Public Disclosure Authorized

Po-lcy, Planning,
WORKING

and Research

PAPERS

Healthand Nutrition Population HumanResources and Department TheWor;dBank July 1989 WPS173

Public Disclosure Authorized

Public Disclosure Authorized

Hospital Management Staffing and Training Issues

Julio Frenk, Enrique Ruelas, and Avedis Donabedian

Public Disclosure Authorized

Hospitals dominate health care, se making hospitals more efficient is crucial to better health care delivery. The authors suggest an agenda for research.

The Policy, Planning, and Research Cxrnplcx dasinbutes PPR WorkingPapers to disseminatc the finidngs of work in progrcss and to enonurage the exchange of ideas amoangBank staff and aUothers interested in devclopmcnt issues Thesc papers carry the names of the authors, reflect only their views, and should be used and cited accordingly Thc findings interpretauons, and conclusions arc the authors' own.They should not be attrbuted to the World Bank, its Board of lirectors, its management, or any of its member countrcs

and Research Plc,Planning,

Healthand NutrtitonI

Hospitals dominate health care in most parts of the world and for a variety of reasons are likely to continue being a key factor in the overall performance of the health care system. Any efforts to improve this performance must therefore give greater hospital efficiency the highest priority. Aftcr discussing key issues of managerial, clinical, and production efficiency, Frenk, Ruelas, and Donabedian suggest an agenda for research, which would include two types of research: Observational studies that document levels of hospital performnanceand correlate them with variorganizational design and environrmental ables. It is especially imponant to devise and test sensitive, specific indicators of managerial, clinical, and service production efficiency. - Comparative intervention studies that would introduce planned change in hospitals and assess the consequences - using control groups as well as cost-benefit and cost-effectiveness analyses.

The mostly highly recommended subjects for research! in order of priority, are:
* Good descriptive studies of the hospital system and the main aspects of organization design - to chