A., & Oriscello, R. (1996). The relationship of hospital process variables to patient outcome post-myocardial infarction. International Journal of Nursing Studies, 33(2), 121-130. Rohrer, J. E., Momany, E. T., & Chang, W. (1993). Organizational predictors of outcomes of long-stay nursing home residents. Social Science & Medicine, 37(4), 549-54. Roseman, C., & Booker, J. (1995). Workload and environmental factors in hospital medication errors. Nursing Research, 44(4), 226-230. Silber, J. H., Rosenbaum, P. R., Sanford Schwartz, J., Ross, R. N., & Williams, S. V. (1995). Evaluation of the complication rate as a measure of quality care in coronary artery bypass graft surgery. JAMA: The Journal of the American Medical Association, 274(4), 317-323.
95. 96. 97. 98.
Nurse Outcomes 99. Adams, A., & Bond, S. (2000). Hospital nurses job satisfaction, individual and organizational characteristics. Journal of Advanced Nursing, 32(3), 536-543.
81
100. Aiken, L, & Sloane, D. (1997). Effects of organizational innovations in AIDS care on burnout among urban hospital nurses. Work and Occupations, 24(4), 453-477. 101. Aiken, L., Clarke, S., & Sloane, D. (2000). Hospital restructuring: Does it adversely affect care and outcomes. Journal of Nursing Administration, 30(10), 457-465. 102. Bakker, A., Christel, K., Johannes, S., & Wilmar, S. (2000). Effort-reward imbalance and burnout among nurses. Journal of Advanced Nursing, 31(4), 884-891. 103. Baumann, A., Giovannetti, P., OBrien-Pallas, L., Mallette, C., Deber, R., Blythe, J., et al. (2001). Healthcare restructuring: The impact of job change. Canadian Journal Nursing Leadership, 14(1), 14-20. 104. Blegen, M. (1993). Nurses job satisfaction: A meta-analysis of related variables. Nursing Research, 42(1), 36-41. 105. Bourbonnais, R., Comeau, M., Vezina, M., & Dion, G. (1998). Job strain, psychological distress, & burnout in nurses. American Journal of Industrial Medicine, 34, 20-28. 106. Buchan, J. (1999). Still attractive after all these years? Magnet hospitals in a changing healthcare environment. Journal of Advanced Nursing, 30(1), 100-108. 107. Burke, R., & Greenglass, E. (2000). Effects of hospital restructuring on full time and part time nursing staff in Ontario. International Journal of Nursing Studies, 37, 163-171. 108. Cameron, S. J., Horsburgh, M. E., & Armstrong-Stassen, M. (1994). Job satisfaction, propensity to leave and burnout in RNs and RNAs: A multivariate perspective. Canadian Journal of Nursing Administration, 7(3), 43-64. 109. Carey, S. J., Campbell, S. T. (1994). Preceptor, mentor, and sponsor roles: creative strategies for nurse retention. Journal of Nursing Administration, 24(12), 39-48. 110. Clarke, H., Laschinger, H., Giovannetti, P., Shamian, J., Thomson, D., & Tourangeau, A. (2001). Nursing shortages: Workplace environments are essential to the solution. Hospital Quarterly (Summer), 50-56. 111. Davison, H., Folcarelli, P. H., Crawford, S., Duprat, L. J., & Clifford, J. C. (1997). 112. Demerouti, E., Bakker, A., Nachreiner, F., & Schaufeli, W. (2000). A model of burnout and life satisfaction amongst nurses. Journal of Advanced Nursing, 32(2), 454-464. 113. Joseph, J., & Deshpande, S. (1997). The impact of ethical climate on job satisfaction of nurses. Healthcare Management Review, 22(1), 76-81. 114. Kangas, S., Kee, C. C., McKee-Waddle, R. (1999). Organizational factors, nurses job satisfaction, and patient satisfaction with nursing care. Journal of Nursing Administration, 29(1), 32-42. 115. Kovner, C., Hendrickson, G., Knickman, J., & Finkler, S. (1994). Nursing care delivery models and nurse satisfaction. Nursing Administration Quarterly, 19(1), 74-85. 116. Kramer, M., & Hafner, L. (1989). Shared values: Impact on staff nurse job satisfaction and perceived productivity. Nursing Research, 38(3), 172-176. 117. Kramer. M., & Schmalenberg, C. (1990). Job satisfaction and retention: Insights for the 90s, Parts I and II. Nursing, 21, 2-7, 9-13. 118. Kutzscher, L. I. T., Sabiston, J. A., Laschinger-Spence, H. K., & Nish, M. (1997). The effects of teamwork on staff perception and empowerment and job satisfaction. Healthcare Management Forum, 10(2), 12-17. 119. Laschinger, H., Finegan J., & Shamian, J. (2001). The impact of workplace empowerment, organizational trust, on staff nurses work and organizational commitment. Healthcare Management Review, 26(3), 7-23.
82
120. Leiter, M. P., Harvie, P., & Frizzell, C. (1998). The correspondence of patient satisfaction and nurse burnout. Social Science & Medicine, 47(10), 1611-1617 121. McGillis Hall, L., Doran, D., Baker, G., Pink, G., Sidani, S., OBrien-Pallas, L., et al. (2001). A study of the impact of nursing staff mix models and organizational change strategies on patient, system and nurse outcomes. Toronto, ON: Faculty of Nursing, University of Toronto and Canadian Health Services Research Foundation/Ontario Council of Teaching Hospitals. 122. McNeese-Smith, D., & Crook, M. (2003). Nursing values and a changing nurse workforce: Values, age, and job stages. Journal of Nursing Administration, 33(5), 260-270. 123. Moss, R., & Rowles, C. J. (1997). Staff nurse job satisfaction and management style. Nursing Management, 28(1), 32-34. 124. Munro, B. H. (1983). Job satisfaction among recent graduates of schools of nursing. Nursing Research, 32, 350-355. 125. Nakata, J.A., & Saylor, C. (1994). Management style and staff nurse satisfaction in a changing environment. Nursing Administration Quarterly, 18(3), 51-57. 126. Registered Nurses Association of Ontario. (2002). Tracking the Nursing Task Force (1999): RNs rate their nursing work life. Toronto, ON: RNAO. 127. Roedel, R., & Nystrom, P. (1988). Nursing jobs and satisfaction. Nursing Management, 19(2), 34-38. 128. Shullanberger, G. (2000). Nurse staffing decisions: An integrative review of the literature. Nursing Economic$, 18, (13), 124-32, 146-8. 129. Stordeur S, Dhoore W, & Vandenberghe, C. (2001). Leadership, organizational stress, & emotional exhaustion among hospital nursing staff. Journal of Advanced Nursing 35(4), 533-542. 130. Tonges, M., Rothsein, H., & Carter, H. (1998). Sources of satisfaction in hospital nursing practice. Journal of Nursing Administration, 28(5) 47-61. 131. Tzeng, H-M., & Ketefian, S. (2002). The relationship between nurses job satisfaction and inpatient satisfaction: An exploratory study in a Taiwan teaching hospital. Journal of Nursing Care Quality, 16(2), 39-49. 132. Vernarec, E. (2000). Just say no to mandatory overtime? RN, 63(12), 69-70, 72, 74. System Outcomes 133. Anderson, B., Higgins, L., & Rozmus, C. (1999). Critical pathways: application to selected patient outcomes following coronary artery bypass graft. Applied Nursing Research, 12(4), 168-174. 134. Bourbonnais, R., & Mondor Myrto (2001). Job strain & sickness absence among nurses in the Province of Quebec. American Journal of Industrial Medicine, 39, 194-202. 135. Liu, C. F., Subramanian, S., & Cromwell, J. (2001). Impact of global bundled payments on hospital costs of coronary artery bypass grafting. Journal of Healthcare Finance, 27(4), 39-54. 136. Pierce, S. (1997). Nurse-sensitive healthcare outcomes in acute care settings: An integrative analysis of the literature. Journal of Nursing Care Quality. 11(4), 60-72.
83
Appendix B. Patient Care Delivery Model
INPUTS
Patient Characteristics
Demographics Significant other support Medical diagnoses Nursing diagnoses OMAHA knowledge, behaviour, status Admission type Pre-operative clinic Education booked post-op/post discharge SF-12 health status
Patient Care Delivery Model

(OBrien-Pallas et al., 2003)
OUPUTS
Patient Outcomes
Medical consequences OMAHA knowledge, behaviour, status SF-12 health status Resource intensity weight Mortality
THROUGHPUTS
Interventions Perceived Work Environment
Nurse Characteristics
Demographics Professional status Employment status Education Clinical expertise Experience
Patient Care Delivery System in Cardiac & Cardiovascular Units
INTERMEDIATE OUTPUTS
Worked hours Productivity/ Utilization
Nurse Outcomes
Burnout Effort & reward imbalance Autonomy & control Job satisfaction Relationships with MDs SF-12 health status Violence at work
System Characteristics
Geographic location Hospital size Unit size, type, patient mix Occupancy
System Outcomes Environmental Complexity Factors

Resequencing of work in response to others Unanticipated delays due to changes in patient acuity Characteristics & composition of caregiving team
System Behaviours
Workload Nurse-to-patient ratios Proportion of RN worked hours Continuity of care/shift change Unit instability Overtime Use of agency & relief staff # of units nurse works on Non-nursing tasks
Length of stay Cost per resource intensity weight Quality of patient care Quality of nursing care Interventions delayed Interventions not done Absenteeism Intent to leave
Feedback
84
Appendix C. Tables
Table 1. Key Variables and Data Sources ......................................................................................................86 Table 2. Estimates for Patient Outcomes from the Hierarchical Linear Models .................................87 Table 3. Odds Ratios for Patient Outcomes from the Hierarchical Linear Models ............................88 Table 4: Hierarchical Logistic Regression for Medical Consequences Developed During Hospital Stay .......................................................................................................................................................89 Table 5: Hierarchical Logistic Regression for Improvement in Patients Physical Health ..............90 Table 6: Hierarchical Logistic Regression for Improvement in Patients Mental Health ................91 Table 7: Hierarchical Logistic Regression for Patients Knowledge Improvement at Discharge/Diagnoses Resolved ....................................................................................................92 Table 8: Hierarchical Logistic Regression for Patients Behaviour Improvement at Discharge/Diagnoses Resolved ....................................................................................................93 Table 9: Hierarchical Logistic Regression for Patients Status Improvement at Discharge/Diagnoses Resolved ....................................................................................................94 Table 10. Estimates for Nurse Outcomes in the Hierarchical Linear Models ......................................95 Table 11. Odds Ratios for Nurse Outcomes in the Hierarchical Linear Models .................................96 Table 12: Hierarchical Linear Regression for Nurse-Physician Relationship .....................................97 Table 13: Hierarchical Linear Regression for Autonomy ..........................................................................98 Table 14: Hierarchical Logistic Regression for Job Satisfaction .............................................................99 Table 15: Hierarchical Logistic Regression for Emotional Exhaustion...............................................100 Table 16: Hierarchical Linear Regression for Nurses Physical Health ..............................................101 Table 17: Hierarchical Linear Regression for Nurses Mental Health .................................................102 Table 18. Estimates for System Outcomes from the Hierarchical Linear Models............................103 Table 19. Odds Ratios for System Outcomes in the Hierarchical Linear Models ............................104 Table 20: Hierarchical Logistic Regression for Patients with Shorter Than Expected Length of Stay .....................................................................................................................................................105 Table 21: Hierarchical Logistic Regression for Interventions Not Done ............................................106 Table 22: Hierarchical Logistic Regression for Interventions Delayed ...............................................107 Table 23: Hierarchical Logistic Regression for Quality of Nursing Care ...........................................108 Table 24: Hierarchical Logistic Regression for Quality of Patient Care .............................................109 Table 25: Hierarchical Logistic Regression for Absenteeism ................................................................110 Table 26: Hierarchical Logistic Regression for Intent to Leave ............................................................111 Table 27: Hierarchical Linear Regression for Productivity/Utilization ...............................................112 Table 28: Hierarchical Linear Regression for Cost per Resource Intensity Weight (Log Scale) .113 Table 29: Hierarchical Linear Regression for Worked Hours per Patient (Log Scale) ...114 Table 30: Hierarchical Linear Models for Patient, Nurse, and System Outcomes on Congruence Between PRN Hours and Actual Worked Hours per Patient.............................................115 Table 31: Summary Table of the Effect of Nursing Hours, Proportion of RN Worked Hours, Nurse-Patient Ratio, and Productivity/Utilization on Patient, Nurse and System Outcomes, in Odds Ratio, Coefficient, and Cut point .........................................................116
85
Table 1. Key Variables and Data Sources

Measure SF-12 Health Status Survey (physical and mental health status) NANDA Nursing Diagnoses and OMAHA Problem Rating Scale When Administered Admission or in pre-op clinic Discharge Admission Discharge Daily to identify new or resolved diagnoses Patient Data Form PRN Workload Tool Once over patient stay Daily Data collector from patient chart/kardex/interview Data collector from patient chart/kardex, unit workload tool, and nurse Electronic file submitted by Heath Records Department Electronic file submitted by Heath Records Department Nurse self-report Data collector from unit assignment sheet and ward clerk Nurses Data collector from patient chart/kardex and nurse Method/Source Patient self-report
Case Mix Group Resource Intensity Weight Nurse Survey Daily Unit Staffing Form
After discharge After discharge Once at beginning of data collection Daily
Environmental Complexity Scale
Daily
86
Table 2. Estimates for Patient Outcomes from the Hierarchical Linear Models
Predictor Medical Conseq. Physical Health Mental Health Omaha Knowledge Omaha Behaviour Omaha Status Patient Level Pre-Operative Clinics Referral for Homecare Medical Consequences Resource Intensity Weight Number of Nursing Diagnoses Physical Health at Admission Mental Health at Admission Knowledge at Admission Behaviour at Admission Status at Admission Worked Hours per Patient Length of Stay Nurse Level Education (ref: Diploma) Overtime Hours Unit Instability Interventions Not done Interventions Delayed Autonomy Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Nurse-Patient Ratio Unit Level Proportion of RN Worked Hours Productivity/Utilization Productivity/Utilization (Quadratic) Productivity/Utilization (beyond 88%) Productivity/Utilization (beyond 80%) Productivity/Utilization (beyond 85%) Proportion of Full-time Employment Proportion of Nurses Reporting Shift Changes
1.43 * 0.02 0.43 * -0.01 -0.03 * -0.10 * -0.12 * -0.13 * 0.00 -0.01 -0.06 0.01 * -0.09 * 0.01 0.05 0.01 0.00 -1.33 * 0.08 0.06 0.01 0.00 -2.14 * 0.13 * 0.03 0.12 -0.02 -0.97 1.08 0.03 -0.02 0.25 3.72 0.00 0.01 0.11 -0.08 * 0.08 -0.02 -0.01 -0.01 0.05 -0.94 -0.06 * -0.04 * 0.05 0.04 0.27 -0.10 0.01 0.01 -0.14 0.53 0.04 0.01 -0.29 0.02 -0.47 0.10 0.17 * -0.02 -0.01 0.11 5.55 * 7.40 * -1.49 * -0.60 * -0.67 0.30 2.13 * -5.75 * -0.02 -0.01 -0.08 -0.03 -0.75 * 0.03 -0.01 -0.01 -0.01 1.02 * -0.03 0.42 17.83 * -10.11 * -1.48 * 0.00 0.00 0.10 0.02 -0.27 -0.42 -0.02 0.02 0.16 2.06 4.94 * -0.80 * 0.05 0.08 0.01 0.00
* for p-value at 0.05 or less Notes: (1) All patient outcome variables were dichotomized and modeled in hierarchical logistic regressions. For medical consequences, 1 = development of complications, falls with injury, or death; for physical and mental health, 1 = improved at discharge; for Omaha knowledge, behaviour, and status, 1 = improvement at discharge or diagnoses resolved. (2) The productivity/utilization cut point is 88.2% for Omaha behaviour.
87
Table 3. Odds Ratios for Patient Outcomes from the Hierarchical Linear Models
Predictor Medical Conseq. Physical Health Mental Health Omaha Knowledge Omaha Behaviour Omaha Status Patient Level Pre-Operative Clinics Referral for Homecare 4.19 Medical Consequences Resource Intensity Weight 1.02 Number of Nursing Diagnoses 1.53 Physical Health at Admission 0.99 Mental Health at Admission 0.97 Knowledge at Admission Behaviour at Admission Status at Admission Worked Hours per Patient 1.13 Length of Stay 1.03 Nurse Level Education (ref: Diploma) 1.13 Overtime Hours 0.98 Unit Instability Interventions Not Done 0.38 Interventions Delayed 2.94 Autonomy Physical Health 1.03 Mental Health 0.98 Satisfaction with Current Job (ref: Dissatisfied) Nurse-Patient Ratio 1.28 Unit Level Proportion of RN Worked Hours 1.45 Productivity/Utilization Productivity/Utilization (Quadratic) Productivity/Utilization (beyond 88%) Productivity/Utilization (beyond 80%) Productivity/Utilization (beyond 85%) 0.51 Proportion of Full-time Employment Proportion of Nurses Reporting Shift Changes
* 0.90 * 0.89 * 0.88 * 1.00 0.99 0.94 1.01 * 0.92 * 1.01 1.05 1.01 1.00 0.26 * 1.08 1.07 1.01 1.00 0.12 * * 1.00 1.01 1.11 0.93 * 1.08 0.98 1.00 0.99 1.05 0.91 0.94 * 0.96 * 1.05 1.04 1.32 0.91 1.01 1.01 0.87 1.05 1.04 1.01 0.75 1.02 0.62 1.11 1.19 * 0.98 0.99 1.11 1.74 * n/a * n/a * 0.55 * 1.35 1.24 * 0.56 * 0.98 0.99 0.92 0.97 0.47 * 1.03 0.99 0.99 0.99 2.76 * 0.97 1.04 n/a * n/a * 0.23 * 1.00 1.00 1.10 1.02 0.76 0.66 0.98 1.02 1.17 1.23 n/a * n/a * 1.05 1.08 1.01 1.00
* *
* for p-value at 0.05 or less Notes: (1) All patient outcome variables were dichotomized and modeled in hierarchical logistic regressions. For medical consequences, 1 = falls with injury, medication errors, death, or development of complications; for physical and mental health, 1 = Improved at discharge; for Omaha knowledge, behaviour and status, 1 = improvement at discharge or diagnoses resolved. (2) The productivity/utilization cut point is 88.2% for Omaha behaviour. (3) The odds ratios for proportion of RN worked hours, proportion of full-time employment and proportion of nurses reporting shift changes are based on a 10% increase. (4) Odds ratio for quadratic transformation of productivity/utilization is not reported.
88
Table 4: Hierarchical Logistic Regression for Medical Consequences Developed During Hospital Stay
Predictor Patient Level Referral for Homecare Resource Intensity Weight Number of Nursing Diagnoses Physical Health at Admission Mental Health at Admission Worked Hours per Patient Length of Stay Nurse Level Education (ref: Diploma) Overtime Hours Interventions Not Done Interventions Delayed Physical Health Mental Health Nurse-Patient Ratio Unit Level Proportion of RN Worked Hours Productivity/Utilization (beyond 85%) Beta 1.43 0.02 0.43 -0.01 -0.03 0.13 0.03 0.12 -0.02 -0.97 1.08 0.03 -0.02 0.25 3.72 -0.67 SE 0.36 0.06 0.08 0.01 0.01 0.05 0.02 0.63 0.07 0.70 0.64 0.04 0.03 0.23 2.59 0.53 Odds Ratio 4.19 1.02 1.53 0.99 0.97 1.13 1.03 1.13 0.98 0.38 2.94 1.03 0.98 1.28 1.45 0.51 * * * *
* p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratio for proportion of RN worked hours is based on a 10% increase.
89
Table 5: Hierarchical Logistic Regression for Improvement in Patients Physical Health

Predictor Beta SE Odds Ratio * * * Patient Level Resource Intensity Weight -0.10 0.04 0.90 Number of Nursing Diagnoses -0.12 0.04 0.89 Physical Health at Admission -0.13 0.01 0.88 Mental Health at Admission 0.00 0.01 1.00 Worked Hours per Patient 0.00 0.02 1.00 Length of Stay 0.01 0.01 1.01 Nurse Level Education (ref: Diploma) 0.11 0.25 1.11 Overtime Hours -0.08 0.04 0.93 Interventions Not Done 0.08 0.40 1.08 Interventions Delayed -0.02 0.33 0.98 Physical Health -0.01 0.02 1.00 Mental Health -0.01 0.01 0.99 Nurse-Patient Ratio 0.05 0.08 1.05 Unit Level Proportion of RN Worked Hours -0.94 1.15 0.91 Productivity/Utilization (beyond 80%) -0.60 0.26 0.55 * p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratio for proportion of RN worked hours is based on a 10% increase.
90
Table 6: Hierarchical Logistic Regression for Improvement in Patients Mental Health

Predictor Patient Level Resource Intensity Weight Number of Nursing Diagnoses Physical Health at Admission Mental Health at Admission Worked Hours per Patient Length of Stay Nurse Level Education (ref: Diploma) Overtime Hours Interventions Not Done Interventions Delayed Physical Health Mental Health Nurse-Patient Ratio Unit Level Proportion of RN Worked Hours Productivity/Utilization (beyond 85%) Beta -0.01 -0.06 0.01 -0.09 -0.06 -0.04 0.05 0.04 0.27 -0.10 0.01 0.01 -0.14 0.53 0.30 SE 0.04 0.04 0.01 0.01 0.03 0.02 0.24 0.03 0.34 0.29 0.01 0.01 0.08 1.08 0.27 Odds Ratio 0.99 0.94 1.01 0.92 0.94 0.96 1.05 1.04 1.32 0.91 1.01 1.01 0.87 1.05 1.35
* * * *
91
Table 7: Hierarchical Logistic Regression for Patients Knowledge Improvement at Discharge/Diagnoses Resolved
Predictor Patient Level Resource Intensity Weight Number of Nursing Diagnoses Physical Health at Admission Mental Health at Admission Knowledge at Admission Worked Hours per Patient Length of Stay Nurse Level Education (ref: Diploma) Overtime Hours Interventions Not Done Interventions Delayed Autonomy Physical Health Mental Health Nurse-Patient Ratio Unit Level Proportion of RN Worked Hours Productivity/Utilization Productivity/Utilization (beyond 88%) Proportion of Full-Time Employment Proportion of Nurses Reporting Shift Changes Beta 0.01 0.05 0.01 0.00 -1.33 0.04 0.01 -0.29 0.02 -0.47 0.10 0.17 -0.02 -0.01 0.11 5.55 7.40 -1.49 2.13 -5.75 SE 0.04 0.04 0.01 0.01 0.13 0.02 0.02 0.25 0.03 0.35 0.30 0.04 0.02 0.01 0.08 1.36 1.16 0.32 0.61 0.82 Odds Ratio 1.01 1.05 1.01 1.00 0.26 1.04 1.01 0.75 1.02 0.62 1.11 1.19 0.98 0.99 1.11 1.74 n/a n/a 1.24 0.56
* * * * *
* p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratios for proportion of RN worked hours, proportion of full-time employment, and proportion of nurses reporting shift changes are based on a 10% increase.
92
Table 8: Hierarchical Logistic Regression for Patients Behaviour Improvement at Discharge/Diagnoses Resolved
Predictor Patient Level Resource Intensity Weight Number of Nursing Diagnoses Physical Health at Admission Mental Health at Admission Behaviour at Admission Worked Hours per Patient Length of Stay Nurse Level Education (ref: Diploma) Overtime Hours Unit Instability Interventions Not Done Interventions Delayed Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Nurse-Patient Ratio Unit Level Proportion of RN Worked Hours Productivity/Utilization Productivity/Utilization (Quadratic) Beta 0.08 0.06 0.01 0.00 -2.14 -0.02 -0.01 -0.08 -0.03 -0.75 0.03 -0.01 -0.01 -0.01 1.02 -0.03 0.42 17.83 -10.11 SE 0.05 0.04 0.01 0.01 0.18 0.04 0.02 0.28 0.04 0.33 0.36 0.35 0.02 0.01 0.27 0.08 1.53 1.41 0.99 Odds Ratio 1.08 1.07 1.01 1.00 0.12 0.98 0.99 0.92 0.97 0.47 1.03 0.99 0.99 0.99 2.76 0.97 1.04 n/a n/a
* *
* p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratio for proportion of RN worked hours is based on a 10% increase. (3) productivity/utilization cut point is 88.2%.
93
Table 9: Hierarchical Logistic Regression for Patients Status Improvement at Discharge/Diagnoses Resolved
Predictor Patient Level Resource Intensity Weight Number of Nursing Diagnoses Physical Health at Admission Mental Health at Admission Status at Admission Worked Hours per Patient Length of Stay Nurse Level Education (ref: Diploma) Overtime Hours Interventions Not Done Interventions Delayed Physical Health Mental Health Nurse-patient Ratio Unit Level Proportion of RN Worked Hours Productivity/Utilization Productivity/Utilization (beyond 88%) Beta 0.05 0.08 0.01 0.00 -1.48 0.00 0.00 0.10 0.02 -0.27 -0.42 -0.02 0.02 0.16 2.06 4.94 -0.80 SE 0.05 0.05 0.01 0.01 0.18 0.03 0.02 0.29 0.04 0.39 0.33 0.02 0.01 0.09 1.26 1.16 0.39 Odds Ratio 1.05 1.08 1.01 1.00 0.23 1.00 1.00 1.10 1.02 0.76 0.66 0.98 1.02 1.17 1.23 n/a n/a
* *
94
Table 10. Estimates for Nurse Outcomes in the Hierarchical Linear Models
Predictor N-P relationship Autotonomy Satisfaction Emotional Exhaust'n
Physical Health -0.90 * -0.01 0.02 -0.11 0.02 0.14 -0.22 -0.26 -0.10 0.37 -0.67 * -0.53 * 0.05 -0.10
Mental Health -0.74 * 0.03 * -0.21 -0.03 0.02 0.08 -0.27 -0.05 -0.05 -0.14 -0.51 -1.56 * 0.04 0.05 -0.04 *
Nurse Level
Gender (ref: Male) Age Education (ref: Diploma) Full-time Employment (ref: PT/Casual) Over Time Hours Clinical Expertise Unit Instability Shift Change Interventions Not Done Interventions Delayed Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-physician Relationship Absenteeism Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Nurse-patient Ratio
0.16 0.16
-0.44 * 0.03 0.01 -0.33 * -0.03 -0.01 -0.33 -0.10 -0.89 * -0.07 0.62 * -0.36 0.02 0.02 1.04 * 0.67 * 0.28 0.67 * 0.83 * 0.04 -0.05 -2.07 * 0.02 1.69 1.58 *
0.30 *
0.44 * 0.12 -0.02 0.09 -0.31 0.01 -0.13 0.01 -0.49 -1.23 * 0.21 * 0.06 0.02 0.02 0.37 1.39 * 0.31
-0.03 0.11 1.03 * 0.01 0.21 0.03 -0.12 1.23 * -0.02 0.00 -0.04 * -0.11 * -1.09 * 0.28 -0.57 0.49
-0.01 0.01 0.29 0.59 * -0.37 *
-0.02 * 0.46 * 0.13 -0.24 -0.38
0.55 * 0.14 0.00 -0.29
Patient Level
Proportion of Patients Attended Pre-operative Clinics Average Resource Intensity Weight Average Number of Nursing Diagnoses
0.02 -0.10 *
0.05 -0.10
0.05 -0.07
Unit Level
Unit Occupancy Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization (beyond 85%) Productivity/Utilization (beyond 80%) Average Age of Nurses Proportion of Nurses Reporting Shift Changes Proportion of Emotionally Exhausted Nurses Average Nurse-physician Relationship Proportion of Physically Healthy Nurses Proportion of Satisfied Nurses * for p-value at 0.05 or less
-0.02 2.89 * -0.40 *
0.10 * 0.78 -0.85 *
0.08 1.63 -0.53
-0.10 * -3.28 * -0.20 -0.15 *
-0.07 * -1.37 -0.16
-2.08 * -0.43 0.88 * 4.62 * -3.88 *
Notes: (1) Except for nurse-physician relationship and autonomy, all nurse outcomes were dichotomized and modelled in hierarchial logistic regression. For satisfaction, 1 = satisfied with current job; for emotional exhaustion, 1 = at risk; for physical and mental health, 1 =healthier than average of female population.
95
Table 11. Odds Ratios for Nurse Outcomes in the Hierarchical Linear Models
Predictor N-P relationship Autotonomy Satisfaction Emotional Exhaust'n
Physical Health 0.41 * 0.99 1.02 0.90 1.02 1.15 0.80 0.77 0.90 1.45 0.51 * 0.59 * 1.06 0.91
Mental Health 0.48 * 1.03 * 0.81 0.97 1.02 1.09 0.76 0.96 0.95 0.87 0.60 0.21 * 1.04 1.06 0.96 *
Nurse Level
Gender (ref: Male) Age Education (ref: Diploma) Full-time Employment (ref: PT/Casual) Over Time Hours Clinical Expertise Unit Instability Shift Change Interventions Not Done Interventions Delayed Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-physician Relationship Absenteeism Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Nurse-patient Ratio
n/a n/a
n/a * n/a n/a n/a * n/a n/a n/a n/a n/a * n/a n/a * n/a n/a n/a n/a * n/a * n/a n/a * n/a * n/a n/a n/a * n/a n/a n/a *
n/a *
1.56 * 1.12 0.98 1.09 0.73 1.01 0.88 1.01 0.61 0.29 * 1.24 * 1.06 1.02 1.02 1.45 4.01 * 1.36
0.97 1.12 2.79 * 1.01 1.24 1.03 0.89 3.42 * 0.98 1.00 0.96 * 0.90 * 0.34 * 1.32 0.56 1.64
n/a n/a n/a n/a * n/a *
0.98 * 1.58 * 1.14 0.79 0.69
1.74 * 1.15 1.00 0.75
Patient Level
n/a n/a *
1.05 0.90
1.05 0.93
Unit Level
Unit Occupancy Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization (beyond 85%) Productivity/Utilization (beyond 80%) Average Age of Nurses Proportion of Nurses Reporting Shift Changes Proportion of Emotionally Exhausted Nurses Average Nurse-physician Relationship Proportion of Physically Healthy Nurses Proportion of Satisfied Nurses * for p-value at 0.05 or less
n/a n/a * n/a * n/a *
1.10 * 1.08 0.43 *
1.08 1.18 0.59
0.90 * 0.72 * 0.82 0.86 *
0.93 * 0.87 0.85
0.65 2.40 * n/a * 0.68 *
Notes: (1) Except for nurse-physician relationship and autonomy, all nurse outcomes were dichotomized and modelled in hierarchial logistic regression. For satisfaction, 1 = satisfied with current job; for emotional exhaustion, 1 = at risk; for physical and mental health, 1 =healthier than average of female population. (2) The odds ratio for proportion of RN worked hours, proportion of nurses with BScN or above, proportion of nurses reporting shift changes, proportion of emotionally exhaused nurses, proportion of physically healthy nurses, proportion of satisfied nurses, proportion of nurses rating good nurse care quality are based on a 10% increase.
96
Table 12: Hierarchical Linear Regression for Nurse-Physician Relationship

Predictor Nurse Level Education (ref: Diploma) Full-Time Employment (ref: PT/Casual) Autonomy Physical Health Mental Health Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Patient Level Average Resource Intensity Weight Average Number of Nursing Diagnoses Nurse-Patient Ratio Unit Level Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization (beyond 85%) Proportion of Nurses Reporting Shift Changes Proportion of Physically Healthy Nurses Beta 0.16 0.16 0.30 -0.01 0.01 0.29 0.59 0.02 -0.10 -0.37 -0.02 2.89 -0.40 -2.08 4.62 SE 0.14 0.14 0.03 0.01 0.01 0.15 0.22 0.02 0.05 0.18 0.03 1.39 0.20 0.78 1.01 Odds Ratio n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a
* *
* * * *
* p-value at 0.05 or less Note: For measurements of predictor and outcome variables, see Appendix F.
97
Table 13: Hierarchical Linear Regression for Autonomy

Predictor Nurse Level
Education (ref: Diploma) Full-Time Employment (ref: PT/Casual) Overtime Hours Clinical Expertise Unit Instability Shift Change Interventions Not Done Interventions Delayed Effort and Reward Imbalance Emotional Exhaustion Nurse-Physician Relationship Absenteeism Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Nurse-Patient Ratio
Beta -0.44 0.03 0.01 -0.33 -0.03 -0.01 -0.33 -0.10 -0.89 -0.07 0.62 -0.36 0.02 0.02 1.04 0.67 0.28 0.67 0.83 0.04 -0.05 -2.07 0.02 1.69 1.58
SE 0.20 0.22 0.02 0.16 0.25 0.20 0.26 0.23 0.29 0.26 0.06 0.25 0.01 0.01 0.22 0.23 0.32 0.24 0.37 0.03 0.07 1.03 0.04 2.01 0.29
Odds Ratio n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a *
* *
* * * *
Patient Level
Unit Level
Unit Occupancy Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization (beyond 85%)
* p-value at 0.05 or less Note: For measurements of predictor and outcome variables, see Appendix F.
98
Table 14: Hierarchical Logistic Regression for Job Satisfaction

Education (ref: Diploma) Full-Time Employment (ref: PT/Casual) Overtime Hours Clinical Expertise Unit Instability Shift Change Interventions Not Done Interventions Delayed Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-Physician Relationship Physical Health Mental Health Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Nurse-Patient Ratio
Beta 0.44 0.12 -0.02 0.09 -0.31 0.01 -0.13 0.01 -0.49 -1.23 0.21 0.06 0.02 0.02 0.37 1.39 0.31 0.05 -0.10 0.10 0.78 -0.85
SE 0.22 0.23 0.02 0.19 0.27 0.22 0.27 0.25 0.39 0.30 0.05 0.07 0.01 0.01 0.25 0.46 0.29 0.04 0.08 0.04 1.99 0.32
Odds Ratio 1.56 1.12 0.98 1.09 0.73 1.01 0.88 1.01 0.61 0.29 1.24 1.06 1.02 1.02 1.45 4.01 1.36 1.05 0.90 1.10 1.08 0.43 * * *
* *
Patient Level
Average Resource Intensity Weight Average Number of Nursing Diagnoses
Unit Level
Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization (beyond 80%)
99
Table 15: Hierarchical Logistic Regression for Emotional Exhaustion

Age Education (ref: Diploma) Full-Time Employment (ref: PT/Casual) Overtime Hours Clinical Expertise Shift Change Interventions Not Done Effort and Reward Imbalance Autonomy Nurse-Physician Relationship Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Nurse-Patient Ratio
Beta -0.03 0.11 1.03 0.01 0.21 0.03 -0.12 1.23 -0.02 0.00 -0.04 -0.11 -1.09 0.28 -0.57 0.49 0.05 -0.07 0.08 1.63 -0.53 -3.88
SE 0.02 0.26 0.28 0.03 0.22 0.27 0.34 0.33 0.05 0.08 0.02 0.01 0.31 0.29 0.40 0.33 0.05 0.10 0.06 2.36 0.35 1.37
Odds Ratio 0.97 1.12 2.79 1.01 1.24 1.03 0.89 3.42 0.98 1.00 0.96 0.90 0.34 1.32 0.56 1.64 1.05 0.93 1.08 1.18 0.59 0.68
* * *
Patient Level
Unit Level
Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization (beyond 85%) Proportion of Satisfied Nurses
Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratios for proportion of RN worked hours and proportion of satisfied nurses are based on a 10% increase.
100
Table 16: Hierarchical Linear Regression for Nurses Physical Health

Gender (ref: Male) Age Education (ref: Diploma) Full-Time Employment (ref: PT/Casual) Overtime Hours Clinical Expertise Unit Instability Shift Change Interventions Not Done Interventions Delayed Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-Physician Relationship Mental Health Satisfaction with Current Job (ref: Dissatisfied) Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Nurse-Patient Ratio
Beta -0.90 -0.01 0.02 -0.11 0.02 0.14 -0.22 -0.26 -0.10 0.37 -0.67 -0.53 0.05 -0.10 -0.02 0.46 0.13 -0.24 -0.38 -0.10 -3.28 -0.20 -0.15 0.88
SE 0.42 0.01 0.18 0.18 0.02 0.15 0.21 0.17 0.23 0.20 0.25 0.23 0.04 0.05 0.01 0.20 0.20 0.28 0.21 0.04 1.20 0.21 0.04 0.17
Odds Ratio 0.41 0.99 1.02 0.90 1.02 1.15 0.80 0.77 0.90 1.45 0.51 0.59 1.06 0.91 0.98 1.58 1.14 0.79 0.69 0.90 0.72 0.82 0.86 2.40 *
* *
* *
Unit Level
Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization (beyond 85%) Average Age of Nurses Average Nurse-Physician Relationship
* * * *
101
Table 17: Hierarchical Linear Regression for Nurses Mental Health

Gender (ref: Male) Age Education (ref: Diploma) Full-Time Employment (ref: PT/Casual) Overtime Hours Clinical Expertise Unit Instability Shift Change Interventions Not Done Interventions Delayed Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-Physician Relationship Physical Health Satisfaction with Current Job (ref: Dissatisfied) Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Nurse-Patient Ratio
Beta -0.74 0.03 -0.21 -0.03 0.02 0.08 -0.27 -0.05 -0.05 -0.14 -0.51 -1.56 0.04 0.05 -0.04 0.55 0.14 0.00 -0.29 -0.07 -1.37 -0.16 -0.43
SE 0.38 0.01 0.18 0.18 0.02 0.16 0.23 0.18 0.23 0.20 0.28 0.23 0.04 0.05 0.01 0.20 0.20 0.28 0.23 0.03 1.25 0.23 0.88
Odds Ratio 0.48 1.03 0.81 0.97 1.02 1.09 0.76 0.96 0.95 0.87 0.60 0.21 1.04 1.06 0.96 1.74 1.15 1.00 0.75 0.93 0.87 0.85 0.65 * *
* *
Unit Level
Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization (beyond 85%) Proportion of Emotionally Exhausted Nurses
* p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratios for proportion of RN worked hours and proportion of emotionally exhausted nurses are based on a 10% increase.
102
Table 18. Estimates for System Outcomes from the Hierarchical Linear Models
Age Dependent Children (ref: No) Education (ref: Diploma) Work on Multiple Units Full-time Employment (ref: PT/Casual) Over Time Hours Clinical Expertise Unit Instability Shift Change Prevalence of Violence Interventions Not Done Interventions Delayed Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-physician Relationship Intent to Leave Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Nurse-patient Ratio Re-sequencing of Work Unanticipated Changes in Patient Acuity More Time Needed -0.04 * 0.32 -0.06 0.48 0.09 * 0.11 0.06 -0.48 0.82 * -0.02 -0.01 0.55 * 0.03 -0.09 0.63 * -0.13 0.43 *
LOS
Not Done
Delay
Quality of Nursing Care
Quality of Patient Care
Absenteeism -0.02 0.56 -0.06 0.93 * 0.01 -0.20 0.27 0.31
Leave
Productivity/ Utilization
Cost per RIW
Worked hours
-0.03
0.00 -0.49 0.48 0.02 0.58 * -0.04 -0.69 * -0.59 -0.11 0.33 -0.61 0.07 0.22 * 0.00 0.01 0.95 * 1.95 * -0.69
0.15 0.29 -0.17 -0.02 -0.84 * 0.23 0.33 -0.25 -0.62 * -0.44 0.30 0.16 * 0.03 0.01 0.01 0.33 2.32 * -0.54
0.00 -0.03 0.70 * -0.72 * -0.01 0.13 1.09 * 0.42
0.003
0.047
0.015
-0.001
0.004 -0.163 *
-0.002
-0.10 0.17 1.28 * -0.12 -0.13 * -0.14 -0.04 -0.02 -0.01 -0.17 0.80 * 0.01 -0.07 0.06 0.37 0.01 -0.02 -0.21
0.024 0.000 -0.03 -0.04 -0.08 0.11 -0.05 * -0.02 0.01 -0.50 0.54 0.38 -0.35 0.37 -0.08 0.07 0.02 -0.01 -0.87 * -0.35 0.09 -0.08
0.069 -0.025
-0.012 0.018
0.007 *
-0.02 0.02
0.001 0.001
-0.011 * 0.000
-0.001 0.000
-0.17
0.43 -1.44 * 2.97 *
0.50
0.045 *
-0.110 *
-0.010
0.001 *
1.10 * 1.05 *
Patient Level
Proportion of Patients Employed Full-time Pre-operative Clinics Post-operative/-discharge Education Medical Consequences Emergency Admission (ref: Elective) Resource Intensity Weight Number of Nursing Diagnoses Physical Health at Admission Mental Health at Admission Worked Hours per Patient Length of Stay
0.241 * 0.128 * -0.153 *
-0.85 *
-0.31 * -0.14 * 0.02 * 0.01 0.01 0.11 * -0.22 * 0.01 -0.23 *
0.000 -0.003
0.030 * 0.001 0.006 * 0.507 *
0.000 0.006 * 0.000 0.000 -0.001
Unit Level
Pure Cardiology (ref: Mix) Step Down Unit (ref: Other Types of Units) Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization Productivity/Utilization (Quadratic) Productivity/Utilization (beyond 85%) Proportion of Nurses with BScN or Above Proportion of Full-time Employment Average Overtime Hours Average Clinical Expertise Prevalence of Violence at Unit Proportion of Emotionally Exhausted Nurses Average Nurse-physician Relationship Proportion of Nurses Reporting Sick Leave Proportion of Mentally Healthy Nurses Proportion of Nurses Rating Good Nursing Care Quality * for p-value at 0.05 or less
-0.196 * -2.262 *
0.49 8.72 * -4.77 * -0.01 0.56 0.07 -1.40
-0.12 -1.66
-0.05 -5.22 *
0.06 -0.01 -0.25 -1.24 -10.32 * -12.71 * 6.47 * 7.68 *
0.532 *
0.090 -10.557 * 5.901 *
0.597 * 11.872 * -6.619 *
-0.38
-0.24 -3.21 *
-0.25
-0.11 3.34 * -0.297 *
-0.29 * -2.77 * 2.10 *
-0.260 * -0.557 * -0.75 *

5.35 *
-0.445 * 6.57 * -3.62 *
Notes: (1) Length of stay, tasks not done or delayed, quality of nursing care and quality of patient care were dichotomized and modeled in hierarchical logistic regressions. For length of stay, 1 = hospital stay shorter than expected; for not done, 1 = at least one task not done on last shift; for delay, 1 = at least one task delayed on last shift; quality of nursing care, 1 = excellent/good; for quality of patient care, 1 = improved; for absenteeism, 1 = more than one occasion in past year; for leave, 1 = intent to leave within next year. (2) The productivity/utilization cut points are 91.4% for LOS, 79.7% for absenteeism, 82.8% for intent to leave, 89.5% for cost per RIW and 89.7% for worked hours.
103
Table 19. Odds Ratios for System Outcomes in the Hierarchical Linear Models
Age Dependent Children (ref: No) Education (ref: Diploma) Work on Multiple Units Full-time Employment (ref: PT/Casual) Over Time Hours Clinical Expertise Unit Instability Shift Change Prevalence of Violence Interventions Not Done Interventions Delayed Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-physician Relationship Intent to leave Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Nurse-patient Ratio Re-sequencing of Work Unanticipated Changes in Patient Acuity More Time Needed 0.96 * 1.37 0.94 1.62 1.09 * 1.12 1.06 0.62 2.27 * 0.98 0.99 1.74 * 1.03 0.92 1.87 * 0.88 1.53 *
LOS
Not Done
Delay
Quality of Nursing Care
Quality of Patient Care
Absenteeism 0.98 1.75 0.94 2.52 * 1.01 0.82 1.31 1.37
Leave
Productivity/ Utilization
Cost per RIW
Worked hours
0.97
1.00 0.61 1.62 1.02 1.79 * 0.97 0.50 * 0.56 0.90 1.40 0.54 1.08 1.25 * 1.00 1.01 2.59 * 7.06 * 0.50
1.16 1.33 0.84 0.98 0.43 * 1.26 1.39 0.78 0.54 * 0.64 1.35 1.17 * 1.03 1.01 1.01 1.39 10.15 * 0.59
1.00 0.97 2.01 * 0.49 * 1.00 1.14 2.97 * 1.52
n/a
n/a
n/a
n/a
n/a n/a *
n/a
0.91 1.18 3.60 * 0.89 0.88 * 0.87 0.96 0.98 0.99 0.85 2.23 * 1.01 0.93 1.06 1.45 1.01 0.98 0.81
n/a n/a 0.97 0.96 0.92 1.12 0.95 * 0.98 1.01 0.60 1.72 1.47 0.70 1.45 0.92 1.08 1.02 0.99 0.42 * 0.71 1.10 0.92
n/a n/a
n/a n/a
n/a *
0.98 1.02
n/a n/a
n/a * n/a
n/a n/a
0.85
1.53 0.24 * 19.47 *
1.65
n/a *
n/a *
n/a
n/a *
3.00 * 2.85 *
Patient Level
Proportion of Patients Employed Full-time Pre-operative Clinics Post-operative/-discharge Education Medical Consequences Emergency Admission (ref: Elective) Resource Intensity Weight Number of Nursing Diagnoses Physical Health at Admission Mental Health at Admission Worked Hours per Patient Length of Stay
n/a * n/a * n/a *
0.43 *
0.74 * 0.87 * 1.02 * 1.01 1.01 1.12 * 0.80 * 1.01 0.80 *
n/a n/a
n/a * n/a n/a * n/a *
n/a n/a * n/a n/a n/a
Unit Level
Pure Cardiology (ref: Mix) Step Down Unit (ref: Other Types of Units) Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization Productivity/Utilization (Quadratic) Productivity/Utilization (beyond 85%) Proportion of Nurses with BScN or Above Proportion of Full-time Employment Average Overtime Hours Average Clinical Expertise Prevalence of Violence at Unit Proportion of Emotionally Exhausted Nurses Average Nurse-physician Relationship Proportion of Nurses Reporting Sick Leave Proportion of Mentally Healthy Nurses Proportion of Nurses Rating Good Nursing Care Quality * for p-value at 0.05 or less
n/a * n/a *
1.05 n/a * n/a * 0.99 1.06 1.07 0.87
0.89 0.85
0.95 0.59 *
1.07 0.97 n/a * n/a *
0.99 0.88 n/a * n/a *
n/a *
n/a n/a * n/a *
n/a * n/a * n/a *
0.69
0.78 0.73 *
0.78
0.89 1.40 * n/a *
0.75 * 0.06 * 8.14 *
n/a * n/a * 0.47 *

1.71 *
n/a * 1.93 * 0.03 *
Notes: (1) Length of stay, tasks not done or delayed, quality of nursing care and quality of patient care were dichotomized and modeled in hierarchical logistic regressions. For length of stay, 1 = hospital stay shorter than expected; for not done, 1 = at least one task not done on last shift; for delay, 1 = at least one task delayed on last shift; for quality of nursing care, 1 = excellent/good; for quality of patient care, 1 = improved; for absenteeism, 1 = more than one occasion in past year; for leave, 1 = intent to leave within next year. (2) The productivity/utilization cut points are 91.4% for LOS, 79.7% for absenteeism, 82.8% for intent to leave, 89.5% for cost per RIW, and 89.7% for worked hours. (3) The odds ratios for proportion of RN worked hours, proportion of nurses with BScN or above, proportion of nurses reporting sick leave and proportion of nurses rating good nursing care quality are based on a 10% increase. (4) Odds ratio for quadratic transformation of productivity/utilization is not reported.
104
Table 20: Hierarchical Logistic Regression for Patients with Shorter Than Expected Length of Stay
Predictor Patient Level Pre-operative Clinics Medical Consequences Resource Intensity Weight Number of Nursing Diagnoses Physical Health at Admission Mental Health at Admission Worked Hours per Patient Nurse Level Education (ref: Diploma) Overtime Hours Interventions Not Done Interventions Delayed Physical Health Mental Health Nurse-Patient Ratio Unit Level Proportion of RN Worked Hours Productivity/Utilization Productivity/Utilization (Quadratic) Beta 1.05 -0.85 -0.31 -0.14 0.02 0.01 0.01 0.32 -0.03 -0.10 0.17 -0.02 0.02 -0.17 0.49 8.72 -4.77 SE 0.26 0.31 0.04 0.04 0.01 0.01 0.03 0.29 0.04 0.36 0.29 0.02 0.02 0.10 1.78 2.10 1.34 Odds Ratio 2.85 0.43 0.74 0.87 1.02 1.01 1.01 1.37 0.97 0.91 1.18 0.98 1.02 0.85 1.05 n/a n/a * * * * *
* *
* p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratio for proportion of RN worked hours is based on a 10% increase. (3) Productivity/utilization cut point is 91.4%.
105
Table 21: Hierarchical Logistic Regression for Interventions Not Done

Age Education (ref: Diploma) Full-Time Employment (ref: PT/Casual) Overtime Hours Clinical Expertise Unit Instability Shift Change Prevalence of Violence Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-Physician Relationship Intent to Leave Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Nurse-Patient Ratio Re-sequencing of Work Unanticipated Changes in Patient Acuity
Beta -0.044 -0.061 0.480 0.085 0.109 0.058 -0.477 0.821 1.281 -0.115 -0.126 -0.143 -0.040 -0.020 -0.009 -0.168 0.428 -1.443 2.969 1.099 0.113 -0.218 -0.011 0.557 -0.376 -0.292 -2.771
SE 0.017 0.246 0.260 0.025 0.210 0.318 0.254 0.219 0.557 0.347 0.051 0.077 0.384 0.016 0.015 0.263 0.481 0.445 0.574 0.399 0.038 0.085 0.067 2.510 0.403 0.139 0.937
Odds Ratio 0.96 0.94 1.62 1.09 1.12 1.06 0.62 2.27 3.60 0.89 0.88 0.87 0.96 0.98 0.99 0.85 1.53 0.24 19.47 3.00 1.12 0.80 0.99 1.06 0.69 0.75 0.06
*
* * *
* * * * *
Patient Level
Proportion of Patients Employed Full-Time Average Resource Intensity Weight Average Number of Nursing Diagnoses
Unit Level
Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization (beyond 85%) Average Overtime Hours Average Clinical Expertise
* *
106
Table 22: Hierarchical Logistic Regression for Interventions Delayed

Predictor Patient Level
Beta 0.01 -0.23 -0.02 -0.01 0.55 0.03 -0.09 0.63 -0.13 0.43 0.80 0.01 -0.07 0.06 0.37 0.01 -0.02 -0.21 0.50 0.07 -1.40 -0.24 -3.21 2.10 5.35
SE 0.03 0.08 0.01 0.22 0.22 0.02 0.19 0.28 0.22 0.17 0.37 0.29 0.04 0.07 0.32 0.01 0.01 0.23 0.32 0.05 2.20 0.30 1.00 0.81 1.63
Odds Ratio 1.01 0.80 0.98 0.99 1.74 1.03 0.92 1.87 0.88 1.53 2.23 1.01 0.93 1.06 1.45 1.01 0.98 0.81 1.65 1.07 0.87 0.78 0.73 8.14 1.71
Nurse Level
Age Education (ref: Diploma) Full-Time Employment (ref: PT/Casual) Overtime Hours Clinical Expertise Unit Instability Shift Change Prevalence of Violence Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-Physician Relationship Intent to Leave Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Nurse-Patient Ratio
* * *
Unit Level
Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization (beyond 85%) Proportion of Nurses with BScN or Above Prevalence of Violence at Unit Proportion of Nurses Reporting Sick Leave
* * *
* p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratios for proportion of RN worked hours, proportion of nurses with BScN or above, and proportion of nurses reporting sick leave are based on a 10% increase.
107
Table 23: Hierarchical Logistic Regression for Quality of Nursing Care

Predictor Nurse Level Education (ref: Diploma) Work on Multiple Units Full-Time Employment (ref: PT/Casual) Overtime Hours Clinical Expertise Unit Instability Shift Change Interventions Not Done Interventions Delayed Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-Physician Relationship Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Improved Quality of Patient Care (ref: Deteriorated) Nurse-Patient Ratio Beta 0.00 -0.49 0.48 0.02 0.58 -0.04 -0.69 -0.59 -0.11 0.33 -0.61 0.07 0.22 0.00 0.01 0.95 1.95 -0.69 SE 0.29 0.30 0.30 0.02 0.22 0.35 0.30 0.48 0.36 0.35 0.34 0.06 0.09 0.02 0.02 0.40 0.36 0.45 Odds Ratio 1.00 0.61 1.62 1.02 1.79 0.97 0.50 0.56 0.90 1.40 0.54 1.08 1.25 1.00 1.01 2.59 7.06 0.50
* *
* *
Unit Level Average Worked Hours -0.12 0.07 0.89 Proportion of RN Worked Hours -1.66 2.63 0.85 Productivity/Utilization (beyond 85%) -0.25 0.35 0.78 Average Nurse-Physician Relationship -0.75 0.23 0.47 Proportion of Nurses Rating Good Nursing Care 6.57 1.62 1.93 Quality * p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratios for proportion of RN worked hours and proportion of nurses rating good nursing care quality are based on a 10% increase.
* *
108
Table 24: Hierarchical Logistic Regression for Quality of Patient Care

Predictor Nurse Level Education (ref: Diploma) Work on Multiple Units Full-Time Employment (ref: PT/Casual) Overtime Hours Clinical Expertise Unit Instability Shift Change Interventions Not Done Interventions Delayed Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-Physician Relationship Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Good Quality of Nursing Care (ref: Deteriorated) Nurse-Patient Ratio Unit Level Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization (beyond 85%) Proportion of Nurses with BScN or Above Beta 0.15 0.29 -0.17 -0.02 -0.84 0.23 0.33 -0.25 -0.62 -0.44 0.30 0.16 0.03 0.01 0.01 0.33 2.32 -0.54 -0.05 -5.22 -0.11 3.34 SE 0.20 0.22 0.21 0.02 0.18 0.27 0.20 0.26 0.23 0.29 0.26 0.04 0.06 0.01 0.01 0.23 0.38 0.41 0.06 1.33 0.47 1.64 Odds Ratio 1.16 1.33 0.84 0.98 0.43 1.26 1.39 0.78 0.54 0.64 1.35 1.17 1.03 1.01 1.01 1.39 10.15 0.59 0.95 0.59 0.89 1.40
* *
* p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratios for proportion of RN worked hours and proportion of nurses with BScN or above are based on a 10% increase.
109
Table 25: Hierarchical Logistic Regression for Absenteeism

Age Dependent Children (ref: No) Education (ref: Diploma) Full-Time Employment (ref: PT/Casual) Overtime Hours Clinical Expertise Unit Instability Shift Change Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-Physician Relationship Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Nurse-Patient Ratio
Beta -0.02 0.56 -0.06 0.93 0.01 -0.20 0.27 0.31 -0.03 -0.04 -0.08 0.11 -0.05 -0.02 0.01 -0.50 0.54 0.38 0.06 -0.25 -10.32 6.47
SE 0.02 0.45 0.24 0.24 0.02 0.22 0.30 0.24 0.37 0.33 0.05 0.07 0.02 0.01 0.27 0.28 0.38 0.29 0.05 1.84 4.96 2.87
Odds Ratio 0.98 1.75 0.94 2.52 1.01 0.82 1.31 1.37 0.97 0.96 0.92 1.12 0.95 0.98 1.01 0.60 1.72 1.47 1.07 0.97 n/a n/a
Unit Level
Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization Productivity/Utilization (Quadratic)
* *
* p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratio for proportion of RN worked hours is based on a 10% increase. (3) Productivity/utilization cut point is 79.7%.
110
Table 26: Hierarchical Logistic Regression for Intent to Leave

Age Dependent Children (ref: No) Education (ref: Diploma) Full-Time Employment (ref: PT/Casual) Overtime Hours Clinical Expertise Unit Instability Shift Change Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-Physician Relationship Physical Health Mental Health Satisfaction with Current Job (ref: Dissatisfied) Improved Quality of Patient Care (ref: Deteriorated) Good Quality of Nursing Care (ref: Poor) Nurse-Patient Ratio
Beta 0.00 -0.03 0.70 -0.72 -0.01 0.13 1.09 0.42 -0.35 0.37 -0.08 0.07 0.02 -0.01 -0.87 -0.35 0.09 -0.08 -0.01 -1.24 -12.71 7.68 -3.62
SE 0.02 0.37 0.23 0.24 0.02 0.19 0.25 0.24 0.31 0.28 0.05 0.07 0.01 0.01 0.29 0.26 0.34 0.36 0.06 1.97 5.00 3.00 1.14
Odds Ratio 1.00 0.97 2.01 0.49 1.00 1.14 2.97 1.52 0.70 1.45 0.92 1.08 1.02 0.99 0.42 0.71 1.10 0.92 0.99 0.88 n/a n/a 0.03
* *
Unit Level
Average Worked Hours Proportion of RN Worked Hours Productivity/Utilization Productivity/Utilization (Quadratic) Proportion of Nurses Rating Good Nursing Care Quality
* * *
Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Odds ratios for proportion of RN worked hours and proportion of nurses rating good nursing care quality are based on a 10% increase. (3) Productivity/utilization cut point is 82.8%.
111
Table 27: Hierarchical Linear Regression for Productivity/Utilization

Predictor Patient Level Resource Intensity Weight Number of Nursing Diagnoses Nurse Level Education (ref: Diploma) Overtime Hours Interventions Not Done Interventions Delayed Autonomy Physical Health Mental Health Nurse-Patient Ratio More Time Needed Unit Level Pure Cardiology (ref: Mix) Proportion of RN Worked Hours Proportion of Emotionally Exhausted Nurses Proportion of Mentally Healthy Nurses Beta 0.000 -0.003 0.003 -0.001 0.024 0.000 0.007 0.001 0.001 0.045 0.001 -0.196 0.532 -0.557 -0.445 SE 0.003 0.005 0.018 0.001 0.024 0.020 0.003 0.001 0.001 0.007 0.000 0.041 0.123 0.140 0.170 Odds Ratio n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a
* * * * * *
* p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F.
112
Table 28: Hierarchical Linear Regression for Cost per Resource Intensity Weight (Log Scale)
Predictor Patient Level Pre-operative Clinics Post-operative/-discharge Education Emergency Admission (ref: Elective) Number of Nursing Diagnoses Physical Health at Admission Mental Health at Admission Length of Stay Nurse Level Education (ref: Diploma) Overtime Hours Clinical Expertise Interventions Not Done Interventions Delayed Physical Health Nurse-Patient Ratio More Time Needed Unit Level Step Down Unit (ref: Other Types of Units) Proportion of RN Worked Hours Productivity/Utilization Productivity/Utilization (Quadratic) Beta 0.241 0.128 -0.153 0.030 0.001 0.006 0.507 0.047 0.004 -0.163 0.069 -0.025 -0.011 -0.110 0.000 -2.262 0.090 -10.557 5.901 SE 0.056 0.043 0.044 0.011 0.002 0.002 0.029 0.068 0.009 0.052 0.086 0.076 0.004 0.027 0.004 0.681 0.558 0.514 0.290 Odds Ratio n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a * * * * * *
* *
* * *
* p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Productivity/utilization cut point is 89.5%.
113
Table 29: Hierarchical Linear Regression for Worked Hours per Patient (Log Scale)
Predictor Patient Level Resource Intensity Weight Number of Nursing Diagnoses Physical Health at Admission Mental Health at Admission Length of Stay Nurse Level Education (ref: Diploma) Overtime Hours Interventions Not Done Interventions Delayed Physical Health Mental Health Nurse-Patient Ratio Unit Level Proportion of RN Worked Hours Productivity/Utilization Productivity/Utilization (Quadratic) Proportion of Full-Time Employment Average Clinical Expertise Beta 0.000 0.006 0.000 0.000 -0.001 0.015 -0.002 -0.012 0.018 -0.001 0.000 -0.010 0.597 11.872 -6.619 -0.297 -0.260 SE 0.002 0.002 0.000 0.000 0.015 0.015 0.002 0.020 0.017 0.001 0.001 0.005 0.180 0.353 0.189 0.075 0.043 Odds Ratio n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a * * * * *
* p-value at 0.05 or less Notes: (1) For measurements of predictor and outcome variables, see Appendix F. (2) Productivity/utilization cut point is 89.7%.
114
Table 30: Hierarchical Linear Models for Patient, Nurse, and System Outcomes on Congruence Between PRN Hours and Actual Worked Hours per Patient
Coefficient
Patient Outcome Medical Consequences Physical Health Improvement Mental Health Improvement Omaha Knowledge Improvement Omaha Behaviour Improvement Omaha Status Improvement Nurse Outcome Nurse-Physician Relationship Autonomy Satisfaction Emotional Exhaustion Mentally Healthy Physically Healthy System Outcome LOS Shorter than Expected LOS Interventions Not Done Interventions Delayed Improved Rating for Quality of Patient Care Good Rating for Quality of Nursing Care Absenteeism Intent to Leave Cost per Resource Intensity Weight Productivity/Utilization
SE
p-value
OR
-0.0026 -0.0045 0.0006 -0.0055 -0.0007 0.0024 0.1476 0.0205 -0.0024 -0.0006 0.0008 0.0022 0.0018 0.0004 0.0047 -0.0041 -0.0002 -0.0016 0.0055 -0.0093 0.0117
0.0022 0.0043 0.0044 0.0041 0.0042 0.0037 0.0108 0.1747 0.0028 0.0025 0.0027 0.0026 0.0040 0.0024 0.0027 0.0026 0.0019 0.0023 0.0020 0.0080 0.0025
0.2209 0.2927 0.8930 0.1779 0.8687 0.5102 0.1721 0.2405 0.3908 0.8247 0.7751 0.4016 0.6546 0.8766 0.0847 0.1247 0.9076 0.4864 0.0067 0.2433 0.0000
1.00 1.00 1.00 0.99 1.00 1.00 n/a n/a 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.01 n/a n/a
* *
* p-value at 0.05 or less Note: (1) For measurements of outcome variables, see Appendix F. (2) The predictor is the PRN hours less the worked hours per patient.
115
Table 31: Summary Table of the Effect of Nursing Hours, Proportion of RN Worked Hours, Nurse-Patient Ratio, and Productivity/Utilization on Patient, Nurse and System Outcomes, in Odds Ratio, Coefficient, and Cut point
Odds ratio Patient outcome Medical Consequences Physical Health Mental Health OMAHA Knowledge OMAHA Behaviour OMAHA Status Nurse outcome Satisfaction Emotional Exhaustion Physical Health Mental Health System outcome Length of Stay Interventions Not Done Interventions Delayed Quality of Patient Care Quality of Nursing Care Absenteeism Intent to Leave NursePatient Ratio ns ns ns ns ns ns ns ns ns ns ns ns ns ns Ns ns ns NursePatient Ratio -0.37 0.67 0.05 -0.11 ns Worked Hours 1.13a ns 0.94a ns ns ns 1.10b ns 0.90b 0.93b ns ns ns ns Ns ns ns Average Worked Hours ns ns n/a n/a n/a Proportion of RN Worked Hours (10% Increase) ns ns ns 1.74 ns ns ns ns 0.72 ns ns ns ns 0.59 Ns ns ns Proportion of RN Worked Hours (10% Increase) 0.29 ns 0.05 ns 0.60 Productivity/ Utilization (Cut point) ns 80.0% ns n/a 88.2% n/a 80.0% ns ns ns 91.4% ns ns ns ns 79.7% 82.8% Productivity/ Utilization (Cut point) 85% 85% n/a 89.5% 89.7%
Coefficient Nurse outcome Relationship with Physician Autonomy System outcome Productivity/Utilization Cost per RIW Worked Hours per Patient
a. Worked hours per patient. b. Average worked hours on unit. Notes: (1) Only significant predictors are presented for odds ratios and cut points. (2) "ns" stands for not significant. Predictors with ns have no impact on the outcome variables. (3) "n/a" stands for not applicable. (4) Cost per RIW and worked hours were modeled in logarithm scale, therefore by transforming back to the original scale, a 10% increase in proportion of RN worked hours would lead to an exponential increase of 0.60 (or 1.06 times) in the worked hours per patient which is a 6% increase in worked hours per patient.
116
Appendix D. Instruments, Psychometric Properties, and Variables at Individual and Unit Levels
Measure Patient Characteristics: These variables were collected from the patients kardex and chart. The actual values of these variables were used at the individual level of analysis. Nursing diagnoses identify the conditions in patients that create the demand for nursing services. Content validity of the taxonomy of nursing diagnoses is inferred from the judgment and agreement of nurse experts meetings held bi-annually for several years1. The number of different nursing diagnoses the patient had over the hospital stay was used as an independent variable at the individual level. Patient medical condition was measured by the Case Mix Groups Patient Input Medical Diagnoses (CMGs) TM developed by the Canadian Institute for Health Information). The CMG methodology has been refined a number of times over the last several years to improve the content validity of the measure. The Resource Intensity Weight (RIW) assigned to an individual CMG was used in the analysis at the individual level. The average RIW was used where aggregation was applied. Patient Input and Output Each nursing diagnosis selected is evaluated on three dimensions OMAHA Problem Rating (knowledge, behaviour, and status) on a 5-point Likert scale at two Scale2 points in time: at admission or when a new health problem is identified (Time 1), and when the health problem is resolved or at discharge (Time 2)2. Knowledge involves what a client knows and understands about a specific health-related problem. Behaviour involves what a client does - the clients practices, performances, and skills. Status involves what a client is and how the clients conditions or circumstances improve, remain stable, or deteriorate2. While this rating scale has been used primarily in the community setting, the actual measurement scale is non sectorial in nature and appropriate for use in the hospital environment. In a previous study, the inter-rater reliability for both nursing diagnoses and the OMAHA outcomes rating scale was maintained at 91% among nurse participants. The admission score ratings for knowledge, behaviour, and status were each entered as an independent variable in the analysis. When used as a dependent variable each variable was dichotomized as improved over hospital stay or as having no change or deteriorated over hospital stay. Patient Inputs -age, sex, significant other support Patient Input NANDA Form: Nursing Diagnoses1 Description and Psychometric Properties Inputs
117
Measure Patient Input/Output, Outcomes Measure; Nurse Survey Outcome Measure; Medical Outcomes Study Short Form 12 measure of health status3
Description and Psychometric Properties The Medical Outcomes Study SF-123 is a 12-item scale measuring 8 health domains: physical functioning, vitality, role functioning, physical problems, social functioning, bodily pain, mental health, and general health perceptions 4, 5, 6, 7. The SF12 has demonstrated excellent psychometric properties and is currently the most widely used generic measure of health status, having been employed in hundreds of studies across a broad spectrum of disease states5. When treated as a dependent variable, both physical health and mental health scores were dichotomized into healthy and not healthy using the average score for the US population as the cut point. When treated as a unit level independent variable, the proportion of nurses on the unit with physical and mental health scores over US population norm was used. These data were collected in the nurse survey. Each survey was assigned a code number which was known to the investigators only in order to link nurse characteristics to specific patient assignments. The variables created at the individual level were age, gender, number of occasions absent and number of shifts missed, professional designation, level of education, employment status, work on multiple units, clinical expertise, voluntary and involuntary overtime worked, job stability, prevalence of violence, frequency of shift change, planning to leave in the next 12 months intervention not completed on a shift and interventions delayed on a shift. Unit level variables created include proportions of: nurses on unit with a bachelors degree or higher, nurses reporting shift changes, nurses who work on more than one unit, nurses experiencing job insecurity, intending to leave in the next twelve months, nurses with interventions not completed or delayed, nurses absent from the unit daily, and full time positions on the unit. Unit level variables also included the mean age of nurses, mean years of experience, mean ratings of clinical expertise, prevalence of violence on the unit and average overtime hours. Number of beds in the unit, unit type (In Patient Unit, Critical Care Unit, Step Down Unit, and Day Surgery Unit), patient composition (pure cardiac or mix), and care delivery system were collected from the nurse manager for each unit. These data were used as independent variables at the unit level.
Nurse Characteristics: Nurse Survey -age, sex, professional designation, education, years of experience (this unit, this hospital, nursing in general), usual shift rotation, usual number of units worked, etc.
System Characteristics: System Characteristics Hospital Profile and Unit Profile
118
Measure System Behaviours: System behaviours Daily Unit Staffing Form
Description and Psychometric Properties The nurse/patient ratio, the daily number of nursing personnel, daily patient census, admission and discharges, actual use of agency and relief staff were collected from the nurse manager or charge nurse on a daily basis. As a measure of continuity, average number of nurses per day over a patients hospital stay was used as an independent variable in the models. The actual worked hours per patient was estimated by total work hours divided by the number of patients at midnight census at unit on daily basis. The proportion of worked hours contributed by Registered Nurses was calculated. Number of patients per unit bed was computed as a measure of unit occupancy. The daily data were aggregated to either unit or individual level by taking the average in order to model their effect on outcome variables. The difference between PRN workload hours from actual worked hours per patient was computed in order to answer research question 2. When the actual worked hours at patient level was used as a dependent variable, a logarithm transformation was applied to assume normality. This instrument lists 214 indicators or interventions that nurses complete on behalf of patients during a 24-hour period. Each indicator has a standard given point value which reflects the time involved to complete interventions for patients; each point represents five minutes8. A higher point value indicates greater amounts of nursing care required. The PRN methodology has had extensive testing and has gone through several iterations since it was first developed in 1972. Content validity was established by nurse experts during a series of meetings held over this time. In 1978, Chagnon, Audette, Lebrun, and Tilquin9 established the construct and predictive validity of the tool. Work measurement studies demonstrated that the time estimates predicted by the tool corresponded to the degree of work actually done. In this study the PRN estimates served as the gold standard for care required. In this study, every patient was rated every day using the PRN form in order to determine the direct care and time associated with direct care activities. The method proposed by Tilquin in1980, which is still in used today was used to determine total hours of care per patients (Charles Tilquin, personal communications, October, 2003).
Workload PRN8
119
Measure Workload GRASP/Medicus
Description and Psychometric Properties GRASP or MEDICUS hours were collected daily for study patients and for the unit as a whole, including non-study patients as well. GRASP captures workload using a standard time methodology. Each site develops a list of tasks based on the activities they perform, and times are assigned to each of these tasks. The times are based on time and/or frequency or are established by staff nurse consensus. These times reflect the average time to complete the task, by an average nurse, on an average day, for an average patient in the individual facility. This reflects the physical and organizational characteristics of the individual facility. The MEDICUS system captures workload by multiplying a pre-set relative value per level of care by the target hours per unit of workload. Throughputs
Environmental Complexity10
Environmental complexity measures the push and pull that nurses experience in providing care to individual patients at the standard outlined in the nursing care plan. Factor analysis in a number of preliminary studies has revealed that the Environmental Complexity measure taps three main domains: unanticipated delays and re-sequencing of work in response to others, unanticipated delays due to changes in patient acuity, and characteristics and composition of the caregiver10. Factor analysis was completed for this study and again the same three factors emerged. The alpha reliabilities were 0.81, 0.84, and 0.84 respectively. Intermediate System Outputs: Worked hours, productivity/utilization Worked hours were collected from the retrospective application of the workload measurement system used at the hospital. The daily unit productivity/utilization for each unit was computed by examining the workload of patients on the unit divided by worked hours. In the analysis of productivity/utilization, linear and quadratic terms were first tried to test the bell or U shape of the effect of productivity/utilization. If the bell or U shape was not supported by data, piecewise linear was tried next. If piecewise linear is significant, it means that the direction at a certain cut point will change. If both strategies failed, dichotomized productivity/utilization at various cut points was tried and the one shown significance was used in the final model. If all these failed, dichotomized productivity/utilization at 85% was tested.
120
Measure Patient Outcomes: Patient Outcome Measure Medical Outcomes Study Short Form 12 measure of health status3 The Patient Data Form collected information about specific patients over their stay.
Description and Psychometric Properties Outputs The psychometric properties of these measures are explained above. Patient SF-12 physical and mental health status was measured at admission and discharge. Patients individual admission scores were used as independent variables in many of the models. When treated as a dependent variable, improvement at discharge or no improvement at discharge was used for both physical health and mental health scores. Patient medical consequences data were gathered from the patients chart on an ongoing basis, including deep or shallow post-operative wound infections, fall with injury, medication errors, urinary tract infections, bedsores, pneumonia were tracked on this form as well as death, transfers back to ICU and whether the patient was re-admitted with the same diagnoses within three months. Patient mortality was obtained from medical records. Since there were numerous medical consequences with very small frequencies, medical consequences including falls with injury, medication errors, death, and complications such as urinary tract infections, pneumonia, wound infections, bed sores, and thrombosis were summed for each patient and that value was used in the analysis. Data were collected about type of admission, presence of a family doctor, attendance at pre-operative clinics, bookings for post-operative or post-discharge education, referrals to home care and support in the home. These variables were used as patient-level independent variables in the analysis. The Maslach10 Burnout Inventory has 25 items and measures 3 dimensions: emotional exhaustion (alpha = 0.90, test-retest reliability = 0.82), depersonalization (alpha = 0.79, test-retest reliability = 0 .60) and personal accomplishment (alpha = 0.71, test retest reliability = 0 .80). All coefficients were significant at pvalue < 0.001. This measure has been used in numerous studies and has proven robust over time. In the current study, only the emotional exhaustion scores for nurses at the individual level and for the proportion of nurses with high levels of emotional exhaustion at the unit level were entered.
Nurse Outcomes: Nurse Survey Outcome Measure Maslach Burnout Inventory10
121
Measure Nurse Survey Outcome Measure Siegrists Effort and Reward Imbalance11
Description and Psychometric Properties The 17-item Effort and Reward Imbalance scale identifies the imbalance between high effort spent and low reward received at work, and is assumed to be particularly stressful as this imbalance violates core expectations about reciprocity and adequate exchange in a crucial area of social life.11 This measure has been used extensively over the past ten years in many work settings to measure the effect of an imbalance on the physical and mental health of workers and more recently in the nursing population. Combined variable odds ratios are reported as 8.241 and 95% confidence interval. At the individual level we entered the score for nurses, and the proportion of nurses at risk of effort and reward imbalance was used at the unit level. The Nursing Work Index, first developed by Kramer and Hafner12 has been used extensively in the US over the past ten years in research related to magnet hospitals, which are hospitals that are known to attract and retain nurses13,14,15,16 and to have better patient outcomes17. Work attributes that have been demonstrated as important to nurses include autonomy (7 items), control over the work environment (18 items) and nurse-physician relations (2 items). In the Aiken et al. (1994) mortality study using ANOVA, the mean difference in observed mortality between magnet and non-magnet hospitals was significantly different ( p = .01). Laschinger & colleagues recomputed the factor analysis on the scale and identified a 5 factor solution. The five factors were autonomy, control over practice, nurse physician relations, leadership, and resources. The alpha reliabilities of each of these new subscales were 0.69, 0.74, 0.83, 0.80, and 0.80 respectively. At the individual level, the scores on each of these subscales were entered into the model, while at the unit level the mean unit score for each variable was entered into the models. These data were retrieved from medical records. The expected length of stay was derived from CIHIs inpatient database. The actual length of stay or its logarithm transformation was used as an independent variable at the individual level. When used as a dependent variable, it was dichotomized as shorter than expected length of stay versus the same as, or longer than, the expected length of stay. The actual cost per equivalent weighted case by hospital was extracted from the Ontario Case Costing Data Base and multiplying it by the RIW for each patient. The out of province hospital was not used in this analysis. Logarithm transformation was applied to assume normality when it was modeled as a dependent variable.
Nurse Survey Process Variables Revised Nursing Work Index (Autonomy Control over Practice, Nurse-MD Relations)12
System Outcomes: System Outcomes -length of stay
System Outcomes -cost per case
122
References 1. Kim, M. J., McFarland, G.K., & McLane, A. M. (1991). Pocket guide to nursing diagnoses (4th edition). St. Louis, MO: Mosby. 2. Martin, K. S., & Scheet, N. J. (1992). The OMAHA system: Application for community health nursing. Philadelphia, PA: WB Saunders. 3. Ware, Jr., J. E, Kosinski, M., & Keller, A. D. (2002). SF-12: How to score the SF-12 physical and mental health summary scales (4th Ed.). Lincoln, RI: QualityMetric Incorporated. 4. McHorney, C. A., Ware, Jr., J. E., Rogers, W., & Raczek, A. E. (1992). The validity and relative precision of MOS Short and Long Term Status Scales and Dartmouth COOP. Medical Care, 30, 253-265. 5. Ware, Jr., J.E., & Sherbourne, C.D. (1992). The MOS 36-Item short form health survey (SF36): Conceptual framework and item selection. Medical Care, 30, 473-483. 6. Ware, Jr., J. E., Snow, K., Kosinski, M., & Gandek, B. (1993). SF-36 Health survey manual and interpretation guide. Boston: The Health Institute. 7. Wu, A. W. (1991). A health status questionnaire using 30 items from the medical outcomes study: Preliminary validation in persons with early HIV infection. Medical Care, 29, 786. 8. Tilquin, C., Carle, J., Saulnier, D., Lambert, P., & Collaborators. (1981). PRN 80: Measuring the level of nursing care required. Equipe de Recherche Oprationnelle en Sant, Institut National de Systmatique Applique, Universit de Montral: Montral, QC. 9. Chagnon, M., Audette, L. M., Lebrun, L., & Tilquin, C. (1978). Validation of a patient classification through evaluation of the nursing staff degree of occupation. Medical Care, 16(6), 465-475. 10. OBrien-Pallas, L. L., Irvine, D., Peereboom, E., & Murray, M. (1997). Measuring nursing workload: Understanding variability. Nursing Economics, 15(4), 172-182. 11. Maslach, C., & Jackson, S. E. (1982). Burnout in health professions: A social psychological analysis. In G. S. Snaders & J. Suls (Eds.), Social psychology of health and illness (pp. 227-251). Hillsdale: Lawrence Erlbaum Associates. 12. Siegrist, J. (1996). Adverse health effects of high-effort/low rewards conditions. Journal of Occupational Health Psychology, 1(1), 27-41. 13. Kramer, M., & Hafner, L. P. (1989). Shared values: Impact on staff nurse satisfaction and perceived productivity. Nursing Research, 38, 172-177. 14. McClure, M. L., Poulin, M. A., Sovie, M. D., & Wandelt, M. A. (1982). Magnet hospitals: Attraction and retention of professional nurses. Kansas City: American Academy of Nurses. 15. Kutzscher, L. I. T., Sabiston, J. A., Laschinger-Spence, H. K., & Nish, M. (1997). The effects of teamwork on staff perception and empowerment and job satisfaction. Healthcare Management Forum, 10(2), 12-17. 16. Kramer, M., & Schmalenberg, C. (1988). Magnet hospitals: Institutions of excellence, Parts I & II. Journal of Nursing Administration, 18(1), 13-24. 17. Kramer. M., & Schmalenberg, C. (1990). Job satisfaction and retention: Insights for the 90s, Parts I and II. Nursing, 21, 2-7 & 9-13. 18. Aiken, L., Smith, H., & Lake, E.T. (1994). Lower Medicare mortality among a set of hospitals known for good nursing care. Medical Care, 32(8), 771-787.
123
Appendix E. Data Collection Forms

Nursing Diagnoses (NANDA).................................................................................................... 125 OMAHA Problem Rating Scale.................................................................................................. 126 NANDA and OMAHA Summary Sheet..................................................................................... 128 The SF-12 Health Survey ........................................................................................................ 129 SF-12 Health Survey (French Version) ...................................................................................... 131 Nurse Survey............................................................................................................................... 134 Hospital Profile ........................................................................................................................... 140 Unit/Program Profile................................................................................................................... 141 Daily Unit Staffing Form ............................................................................................................ 143 Daily Environmental Complexity Scale ..................................................................................... 148 PRN 80........................................................................................................................................ 151 PRN Daily Workload and Grasp Patient Care Hours ................................................................. 153 Patient Data Form ....................................................................................................................... 154 Maslach=s Burnout Inventory...................................................................................................... 156 Effort-Reward Imbalance............................................................................................................ 158 Nursing Work Index ................................................................................................................... 159
124
Hospital: ______________Unit: ______________Patient: _________________ Hospital Day: __________Date: ______________Data collector: __________
Evidence-based Standards for Measuring Nurse Staffing and Performance
Nursing Diagnoses (NANDA)

Please circle all numbers representing the patients problems that require nursing care. 1. Activity intolerance 2. Activity intolerance, high risk 3. Adjustment, impaired 4. Anxiety 5. Aspiration, high risk for 6. Body image disturbance 7. Body temperature, altered, high risk for 8. Bowel incontinence 9. Breathing pattern, ineffective 10. Cardiac output, decreased 11. Communication, impaired verbal 12. Constipation 13. Constipation, colonic 14. Constipation, perceived 15. Coping, defensive 16. Coping, family: potential for growth 17. Coping, ineffective family: compromised 18. Coping, ineffective family: disabling 19. Coping, ineffective individual 20. Decisional conflict 21. Denial, ineffective 22. Diarrhea 23. Disuse syndrome, high risk for 24. Diversional activity deficit 25. Dysreflexia 26. Family processes, altered 27. Fatigue 28. Fear 29. Fluid volume, deficit (1) 30. Fluid volume deficit (2) 31. Fluid volume, high risk for 32. Fluid volume excess 33. Gas exchange impaired 34. Health maintenance, altered 35. Health-seeking behaviours 36. Hopelessness 37. Incontinence, functional 38. Incontinence, reflex 39. Incontinence, stress 40. Incontinence, total 41. Incontinence, urge 42. Infection, high risk for 43. Injury, high risk for 44. Knowledge deficit 45. Management of therapeutic regimen (individuals), ineffective 46. Mobility, impaired physical 47. Non-compliance 48. Nutrition, altered: less than body requirements 49. Nutrition, altered: more than body requirements 50. Nutrition altered: high risk for more than body requirements 51. Oral mucous membrane, altered 52. Pain 53. Pain, chronic 54. Peripheral neurovascular dysfunction, high risk for 55. Post-trauma response 56. Powerlessness 57. Role performance, altered 58. Self-care deficit, bathing/hygiene 59. Self-care deficit, dressing/grooming 60. Self-care deficit, feeding 61. Self-care deficit, toileting 62. Self-esteem disturbance 63. Self-esteem, chronic low 64. Self-esteem, situational low 65. Sensory/perceptual alteration 66. Social interaction, impaired 67. Social isolation 68. Spiritual distress 69. Swallowing, impaired 70. Thought processes, altered 71. Tissue integrity, impaired 72. Tissue perfusion, altered 73. Trauma, high risk for 74. Unilateral neglect 75. Urinary elimination, altered 76. Urinary retention 77. Ventilation, inability to sustain spontaneous 78. Ventilatory, weaning process, dysfunctional
125
Hospital: ______________Unit: ______________Patient: _________________
OMAHA Problem Rating Scale

For each patient, complete at time of: < admission (A) < discharge (D) < new nursing diagnoses after admission (N) < resolution of nursing diagnoses during care (R) Hospital Day Nursing Diagnosis Time Knowledge Ability of the patient to remember and interpret information
1-No knowledge 2-Minimal knowledge 3-Basic knowledge 4-Adequate knowledge 5-Superior knowledge
Behaviour Observable responses, actions or activities of the patient fitting the occasion or purpose
1-Never appropriate 2-Rarely appropriate 3-Inconsistently appropriate 4-Usually appropriate 5-Consistently appropriate
Status Condition of the patient in relation to objective and subjective defining characteristics
1-Extreme signs/symptoms 2-Severe signs/symptoms 3-Moderate signs/symptoms 4-Minimal signs/symptoms 5-No signs/symptoms
enter: A D N R
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
5 5 5 5 5 5 5 5 5 5 5 5 5 5 5
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
5 5 5 5 5 5 5 5 5 5 5 5 5 5 5
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
5 5 5 5 5 5 5 5 5 5 5 5 5 5 5
126
Hospital Day
Nursing Diagnosis
Time
Knowledge Ability of the patient to remember and interpret information 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5
Behaviour Observable responses, actions or activities of the patient fitting the occasion or purpose 1 2 3 4 5 1 1 2 2 3 3 4 4 5 5
Status Condition of the patient in relation to objective and subjective defining characteristics 1 2 3 4 5 1 1 2 2 3 3 4 4 5 5
127
NANDA and OMAHA Summary Sheet

Date Unit NANDA Diagnoses Time OMAHA Ratings Code A D Knowledge Behaviour N R Status NRU Use Only
128
The SF-12 Health Survey

Your Health in General 1. In general, would you say your health is: Excellent {1 Very good {2 Good {3 Fair {4 Poor {5
2. The following questions are about activities you might do during a typical day. Does your health now limit you in these activities? If so, how much? Yes, limited a lot a) Moderate activities, such as moving a table, pushing a vacuum cleaner, bowling, or playing golf b) Climbing several flights of stairs {1 Yes, limited a little {2 No, not limited at all {3
{1
{2
{3
3. During the past week, have you had any of the following problems with your work or other regular daily activities as a result of your physical health? Yes a) Accomplished less than you would like b) Were limited in the kind of work or other activities {1 {1 No {2 {2
4. During the past week, have you had any of the following problems with your work or other regular daily activities as a result of any emotional problems (such as feeling depressed or anxious)? Yes a) Accomplished less than you would like b) Did work or other activities less carefully than usual {1 {1 No {2 {2
129
5. During the past week, how much did pain interfere with your normal work (including both work outside the home and housework)? Not at all {1 A little bit {2 Moderately {3 Quite a bit {4 Extremely {5
6. These questions are about how you feel and how things have been with you during the past week. For each question, please give the one answer that comes closest to the way you have been feeling. How much of the time during the past week... A good All of Most bit of Some A little None the of the the of the of the of the time time time time time time a) have you felt calm and peaceful? b) did you have a lot of energy? c) have you felt downhearted and blue? {1 {1 {1 {2 {2 {2 {3 {3 {3 {4 {4 {4 {5 {5 {5 {6 {6 {6
7. During the past week, how much of the time has your physical health or emotional problems interfered with your social activities (like visiting friends, relatives, etc.)? All of the time {1 Most of the time {2 Some of the time {3 A little of the time {4 None of the time {5
THANK YOU FOR COMPLETING THIS QUESTIONNAIRE!
130
SF-12 Health Survey (French Version)

QUESTIONNAIRE SUR L'TAT DE SANT - SF-12 DIRECTIVES: Les questions qui suivent portent sur votre sant, telle que vous la percevez. Vos rponses permettront de suivre l'volution de votre tat de sant et de savoir dans quelle mesure vous pouvez accomplir vos activits courantes. Veuillez rpondre toutes les questions en cochant une case. En cas de doute, rpondez de votre mieux. 1. En gnral, diriez-vous que votre sant est:
Excellente
Trs bonne
Bonne
Passable
Mauvaise
2. Les questions suivantes portent sur les activits que vous pourriez avoir faire au cours d'une journe normale. Votre tat de sant actuel vous limite-t-il dans ces activits? Si oui, dans quelle mesure? Mon tat Mon tat Mon tat de sant de sant de sant me limite me ne me beaucoup limite un limite pas peu du tout a) Dans les activits modres comme dplacer une table, passer l'aspirateur, jouer aux quilles ou au golf Pour monter plusieurs tages pied
b)
Copyright 1994 Health Assessment Lab. All rights reserved. (SF-12 French (Canadian) Standard Version 1.0)
131
3. Au cours de la dernire semaine, avez-vous eu l'une ou l'autre des difficults suivantes au travail ou dans vos autres activits quotidiennes cause de votre tat de sant physique? OUI a) b) Avez-vous accompli moins de choses que vous l'auriez voulu? Avez-vous t limit(e) dans la nature de vos tches ou de vos autres activits? NON
4. Au cours de la dernire semaine, avez-vous eu l'une ou l'autre des difficults suivantes au travail ou dans vos autres activits quotidiennes cause de l'tat de votre moral (comme le fait de vous sentir dprim(e) ou anxieux(se))? OUI a) b) Avez-vous accompli moins de choses que vous l'auriez voulu? Avez-vous fait votre travail ou vos autres activits avec moins de soin qu' l'habitude? Au cours de la dernire semaine, dans quelle mesure la douleur a-t-elle nui vos activits habituelles (au travail comme la maison)? NON
5.
Pas du tout
Un peu
Moyennement
Beaucoup
normment
132
6. Ces questions portent sur de la dernire semaine. Pour chacune des questions suivantes, donnez la rponse qui s'approche le plus de la faon dont vous vous tes senti(e). Au cours de la dernire semaine, combien de fois: La plupart Souvent Quelquefois du temps Jamais
Tout le temps a) Vous tes-vous senti(e) calme et serein(e)? b) Avez-vous eu beaucoup d'nergie? c) Vous tes-vous senti(e) triste et abattu(e)? 7.
Rarement
Au cours de la dernire semaine, combien de fois votre tat physique ou moral a-t-il nui vos activits sociales (comme visiter des amis, des parents, etc.)?
Tout le temps
La plupart du temps
Parfois
Rarement
Jamais
133
Nurse Survey
Please circle the number of the appropriate response to each question or, where indicated, fill in the blanks. A. A1 Questions About Your Work Life What is your current employment status at this hospital? Is your employment: Is this the unit you normally work on? What is your job title? Full time................................ Part time............................... Casual.................................... Permanent............................. Temporary............................. Yes......................................... No......................................... RN......................................... RPN....................................... Other (specify):___________ _____ years _____ months _____ years _____ months _____ years _____ months _____ years _____ months Yes......................................... No (specify):______________ 1 2 1 2 3 1 2 1 2 1 2 3
A2 A3 A4
A5
How many years have you worked: a) as an RN/RPN b) as an RN/RPN at your present hospital c) as an RN/RPN on your current unit d) as a casual worker Is your immediate supervisor a nurse? If No, please specify the profession of your supervisor. In the past year, how many hours per week did you work, on average: a) in this hospital for paid work? b) for any other paid work? In the past year, how many hours per week, on average, did you work on units other than the one to which you are usually assigned? (that is, the one where you work the most hours): In the past year, how many hours a week, on average, did you work in this hospital the following types of overtime? a) Voluntary Paid b) Voluntary Unpaid c) Involuntary Paid d) Involuntary Unpaid
A6 A7
_______ hours per week _______ hours per week
A8
_______ hours per week
A9
_______ hours per week _______ hours per week _______ hours per week _______ hours per week
134
A10
In the past year, has the amount of overtime required of you:
Increased................................ Remained the same............... Decreased.............................. Not applicable........................ None........................................ Once........................................ Twice....................................... Other (specify):____________ Yes......................................... No..........................................
1 2 3 4 1 2 3 4 1 2
A11
In the past 2 weeks, how often did you change shifts? (e.g., from days to evenings, evenings to nights, nights to days, etc.) In the past year, have you been required to permanently change nursing units due to restructuring/reorganization?
A12
A13
In the next year, do you anticipate having to permanently change nursing units due to restructuring/reorganization?
Yes..................................... No......................................
1 2
B. Questions About Your Job Satisfaction How satisfied are you with the following aspects of your current job?
Very Dissatisfied Very Satisfied
B1 B2 B3 B4 B5 B6 B7
Opportunities for social contact at work Opportunities for social contact with your colleagues after work Opportunities to interact with management/administration Your amount of responsibility
1 1 1 1
2 2 2 2 2 2
3 3 3 3 3 3
4 4 4 4 4 4
5 5 5 5 5 5 1 2 3 4 1 2 3 1 2 3 4
On the whole, how satisfied are you with your present job? 1 Independent of your present job, how satisfied are you with being 1 a nurse? Thinking about the next 12 months, how likely is it that you will lose your job?
Very Likely................................... Fairly Likely.................................. Not too likely................................ Not at all likely............................. Yes, within the next 6 months....... Yes, within the next year............... No plans within the year............... Very easy................................ Fairly easy............................... Fairly difficult........................... Very difficult...........................
B8
Do you plan to leave your present nursing job?
B9
If you were looking for another job, how easy or difficult do you think it would be for you to find an acceptable job in nursing?
135
C. C1 C2 C3
Questions About You What is your gender? What is your age? Do you have any dependent children/others living with you? a) children b) other dependents Female................................................... Male...................................................... _______ years 1 2
Yes, how many children_____________ No......................................................... Yes, how many other dependents______ No......................................................... RPN Diploma......................................... RN Diploma........................................... BScN.................................................... MScN.................................................... PhD Nursing.......................................... Post RN Certificate Cardiac.................... Post RN Certificate Other (specify): _______________________________ Diploma................................................. Baccalaureate........................................ Masters................................................. PhD....................................................... Other (specify):___________________ Not applicable........................................
1 2 1 2 1 2 3 4 5 6 7 1 2 3 4 5 6
C4
What is your highest Nursing educational credential?
C5
What is your highest Non-nursing educational credential?
C6
In the past year: a) On how many occasions (episodes) have you missed work due to illness/disability? b) How many shifts have been missed due to illness/disability? In the past year, what is the most common reason you missed work? (Choose one only)
_________ # occasions _________ # shifts Physical illness....................................... Mental health day................................... Injury (work related)............................... Family illness/crisis/ commitment............. Unable to get requested day off.............. Other (specify):___________________ 1 2 3 4 5 6
C7
136
C8
How often are you selected to be a preceptor for another nurse?
Never.................................................... Rarely................................................... Occasionally........................................... Frequently.............................................. Never.................................................... Rarely................................................... Occasionally........................................... Frequently.............................................. I am a nurse who... (circle only one response) 1) ...relies primarily on standards of care, unit procedures and physicians= and nurses= orders to guide patient care 2) ...has increased clinical understanding, technical and organizational skills and is able to anticipate the likely course of events 3) ...perceives the patient situation as a whole and responds appropriately as conditions change 4) ...is good at recognizing unexpected clinical responses and often provides an early warning of patient changes
1 2 3 4 1 2 3 4
C9
How often do nurses come to you for clinical judgment on a difficult clinical problem?
C10
The following descriptions are intended to represent levels of skill and ability in nursing roles and functions. Which one of the following would you say best describes the way in which you practice on your unit?
1 2
3 4
D.
Questions About Violence In the last 5 shifts you worked, have you experienced any of the following while carrying out your responsibilities as a nurse:
D1
a) Physical assault b) If yes, indicate source of physical assault
Yes..................................... No....................................... Source of physical assault: Patient................................. Family/visitor........................ Physician............................. Nursing co-worker............... Other, specify:___________
1 2 1 2 3 4 5
137
D2
a) Threat of assault b) If yes, indicate source of threat of assault
Yes..................................... No....................................... Source of threat of assault: Patient................................. Family/visitor........................ Physician............................. Nursing co-worker............... Other, specify:___________ Yes..................................... No....................................... Source of emotional abuse: Patient................................. Family/visitor........................ Physician............................. Nursing co-worker............... Other, specify:___________
1 2 1 2 3 4 5 1 2 1 2 3 4 5
D3
a) Emotional abuse b) If yes, indicate source of emotional abuse
E. E1
Questions About Your Perceptions of Quality of Care Overall, in the past year, would you say the quality of patient care in your unit has: How would you describe the quality of nursing care delivered on your last shift? Improved................ Remained the same. Deteriorated........... Excellent................ Good...................... Fair........................ Poor....................... Circle all that apply 1 2 3 4 5 6 7 8 1 2 3 1 2 3 4
E2
E3
Which of the following tasks did you perform during your last shift? 1) Delivering/retrieving trays 2) Ordering, coordinating or performing ancillary services (e.g., physical therapy, ordering labs) 3) Starting IVs 4) Arranging discharge referrals and arranging transportation (including nursing homes) 5) Performing ECGs 6) Routine phlebotomy (venipunctures) 7) Transporting patients (including to nursing homes) 8) Housekeeping duties (e.g., cleaning patient rooms)
138
E4
Which of the following situations occurred on your last shift due to time pressures? 1) Routine vital signs, medications or dressings not done 2) Routine vital signs, medications or dressings not on time 3) Routine mobilization or turns not done 4) Routine mobilization or turns not done on time 5) Delay in administering PRN pain medications 6) Delay in responding to patient bell Which of the following tasks were necessary but left undone during your last shift because you lacked the time to complete them? 1) Routine teaching for patients and families 2) Prepare patient and family for discharge 3) Comforting/talking with patients 4) Adequately documenting nursing care 5) Back rubs and skin care 6) Oral hygiene 7) Develop or update nursing care plan
Circle all that apply 1 2 3 4 5 6 Circle all that apply 1 2 3 4 5 6 7
E5
139
Hospital _________
Date ________________________
Hospital Profile
Interview Questions 1 2 3 What is the title of the nurse responsible for Operating Decisions? What is the title of the nurse responsible for Standards of Practice? If the hospital is structured by Programs, is there any central control of nurse staffing decisions? Who determines the volume and skill mix for nursing? Does your hospital have: a) Student nurses b) Student physicians c) Residents Where are inpatient cardiovascular and cardiology services provided in your hospital? List the names and designations of nursing units. How long do patients routinely stay in ICU? Does your hospital have a formal preoperative process prior to elective cardiovascular surgery? Yes.......................................................... No............................................................ If yes, please provide a description and indicate the proportion of annual cardiovascular patient surgical volume that attend________________________ Yes, please provide copies...................... No............................................................ Yes, provide copies of their role............. No............................................................ 2 Yes, please provide copies...................... No............................................................ 1 2 1 2 Circle all that apply: a b c Yes.......................................................... No............................................................ 1 2
4 5
7 8
9 10
Does your hospital have any policies for violence/abuse against nurses? Does your hospital have advanced practice nurses for cardiovascular and/or cardiology services?
1 2 1
11
Does your hospital have any policies for sheath removal?
140
Hospital _________
Date ________________________
Unit/Program Profile
1 Does your unit use Critical Pathways or Clinical Guidelines for common cardiovascular and cardiology diagnoses? Does your unit use standard nursing care plans for common cardiovascular and cardiology diagnoses? In your unit, do your policies and procedures for cardiovascular and cardiology patients allow nurses, during an emergency situation, to: a) Initiate an IV b) Defibrillate c) Initiate thrombolytic agents d) Initiate oxygen Is there a formal pre-op program for your patients? Yes, please provide copies................ No....................................................... Yes, please provide copies................ No....................................................... Circle all that apply: 1 2 1 2
a b c d Yes...................................................... No....................................................... Provide a description of program goals, length of program, the services provided and the criteria for inclusion Describe and provide an estimate of the percentage of patients that would be involved 1 2
What is the post discharge follow-up procedure for patients?
Do you have a nurse educator: a) For the unit: b) For the program:
a)Yes, indicate #FTEs___________ No....................................................... b)Yes, indicate #FTEs___________ No....................................................... Yes, please provide details............... No.......................................................
1 2 1 2 1 2
7 8
Do you have any outpatient activity on this unit? What are the average daily hours of housekeeping support for this unit?
141
Hospital _________
Date ________________________
Are there any current issues related to the physical environment that we should be aware of? Determining turnover rate: a) How many positions have been posted in the past year? ___________ b) How many positions have been budgeted for in the past year? ______
10
11
Determining vacancy rate:
a) How many days have been vacant in the past year? ______ b) How many budgeted FTEs in the past year? ______
12
Do you have access to the following allied health professionals and are they dedicated to the unit? a) Physiotherapists
Circle only per discipline: a) Access: Yes................................. No................................. Dedicated: Yes................................ No.................................. #FTEs dedicated to unit_________ b) Access: Yes................................. No................................. Dedicated: Yes................................. No................................. #FTEs dedicated to unit_________ c) Access: Yes................................ No................................. Dedicated: Yes................................. No................................. #FTEs dedicated to unit__________ d) Access: Yes.............................. No................................. Dedicated: Yes................................. No................................. #FTEs dedicated to unit__________ e) Access: Yes............................... No............................... Dedicated: Yes................................ No................................. #FTEs dedicated to unit__________ 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2
b) Occupational Therapists
c) Social Workers
d) Nutritionists/Dieticians
e) Other(s):__________________________
142
Daily Unit Staffing Form

Patient Census by Shift Shifts days: 0730-1530 evenings: 1530-1930 evenings: 1930-2330 nights: 2330-0730 other: Number of Staff Working on Unit Shifts days: 0730-1530 evenings: 1530-1930 evenings: 1930-2330 nights: 2330-0730 other: Number of Agency Staff Shifts days: 0730-1530 evenings: 1530-1930 evenings: 1930-2330 nights: 2330-0730 other: Number of Overtime Hours # of Agency Nurses # of Agency Non-Nurses (e.g., Sitters) # RNs FT PT # RPNs/RNAs FT PT Casual # other (UCPs) FT PT Casual # of Patients # Admissions or Transfers In # Discharges or Transfers Out
Casual
143
Hospital _________ Shifts days: 0730-1530 evenings: 1530-1930 evenings: 1930-2330 nights: 2330-0730 other: # RNs FT PT
Date ________________________ # RPNs/RNAs FT PT Casual # other (UCPs) FT PT Casual
Casual
Number of Staff Absent from Unit Due to Illness Shifts days: 0730-1530 evenings: 1530-1930 evenings: 1930-2330 nights: 2330-0730 other: # RNs FT PT # RPNs/RNAs FT PT Casual # other (UCPs) FT PT Casual
Casual
Number of Staff Absent from Unit Due to Reasons Other than Illness Shifts days: 0730-1530 evenings: 1530-1930 evenings: 1930-2330 nights: 2330-0730 other: # RNs FT PT #RPNs/RNAs # other (UCPs) Casual FT PT Casual FT PT Casual
144
Hospital _________ Number of Staff who Floated To Unit Shifts days: 0730-1530 evenings: 1530-1930 evenings: 1930-2330 nights: 2330-0730 other: # RNs FT PT
Date ________________________
#RPNs/RNAs # other (UCPs) Casual FT PT Casual FT PT Casual
Number of Staff who Floated From Unit Shifts days: 0730-1530 evenings: 1530-1930 evenings: 1930-2330 nights: 2330-0730 other: # RNs FT PT #RPNs/RNAs # other (UCPs) Casual FT PT Casual FT PT Casual
Number of Staff on Orientation (either formal or on-the-job training) Shifts days: 0730-1530 evenings: 1530-1930 evenings: 1930-2330 nights: 2330-0730 other: # RNs FT PT #RPNs/RNAs # other (UCPs) Casual FT PT Casual FT PT Casual
145
Hospital _________ Number of Patient Falls or Medication Errors Shifts days: 0730-1530 evenings: 1530-1930 evenings: 1930-2330 nights: 2330-0730 other: Patient Falls
Date ________________________
Medication Errors
Unit Workload Data Patient Care Workload: GRASP Patient Care Hours or Medicus Patient Type ________________________________________ Non-patient Care Workload: GRASP Non-patient Care Hours or Medicus Non-patient care workload ________________________
146
Hospital _________
Date ________________________
Number of Patients Assigned to RNs at Beginning of Day Shifts RN Code FT PT Casual
147
Hospital _________
Date ________________________
Daily Environmental Complexity Scale

For the following, please rate how each item influenced your ability to provide required care for patient(s) on this shift. Please reflect on the workload you anticipated prior to starting your shift and decide the nature of the influence of each item. Did the item increase anticipated workload, decrease anticipated workload or have no influence on anticipated workload? (i=increased work, d=decreased work, or s=same as usual/no change). Then, rate the extent of the items influence on a scale of 1 to 5 (1=low influence, 3=medium influence, 5=high influence). Circle your responses. For those items not applicable to this shift please leave blank.
Decreased workload 1 Increased workload Same as usual/ No change 2 3 4 5
Medium
Students: 1 2
Students on the unit today required supervision and assistance Students wanted access to charts, equipment & supplies
Low
i i
d d
s s
Staffing: 3
Scheduled unit staff absent this shift (includes UM, RNs, RNAs, LVNs, clerical and assistive staff)
Nursing Team Functioning: 4 5 6 7 8 9 10

Staff unable to pull together to complete unit work
i i i i i i i
d d d d d d d
s s s s s s s
Assignment:
Rushing to get work done 1 2 3 4 5
Unanticipated Communication with Doctors:

More than the usual calls to doctors this shift Clarifying doctors' orders 1 1 2 2 3 3 4 4 5 5
Unanticipated Delays:
Doctors not answering pages Multiple delays experienced on the unit Medication, supplies and narcotic keys missing 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5
148
High
Hospital _________ 11 12 13
Language barrier with family and/or patient
Date ________________________ i i i d d d s s s
1 2 3 4 5
Unexpected Change in Patient Condition:

Agitated, confused, or restless patient(s) Unanticipated increase in patient acuity 1 1 2 2 3 3 4 4 5
Unanticipated Time Consuming Interventions for Patient and Family: 14

Stat blood work
i
Increased workload
d
Decreased workload
s
Same as usual/ No change
Medium
15 16 17 18 19 20 21 22
Extra vital signs Extra charting and paperwork Greater demand for routine patient teaching Greater demand for psychosocial support for patient Greater demand for psychosocial support for family
Low
i i i i i i i i
d d d d d d d d
s s s s s s s s
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
Unanticipated and Time Consuming NonPatient Care Activities:

Completing work of others (e.g. dietary, clerical staff, housekeeping, nursing administration) Interruptions (e.g., called back to desk, phone) that influences time with patients and family Participating in nursing research 1 2 3 4 5
Please add and rate any other information regarding events that significantly influenced your ability to provide required care on this shift: 23 24 25
i i i d d d
s s s
1 1 1
2 2 2
3 3 3
4 4 4
Copyright L. OBrien-Pallas
149
High
5 5 5 5 5
5 5 5
Hospital _________
Date ________________________
Please rate the time available to deliver care on THIS SHIFT compared to the last five shifts you have worked. CIRCLE ONLY ONE RESPONSE: 26
Less time than usual About the same amount of time as usual More time than usual
Approximately how much more time do you feel you need to give the type of care stated in the nursing care plan or your assessment of patients needs to day? CIRCLE ONLY ONE RESPONSE: 27
No more time needed < 15 minutes 15-30 minutes 31-45 minutes 46-60 minutes >60 minutes
Copyright L. OBrien-Pallas
150
Hospital __________
Unit __________
Patient __________
PRN 80
151
Hospital __________
Unit __________
Patient __________
152
Hospital __________
Unit __________
Patient __________
PRN Daily Workload and Grasp Patient Care Hours

Date Unit Nurse Code Respiration Feeding Elimination Hygiene Communication Treatment Diagnostic GRASP PCH
153
Hospital __________
Unit __________
Patient __________
Patient Data Form

Please circle the number of the appropriate response to each question or, where indicated, fill in the blanks.
1 2 3 4 5 6 7 8
Date of admission Date of discharge/transfer/death Admission diagnosis Other concurrent diagnosis Sex Age Occupation Highest level of education:
dd/mm/yy _______________________ dd/mm/yy _______________________
Male....................................................... Female................................................... __________years Less than high school diploma............. High school diploma............................ Trade certificate/college/some university University degree................................. Yes............................................................. No............................................................... Yes............................................................ No.............................................................. _______ hours _______ hours Yes, number of hours in ICU___________ No............................................................... Yes............................................................. No...............................................................
1 2
1 2 3 4 1 2 1 2
9 10 11 12 13
Does the patient have a potential caregiver at home? Does patient have a family physician? Actual length of stay in ICU: Actual length of stay in hospital: Has the patient been transferred back to ICU? Did patient attend pre-operative clinic?
1 2 1 2
14
154
Hospital __________ 15 16 17
Unit __________
Patient __________ 1 2 1 2 1 2 1 2
Is patient booked for post-operative/postdischarge education? Did patient spend time in step-down unit? Did patient have any falls: a) Resulting in injury: b) Not resulting in injury:
Yes......................................................... No.......................................................... Yes, number of hours in SDU_______ No........................................................ a)Yes, specify injury_______________ No..................................................... b) Yes..................................................... No...................................................... Circle all that apply a b c d e f Yes......................................................... No.......................................................... Yes......................................................... No.......................................................... Yes......................................................... No.......................................................... Yes, with patient consequences, specify________________________ Yes, without patient consequences No medication errors
18
Did the patient develop any of the following: a) urinary tract infection b) pneumonia c) superficial incisional surgical site infection d) deep incisional surgical site infection e) bedsores f) thrombosis Did patient get a referral for home care? Was this a planned admission? Was the patient hospitalized for the same condition in the past 3 months? Were there any nurse medication errors with this patient?
19 20 21 22
1 2 1 2 1 2 1 2 3
155
Maslach=s Burnout Inventory

This section contains statements of JOB-RELATED FEELINGS. If you have never had this feeling, circle the A0@ (zero) after the statement. Otherwise, indicate how often you feel like this by circling the number (from1 to 6) that best describes how frequently you feel that way.
How Often?
Once a week Never A few times a year or less Once a month or less A few times a month A few times a week Every day
1 2 3 4 5
I feel emotionally drained from my work I feel used up at the end of the workday. I feel fatigued when I get up in the morning and have to face another day on the job. I can easily understand how my patients feel about things. I feel I treat some patients as if they were impersonal objects. Working with people all day is really a strain for me. I deal very effectively with the problems of my patients. I feel burned-out from my work. I feel I=m positively influencing other people=s lives. I=ve become more callous toward people since I took this job. I worry that this job is hardening me emotionally. I feel very energetic. I feel frustrated by my job. I feel I=m working too hard on my job. I don=t really care what happens to some patients. Working directly with people puts too much stress on me. I can easily create a relaxed atmosphere with my patients. I accomplish many worthwhile things in this job.
0 0 0
1 1 1
2 2 2
3 3 3
4 4 4
5 5 5
6 6 6
0 0
1 1
2 2
3 3
4 4
5 5
6 6
6 7 8 9 10 11 12 13 14 15 16 17 18
0 0 0 0 0
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
5 5 5 5 5
6 6 6 6 6
0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2
3 3 3 3 3 3 3 3
4 4 4 4 4 4 4 4
5 5 5 5 5 5 5 5
6 6 6 6 6 6 6 6
156
Never
A few times a year or less
Once a month or less
A few times a month
Once a week
A few times a week
Every day
19 20 21 22
I feel exhilarated after working closely with my patients. I feel like I=m at the end of my rope. In my work, I deal with emotional problems very calmly. I feel patients blame me for some of their problems.
0 0 0 0
1 1 1 1
2 2 2 2
3 3 3 3
4 4 4 4
5 5 5 5
6 6 6 6
157
Effort-Reward Imbalance
For each of the following statements, please indicate first whether you agree or disagree with it. If there is an arrow behind your answer please also indicate how much you are generally distressed by this situation. Thank you for answering all statements. I am very distressed I am distressed I am somewhat distressed I am not at all distressed
1. I have constant time pressure due to a heavy work load. 2. I have many interruptions and disturbances in my job. 3. I have a lot of responsibility in my job. 4. I am often pressured to work overtime 5. My job is physically demanding. 6. Over the past few years, my job has become more and more demanding. 7. I receive the respect I deserve from my superiors. 8. I receive the respect I deserve from my colleagues. 9. I experience adequate support in difficult situations. 10. I am treated unfairly at work. 11. My job promotion prospects are poor. 12. I have experienced or I expect to experience an undesirable change in my work situation. 13. My job security is poor. 14. My current occupational position adequately reflects my education and training. 15. Considering all my efforts and achievements, I receive the respect and prestige I deserve at work. 16. Considering all my efforts and achievements, my work prospects are adequate. 17. Considering all my efforts and achievements, my salary /income is adequate. disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree disagree agree
4 3 2 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
158
Nursing Work Index

For each item in this section, please indicate the extent to which you agree that the following items ARE PRESENT IN YOUR CURRENT JOB. Indicate your degree of agreement by circling the appropriate number.
The following are present in your current job . . . 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Adequate support services allow me to spend time with my patients. Physicians and nurses have good working relationships. A good orientation program for newly employed nurses. A supervisory staff that is supportive of the nurses. A satisfactory salary. Nursing controls its own practice. Active staff development or continuing education programs for nurses. Career development/clinical ladder opportunity. Opportunity for staff nurses to participate in policy decisions. Support for new and innovative ideas about patient care. Enough time and opportunity to discuss patient care problems with other nurses. Enough registered nurses on staff to provide quality patient care. A nurse manager or immediate supervisor who is a good manager and leader. A senior nursing administrator who is highly visible and accessible to staff. Flexible or modified work schedules are available. Enough staff to get work done. Freedom to make important patient care and work decisions. Praise and recognition for a job well done. Strongly Agree 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Somewhat Agree 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 Somewhat Disagree 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 Strongly Disagree 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
159
19 20 21 22 23 24 25 26 27 28 29 30 31 32
The opportunity for staff nurses to consult with clinical nurse specialists or expert nurse clinicians/educators. Good working relationships with other hospital departments or programs. Not being placed in a position of having to do things that are against my nursing judgement. High standards of nursing care are expected by the administration. A senior nursing administrator equal in power and authority to other top level hospital executives. A lot of team work between nurses and physicians. Physicians give high quality medical care. Opportunities for advancement. Nursing staff are supported in pursuing degrees in nursing. A clear philosophy of nursing that pervades the patient care environment. Nurses actively participate in efforts to control costs. Working with nurses who are clinically competent. The nursing staff participates in selecting new equipment. A nurse manager or supervisor who backs up the nursing staff in decision-making, even if the conflict is with a physician. Administration that listens and responds to employee concerns. An active quality assurance program. Staff nurses are involved in the internal governance of the hospital (e.g., practice and policy committees). Collaboration between nurses and physicians. A preceptor program for newly hired RNs. Nursing care is based on a nursing rather than a medical model. Staff nurses have the opportunity to serve on hospital and nursing committees.
1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2 2 2 2 2 2
3 3 3 3 3 3 3 3 3 3 3 3 3 3
4 4 4 4 4 4 4 4 4 4 4 4 4 4
33 34 35 36 37 38 39
1 1 1 1 1 1 1
2 2 2 2 2 2 2
3 3 3 3 3 3 3
4 4 4 4 4 4 4
160
40 41 42 43 44 45
The contributions that nurses make to patient care are publicly acknowledged. Nurse managers consult with staff on daily problems and procedures. A work environment that is pleasant, attractive, and comfortable. Opportunity to work on a highly specialized patient care unit. Written up-to-date nursing care plans for all patients. Patient care assignments that foster continuity of care (i.e., the same nurse cares for the patient from one day to the next). Staff nurses do not have to float from their designated unit. Staff nurse actively participate in developing their own working schedule (i.e., what days they work, days off, etc.). Each patient care unit determines its own policies and procedures. Working with experienced nurses who Aknow@ the hospital nurses.
1 1 1 1 1 1
2 2 2 2 2 2
3 3 3 3 3 3
4 4 4 4 4 4
46 47
1 1
2 2
3 3
4 4
48 49
1 1
2 2
3 3
4 4
161
Appendix F. Methods
Methods
Methods: Multilevel Modeling and MLwin ............................................................................... 163 Multilevel structures ............................................................................................................... 163 Variables ................................................................................................................................. 165 MLwin ..................................................................................................................................... 167
List of Tables Table 1: Unit Characteristics Aggregated from Individual Nurse Level.................................... 164 Table 2: Dichotomy of Outcome Variables ................................................................................ 165
162
Methods: Multilevel Modeling and MLwin

Health science research often concerns problems that have a hierarchical structure, for example, when patients are nested within units, and units are nested within hospitals. In multilevel analysis, data structure of this nature is viewed as a multistage sample from a hierarchical population. By focusing on the level of hierarchy in the population, multilevel modeling enables researchers to understand where and how effects are occurring. It provides better estimates in answering simple questions for which single-level analyses were once used and in addition, allows more complex questions to be addressed. For example, how the effect of unit characteristics might affect the patients outcome? Or more specifically, patients spent more time in a unit with more full-time nurses would more likely result in a sooner discharge than in other units. Failing to recognize the existence of clustering will generally imply standard errors of regression coefficients for higher level variables to be underestimated, which also leads to a higher probability of type I error. For example, if standard errors were underestimated, it might be inferred that there was a real difference between a teaching hospital and a non-teaching hospital, which in fact could not be claimed due to the lack in number of hospitals included in the study. Multilevel structures Three sets of outcome variables, patient, nurse and system outcomes, were modeled in this study. In order to model patient and nurse outcomes, a three-level data structure was originally considered. Level 1 is the individual level, i.e., patient level for patient outcome or nurse level for nurse outcomes. Level 2 is unit and level 3 is hospital. Since there were a small number of hospitals (only 6) recruited in the study, any difference among hospitals would not be detected due to the limited sample size at hospital level. Thus, the hospital level was removed, and only unit level, individual patient and nurse levels were modeled. The removal of hospital level meant that the effect of the hospital level variables was not investigated on patient, nurse, and system outcomes, such as teaching hospital status, hospital size, etc. To study the effect of nurses characteristics and measures on patient outcomes and because one or more nurses may have cared for a patient during their hospital stay, information from all those nurses was aggregated to patient level as well as attached to each patient as patient level variables. Similarly, for those nurses who may have cared for multiple patients, information from those patients, during the study period, was aggregated as well as attached to each nurse as nurse level variables. Note that averaging or proportion was used in the aggregation of variables. Missing values were imputed using either regression imputation, cell mean imputation, or mean of nearby points (for daily data). Since patients have changed their units during their hospital stay, the proportion of days in the unit out of their total length of stay was assigned as weights to each unit they ever stay. For nurse outcomes, the unit indicated in the nurse survey form was used in the analyses. In total, there are
163
24 units at unit level for either patient, nurse or system outcomes. There are 1198 patients in patient outcome models and there are 727 nurses for most of the nurse outcome models. For those nurse outcomes where patient characteristics have immediate impact on, only 555 direct care nurses, during the study period, were included in the models. To answer part of research question 1 and 3, daily productivity/utilization measurements were modeled in a multilevel framework as well. Level 1 is date and Level 2 is the unit. Again, hospital was not included as a level as it was excluded for other outcome models. The nurse and patient characteristics and measurements were aggregated by date and unit such that the impact of patient and nurse variables on the productivity/utilization at unit level can be studied. As a measurement of unit atmosphere or morale at unit, some of individual nurse measurements were aggregated to unit level as unit measurements. Most of them were proportions, such as proportion of full time nurses, proportion of satisfied nurses in the unit. Others were based on average within the unit, for example, nurse age on average. The following variables at unit level, Table 1, have been constructed and considered in models, though not necessary included in the final models. Table 1: Unit Characteristics Aggregated from Individual Nurse Level
Unit Characteristics Aggregated from Individual Nurse Level Average Age of Nurses Proportion of Nurses with BScN or Above Proportion of Nurses Work On Multiple Units Proportion of Full-time Employment Average Overtime Hours Average Clinical Expertise Proportion of Nurses Reporting Job Instability Proportion of Nurses Reporting Shift Changes Prevalence of Violence at Unit Proportion of Nurses with Interventions Not Done Proportion of Nurses with Interventions Delayed Proportion of Nurses with Risk at Effort and Reward Imbalance Proportion of Emotionally Exhausted Nurses Average Nurse Autonomy Average Nurse-physician Relationship Proportion of Nurses Reporting Sick Leave Proportion of Nurses Intending to Leave Current Job Proportion of Physically Healthy Nurses Proportion of Mentally Healthy Nurses Proportion of Satisfied Nurses Proportion of Nurses Rating Good Patient Care Quality Proportion of Nurses Rating Good Nursing Care Quality
164
Variables Most of outcome variables were dichotomized and multilevel logistic regressions were used to model the effects of predictors. How each variable was constructed or dichotomized is shown in Table 2. Only productivity/utilization, averaged nursing hours, cost per RIW, autonomy and nurse-physical relationship were treated as continuous variables. Logarithm transformation was applied to averaged nursing hours and cost per RIW to assume the normality. Length of stay was also logarithm transformed as a control variable when nursing hours and cost were modeled. Only those measurements to answer research questions and predictors relevant conceptually or theoretically to the outcome variables were included in the models. Other predictors will be included in the final models if they are significantly associated with outcome variables. Table 2: Dichotomy of Outcome Variables
Predictor and Dependent Variable Patient variables Medical consequences Length of stay Patients physical health Patients mental health Omaha knowledge Omaha behaviour Omaha status Actual worked hours per patient Cost per RIW Nurse variables Education Work on Multiple Units Clinical Expertise Unit Instability Measurement
Yes to any of the following: fall with injury, medication errors, death, or complications such as UTI, pneumonia, superficial surgical site infection, deep surgical site infection, bedsores, and thrombosis; dichotomized as yes vs. no. Measured by the difference score between length of stay from medical record and expected length of stay from CIHI inpatient data for Ontario; dichotomized as shorter than expected length of stay vs. others Measured with SF-12 scale at admission and discharge; Improved at discharge vs. others Measured with SF-12 scale at admission and discharge; Improved at discharge vs. others Increased at discharge or diagnosis resolved vs. others Improved at discharge or diagnosis resolved vs. others Improved at discharge or diagnosis resolved vs. others Total nursing hours divided by midnight census RIW*actual cost per equivalent weighted case Highest nursing educational credential; dichotomized as BScN or above vs. diploma Work on more than one units vs. one Average scores on 4-point scale on being a preceptor for another nurse, providing clinical advice, level on expertise Reporting any of the following: Forced to change unit in past year, anticipate forced change of units in next year or expect to lose job within the next year; dichotomized as yes vs. no Reporting more than one shift change in the past 2 weeks vs. none Reporting any of the following: physical assault, threat assault, or emotional abuse; dichotomized as yes vs. no.
Shift Change Prevalence of Violence
165
Predictor and Dependent Variable Interventions Not Done
Measurement Reporting interventions not done on the last shift for the following interventions: vital sign/medications/dressings, mobilization/turns, patient/family teaching, discharge prep, comforting/talking with patients, documenting nursing care, back rubs/skin care, oral hygiene, or care plan; dichotomized as one or more interventions not done vs. none Reporting interventions delayed on the last shift for the following: vital signs/medications/dressings, mobilization/turns, response to patient bell, or PRN pain medication; dichotomized as one or more interventions delayed vs. none Dichotomized as at risk of effort and reward imbalance (> 1) vs. not at risk ( 1) Sum score of nine 7-point scale item; dichotomized as at risk (score > 27) vs. not at risk (score 27) Sum score of six autonomy items from NWI; the higher the score, the more autonomy nurses feel about work. Sum score of three nurse-physician relationship items from NWI; the higher the score, the more positive nurses feel about the nurse-physician relationship. Number of occasions missing work due to illness and disability; dichotomized as one or more sick leaves vs. none Plan to leave within the next year vs. no Physical health, measured with SF-12 Mental health, measured with SF-12 Average score of 5-point scale on social contact at work, social contact after work, opportunities to interact with management, amount of responsibility, satisfaction with present Job, and satisfaction with a being a nurse; dichotomized as satisfied/very satisfied vs. dissatisfied/very dissatisfied. Quality of patient care in the unit in the past year; dichotomized as improvement vs. others Quality of nursing care in the last shift; dichotomized as excellent/good vs. fair/poor
Interventions Delayed Effort and Reward Imbalance Emotional Exhaustion Autonomy Nurse-Physician Relationship Absenteeism Intent to Leave Physical Health Mental Health Satisfaction with Current Job Improved Quality of Patient Care Good Quality of Nursing Care
Environmental Complexity Scale Re-sequencing of Work Re-sequencing of work in response to others Unanticipated Changes in Unanticipated changes in patient condition, unanticipated time consuming Patient Acuity interventions for patient and family, and so on. Composition & Supervision and assistance of student nurses Characteristics of Care Team More Time Needed Amount of more time needed to give the type of care stated in the nursing care plan Unit variables Unit Occupancy Proportion of RN worked hours Average worked hours Productivity/utilization
Measured by midnight census divided by beds on unit Proportion of nursing hours contributed by RNs in the unit Average worked hours provided to patients on unit Unit workload divided by total worked hours on unit
166
MLwin MLwin beta version 2.0 was used to analyze the data. In MLwin, the hierarchical structure of the data is identified by variables which label the units at each level1. These are known as level or unit identifiers and must be declared when a model is being set up in the equations window or estimate tables. The data were sorted according to the data hierarchy to ensure MLwin functioned properly. The order of entry of variables was consistent with the theoretical framework at two levels. The level 1 variables were first entered and tested, then moved to the second level. RIGLS/IGLS estimation was used to generate coefficients and their standard errors. In the case of estimation failure from RIGLS/IGLS estimation, MCMC methods were used to continue the estimation. The -2 log likelihood value was used to make comparisons among different models. The test of significance for individual variables was conducted by using the intervals and tests facility in MLwin. Reference 1. Rasbash, J., Browne, W. Goldstein, H., et al. (2000). A users guide to MLWIN. Multilevel Models Project [Computer software and manual], Institute of Education, University of London.
167
Appendix G. Descriptive Analyses

List of Tables .............................................................................................................................. 169 Descriptive Analyses .................................................................................................................. 171
1. System Characteristics ..................................................................................................................... 171 2. Patient Characteristics ..................................................................................................................... 173
2.1 Patient Demographics ...............................................................................................................................173 2.2 Medical Diagnoses....................................................................................................................................176 2.3 Nursing Diagnoses and OMAHA Scores at Admission............................................................................177 2.4 Health Status at Admission .......................................................................................................................178
3. Nurse Characteristics ....................................................................................................................... 180

3.1 Nurse Demographics.................................................................................................................................180 3.2 Professional and Employment Status........................................................................................................181 3.3 Education and Clinical Expertise ..............................................................................................................181 3.4 Experience ................................................................................................................................................182
4. System Behaviours............................................................................................................................ 184

4.1 Workload ..................................................................................................................................................184 4.2 Workload Variation by Patient Medical Diagnosis...................................................................................186 4.3 Overtime and Continuity of Care/Shift Change and Unit Instability ........................................................188 4.4 Non-Nursing Tasks ...................................................................................................................................190
5. Intermediate System Outputs ............................................................................................................ 190

5.1 Worked Hours...........................................................................................................................................190 5.2 Productivity/Utilization.............................................................................................................................193
6. Environmental Complexity ............................................................................................................... 195 7. Patient Outcomes.............................................................................................................................. 197

7.1 Medical Consequences..............................................................................................................................197 7.2 OMAHA Scores at Discharge and Change from Admission ....................................................................198 7.3 Health Status at Discharge and Change from Admission .........................................................................198
8. Nurse Outcomes................................................................................................................................ 199

8.1 Burnout and Effort & Reward Imbalance .................................................................................................199 8.2 Autonomy and Control .............................................................................................................................201 8.3 Job Satisfaction .........................................................................................................................................201 8.4 Health Status .............................................................................................................................................202 8.5 Violence at Work ......................................................................................................................................203
9. System Outcomes .............................................................................................................................. 204

9.1 Quality of Care..........................................................................................................................................204 9.2 Absenteeism..............................................................................................................................................205 9.3 Intent to Leave ..........................................................................................................................................206
References ................................................................................................................................................. 207
168
List of Tables
Table 1: Hospital Characteristics............................................................................................... 171 Table 2: Unit Characteristics ..................................................................................................... 172 Table 3: Aspects of Care Process Percent of Patients Reporting Yes to Items in the Table, by Hospital..................................................................................................................... 173 Table 4: Percent of Surgical Patients who Attended Pre-op Clinic and Post-op Education, by Hospital..................................................................................................................... 173 Table 5: Patient Demographics, by Hospital ............................................................................. 174 Table 6: Patient Age, by Gender and Hospital........................................................................... 174 Table 7: Percent Distribution of Patient Occupation, by Category, by Hospital....................... 175 Table 8: Patient Employment Status Percent Distribution by Hospital .................................. 175 Table 9: Patient Educational Status Percent Distribution by Hospital................................... 175 Table 10: Percent Distribution of the Number of CMGs, by Hospital ....................................... 176 Table 11: Mean of Number of Nursing Diagnoses, by Unit Type .............................................. 177 Table 12: OMAHA Scores at Time 1 (Admission and Appearance of New Diagnosis), by Hospital ................................................................................................................................... 178 Table 13: Patient Health Status at Admission, by Hospital........................................................ 179 Table 14: Patient Health Status at Admission, Percent Less than US Norm, by Hospital......... 179 Table 15: Patient Health Status at Discharge, by Hospital........................................................ 180 Table 16: Patient Health Status at Discharge, Percent Less than US Norm, by Hospital......... 180 Table 17: Nurse Demographics, by Hospital ............................................................................. 181 Table 18: Nurse Employment Status, by Hospital ...................................................................... 181 Table 19: Nurse Education and Expertise, by Hospital ............................................................. 182 Table 20: Nurse Experience, N and Percent of Total Respondents, by Hospital ....................... 182 Table 21: Mean (SD) of PRN Workload (in Minutes) by Category, by Hospital ....................... 185 Table 22: PRN Workload Category as Percent of Total PRN Workload, by Hospital .............. 185 Table 23: Comparison of PRN to GRASP/Medicus Workload (in Hours), by Hospital ............ 186 Table 24: Percent Distribution of Work (in Minutes) by Workload Category, by CMG Type... 187 Table 25: Percent of Nurses Reporting Overtime in Average Hours per Week, by Hospital..... 188 Table 26: Percent Change of Nurse Overtime Hours in the Past Year, by Unit Type ............... 188 Table 27: Percent of Overtime Unpaid or Involuntary, if Working Overtime, by Hospital ....... 189 Table 28: Continuity of Care and Amount of Change, by Hospital ........................................... 189 Table 29: Percent of Nurses Reporting Performing Non-Nursing Tasks for Items in the Table, by Hospital..................................................................................................................... 190 Table 30: Actual Staffing Hours, by Unit, by Day...................................................................... 191 Table 31: Percent of Actual Staffing, by Unit, by Day ............................................................... 192 Table 32: Daily Patient Census, Admissions, and Discharges, by Unit..................................... 193 Table 33: Number of Days When Unit GRASP/Medicus is Greater than 85% and 93% of Total Nurse Hours, by Unit ................................................................................................ 194 Table 34: Percent of Nurses Reporting Average Hours Worked Per Week in the Past Year, by Hospital..................................................................................................................... 195 Table 35: Mean of Three Subscales from ECS, by Hospital....................................................... 196 Table 36: Percent of Nurses Reporting Additional Time Needed to Provide Quality of Care, by Hospital Unit............................................................................................................. 196
169
Table 37: Medical Consequences Percent Reporting Yes to the Items in the Table, by Hospital ................................................................................................................................... 197 Table 38: OMAHA Scores at Time 2 (Resolution of Diagnosis or at Discharge), by Hospital . 198 Table 39: Differences in OMAHA Scores Between Time 1 and Time 2, by Hospital................. 198 Table 40: Change in Patient Physical Health Status (SF-12) from Admission to Discharge .... 199 Table 41: Change in Patient Mental Health Status (SF-12) from Admission to Discharge....... 199 Table 42: Burnout Mean Scores of MBI Subscales, by Hospital............................................. 200 Table 43: Burnout Percent of Nurses at Risk for Emotional Exhaustion and ERI, by Hospital ................................................................................................................................... 200 Table 44: Nurse Work Index Subscales, by Hospital.................................................................. 201 Table 45: Job Satisfaction Percent of Nurses Dissatisfied, by Hospital ................................. 202 Table 46: Nurse Health Status, by Hospital ............................................................................... 202 Table 47: Nurse Health Status, Percent of SF-12 Scores Less than US Norm for Females, by Hospital..................................................................................................................... 203 Table 48: Prevalence of Violence Percent of Nurses Reporting Yes to the Items in the Table, by Hospital..................................................................................................................... 203 Table 49: Source of Emotional Abuse, by Hospital.................................................................... 204 Table 50: Quality Issues Percent of Nurses Reporting Yes to Items in the Table, by Hospital205 Table 51: Absenteeism Percent of Episodes Absent and Mean Shifts per Episode in the Past Year, by Hospital ...................................................................................................... 206 Table 52: Absenteeism Most Common Reason to Miss Work in the Past Year, by Hospital .. 206 Table 53: Intent to Leave Percent of Nurses Reporting Yes to the Items in the Table, by Hospital..................................................................................................................... 206
170
Descriptive Analyses
The results of descriptive analyses are presented at hospital or unit level. Hospital names are suppressed to ensure confidentiality. All comparisons between hospitals and units are merely crude rate comparisons that do not take into account differences in characteristics of patients, nurses, or organizations.
1. System Characteristics
Tables 1 and 2 outline the profiles of six hospitals and 24 nursing cardiac and cardiovascular units. Six characteristics describe the hospitals. The total number of inpatient beds denotes the overall size of individual hospitals. Hospital 6 had the largest number of beds whereas Hospital 5 had the smallest. Four of the six hospitals were teaching hospitals. The survey period varied at each site because the volume of eligible patients in each hospital influenced the number of study days. Each hospital had a target of 200 patients. Due to staffing problems, Hospitals 1 - 5 agreed to extend their data collection period to ensure that a sufficient number of patients were included in the analysis. Hospital 6, however, was not able to participate fully and thus had fewer patients completing the survey form and finished the study in a much shorter period of time than the other hospitals. The ability to capture patient level data was limited in some organizations due to the length of time required each day to collect staff data from non computerized systems. Table 1: Hospital Characteristics Hospital Number of Beds Teaching Survey Patients Number of Study Days Patient Midnight Census Number of Units 1 567 N 189 136 14 3 2 778 Y 243 121 23 3 3 507 N 259 114 33 2 4 777 Y 195 184 19 5 5 121 Y 285 136 18 6 6 1060 Y 59 64 13 5 Total 3243 n/a 1230 755 19 24
Note: Y=Yes, teaching hospital; N=No, non-teaching hospital; n/a=Not applicable Number of Beds=Total inpatient beds Surveyed Patients=Number of patients who completed Patient Data Form Patient Midnight Census=Average number of patients in surveyed units
Hospitals in the sample provided cardiac and cardiology nursing care using a variety of organizational structures. Surgical patients generally received a portion of their care in a critical care unit (CCU) but pre- and post-operative care was provided on an inpatient (IP) unit. In some organizations (hospitals 1 and 6), step-down units (SDU) were used in addition to the CCU. Some patients also received care in a CCU or SDU but many patients did not use critical care services. The structure and organization of health delivery can affect patient, nurse, and system outcomes. For example, attendance at pre admission or post operative education may have an effect on the resources required during the hospital stay and on the overall length of stay.
171
Table 2: Unit Characteristics

Hospital 1 2 3 4 Unit Name Coronary care unit Step-down unit Cardiac post-surgery/pre-op Intensive care unit Coronary care unit Combined Cardiology/Cardiovascular Critical care area Combined Cardiology/Cardiovascular 6 Eaton South + 6 NU Cardiovascular intensive care units (2) Interventional short stay unit Inpatient cardiology Interventional cardiology unit Cardiac surgical unit/Recovery Coronary Care unit Cardiovascular Cardiology/surgical Cardiology Coronary Care units (3) Cardiology Cardiology Step-down
Unit Type Pure Cardiology* Number of Beds Number of Study Patients
CCU SDU IP CCU CCU IP CCU IP IP CCU DS IP DS CCU CCU IP IP IP CCU IP SDU
Y Y Y N Y N N Y Y Y Y Y Y Y Y Y N Y Y Y Y
12 18 27 19 11 40 15 48 56 19 29 26 17 16 6 28 27 33 38 32 32
39 48 102 7 134 102 36 223 19 40 66 70 35 9 2 38 79 122 51 7 1
Note: CCU = Critical Care Unit DS = Day Surgery IP = Inpatient SDU = Step-Down Unit *Pure cardiology = units that provide care exclusively for cardiac and cardiovascular patients as opposed to patients with other medical or surgical conditions.
Tables 3 and 4 show various aspects of the care process in planned admission, pre-op and postop clinics, referrals to home care, time in SDU, and transfer to ICU. On average, almost half of survey patients reported that their admission was a planned readmission. More than one-fifth (22%) of the patients attended a pre-op clinic and more than half (53%) had post-admission education. About one in ten patients (10.9%) were referred to home care. There were 11.3% of patients who spent time in a SDU. Only 2% of the patients were transferred back to ICU. Hospital 5 had the largest proportion of patients with planned admission (65.4%), which was almost six times that of Hospital 6 (11.9%). Hospital 1 had more surgical patients attending preoperation clinics than all other hospitals, while Hospital 3 provided post-admission education for more cardiac and cardiovascular patients than any of the other hospitals. Hospital 2 referred 37.7% of patients to home care which was higher than other hospitals (4.3-7.7%). Hospitals 1 and 6 had a relatively larger proportion of patients spending time in SDU. Few if any patients were transferred back to ICU in Hospitals 2, 5, and 6.
172
Table 3: Aspects of Care Process Percent of Patients Reporting Yes to Items in the Table, by Hospital
Hospital Number of cases Planned admission Referred to home care Spent time in SDU Transferred back to ICU 1 189 35.4 4.3 28.0 3.7 2 243 32.2 37.7 6.2 0.4 3 259 38.2 6.5 7.6 3.1 4 195 56.3 6.6 10.0 3.6 5 285 65.4 7.1 0.5 0.7 6 59 11.9 7.7 28.3 0.0 Total 1230 44.5 10.9 11.3 2.0
Note: Due to missing values in each category the denominators to generate percentages are slightly different from N.
Table 4: Percent of Surgical Patients who Attended Pre-op Clinic and Post-op Education, by Hospital
Hospital Number of cases Attended pre-op clinic Post-op education 1 66 59.1 65.2 2 69 13.0 10.1 3 82 31.7 72.0 4 78 48.7 70.5 5 129 20.9 63.6 6 16 37.5 43.8 Total 440 33.0 57.5
Note: Due to missing values in each category the denominators to generate percentages are slightly different from N.
2. Patient Characteristics
Patient characteristics were captured from a variety of data sources. 1) Patients provided information about themselves and their care process in a survey. 2) Each hospitals Health Records Department provided health records data that included medical diagnosis at discharge, resource intensity weight, length of stay, admission type, etc. 3) Patients completed a SF-12 Health Survey indicating their functional status at the time of admission and discharge. 4) Data collectors collected nursing diagnoses (NANDA) and ratings of patient OMAHA knowledge, behaviour, and status concerning each nursing diagnosis from the chart, Kardex, and in consultation with the nurse. In total, 1,230 patients were entered into the study. 2.1 Patient Demographics As shown in Tables 5 and 6, the average age of patients was 63.5 years and two-thirds were male (66.7%). Hospital 2 had the largest proportion of females, and Hospitals 1 and 6 had female proportions well below the average. Hospitals 2 and 3 had high proportions (40.2 and 40.0% respectively) of patients over the age of 70, whereas Hospital 4 has the highest proportion (24.6%) of patients under the age of 50. Patients at Hospitals 2 and 3 were less likely to have a caregiver at home. This may be explained by the higher average age of patients at these sites. On average, over 95% of patients had a family doctor. Evidence-based Staffing 173
Table 5: Patient Demographics, by Hospital

Hospital Number of patients % Male % Female % Age >=70 % Age < 50 % Caregiver at home % Family Physician 1 189 74.1 25.9 30.7 14.8 87.3 94.7 2 243 60.1 39.9 40.2 11.9 79.1 95.9 3 259 63.3 36.7 40.0 13.8 75.9 93.1 4 195 70.8 29.2 29.2 24.6 85.1 93.8 5 285 66.7 33.7 37.0 14.2 84.2 98.6 6 59 74.6 25.4 33.9 11.9 83.1 91.5 Total 1230 66.7 33.3 35.9 15.3 82.0 95.2
Note: The denominators used to generate percentages for each demographic may be slightly different from the number of patients presented in the table.
Table 6: Patient Age, by Gender and Hospital

Hosp 1 2 3 4 5 6 Total N 140 146 164 138 189 44 821 Male Mean 62.0 63.9 63.4 60.1 62.5 61.5 62.4 SD 11.64 11.71 12.66 13.68 12.58 11.24 12.44 N 49 97 95 57 96 15 409 Female Mean 64.8 68.1 65.8 59.0 66.6 70.1 65.6 SD 13.30 13.37 11.81 16.51 14.08 11.70 13.85 N 189 243 259 195 285 59 1230 Total Mean 62.8 65.6 64.2 59.8 63.9 63.7 63.5 SD 12.12 12.54 12.39 14.53 13.22 11.86 13.01
Patient occupation was originally collected as an open-ended question. Occupations were subsequently classified into 14 categories according to work environments, knowledge, skill, and level of control1 As seen in Table 7, the occupational distribution of patients varied greatly by hospital. Hospitals 4, 5 and 6 had higher proportions of patients reporting a professional occupation, whereas Hospital 1 had mostly service, outdoor physical, and professional occupations. Hospital 3 had a large proportion of housewives and patients with outdoor physical occupations. In contrast, Hospital 2 had an overall even distribution across all occupation categories.
174
Table 7: Percent Distribution of Patient Occupation, by Category, by Hospital

Hospital Number of Cases Self-employed Professional Outdoor Physical* Healthcare Provider Government Housewife Clerical Management Retail/Sales Factory Trades Business Service Not working 1 176 5.7 10.8 10.8 5.1 2.8 8.0 0.6 8.5 3.4 9.1 5.1 4.5 23.3 2.3 2 152 0.7 8.6 16.4 3.9 3.9 8.6 6.6 6.6 5.3 7.9 5.3 3.9 11.2 11.2 3 220 6.4 8.6 15.0 1.8 2.3 18.6 0.9 5.9 5.5 6.4 3.2 6.4 12.7 6.4 4 153 5.2 26.8 3.9 2.6 0.7 7.8 5.2 5.2 3.3 8.5 3.3 3.9 16.3 7.2 5 276 4.7 17.8 4.7 6.5 7.2 5.8 6.2 13.4 4.3 7.2 1.8 6.5 12.3 1.4 6 57 3.5 19.3 1.8 3.5 0.0 8.8 3.5 10.5 7.0 14.0 12.3 3.5 10.5 1.8 Total 1034 4.6 14.7 9.4 4.2 3.6 9.8 3.9 8.6 4.5 8.0 4.0 5.2 14.6 4.9
*Mostly farmers for Hospital 2 and miners for Hospital 3 in the outdoor physical occupation.
More than 60% of patients from Hospitals 1 and 5 were employed, but merely one-third of patients in Hospitals 2 and 3 were working at the time of the survey. Hospital 3 also had the largest proportion (20.0%) of patients not employed (Table 8). The not-employed group consists of housewives, disabled persons, and students. Hospital 2 had primarily retired patients (53.1%) as patients in Hospital 2 were much older than patients in other hospitals. In contrast, less than one-third of patients in Hospitals 1 and 5 fell into the retired group. Table 8: Patient Employment Status Percent Distribution by Hospital
Hospital Employed Not employed Retired Number of cases 1 60.8 9.5 29.6 189 2 34.4 12.5 53.1 224 3 34.9 20.0 45.1 255 4 47.2 11.4 41.5 193 5 65.8 6.1 28.1 278 6 52.6 7.0 40.4 57 Total 48.9 11.7 39.4 1196
Note: Not employed includes not working and housewives categories in Table 7.
Table 9 shows the educational status of the patients. The education level was lower in Hospitals 1 and 3, with less than one third reporting more than high school education. This may reflect the higher proportion of service and outdoor workers in Hospital 1 and the high proportion of housewives in Hospital 3. Table 9: Patient Educational Status Percent Distribution by Hospital
Hospital More than high school Number of cases 1 28.2 177 2 53.5 185 3 30.6 258 4 45.0 188 5 51.3 281 6 41.4 58 Total 41.9 1147
175
2.2 Medical Diagnoses Table 10 shows the percent distribution of CMGs by hospital and the dominant CMG group in each hospital is bolded. As can be seen, Percutaneous Transluminal Coronary Angioplasty without Cardiac Catheterization accounts for the largest proportion for Hospitals 3, 4 and 5 whereas Cardiac Catheterization without Specified Cardiac Condition accounts for the largest proportion in Hospital 1, Percutaneous Transluminal Coronary Angioplasty with Complicated Cardiac Condition has the largest percentage for Hospital 6, and Major Cardio Procedure no Pump no Catheterization is the dominant CMG group for patients in Hospital 2. Table 10: Percent Distribution of the Number of CMGs, by Hospital
CMG 1
PTCA WO CARD CATH PERM PACE IMPL WO SPEC CARD C C BYPASS W PUMP WO CARD CATH PTCA W CARD COMP C VALVE REP W PUMP WO CARDIAC CARD CATH NO SPEC CARD COND C BYPASS W PUMP W CARD CATH MAJ CARDIO PR NO PUMP NO CATH ARRTHYMIA HEART FAILURE AMI NO CARD CATH NO SPEC COND OTHER CIRCULATORY DIAGNOSIS CHEST PAIN CARDIAC CATH W SECIFIED CARD UNS ANGINA W CATH NO SPEC CON AMI WITH CARD CATH NO SP COND PERM PACEMAKER/SPEC CARD COND AMI WO CARD C W CHF UNSTABLE ANGINA WO CATH WO SP UNS ANGINA W CATH NO CRD COND CARDIAC CATH WITH VENT TACH CRD VLV REP W PUMP W CRD CATH AMI W CARD CATH W & WO ANGINA SYNCOPE AND COLLAPSE ANGINA PECTORIS AMI NO CARD CATH W VENT TACH AMI W CAD CATH WITH CHF CAR CATH W CHF CARD CATH W VENTR TACH MAJOR CARDIO THORA PROCE WO P OTHER CARDIOTORACIC PROC W PU PROCEDURE CANCELLED MNRH ATHEROSCLEROSIS MNRH EXTNSIVE UNREL OR PROCESUDRE PERRIPHERAL VASCULAR DISEASE RENAL FAILURE NO DIALYSIS UNS ANGINA W CATH W SPEC COND Total 15.3 6.3 11.1 4.2 10.6 19.6 5.3 0.0 4.2 2.6 2.1 0.0 1.6 2.1 1.1 0.0 0.5 1.1 0.5 2.6 0.0 1.1 0.0 0.5 0.0 0.0 0.5 0.5 1.6 1.1 0.5 1.6 0.5 0.5 0.5 0.0 0.0 189 2 1.6 5.3 6.6 9.9 2.9 0.4 11.1 13.6 4.9 3.7 4.9 4.9 1.2 4.5 3.7 4.1 2.1 3.3 3.7 0.0 1.6 0.4 1.6 0.8 1.2 0.8 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.4 243 Hospital 3 4 15.4 27.2 11.6 18.5 9.7 5.1 8.9 4.6 6.9 4.6 0.0 2.1 2.7 3.6 4.2 4.1 6.6 4.1 4.6 1.5 4.2 1.0 5.0 4.1 8.9 0.0 0.4 2.6 1.9 1.0 1.2 2.1 0.4 3.1 1.2 1.0 1.9 0.5 1.2 0.0 0.8 1.5 0.4 0.5 0.8 0.5 0.4 0.0 0.4 0.0 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 259 182 Total 5 13.0 12.6 13.0 8.1 9.1 10.9 5.6 1.8 3.2 3.2 1.4 1.8 2.1 2.1 1.4 1.1 2.8 0.4 0.4 2.1 0.4 2.1 0.4 0.4 0.0 0.0 0.0 0.4 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0.4 0.0 285 6 5.1 8.5 0.0 22.0 8.5 0.0 5.1 5.1 5.1 3.4 10.2 1.7 3.4 0.0 1.7 3.4 1.7 3.4 0.0 0.0 1.7 0.0 1.7 1.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 55 13.5 10.7 8.9 8.1 6.9 5.9 5.7 4.9 4.6 3.3 3.2 3.2 3.0 2.2 1.9 1.8 1.8 1.5 1.4 1.1 0.9 0.9 0.7 0.5 0.3 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.1 1213
Note: Data are sorted by the percent in the Total column.
176
2.3 Nursing Diagnoses and OMAHA Scores at Admission Nursing diagnosis is a clinical judgment about an individual or family response to an actual or potential health problem or life process. Nursing diagnoses provide the basis for selecting nursing interventions to achieve outcomes for which nurses are responsible. Nursing diagnoses were recorded from admission until discharge. Table 11 shows that the number of nursing diagnoses averaged nearly five (SD=2.6) across all units. The highest mean number of diagnoses was observed in Hospital 5s CCU (8.44). The lowest average numbers of nursing diagnoses were observed in the DS unit in Hospital 4 (2.58), the DS unit in Hospital 5 (3.42), and IP unit in Hospital 3 (3.47). (Note that only 1,189 patients were included due to missing nursing diagnoses for some patients.) It should be noted that patients could change units during their hospital stay. Because patients may be admitted to one unit and discharged from another, the number of patients admitted to a unit may be different from the number of patients discharged from that unit. Table 11: Mean of Number of Nursing Diagnoses, by Unit Type
Unit at Discharge
Hospital 1
Hospital 2 Hospital 3 Hospital 4
Hospital 5
Hospital 6
CCU IP SDU CCU IP CCU IP CCU DS IP CCU DS IP CCU IP SDU
Total
# of Patients Grouped by Unit at Discharge 20 110 59 105 132 18 214 17 59 115 9 53 219 45 11 2 1189
Total Number of Diagnoses Mean SD 5.55 2.31 6.80 3.36 5.19 2.07 6.48 3.06 4.78 2.50 4.44 1.89 3.47 1.51 5.12 3.55 1.63 2.58 4.46 1.96 3.28 8.44 3.42 1.71 4.34 2.41 5.27 1.51 4.36 .81 6.50 .71 2.60 4.68
OMAHA Problem Scales for Outcomes is an evaluation tool developed by the Omaha Visiting Nurse Association2. The tool measures clinical progress of patients in relation to specific problems or nursing diagnoses at two points in time: at admission or when a new health problem was identified (time 1), and when the health problem was resolved or at discharge (time 2). The three essential dimensions, knowledge, behaviour, and status, are each rated on a 5-point Likert scale ranging from very negative to very positive. A mean score was computed for each person on each dimension and then averaged for the sample. Data about OMAHA at time 1 are presented in Table 12. The data for OMAHA at time 2 and difference scores are presented under section 7.2 in Patient Outcomes.
177
Table 12 shows that for all survey patients at time 1, the mean scores were 3.4 (SD=0.75) for knowledge, 4.0 (SD=0.58) for behaviour, and 3.3 (SD=0.62) for status. Patients from Hospital 4 had the highest ratings across all three dimensions. Table 12: OMAHA Scores at Time 1 (Admission and Appearance of New Diagnosis), by Hospital
Hospital 1 2 3 4 5 6 Total N 188 237 232 192 282 58 1189 Knowledge Mean SD 2.9 0.41 3.5 0.75 3.0 0.52 0.54 3.8 0.80 3.8 2.6 0.67 3.4 0.75 Behaviour Mean SD 3.8 0.38 4.0 0.46 3.6 0.54 0.52 4.3 4.2 0.63 0.49 4.3 4.0 0.58 Status Mean 3.0 3.6 3.0 3.8 2.9 3.5 3.3 SD 0.26 0.43 0.63 0.47 0.55 0.57 0.62
Note: Measured on a 5-point Likert scale for Knowledge (1=No knowledge and 5=Superior knowledge), Behaviour (1=Never appropriate in behaviour and 5=Consistently appropriate in behaviour), and Status (1=Extreme symptoms and 5=No symptoms).
2.4 Health Status at Admission The SF-12 is a widely used measure of general health. The questions tap eight health concepts that are reported in two categories: physical and mental health. The eight categories are physical functioning, role physical, bodily pain, general health, energy/fatigue, social functioning, role emotional, and mental health. Although the SF-12 is not quite as reliable and comprehensive as the SF-36, the results of the two measurement tools are highly correlated (r=0.95). Confidence intervals are largely determined by sample size; with large samples, the results will be almost the same regardless of the tool used3. Table 13 shows the two SF-12 subscales measured at admission at the interval level. The mean physical health score of 35.2 (SD=11.2) ranged from, 11.0 to 64.8. The mean mental health score of 48.2 (SD=11.0) ranged from 15.6 to 70.3. The least physically and mentally healthy patients were found in Hospitals 5 and 1 respectively.
178
Table 13: Patient Health Status at Admission, by Hospital

Hospital 1 2 3 4 5 6 Total N 186 207 256 191 277 59 1176 Physical Health Scale Mean 35.7 35.6 34.1 37.4 33.4 37.6 35.2 SD 10.3 11.9 10.2 12.1 11.1 11.5 11.2 Range 13.7-63.0 11.2-63.4 14.2-60.1 11.9-62.4 11.0-64.8 16.8-60.6 11.0-64.8 Mental Health Scale Mean 46.9 48.1 47.4 49.6 48.8 49.3 48.2 SD 10.7 10.8 12.0 10.5 10.5 11.7 11.0 Range 18.4-67.2 20.4-68.6 16.9-69.0 19.1-70.3 15.6-66.9 16.8-66.2 15.6-70.3
These two SF-12 subscales can be dichotomized using US norms for the general population as the cut-point. Dichotomized data are presented in Table 14. Nearly nine in ten patients (87.0%) were below the US population norm in physical health upon admission. About half (49.2%) of patients had mental health scores below the US population norm. The percentage of patients that scored below the norm for physical and mental functional status varied across hospitals (2=14.8 and 15.2 respectively, df=5, p<0.05). Generally, patients in Hospital 3 were the least physically and mentally healthy, on average. Table 14: Patient Health Status at Admission, Percent Less than US Norm, by Hospital
Hospital 1 2 3 4 5 6 Total % Not Physically Healthy 88.7 82.6 91.0 82.2 89.1 84.7 87.0 % Not Mentally Healthy 53.2 47.3 50.8 47.1 49.3 42.4 49.2 N 186 207 256 191 276 59 1176
US Norm3 Physically not healthy: < 50.12 Mentally not healthy: < 50.04
Tables 15 and 16 display two SF-12 subscale scores measured at discharge at the interval and dichotomous levels. Patients in Hospitals 2 and 5 tended to be the least physically healthy at discharge while patients in Hospitals 1 and 5 had the lowest average mental health scores at discharge. In contrast, the highest average physical and mental health scores for patients were observed in Hospitals 6 and 4 respectively. The comparison of changes in patient functional status from admission to discharge at the individual level will be presented in section 7.2.
179
Table 15: Patient Health Status at Discharge, by Hospital

Hospital
N 163 185 250 155 249 39 1041
Physical Health Scale Mean 34.7 31.2 35.3 36.0 32.8 36.1 34.0 SD 9.3 8.7 9.7 10.0 9.4 11.0 9.6 Range 14.0-58.4 15.2-56.6 14.2-56.8 16.4-60.8 12.1-56.8 18.2-59.4 12.1-60.8
Mental Health Scale Mean 46.0 48.0 48.9 50.7 47.3 48.2 48.1 SD 10.0 9.6 11.7 10.7 11.0 12.9 10.9 Range 25.4-66.3 20.6-65.9 17.3-69.0 23.7-68.4 17.5-69.8 15.6-65.1 15.6-69.8
1 2 3 4 5 6 Total
Table 16: Patient Health Status at Discharge, Percent Less than US Norm, by Hospital
Hospital 1 2 3 4 5 6 Total % Not Physically Healthy 94.5 95.7 90.0 89.0 95.2 84.6 92.6 % Not Mentally Healthy 60.7 54.6 47.6 42.6 55.4 43.6 51.9 N 163 185 250 155 249 39 1041
US Norm3 Physically not healthy: PHYSICAL HEALTH SCALE < 50.12 Mentally not healthy: Mental Health Scale < 50.04
3. Nurse Characteristics
The nursing information was collected from the Nurse Survey, and all the data are based on self report. The Nurse Survey was a very comprehensive survey, covering mental and physical health, job satisfaction, workload and violence during work using many reliable measures such as functional status of health (SF-12), Maslach Burnout Inventory (MBI), Revised Nursing Work Index (R-NWI), and Effort/Reward Imbalance (ERI). A total of 727 nurses participated in the study. 3.1 Nurse Demographics Table 17 demonstrates the gender and age distributions for nurses in each of the participating hospitals. The vast majority of nurses completing the survey were female (93.9%), which is consistent with the female to male ratio of the Canadian nursing workforce4. Only Hospital 1 reported a slightly higher proportion of males than the other sites. This site also reported the highest proportion of nurses less than 30 years of age and the lowest number over the age of 50. Hospitals 3 and 4 had higher proportions of nurses who were less than 30 than the remaining three hospitals but significantly less then Hospital 1. Unlike Ontarios profile5, some of these organizations have more nurses under 30 than over 50.
180
Table 17: Nurse Demographics, by Hospital

Hospital Gender % Female Number of Cases Age % < 30 % > 50 Number of Cases 1 87.9 58 23.2 5.4 56 2 94.6 186 7.7 12.1 182 3 93.2 74 13.9 11.1 72 4 94.2 86 14.3 11.9 84 5 95.7 210 7.8 9.8 205 6 92.9 112 6.5 17.6 108 Total 93.9 726 10.2 11.6 707
3.2 Professional and Employment Status Table 18 shows the percentages of nurse respondents who were employed full-time, employed on a permanent basis, and were Registered Nurses. The proportion of full-time nurses varied from 54.3% to 70.9%, with three of the hospital under 60% and three over 65%. Most nurses worked in permanent positions (97.8% on average), with the exception of Hospital 1 (89.8%). The percentage of full-time and/or permanent employees may have been inflated because fulltime and/or permanent nurses were more likely to participate in the survey. In terms of nursing composition, respondents were comprised of Registered Nurses (96.6%), Registered Practical Nurses (0.8%), and charge nurses and other personal care workers such as orderlies, Registered Nurse Technicians, clinical leaders, and healthcare aids (2.5%). All respondents from Hospital 2 were Registered Nurses and the respondents from Hospital 3 were all in permanent positions. Table 18: Nurse Employment Status, by Hospital
Hospital Number of Cases % FT % Permanent % RN 1 59 67.8 89.8 91.4 2 186 54.3 98.9 100.0 3 74 67.6 100.0 93.2 4 86 70.9 98.8 95.1 5 210 56.2 97.1 98.6 6 112 58.0 99.1 93.6 Total 727 59.8 97.8 96.6
3.3 Education and Clinical Expertise Table 19 gives a breakdown of the educational background of the nurses and their level of expertise, as measured by how often peers sought their advice, how often they filled the role of preceptor, and to what degree individuals possessed clinical expertise. The percentage of nurses with BScN and higher degree preparation varied across hospitals, ranging from 29.3% to 61.6%. Despite educational levels, a higher proportion of nurses in Hospital 6 acted as preceptors and 45.0% were frequently asked for clinical advice and considered themselves to be expert clinicians. Nurses at Hospital 1 reported the lowest levels of confidence in their clinical ability and fewer nurses were asked by their peers for clinical advice. Some of the variation in preceptor reporting may be due to variations in the number of student nurses at these hospitals or rate of staff nurse turnover. If turnover rates were high, larger number of staff nurses might be preceptors regardless of their knowledge or skill level.
181
Table 19: Nurse Education and Expertise, by Hospital

Hospital N Education - % with BScN and higher Preceptor - % Frequent Clinical advice - % Frequent Expert - % Recognized expected clinical responses 1 59 29.3 7.0 32.8 19.0 2 186 34.9 10.2 40.3 39.7 3 74 40.5 14.9 59.5 35.7 4 86 61.6 14.1 43.0 36.5 5 210 50.5 9.1 46.2 45.9 6 112 32.1 22.3 45.0 45.0 Total 727 42.3 12.4 44.4 39.8
3.4 Experience Table 20: Nurse Experience, N and Percent of Total Respondents, by Hospital
Hospital 1 Yrs as a nurse Yrs in hospital Yrs on unit 2 Yrs as a nurse Yrs in hospital Yrs on unit 3 Yrs as a nurse Yrs in hospital Yrs on unit 4 Yrs as a nurse Yrs in hospital Yrs on unit 5 Yrs as a nurse Yrs in hospital < 1 yr 0 0.0% 1 2.2% 6 14.3% 0 0.0% 3 2.1% 7 5.3% 0 0.0% 1 2.2% 1 2.3% 2 2.4% 1 1.6% 1 1.8% 0 0.0% 2 1.4% 1- 5 yrs 13 23.2% 19 42.2% 22 52.4% 13 7.1% 37 26.2% 50 38.2% 8 11.1% 12 26.1% 24 54.5% 19 22.9% 25 41.0% 35 63.6% 22 10.6% 39 27.1% 6 -10 yrs 12 21.4% 8 17.8% 7 16.7% 23 12.5% 13 9.2% 19 14.5% 8 11.1% 4 8.7% 4 9.1% 14 16.9% 7 11.5% 4 7.3% 17 8.2% 16 11.1% 11-15 yrs 14 25.0% 9 20.0% 6 14.3% 40 21.7% 45 31.9% 34 26.0% 21 29.2% 18 39.1% 9 20.5% 10 12.0% 15 24.6% 11 20.0% 49 23.7% 41 28.5% > 15 yrs 17 30.4% 8 17.8% 1 2.4% 108 58.7% 43 30.5% 21 16.0% 35 48.6% 11 23.9% 6 13.6% 38 45.8% 13 21.3% 4 7.3% 119 57.5% 46 31.9% Total 56 100% 45 100% 42 100% 184 100% 141 100% 131 100% 72 100% 46 100% 44 100% 83 100% 61 100% 55 100% 207 100% 144 100%
182
Hospital Yrs on unit 6 Yrs as a nurse Yrs in hospital Yrs on unit Total Yrs as a nurse Yrs in hospital Yrs on unit
< 1 yr 6 4.3% 2 1.9% 3 4.6% 5 8.3% 4 0.6% 11 2.2% 26 5.5%
1- 5 yrs 71 50.4% 6 5.7% 11 16.9% 23 38.3% 81 11.4% 143 28.5% 225 47.6%
6 -10 yrs 22 15.6% 12 11.3% 7 10.8% 11 18.3% 86 12.1% 55 11.0% 67 14.2%
11-15 yrs 26 18.4% 25 23.6% 21 32.3% 14 23.3% 159 22.5% 149 29.7% 100 21.1%
> 15 yrs 16 11.3% 61 57.5% 23 35.4% 7 11.7% 378 53.4% 144 28.7% 55 11.6%
Total 141 100% 106 100% 65 100% 60 100% 708 100% 502 100% 473 100%
Table 20 compares overall nursing experience of the nurses, their experience within the same institution and within the same unit. Hospital 1 nurses reported the least experience in all three categories. In all six hospitals, 53.4% of nurses reported more than 15 years of nursing experience. Years as a nurse varied more than years in hospital or years on the nursing unit. Nurses with greater than 15 years experience reported less years on unit than nurses with fewer years of experience. This may reflect hospital restructuring activities in recent years. However, Hospital 1 nurses worked the shortest time on unit compared to any of the other hospitals which may explain lower levels of expertise as reported in Table 19. Age and experience were highly correlated. The correlation coefficients between age and years worked as RN/RPN, years worked as RN/RPN at the current hospital and years worked as RN/RPN at current unit were 0.83, 0.61 and 0.45 respectively (p<0.001). Years worked as RN/RPN was highly associated with all three expert statuses: acting as a preceptor (F=5.27, p<0.05), providing clinical advice (F=39.75, p <0.001), and acting as a clinical expert (F=35.73, p <0.001). A nurse with more years experience as RN/RPN at the current hospital was more likely to be asked for clinical advice (F=23.65, p<0.001) or act as a clinical expert (F=8.51, p <0.01). However, greater years of experience in current hospital were not significantly related to acting as a preceptor. A similar association was found between years on current unit and expert status. A nurse with more years as RN/RPN on his/her current unit was more likely to be asked for clinical advice (F=12.94, p<0.001) or act as a clinical expert (F=42.97, p <0.001). However, more years on current unit was not significantly related to role as a preceptor. Higher education was positively associated with expert status. This association was only statistically significant nurses responded that other nurses sought their clinical judgment Evidence-based Staffing 183
(2=4.26, p<0.05). The associations between education and the status of providing clinical advice or acting as a clinical expert were significant at the 0.1 level.
4. System Behaviours
4.1 Workload Workload data (measured by PRN 806) were collected for each study patient on each study day by the site data collectors. The PRN tool measures the volume of nursing work in minutes by selecting the tasks that need to be completed for that day. Each task is assigned a value based on studies completed by the PRN system in numerous facilities. Workload values are presented on the measurement tool in five minute increments. This value reflects the average time to complete the task, by an average nurse, on an average day, for an average patient. This methodology is referred to as an average time methodology. The PRN tool does not directly capture the workload associated with activities that are not patient specific, however, the PRN value can be adjusted to account for indirect patient care. The PRN 80 values for indirect care time provided by Tilquin were included to determine the total hours of care patients required in the next 24 hour period (Tilquin, personal communications, August, 2003). Patient care workload for each study patient was also recorded from the unit workload tool. Hospital 2 used MEDICUS while all other study hospitals used GRASP. Both workload tools measure nursing hours including direct and indirect services related to patients. These values were compared to the PRN workload value adjusted for indirect patient care. In this study, GRASP or MEDICUS hours were collected daily for study patients and for the unit as a whole, including non-study patients as well. The GRASP methodology captures workload using a standard time methodology. Each site develops a list of tasks based on the activities they perform, and times are assigned to each of these tasks. The times are based on time and/or frequency or are established by staff nurse consensus. These times reflect the average time to complete the task, by an average nurse, on an average day, for an average patient in the individual facility. This reflects the physical and organizational characteristics of the individual facility. The MEDICUS system captures workload by multiplying a pre-set relative value per level of care by the target hours per unit of workload. Table 21 shows PRN patient care time (in minutes) by workload category. Total workload for each day is the sum of seven categories of activity including respiration, feeding and hydration, elimination, hygiene and comfort, communication, treatment, and diagnostic procedures. The average total PRN value for six hospitals was 274.5 minutes, or about 4.5 hours, for each patient day, with wide variations (SD=227.1). In descending order, average minutes for PRN activities are as follows: diagnostic procedures (109.0), treatment (48.6), hygiene and comfort (42.6), communication (32.4), respiration (16.6), feeding and hydration (15.1), and elimination (10.2). Hospital 1 averaged the most PRN minutes in total by category, except for diagnostic procedures. Hospital 3 averaged the fewest PRN in total and tended to rank low across categories.
184
Table 21: Mean (SD) of PRN Workload (in Minutes) by Category, by Hospital
Hospital Feed. & Elim. Hyg. & Hyd. Comfort 42.3 21.2 16.6 66.6 1 (48.4) (8.7) (12.5) (39.5) 15.3 17.5 12.2 62.0 2 (17.2) (12.2) (12.2) (39.4) 9.6 13.0 6.5 26.3 3 (15.0) (9.1) (10.1) (28.2) 9.3 14.4 8.6 30.0 4 (15.8) (9.7) (11.0) (31.0) 15.0 12.2 8.2 31.4 5 (25.3) (7.5) (12.0) (26.1) 11.8 10.9 10.2 32.5 6 (24.6) (8.6) (11.8) (26.9) 16.6 15.1 10.2 42.6 Total (27.2) (10.2) (12.1) (36.8) Note: Overall means based on daily patient data. Resp. Comm. 66.7 (18.8) 14.1 (11.2) 31.8 (21.9) 38.0 (14.1) 28.9 (11.6) 39.2 (13.8) 32.4 (21.9) Treatment 58.5 (40.1) 42.9 (43.2) 46.3 (50.1) 53.4 (37.2) 48.5 (51.8) 43.1 (32.6) 48.6 (45.4) Diag. Proc. 132.4 (89.2) 154.9 (156.4) 49.9 (95.6) 96.4 (139.7) 90.3 (149.3) 171.1 (143.8) 109.0 (140.0) Total 404.2 (179.6) 318.9 (238.3) 183.5 (179.8) 249.7 (213.2) 234.6 (239.2) 318.3 (201.2) 274.5 (227.1)
Table 22 shows the total PRN minutes accounted for by each workload category. Diagnostic procedures comprised almost one-third (31.1%) of the total PRN minutes, followed by treatment and communication as the second and third highest proportions. Activities related to respiration, feeding and hydration, and elimination accounted for the smallest proportion, less than five percent, of total PRN patient care. Large amounts of variation were observed across hospitals. Hospitals 6 and 2 had the highest proportion of diagnostic procedures, whereas Hospital 3 has the smallest proportion. The proportion of treatment time was much higher in Hospitals 3 and 4 (23.9%) than in other hospitals (14.5%-20.7%). Hygiene and comfort in Hospital 2 was much higher (23.0%) than in other hospitals (10.4%-16.4%). The proportion of time spent in communication with patients was extremely low in Hospital 2 (6.7% vs. 17.3%-23.6% for other hospitals). Table 22: PRN Workload Category as Percent of Total PRN Workload, by Hospital
Hospital Feed. & Elim. Hyg. & Comm. Hyd. Comfort 1 5.7 15.8 19.4 8.8 4.0 2 3.9 8.2 6.7 4.0 23.0 3 4.3 3.1 15.0 10.4 23.6 4 2.8 7.7 3.5 11.9 21.7 5 4.7 8.3 3.3 15.0 19.9 6 2.7 4.6 2.8 10.4 16.7 Total 4.6 8.0 3.6 16.4 17.3 Note: Differences by hospital are statistically significant at p<0.000 Resp. Treatment 14.5 13.8 23.9 23.9 20.7 14.9 19.0 Diag. Proc. 31.7 40.4 19.8 28.7 28.2 48.1 31.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Table 23 demonstrates the gap between patient care time PRN and patient workload measured in GRASP or MEDICUS across hospitals. The PRN scores in the last column of Table 21 were adjusted for indirect care time using the method developed by Charles Tilquin. Workload measured by GRASP or MEDICUS was on average 1.6 hours greater than workload measured by PRN. Hospital 2, which used MEDICUS, showed the largest workload value (10.77 hours) among all hospitals. With a mean PRN value of 6.78 hours, Hospital 2 had the largest discrepancy (3.99 hours) with PRN. The discrepancy is also large for Hospital 4 (2.09 hours).
185
Hospital 1 was high in both GRASP and PRN scores, with only a small discrepancy (0.63 hours). The GRASP values, which were generally higher than PRN, were lower for Hospital 3 (-0.56 hours). Table 23: Comparison of PRN to GRASP/Medicus Workload (in Hours), by Hospital
Hospital 1 2 3 4 5 6 Total WL(GM) N Mean SD N PRN Mean SD WL(GM)-PRN N Mean SD
1094 9.29 5.49 1038 8.65 3.50 1038 0.63 4.83 2029 7.06 2018 6.78 4.68 2018 3.99 4.33 10.77 1422 3.50 3.12 1421 4.06 3.49 1421 -0.56 2.43 1295 7.52 5.40 1280 5.44 4.19 1280 2.09 3.85 1939 6.20 5.47 1932 5.09 4.70 1932 1.12 3.31 334 7.13 3.94 308 6.99 4.11 308 0.23 3.90 8113 7.54 6.08 7997 5.92 4.48 7997 1.60 4.12 Note: (1) WL(GM) stands for workload measured by GRASP (for Hospitals 1, 3, 4, 5, 6) or MEDICUS (for Hospital 2). (2) Based on overall means using daily patient data.
4.2 Workload Variation by Patient Medical Diagnosis The variation in the distribution of work across various sites may be due to variations in the mix of patients at each site. Table 24 presents the distribution of work by Case Mix Groups (CMG) in descending order of average PRN workload in minutes. This analysis shows that PRN workload varies significantly by CMG. The workload for CMG Extensive Unrelated Operating Room Procedure and CMG group Major Cardio-thoracic Procedures Without Pump was nearly 10 hours (>588.8 minutes), the highest among all CMGs, and six times the size of workload for Cardiac Catheterization with Congestive Heart Failure (1.5 hours or 90 minutes). Table 24 also showed that not all CMGs had the same percentage distribution of workload category as total workload. For example, diagnosis procedures had a much higher percentage for CMG Acute Myocardial Infarction with Cardiac Catheterization with Congestive Heart Failure than for others CMG subgroups.
186
Table 24: Percent Distribution of Work (in Minutes) by Workload Category, by CMG Type
N* Resp 17.35 7.35 15.10 8.57 4.72 4.28 5.57 6.97 4.94 2.10 6.91 4.82 2.89 6.16 5.88 3.55 3.12 2.48 1.94 1.92 3.17 2.63 2.12 4.15 4.63 1.79 2.37 4.14 5.86 3.31 3.45 2.39 1.32 3.52 3.15 8.60 0.00 4.60 4.06 4.71 4.79 5.74 7.04 5.94 6.72 6.52 6.41 5.29 6.86 8.26 8.39 3.41 9.80 8.47 9.38 9.70 9.03 8.08 9.69 9.78 8.65 8.22 10.62 9.70 8.61 10.08 7.03 9.49 10.43 9.92 10.64 10.40 11.47 9.40 12.04 8.06 2.23 2.99 2.30 3.48 3.32 4.36 3.00 3.32 4.44 1.22 4.34 6.23 3.53 0.00 10.49 3.49 3.93 6.20 3.15 2.79 3.46 3.07 4.17 5.36 4.85 3.99 2.92 2.59 0.71 2.73 5.00 3.50 1.91 2.32 1.64 0.72 0.00 3.55 21.11 18.08 14.92 15.57 19.10 17.97 17.15 15.88 17.68 11.61 17.13 20.57 15.48 9.51 17.23 14.66 17.60 22.32 12.38 18.25 19.43 13.52 14.38 19.88 19.00 17.67 15.47 14.69 18.93 15.14 20.21 18.07 17.20 18.85 14.74 14.26 12.04 16.48 12.33 13.40 16.62 15.64 11.35 12.14 14.19 17.13 19.87 31.53 14.47 13.12 15.79 30.49 19.97 17.63 14.98 13.29 22.28 5.76 15.49 28.11 22.28 13.21 18.06 18.55 18.22 23.02 24.58 20.72 16.89 17.50 18.54 16.55 26.01 19.50 29.17 17.30 17.44 15.42 17.38 21.50 19.13 11.17 18.85 21.94 11.39 19.07 20.00 15.53 15.95 27.08 10.56 17.77 14.71 14.75 20.93 10.40 15.67 16.67 20.40 12.84 17.57 15.70 19.35 21.35 11.35 15.62 17.52 18.52 15.52 19.93 16.70 13.43 6.02 18.99 25.48 38.05 28.88 29.42 35.35 44.15 34.64 28.25 35.28 29.18 30.29 31.46 37.97 23.34 26.07 34.43 36.26 31.39 30.37 52.79 33.08 26.22 28.10 36.34 24.95 32.60 33.06 24.12 31.54 33.00 26.52 30.10 34.87 28.43 26.29 34.08 40.74 31.02 Feed & Hyd Elim. Hyg & Comfort Comm. Treatm ent Diag. Proc. Average Total Workload 596.1 588.8 411.4 389.9 361.1 351.5 344.4 341 313 310 307.1 273.1 272.6 259 256.7 247.4 234.7 229.7 212 209.2 207.6 205.7 200.1 191.5 191.3 186.9 184.3 176.4 176.3 174.4 169 166.1 160.7 158.1 138.8 130 97.5 274 Percentage Distribution of Workload
CMG
EXTNSIVE UNREL OR PROCESUDRE MAJOR CARDIO THORA PROCE WO P OTHER CARDIOTORACIC PROC W PU C VALVE REPLACE W PUMP WO CAR MAJ CARDIO PR NO PUMP NO CATH AMI W CAD CATH W CHF C BYPASS W PUMP W CARD CATH C BYPASS W PUMP WO CARD CATH CAD CATH W VENTR FIB ATHEROSCLEROSIS MNRH CAR VALVE REPL W PUMP W CARD AMI WO CARD C W CHF AMI W CARD CATH NO SPEC COND PROCEDURE CANCELLED MNRH PERRIPHERAL VASCULAR DISEASE PTCA W CARD COMP UNSTABLE ANGINA WO CATH WO SP AMI NO CARD CATH NO SPEC COND PTCA WO CARD CATH UNS ANGINA W CATH W SPEC COND AMI NO CARD CATH W VENT TACH CARD CATH WO SPECIFIED CONDITI PERM PACE IMPL WO SPEC CARD C ANGINA PECTORIS HEART FAILURE AMI W CARD CATH WITH ANGINA PERM PACEMAKER IMPLANT W UNS ANGINA W CATH NO CRD COND CARDIAC CATH W CHF ARRYTHMIA OTHER CIRCULATORY DIAGNOSIS CARDIAC CATH W SECIFIED CARD CARDIAC CATH WITH VENT TACH UNSTABLE ANGINA W CARD CATH W CHEST PAIN SYNCOPE AND COLLAPSE RENAL FAILURE NO DIALYSIS Total
14 38 22 747 704 13 848 993 22 3 171 158 168 5 3 505 78 264 579 13 19 228 545 13 324 53 206 70 8 281 250 189 90 133 124 21 2 7904
Note: (1) Based on patient daily entry data for 1,198 patients. (2) Data are sorted by Average Total Workload. 187
4.3 Overtime and Continuity of Care/Shift Change and Unit Instability Table 25 shows the worked overtime hours in three categories by hospital (on average, 7% of the nurses worked more than 40 hours per week in their own hospital; see Table 34). Close to half of the nurses (45.1%) in all the hospitals reported either no overtime or less than one hour overtime per week. Another 32.2% worked 1-4 hours of overtime per week. The rest of the nurses (22.7%) worked greater than four hours overtime per week. Within the latter category, the highest and lowest proportions were achieved by Hospitals 6 (38.5%) and 4 (16.7%). Table 25: Percent of Nurses Reporting Overtime in Average Hours per Week, by Hospital Hospital
1 2 3 4 5 6 Total
N
52 153 64 72 184 96 621
0-1
36.5 47.7 37.5 51.4 47.8 40.6 45.1
1-4
32.7 33.3 45.3 31.9 32.6 20.8 32.2
>4
30.8 19.0 17.2 16.7 19.6 38.5 22.7
Table 26 presents the percent change in overtime in the past year by unit type. Nurses were asked whether the amount of overtime required had increased, remained the same, or decreased in the past year. For most nurses (64.3%), overtime work increased. The CCU in Hospital 4 reported the highest proportion of no change in overtime hours. Nurses in Hospitals 4 and 5 more frequently reported decreases in overtime in the past year than the nurses in the other institutions. Table 26: Percent Change of Nurse Overtime Hours in the Past Year, by Unit Type
Unit Type Hospital 1
N
Increased
Remained Same
Decreased
3.6 5.6 0.0 1.0 0.0 6.5 2.9 18.5 0.0 13.0 21.2 3.4 4.3 0.0 9.1 0.0 3.6
28 64.3 32.1 18 72.2 22.2 3 0.0 100.0 Hospital 2 97 62.9 36.1 37 89.2 10.8 Hospital 3 31 54.8 38.7 35 68.6 28.6 Hospital 4 27 18.5 63.0 5 80.0 20.0 23 39.1 47.8 Hospital 5 52 38.5 40.4 29 69.0 27.6 70 54.3 41.4 Hospital 6 58 75.9 24.1 11 63.6 27.3 15 60.0 40.0 Total 539 64.3 32.1 Note: Significance test is not available because of small N in some of the cells.
The DS unit in Hospital 4 also had a very low number of nurses reporting an increase in levels of overtime while the other units in the same hospital had high numbers reporting a decrease (Table
188
26). These nurses were more likely (51.4%) to report an average of less than one hour of overtime as compared to other hospitals (Table 25). Table 27 contrasts the prevalence of unpaid overtime and involuntary overtime. Unpaid overtime includes both voluntary and involuntary; involuntary includes both paid and unpaid overtime. Hospital 4 reported the highest percentage of unpaid overtime but below average involuntary overtime. Overtime may not by itself be a contributing factor to high stress levels, but rather the involuntary nature of overtime in some organizations may lead to high levels of stress. Some nurses choose to work overtime and therefore overtime may not be a source of stress. However, we hypothesize that when overtime is involuntary in nature it may serve as a source of stress. Table 27: Percent of Overtime Unpaid or Involuntary, if Working Overtime, by Hospital
Hospital 1 2 3 4 5 6 Total
N
37 84 46 37 114 65 383
% Unpaid
13.0 28.6 36.1 42.1 25.7 18.6 26.7
% Involuntary
27.7 25.6 29.3 21.3 18.4 20.5 22.8
Continuity of care was operationalized as the proportion of shift changes more frequent than once in the last two weeks as well as the proportion of nurses forced to change units in the past year and of those who anticipated forced changes in their unit in the coming year (Table 28). Nurses in Hospital 5 experienced higher levels of forced change in the last year, but Hospital 1 had the highest percentage anticipating a change in the next year. The nurses in Hospital 5 also reported the highest number of shift changes per week. Nurses who were forced to change unit in the past year had a higher nurse-patient ratio than nurses not experiencing a forced unit change (F=12.7, p<0.001). The proportion of shift changes and anticipation of forced changes in unit was associated with emotional exhaustion (measured by Maslachs Burnout Inventory). Those who anticipated forced change of unit were more likely to rank high on the emotional exhaustion index (thus not healthy) than those who did not anticipated forced change of unit (F=8.7, p<0.01). Table 28: Continuity of Care and Amount of Change, by Hospital
Hospital % More than 1 shift change in the past 2 weeks N % Forced to change nursing units in the past year N % Anticipating forced change of unit N 1 36.2 58 10.3 58 24.6 57 2 36.3 182 8.7 184 14.1 184 3 17.6 74 0.0 74 6.8 73 4 30.2 86 2.4 85 7.1 85 5 41.3 206 13.8 210 20.8 207 6 18.5 108 4.5 111 17.0 112 Total 32.4 714 8.0 722 15.7 718
189
4.4 Non-Nursing Tasks Despite the nurses reports of poor quality nursing care and the high percentage of nurses reporting that essential nursing tasks are not being completed or are delayed, nurses continue to report a high volume of tasks that could be delegated to non-nursing personnel. Table 29 shows how likely a nurse is to perform these tasks. Among eight non-nursing tasks listed in Table 29, nurses more likely reported ancillary service, venipunctures, housekeeping, delivering trays, and starting IVs. Each hospital varied in which tasks were more likely to be performed by the nurse. Hospital 1 had the largest proportion of nurses who reported delivering trays, venipunctures, and performing ECGs. Hospital 2 was more likely to mention transporting patients than the others. Hospital 3 had the largest proportion reporting housekeeping and arranging transportation. Hospital 5 was most likely to report starting IVs. Hospital 6 had the largest proportion reporting ancillary service. Table 29: Percent of Nurses Reporting Performing Non-Nursing Tasks for Items in the Table, by Hospital
Non nursing tasks Number of Cases Ancillary service Venipunctures Delivering trays Housekeeping Starting IVs Performing ECGs Transporting patients Arranging transportation 1 58 87.9 82.8 72.4 63.8 67.2 51.7 22.4 32.8 2 184 82.6 72.3 34.2 44.6 17.4 15.8 42.4 27.2 3 73 90.4 11.0 54.8 87.7 63.0 9.6 24.7 38.4 4 86 89.5 64.0 51.2 47.7 62.8 37.2 20.9 30.2 5 210 74.3 73.8 65.2 64.3 71.0 50.0 26.2 23.3 6 111 91.0 62.2 64.9 35.1 43.2 48.6 36.0 31.5 Total 722 83.5 64.8 55.1 55.1 51.0 35.6 30.7 28.7
Note: Non-nursing tasks are sorted by percent in the Total column
5. Intermediate System Outputs

Staffing information was collected by the site data collectors for each day of the study and for each nursing unit. Data were collected by shift, by skill level, and by employment status. Data about unit staffing, patient census, and unit daily workload are displayed in Tables 30-34. 5.1 Worked Hours Table 30 shows that the actual staffing hours varied from hospital to hospital and from unit to unit. In general, CCUs had a higher average daily worked hours and workload with the highest for CCU in Hospital 3 (329.3) and lowest (88.0) for the DS unit in Hospital 4. The IP unit in Hospital 4 averaged the highest number of agency worked hours (23.9). In contrast, Hospital 3 used no agency worked hours. The SDU in Hospital 6 averaged the highest number of overtime hours (6.7) and absent hours (27.4).
190
Table 30: Actual Staffing Hours, by Unit, by Day

Unit Type Worked Hours All Staff
N
Agency Worked Hours
Overtime Hours
Absent Hours
All GRASP Workload
Mean N Mean N Mean N Mean N Mean (A) (B) (C) (D) (E) Hospital 1 CCU 87 146.4 87 3.2 87 4.4 87 14.7 79 148.1 IP 117 160.2 117 3.5 117 1.4 117 8.4 103 149.6 SDU 87 104.9 87 0.0 87 1.3 87 5.4 81 101.1 CCU 227 298.4 227 0.9 227 0.3 228 12.7 228 Hospital 2 300.8 IP 122 206.0 122 5.2 122 0.8 122 9.1 122 213.5 108 329.3 108 0.0 108 0.4 108 9.9 107 299.1 Hospital 3 CCU IP 112 264.5 112 0.0 112 5.4 112 5.6 112 200.7 Hospital 4 CCU 219 284.4 219 0.1 219 0.4 219 9.4 217 252.4 DS 109 88.0 109 0.1 110 0.1 109 4.8 107 77.2 IP 220 249.1 220 220 0.6 220 9.2 217 272.3 23.9 200 203.5 200 0.1 200 2.7 200 7.4 200 171.9 Hospital 5 CCU DS 101 134.5 101 0.1 101 1.4 101 6.9 100 115.1 IP 229 143.4 228 2.3 228 0.7 229 6.3 226 141.0 142 141.9 142 1.5 142 4.7 142 13.9 140 90.1 Hospital 6 CCU IP 62 136.2 62 1.7 62 3.6 62 13.2 60 88.6 SDU 19 222.9 18 7.6 18 18 19 107.1 6.7 27.4 Total 2161 206.1 2159 3.6 2160 1.6 2161 9.2 2118 190.9 Note: (1) N is the number of unit days. (2) Letters A, B, C, D, and E will be used in the following table. (3) GRASP hours are MEDICUS hours for Hospital 2. (3) Agency worked hours includes hours worked by agency nurses and/or agency non-nursing staff (e.g., sitters).
The hospital workload measure, agency worked hours, overtime worked hours, and absent hours as a percent (or ratio) of total worked hours are displayed in Table 31. The hospital workload measure averaged 92.6% of total worked hours across all units and hospitals. Of 16 hospital units, four had ratios over 100% and three had ratios over 93%. Hospital 6 has the lowest ratio for all unit types (less than 65%). Units in Hospital 6, on the other hand, reported more overtime hours and absent hours than other units. Agency hours comprised 9.6% of worked hours on Hospital 4s IP unit as compared to 3.4% or less for other units.
191
Table 31: Percent of Actual Staffing, by Unit, by Day

Unit Type Hospital 1 CCU IP SDU CCU IP CCU IP CCU DS IP CCU DS IP CCU IP SDU
WL(GM)/WH (E/A)
Agency/WH (B/A)
Overtime/WH (C/A)
Absent/WH (D/A)
101.2 2.2 3.0 10.0 93.4 2.2 0.9 5.2 96.4 0.0 1.2 5.1 Hospital 2 100.8 0.3 0.1 4.3 103.6 2.5 0.4 4.4 Hospital 3 90.8 0.0 0.1 3.0 75.9 0.0 2.0 2.1 Hospital 4 88.7 0.0 0.1 3.3 87.7 0.1 0.1 5.5 0.2 3.7 109.3 9.6 Hospital 5 84.5 0.0 1.3 3.6 85.6 0.1 1.0 5.1 98.3 1.6 0.5 4.4 Hospital 6 63.5 1.1 9.8 3.3 65.1 1.2 2.6 9.7 48.0 3.4 3.0 12.3 Total 92.6 1.7 0.8 4.5 Notes: (1) WH denotes worked hours. (2) See Table 30 for designations of A-E. (3) Agency worked hours includes hours worked by agency nurses and/or agency non-nursing staff (e.g., sitters).
As shown in Table 32, the overall daily averages per unit were 19.8 patients on census, 6.1 admissions, and 6.1 discharges. Patient census, daily admission, and daily discharge, however, varied greatly by unit type and by hospital. For all hospitals, IP units averaged the most patients, ranging from 15.5 to 49.7 patients per day, and the most admissions and discharges, except in the case of DS units. The IP unit in Hospital 3 averaged the highest number of patients, admissions, and discharges per day. The IP unit for Hospital 2 also tended to average a higher number of patients. The lowest and highest mean censuses were found in Hospitals 1 and 3 respectively.
192
Table 32: Daily Patient Census, Admissions, and Discharges, by Unit

Midnight Census N Mean

87 117 87 228 122 108 112 219 117 221 200 101 228 142 62 19 2170 10.0 17.0 15.1 15.5 37.8 16.2 49.7 10.4 8.3 34.5 9.0 12.6 28.1 10.3 15.5 29.5 19.8
Daily Admissions N Mean

87 117 87 228 122 108 111 219 109 221 200 101 228 142 62 19 2161 2.7 6.3 3.7 4.5 9.3 6.6 15.6 3.4 7.0 6.5 2.8 7.6 6.5 4.0 13.0 5.3 6.1
Daily Discharges N Mean

87 117 87 228 122 108 112 219 117 221 200 101 229 142 62 19 2171 2.7 6.4 3.6 4.6 9.0 6.6 15.2 3.4 7.8 6.7 2.7 7.4 6.4 4.0 12.9 5.3 6.1
Hospital 2 Hospital 3 Hospital 4
Hospital 5
Hospital 6
Total
5.2 Productivity/Utilization Nursing unit productivity/utilization is measured as the ratio of GRASP/Medicus workload hours to worked hours. The maximum work capacity of any employee is 93%. Seven percent of worked hours are allocated to paid breaks during which time no workload is contractually expected. At 93% nurses are working flat out with no flexibility to meet unanticipated demands or rapidly changing patient acuity. This study hypothesized that a value of 85% is an appropriate productivity/utilization level to ensure high quality cost effective care. As shown in Table 33, in a large proportion of units, nurses were working beyond 93% productivity/utilization levels. On 61.5% of the study days, productivity/utilization levels were higher than 85%. Generally, IP units had higher productivity/utilization levels than CCUs, SDUs, and DS units. This finding is not unexpected since these units must staff differently to monitor patients. The productivity/utilization levels were lowest in Hospital 6. On 46.5% of the study days, productivity/utilization levels were higher than 93%. CCUs in Hospitals 1, 3, and 6 as well as IP units in Hospitals 2, 4, and 5 were most frequently above 93% productivity/utilization levels within their respective organizations.
193
Table 33: Number of Days When Unit GRASP/Medicus is Greater than 85% and 93% of Total Nurse Hours, by Unit
Unit Type
Number of Days Reported

% Days with WL(GM) > 85% of Total Nursing Hours
% Days with WL(GM) > 93% of Total Nursing Hours
79 79.7 62.0 103 63.1 49.5 81 67.9 58.0 Hospital 2 225 60.0 51.1 121 99.2 90.1 Hospital 3 107 60.7 48.6 112 29.5 11.6 Hospital 4 216 67.1 35.2 99 48.5 34.3 216 96.8 88.0 Hospital 5 199 41.2 21.6 100 51.0 37.0 226 86.7 63.7 Hospital 6 140 13.6 9.3 60 10.0 8.3 19 5.3 0.0 Total 2103 61.5 46.5 Note: (1) WL(GM) stands for workload measured as GRASP (in Hospital 1, 3, 4, 5, 6) or Medicus (in Hospital 2). (2) Differences by unit are statistically significant (2=744.9, df=23, p<0.01)
Hospital 1
Table 34 displays worked hours of nurses in the study hospital and in other employment. Nearly two thirds of surveyed nurses worked between 31 to 40 hours a week in the past year, which is typical of full-time employee work time. One quarter of surveyed nurses reported working hours between 21 and 30 hours. On average, only 3.8% of nurses worked less than 20 hours a week. In addition, 7% of nurses averaged over 40 worked hours per week. Nearly one in four nurses (22.5%) reported paid work outside the study hospital. Hospital 1 had the highest proportion of nurses working full-time (71.9%), whereas the proportion of nurses working part-time (i.e., less than 30 work hours per week) was highest in Hospital 2 (35.8%) and 5 (38.0%). Hospital 2 also had a relatively high proportion of nurses (8.4%) working less than 20 hours weekly in the study hospital. Nurses in Hospitals 3 and 4 are vulnerable to strenuous work conditions: nearly one in nine (11.8%) nurses in Hospital 3, and one in six (15.9%) nurses in Hospital 4 worked the most overtime. In addition, nurses in Hospital 4 were most likely to hold paid employment outside the study hospital. Some of this variation may be due to the available supply of employment opportunities in individual locations.
194
Table 34: Percent of Nurses Reporting Average Hours Worked Per Week in the Past Year, by Hospital
Hospital
N 1 2 3 4 5 6 Total 57 179 69 85 208 107 705 < 20 hrs 1.8 8.4 1.4 2.4 4.8 1.9 3.8
In this Hospital
21-30 hrs 19.3 27.4 24.6 16.5 33.2 24.3 26.4 31-40 hrs 71.9 59.8 58.0 69.4 57.7 67.3 62.3 > 40 hrs 7.0 4.5 15.9 11.8 4.3 6.5 7.0
Other Employment
N 15 60 21 20 44 26 186 > 10 hrs 6.7 21.6 28.5 35.0 27.2 11.4 22.5
6. Environmental Complexity
The Environmental Complexity Scale (ECS) captures nurses ratings of how daily unit factors influence their ability to provide required care for patients. Twenty-two ECS items were administered daily to all nurses working day shift on participating study units. Missing data were imputed using individual and unit means. Three subscales were constructed to capture different dimensions of nurses work complexity: (1) Re-Sequencing of Work in Response to Others; (2) Unanticipated Changes in Patient Acuity; and, (3) Composition and Characteristics of the Care Team which considers students, staffing, and nurse team functioning. The values range between 0 and 10. As shown in Table 35, the means for the three subscales (resequence, change, team) were 6.1 (SD=0.88), 6.5 (SD=1.04), and 5.7 (SD=1.06), respectively, for all study hospitals. The differences by unit for all subscales are statistically significant at p<0.001, but there are no clear patterns by hospital or by unit type.
195
Table 35: Mean of Three Subscales from ECS, by Hospital

N*
Resequence Mean SD
5.9 6.0 5.5 5.6 5.8 6.1 6.3 5.7 5.9 6.1 6.8 6.6 6.4 6.3 6.3 6.1 6.1 0.75 1.32 0.65 0.72 0.74 0.97 0.92 0.92 1.04 0.81 0.52 0.74 0.76 0.66 0.49 0.73 0.88
Change Mean SD
6.6 6.5 6.3 6.2 6.2 6.7 6.8 6.7 6.0 6.9 6.7 6.6 6.5 6.7 6.8 6.6 6.5 1.17 1.46 1.13 0.98 0.98 1.26 1.17 1.18 0.98 1.05 0.77 1.01 0.94 0.80 0.68 1.05 1.04
Team Mean
5.8 6.1 5.5 5.5 5.3 5.3 5.8 5.6 5.3 6.3 5.7 5.9 5.3 6.6 6.7 6.4 5.7
SD
1.14 1.64 0.93 0.99 0.76 0.81 1.28 1.58 0.49 1.02 0.46 0.38 0.73 1.09 1.01 1.42 1.06
519 765 355 Hospital 2 2574 1035 Hospital 3 438 627 Hospital 4 531 102 242 Hospital 5 1673 621 1732 Hospital 6 827 384 202 Total 12627 *N=Number of day entries by nurses.
On average, 42.9% nurses on study units needed more time on their shift to provide the level of patient care specified in the nursing care plan. These results are presented by hospital unit in Table 36. Table 36: Percent of Nurses Reporting Additional Time Needed to Provide Quality of Care, by Hospital Unit
Unit Type N* Hospital 1 CCU 475 IP 702 SDU 304 Hospital 2 CCU 1976 IP 889 Hospital 3 CCU 393 IP 553 Hospital 4 CCU 478 DS 96 IP 217 Hospital 5 CCU 1655 DS 612 IP 1677 Hospital 6 CCU 576 IP 250 SDU 175 Total 11028 *N=Number of day entries by nurses. no more time needed 31.6 28.5 30.3 60.7 46.2 49.6 25.7 55.9 33.3 27.6 85.1 78.4 71.8 51.9 51.6 13.7 57.1 % <15 min. 11.2 15.8 5.3 5.1 8.8 7.9 14.1 5.0 11.5 3.7 0.8 1.1 2.1 3.3 6.8 11.4 5.6 % 15-30 min. 18.5 27.4 28.6 12.9 20.4 16.0 27.5 12.8 34.4 21.2 6.3 8.3 11.6 16.3 20.0 27.4 15.4 % 31-45 min. 13.9 14.1 13.8 7.9 11.9 8.9 14.6 8.4 7.3 17.1 3.3 5.9 4.9 13.0 12.8 21.7 8.9 % 46-60 min. 13.9 7.0 9.2 6.1 5.6 9.2 9.2 9.0 5.2 14.7 3.3 3.9 5.2 8.3 5.6 12.6 6.6 % >60 min. 10.9 7.3 12.8 7.3 7.1 8.4 8.9 9.0 8.3 15.7 1.2 2.3 4.4 7.1 3.2 13.1 6.3
196
7. Patient Outcomes
Patient outcomes in this section include medical consequences (from the Patient Data Form) and patient health status change (from the Patient SF-12 Health Status Survey). 7.1 Medical Consequences Data on medical consequences were collected by data collection staff via health records, who recorded the number of negative events and medical complications patients experienced during their hospital stay. Negative events included an unexpected return to the Intensive Care Unit (ICU), a fall with injury, a readmission within three months, and a recorded complication in the health record. Medical complications tracked included urinary tract infection, pneumonia, deep or superficial surgical wound infection, bedsore, and thrombosis. As shown in Table 37, medical consequences were rare for all six hospitals, although these tended to be somewhat higher for Hospitals 4 and 2 than for other hospitals. Hospital 4 had the highest incidences of medical complications. No hospitals reported medical errors with patient consequences except for Hospital 1 (one case reported). Nearly one-third (32%) of patients have been hospitalized for the same condition in the past 3 months. The readmission rate is the highest in Hospital 1 (39%) and lowest in Hospital 6 (12%), where the nurses had low workload to worked hours ratio. In addition, Hospital 1 reported higher incidences in returned to ICU post-op than other hospitals. Table 37: Medical Consequences Percent Reporting Yes to the Items in the Table, by Hospital
Hospital Number of cases Returned to ICU post-op Falls with Injury Urinary tract Infection Pneumonia Wound Infection Bedsores Thrombosis Medication Errors with consequences Admitted in past week with same diagnosis Hospitalized for the same condition in the past 3 months 1 189 3.7 0.0 1.1 2.1 0.0 0.0 0.0 0.5
1.6
2 243 0.4 0.8 2.5 2.1 3.3 0.4 0.0 0.0

1.6
3 259 3.1 0.8 0.0 0.4 0.8 0.4 0.0 0.0

1.2
4 195 3.6 0.0 3.6 2.6 2.1 1.5 1.0 0.0

6.0
5 285 0.7 1.4 1.1 0.4 0.7 0.0 0.0 0.0

0.4
6 59 0.0 0.0 0.0 0.0 1.7 0.0 0.0 0.0

1.9
Total 1230 2.0 0.7 1.5 1.3 1.4 0.4 0.2 0.0
1.6
38.6
36.7
29.3
31.7
30.6
12.3
32.0
197
7.2 OMAHA Scores at Discharge and Change from Admission Patient OMAHA scores at admission (or appearance of new diagnosis) were presented earlier in Table 12. Table 38 shows the mean OMAHA scores at time 2. For all study patients, the means were 3.8 (SD=0.67) for knowledge, 4.2 (SD=0.52) for behaviour, and 4.1 (SD=0.62) for status. Patients from Hospital 2 had the highest scores across all three dimensions. Patients from Hospital 1 ranked low for all dimensions at time 2, probably due to low scores at admission. Table 38: OMAHA Scores at Time 2 (Resolution of Diagnosis or at Discharge), by Hospital
Table 39 shows the mean change scores for each hospital for knowledge, behaviour, and status between time 1 (at admission or appearance of new diagnosis) and time 2 (at discharge or resolution of diagnosis). Table 39: Differences in OMAHA Scores Between Time 1 and Time 2, by Hospital
7.3 Health Status at Discharge and Change from Admission Patient physical and mental functional status at admission was presented earlier in Tables 13 and 14. Tables 40 and 41 show the changes in patient functional status from admission to discharge. As can be seen, more than 40% of patients had improved physical and mental health at discharge (41% for physical health and 42% for mental health) and about 45% of patients had physical and mental scores at discharge lower than at admission. For patients who reported an improvement in health, Hospital 1 had the highest change score among all hospitals (9.39 for physical health and 10.81 for mental health). For patients who had a decrease in the score, Hospital 3 had the smallest change (-7.37 for physical health and -4.53 for mental health).
198
Table 40: Change in Patient Physical Health Status (SF-12) from Admission to Discharge
Hospital N 1 2 3 4 5 6 All hospitals Percent
64 53 116 64 102 15 414
Increased Mean (SD)

9.39 (7.72) 6.19 (5.72) 7.06 (5.71) 8.41 (7.48) 7.63 (6.87) 7.13 (5.32) 7.66 (6.65)
Decreased N
83 96 71 81 116 18 465
No Change N
15 12 62 6 28 6 129
Mean (SD)
-9.57 (7.99) -10.69 (9.00) -7.37 (7.22) -10.05 (8.14) -8.77 (7.17) -7.44 (6.24) -9.27 (7.91)
41.1%
46.1%
12.8%
Table 41: Change in Patient Mental Health Status (SF-12) from Admission to Discharge
Hospital N 1 2 3 4 5 6 All hospitals Percent
61 68 112 77 90 18 426
Increased Mean (SD)

10.81 (9.82) 8.78 (8.77) 6.31 (7.09) 8.70 (5.90) 7.20 (7.01) 5.66 (3.72) 7.94 (7.64)
Decreased N
86 81 75 68 128 15 453
No Change N
15 12 62 6 28 6 129
Mean (SD)
-8.81 (6.78) -8.27 (6.75) -4.53 (5.41) -7.66 (6.33) -8.93 (7.65) -10.60 (8.25) -7.93 (6.97)
42.3%
44.9%
12.8%
8. Nurse Outcomes
This section presents results related to nurse burnout, job satisfaction, absenteeism, and intent to leave. 8.1 Burnout and Effort & Reward Imbalance Maslachs Burnout Inventory (MBI) and the Effort-Reward Imbalance (ERI) questionnaire were included in the Nurse Survey to measure burnout. The MBI is a 22-item scale with three subscales: emotional exhaustion (EE), depersonalization (DP) and personal accomplishment (PA). A seven point categorical scale ranging from never, a few times a year or less to everyday is used to examine nurses feelings about their work and patients. Items are split amongst the EE (9 items), DP (5 items), and PA (8 items) subscales. Table 42 displays the means and standard deviations. The highest MBI subscale scores amongst Hospitals 3, 1, and 5 were associated with the three highest proportions of nurses who ranked below the US norms for mental health for females (Table 48). Table 42 shows nurses who worked at Hospital 3 averaged the worst job-related feelings of emotional exhaustion, while those at Hospital 1 scored highest on depersonalization aspects. The Evidence-based Staffing 199
nurses who worked at Hospital 5 were least likely to claim that they had personal accomplishments. Table 42: Burnout Mean Scores of MBI Subscales, by Hospital
Hospital N 1 2 3 4 5 6 Total 58 171 71 82 199 108 689 EE Mean 25.0 20.8 26.3 20.6 23.1 23.3 22.7 SD 12.1 9.6 10.8 10.2 10.2 10.5 10.5 N 58 173 74 79 205 108 697 DP Mean 7.6 5.5 7.1 4.6 5.9 6.0 6.0 SD 6.4 5.3 5.6 4.4 5.1 4.9 5.2 N 56 163 70 82 205 101 677 PA Mean 10.9 12.2 12.5 10.8 13.3 11.6 12.2 SD 7.0 7.1 5.9 5.8 7.0 6.8 6.8
Note: EE=Emotional Exhaustion Index DP=Depersonalization Index PA=Personal Accomplishment Index Difference by hospital is statistically significant at p<0.05 for all three subscales.
The Effort-Reward Imbalance (ERI) scale, a 17-item scale that measures the balance between nurses efforts and their rewards, was also used to assess burnout. The ERI uses a four point scale to measure the extent of stress, from I am not at all distressed to I am very distressed. The greater the ERI score, the greater the individuals distress. To determine which proportions of nurses were at risk for ERI or MBI emotional exhaustion, the ERI and MBI (EE) variables were dichotomized. The ERI already has values of 0 or 1, while the MBI (EE) was dichotomized using values greater than 27 as per instrument guidelines. The proportion of nurses at risk is displayed in Table 43. Nurses in Hospital 3 were most at risk for emotional exhaustion (43.7% vs. 22.2%-36.2%). These results are consistent with the MBI EE scores in Table 42, as well as the mental health data in Table 48 where Hospital 3 had the largest proportion of nurses whose mental health was below the US population norm. Nurses in Hospital 3 were also more likely to feel imbalances in effort and reward than nurses in other hospitals (25.8% vs. 10%-22%). Table 43: Burnout Percent of Nurses at Risk for Emotional Exhaustion and ERI, by Hospital
Hospital 1 2 3 4 5 6 Total N 58 171 71 82 199 108 689 % at Risk for MBI_EE 36.2 22.2 43.7 25.6 29.6 33.3 29.9 N 50 141 62 70 166 88 577 % at Risk for ERI 22.0 17.7 25.8 10.0 16.3 20.5 18
Note: ERI=Effort-Reward Imbalance (>1, at risk) MBI_EE=Maslachs Burnout Inventory-Emotional Exhaustion (>27, at risk) Difference by hospital is not significant (2=6.8, df=5, p>.05)
200
8.2 Autonomy and Control Table 44 shows the subscale response from the Revised Nursing Work Index (R-NWI). The NWI measures nurses job satisfaction and perception of quality of care. For each of the 41 items in the survey, the nurses responded on a 4-point forced choice Likert scale, with choice ranging from 1=strongly disagree to 4=strongly agree. No neutral response options are provided. Five subscales were created and listed by hospital. Higher scores reflect better working conditions and job satisfaction. Hospital 4 showed the best scores for all five subscales: autonomy, control over patient care, nurse-physician (NP) relationship, leadership, and resources available. Table 44: Nurse Work Index Subscales, by Hospital
Hospital
1
Autonomy N Mean (SD)

55
Control N Mean (SD)
NP Relation N Mean (SD)
Leadership N Mean (SD)
Resource N Mean (SD)

58 184 73 84 206 108 713 9.7 (1.44) 10.2 (1.37) 9.5 (1.57) 10.4 (1.74) 10.0 (1.51) 9.9 (1.34) 10.0 (1.49)
16.2 58 17.0 59 7.4 52 28.2 (3.28) (3.31) (2.15) (5.54) 2 173 16.4 181 18.5 182 8.6 167 30.1 (2.83) (3.71) (1.76) (5.72) 3 69 15.2 73 18.1 73 8.1 69 28.7 (3.23) (3.80) (2.01) (5.32) 4 81 83 85 79 20.6 9.2 32.7 17.6 (3.06) (4.22) (1.86) (6.26) 5 203 15.6 204 19.0 208 8.1 201 27.5 (3.11) (3.98) (1.99) (5.65) 6 101 14.7 108 17.5 110 8.4 104 26.2 (3.18) (3.42) (1.92) (6.26) Total 682 15.9 707 18.6 717 8.4 672 28.8 (3.18) (3.81) (1.97) (6.09) Note: Difference by hospital is statistically different at p<0.002 for all nurse work indices.
8.3 Job Satisfaction Table 45 shows the percent of dissatisfied nurses for each hospital. On average, 17.0% of the nurses were dissatisfied with their current jobs. Variation in dissatisfaction in present job across hospitals was minimal with the exceptions of Hospitals 1 (28.9%) and 4 (9.3%). Satisfaction with current job was only moderately correlated with satisfaction of being a nurse (r=0.474, p<0.001). In five hospitals, 37.6-49.0% of the nurses were dissatisfied with their opportunities to interact with management. The one notable exception was Hospital 6 with over 63% of nurses reporting this issue. In four of the six hospitals only 15.3-22.7% of nurses were dissatisfied with the amount of responsibility. Hospitals 1 and 3 reported notably higher scores at 43.9% and 42.3% respectively. In five hospitals, 11.0-15.1% of nurses were dissatisfied with opportunities for social contact at work and 19.2-27.9% with opportunities for social contact after work. The one exception was Hospital 2 where nurses appeared more satisfied with social opportunities. Evidence-based Staffing 201
Table 45: Job Satisfaction Percent of Nurses Dissatisfied, by Hospital

Hospital Opportunities to interact w/ Mgmt Amount of Responsibility Social Contact after work Satisfaction - present job Satisfaction - nurse Social Contact work Number of Cases 1 40.4 43.9 26.3 28.6 21.1 12.3 57 2 37.9 18.7 16.6 14.1 9.2 4.9 181 3 40.8 42.3 19.2 19.7 21.9 13.9 73 4 37.6 15.3 27.9 9.3 10.5 15.1 86 5 49.0 19.8 20.2 18.2 20.2 11.0 208 6 63.6 22.7 22.0 18.2 13.6 13.6 109 Total 45.5 23.6 20.9 17.0 15.4 10.7 714
Note: (1) Items are sorted by percent in the Total column. (2) Number of valid responses varied slightly from item to item.
8.4 Health Status Table 47 shows the physical and mental health levels for nurses in each of the hospitals as measured by the SF-12. The average physical health score for nurses was 50.1, slightly above the US norm for females in the general population (49.11). However, the average mental health score (47.3) was below the US female norm (49.42). Hospital 4 had the highest mean physical health scores (53.0), and Hospital 6 had a slightly better mean mental health score than those of others hospitals (48.2). Since the SF-12 questionnaire was completed by both patients and nurses, comparisons are possible. The overall average mental health score for nurses (47.3) was lower than the average scores for patients, either at admission (48.2) or discharge (48.1) as noted in Tables 13 and 15. Table 46: Nurse Health Status, by Hospital
Hospital 1 2 3 4 5 6 Total N 58 178 73 80 204 106 699 Physical Health Scale Mean SD Range 50.7 8.9 15.3-62.2 50.3 8.7 17.6-65.2 48.9 10.2 21.6-64.7 5.3 36.9-56.0 53.0 49.1 8.6 22.6-65.4 50.2 8.1 18.9-64.6 50.1 8.5 15.3-65.4 Mental Health Scale Mean SD Range 45.1 12.2 23.7-63.9 48.1 10.0 16.3-64.1 45.9 10.5 18.9-63.8 47.9 9.9 18.7-62.1 47.1 10.5 17.4-62.6 10.4 20.7-63.8 48.2 47.3 10.4 16.3-64.1
Table 48 gives the proportion of nurses in each hospital that scored below the norm for females in the general US population, and thus was not considered healthy. Only 34.8% of nurses scored below the population norm, indicating that, in general, nurses are physically healthier than the general US female population. Nearly half of nurses (49.2%) were categorized as not mentally
202
healthy, which was comparable to the categorization of patients at admission (49.2%) and discharge (51.9%) as noted in Tables 14 and 16. Table 47: Nurse Health Status, Percent of SF-12 Scores Less than US Norm for Females, by Hospital
Hospital 1 2 3 4 5 6 Total N 58 178 73 80 204 106 699 % Not Physically Healthy
31.0 34.3 38.4 20.0 39.7 36.8 34.8
% Not Mentally Healthy

53.4 46.1 54.8 47.5 52.0 44.3 49.2
Norm for Female3 Physically not healthy: Physical Health Scale < 49.11 Mentally not healthy: Mental Health Scale < 49.42
8.5 Violence at Work Table 49 presents the prevalence of violence and abuse towards nurses. Among 720 nurses completing the survey, 24.9% experienced emotional abuse, 13.6% experienced the threat of assault, and 10.2% experienced physical assault. Hospital 6 reported the highest proportion of emotional abuse and threats of assault. Hospital 1 reported the highest percentage of physical assaults. The smallest proportion of nurses who reported emotional abuse and threats of assault were in Hospital 4. More than 97% of physical assault or threats of assault were from patients (data not shown). Table 48: Prevalence of Violence Percent of Nurses Reporting Yes to the Items in the Table, by Hospital
Hospital Physical Assault Threats of Assault Emotional Abuse Number of cases 1 3.4 8.5 25.4 59 2 10.8 15.7 17.4 184 3 11.0 13.9 27.4 73 4 2.4 5.9 25.9 85 5 13.8 13.9 26.8 209 6 11.8 18.2 30.9 110 Total 10.2 13.6 24.9 720
Note: Statistically significant for physical assault (2=12.1, df=5, p<0.05), but not for threat assault or emotional abuse.
The sources of emotional abuse varied as evidenced in Table 50. The major source of emotional abuse was from patients (31.1%), followed by abuse from other nurses (21.5%), and then from physicians or patients families.
203
Table 49: Source of Emotional Abuse, by Hospital

Hospital Patient Family MD Nurse Any two combined Any three combined Any four combined Number of cases 1 33.3 0.0 0.0 26.7 33.3 6.7 0.0 15 2 28.1 12.5 12.5 28.1 12.5 3.1 3.1 32 3 26.3 15.8 26.3 10.5 15.8 5.3 0.0 19 4 22.7 13.6 13.6 36.4 13.6 0.0 0.0 22 5 33.9 10.7 19.6 17.7 7.1 7.1 3.6 56 6 36.4 9.1 15.2 15.2 18.2 3.0 0.0 32 Total 31.1 10.7 15.8 21.5 14.1 4.5 1.7 176
9. System Outcomes
9.1 Quality of Care Table 53 presents the quality of care and likelihood that tasks are delayed or accomplished. Quality of nursing care is shown as the percent of nurses reporting fair/poor care delivered to patients; quality of patient care is shown as the percent of nurses reporting quality of care deteriorated. Hospital 1 reports an overall high level of tasks not done or delayed whereas Hospital 4 tends to report low levels in almost all categories. When faced with a shortage of time, different decisions appear to be made about which tasks can be left undone or delayed. Nurses are more likely to complete tasks that are ordered by physicians such as vital signs, medication, and dressings. We hypothesize that delayed actions can also have negative consequences in terms of patient clinical outcomes, patient satisfaction, and system costs.
204
Table 50: Quality Issues Percent of Nurses Reporting Yes to Items in the Table, by Hospital
Hospital Quality of Care Number of Cases Nursing Care: Fair/Poor Care Delivered Patient Care: Quality of Care Deteriorated Not Done Number of Cases Care Plan* Comforting/Talking Back/Skin Care Oral Hygiene Pt/Family Teaching Documentation Mobilization/Turns Discharge Prep VS/Meds/Dressings Mean # Tasks Not Done Delayed Number of Cases VS/Meds/Dressings Mobilization/Turns Call bell Response PRN pain meds # Tasks Delayed 1 56 31.0 58.9 58 84.6 62.1 43.1 60.3 46.6 42.1 19.0 25.9 6.9 3.19 58 51.7 17.2 53.4 37.9 1.58 2 182 10.9 40.1 184 64.9 34.8 27.2 41.8 18.6 31.5 13.0 13.7 5.4 1.97 184 40.8 40.2 20.7 13.9 1.14 3 73 20.5 61.6 73 57.6 45.2 31.5 31.5 26.0 28.8 13.7 12.3 2.7 2.35 73 47.9 35.6 30.1 31.5 1.43 4 85 10.6 15.3 86 37.6 25.6 23.3 12.8 18.6 15.1 9.3 7.0 3.5 1.52 86 24.4 20.9 19.8 14.0 .79 5 209 8.1 39.7 210 42.9 35.7 32.9 17.1 20.0 11.4 6.2 10.5 1.0 1.78 210 28.6 27.1 25.2 9.0 .90 6 109 15.5 47.7 111 51.8 44.1 36.0 22.5 27.0 20.7 27.0 4.5 7.2 2.38 111 43.2 31.5 23.4 17.1 1.14 Total 714 13.4 41.9 722 48.2 38.6 31.4 28.7 23.3 22.6 13.3 11.4 4.0 2.06 722 37.3 30.5 25.9 16.6 1.09
*For care plan, the valid number of cases is low for Hospitals 1 (N=13), 2 (N=37) and 3 (N=59). Note: Items are sorted by percent in the Total column
9.1 Absenteeism The Nurse Survey asked: In the past year: a) On how many occasions (episodes) have you missed work due to illness/disability? b) How many shifts have been missed due to illness/disability? Table 51 shows that the total number of missed work episodes from the six survey hospitals was 1,768. Individual nurses missed from 1 to 56 episodes (data not shown). Of the 683 nurses who reported missed occasions in the past year, 42.9% missed 1-2 episodes, 25.2% missed 3-4 episodes, and 15.5% missed more than four episodes. Another 16% indicated nil work episodes missed. Once the number of episodes was grouped into four categories, the pattern of distribution varied amongst hospitals. Hospitals 6, 2, 3, and 1 had the largest proportions in order of ascending categories from 0 episodes to >4 episodes. On average, Hospital 6 averaged the highest missed shifts per episode (3.71; SD=7.7) shifts.
205
Table 51: Absenteeism Percent of Episodes Absent and Mean Shifts per Episode in the Past Year, by Hospital
Hospital 1 2 3 4 5 6 Total Total # episodes 226 382 172 252 513 225 1768 N 57 174 71 81 193 107 683
%0 episodes % 1-2 episodes % 3-4 episodes %>4 episodes
N 49 145 56 57 162 78 547
8.8 14.4 19.7 18.5 14.5 23.4 16.4
31.6 53.4 32.4 45.7 41.5 39.3 42.9
24.6 24.1 35.2 22.2 22.8 27.1 25.2
35.1 8.0 12.7 13.6 21.2 10.3 15.5
Mean (SD) shifts/episode 1.61 (0.90) 2.47 (4.58) 2.31 (2.77) 1.39 (0.61) 2.41 (5.32) 3.71 (7.70) 2.42 (4.85)
Table 52 presents the common reasons for nurse absenteeism. The primary reason was physical illness (71.4%). Other reasons were mental health days and injury. Nurses at Hospitals 2 and 3 indicated physical illness most frequently, while Hospital 1 nurses tended to report mental health day, injury, and other. Table 52: Absenteeism Most Common Reason to Miss Work in the Past Year, by Hospital
Hospital 1 2 3 4 5 6 Total N 56 167 63 73 194 94 647 % Physical illness 50.0 77.8 77.8 72.6 73.2 63.8 71.4 % Mental health day 12.5 5.4 4.8 4.1 4.1 5.3 5.4 % Injury 8.9 4.8 3.2 2.7 4.1 6.4 4.8 % Other 28.6 12.0 14.3 20.5 18.6 24.5 18.4
9.2 Intent to Leave Table 46 shows the percent of nurses who expected to lose their job or intended to leave within the next 6 months. Hospital 1 nurses not only reported the highest levels of dissatisfaction with their current job (Table 45), but also the highest levels of intent to leave. Surprisingly, nurses in this hospital were most likely to expect difficulties in finding a new job. Hospital 5 had the highest percent of nurses expecting to lose their jobs, while not a single nurse in Hospital 1 expected to lose theirs in the next 12 months. Table 53: Intent to Leave Percent of Nurses Reporting Yes to the Items in the Table, by Hospital
Hospital Number of Cases Expect to lose job Plan to Leave 6 months Difficult finding job 1 58 0 10.3 14.0 2 186 1.6 3.8 1.6 3 74 2.7 6.8 2.8 4 86 3.5 3.5 2.3 5 207 12.1 5.3 10.6 6 110 5.5 3.6 3.6 Total 721 5.4 5.0 5.7
206
References
1. Karasek, R. & Theorell, T. (1990). Healthy Work: Stress, Productivity, and the Reconstruction of Working Life. New York: Basic Books, Inc., Publishers. 2. Martin, K. S., & Scheet, N. J. (1992). The OMAHA System: Application for Community Health Nursing. Philadelphia, PA: WB Saunders. 3. Ware, J., Kosinski, M. & Keller, S. (2002). SF-12: How to Score the SF-12 Physical and Mental Health Summary Scales. Fourth Edition, QualityMetric Incorporated, Lincoln, Rhode Island, and Health Assessment Lab, Boston, Massachusetts. 4. Canadian Institute for Health Information. (2003). Workforce Trends of Registered Nurses in Canada, 2002. Ottawa, ON: Author. 5. OBrien-Pallas, L., Thomson, D., Alksnis, C., Luba, M., Pagniello, A., Ray, K. L., & Meyer, R. (2003). Stepping to Success and Sustainability: An Analysis of Ontario's Nursing Workforce. Toronto, ON: Nursing Effectiveness, Utilization, and Outcomes Research Unit, University of Toronto. 6. Chagnon, M., Audette, L. M., Lebrun, L., & Tilquin, C. (1978). Validation of a patient classification through evaluation of the nursing staff degree of occupation. Medical Care, 16(6), 465-475.
207

Evidence Based Standards For Measuring Nurse Staffing and Performance

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Evidence Based Standards For Measuring Nurse Staffing and Performance

Uploaded by

Copyright:

Available Formats

Evidence-based Standards for Measuring Nurse Staffing and Performance Project # RC1 0621 06 Report for the Canadian

Key Implications for Decision Makers

Appendix A. Annotated Bibliography

2. Bull, Hansen, Gross (2000)

Discharge planning model for elders hospitalized with heart failure.

Predicting resource utilization in patients with unstable angina.

4. Crilley, Farrer (2001)

Impact of a first myocardial infarction on self-perceived health status.

SF-12 patient health status

5. Czar, Engler (1997)

Perceived learning needs of patients with coronary artery disease.

6. Hemingway, Marmot (1999)

Psychosocial factors in the development & prognosis of coronary heart disease.

7. Johnson, Stewart, Hall, Fredlund, Theorell (1996)

Patient occupation, social support, patient demographics education.

8. Marchette, Holloman (1986)

Variable affecting length of stay.

Medical diagnosis, patient demographics, patient education given.

Direct relationships between longer length of

10. Shih, Chu, Yu, Hu, Huang, (1997)

Turning points of recovery from cardiac surgery in an intensive care unit.

11. Siegrist (1996)

12. Siegrist, Peter, Junger, Cremer, Seidel (1990)

13. Silber, Rosenbaum, Ross (1995)

Predictors of hospital outcomes.

Staffing ratios, medical diagnosis, patient demographics, hospital size, Occupation,

Focus the cause of coronary heart disease.

Inputs social support, medical diagnosis

Authors, Year 15. Bruce, Sale, Shamian, O'Brien-Pallas, Thomson (2002).

16. Coutts (2001)

Authors, Year 17. Josephson, Lagerstrom, Hagberg, Hjelm (1997)

Focus Musculoskeletal symptoms & job strain in nurses.

Outputs Nurses health status.

18. Josephson, Vingard, MUSICNorrtalje Study Group (1998)

Comparison of lowback pain in female nurses & other employed women.

Conflicting demands, job strain.

19. Toomingas, Theorell, Michelsen, Nordemar (1997)

20. Wunderlich, Sloan, Davis (1996)

Summary report from Institute of Medicine about adequacy of nursing in hospitals

Focus & nursing homes.

Inputs hours, staffing levels.

System Characteristics and Behaviours

Authors, Year 21. Aiken, Smith, Lake (1994)

22. Aiken, Clarke, Sloane, (2002).

Workload, proportion of nurse worked hours, organizational support.

Nurse burnout, job satisfaction, nurses perceived quality of care.

23. Aiken, Clarke, Sloane, Sochalski, Silber (2002).

24. Arthur, James (1994)

Hospital ownership, performance, & outcomes.

6,097 hospitals in the U.S.

26. Blegen, Vaughn (1998)

Nurse staffing & patient occurrences.

27. Blegen, Goode, Reed, (1998).

42 inpatient units in an 880-bed university hospital.

Patient acuity: (unique Patient Classification System; levels

Inputs range between 1 and 7 most acute/most care)

28. Brown (2001)

Synopsis of ANA study on relationship between staffing & patient outcomes.

Nearly 13 million patients in 1500 hospitals from 9 states

29. Buerhaus (1997)

Mandatory minimum nurse staffing levels in hospitals

Proportion of nurse worked hours, nurse-to-patient ratio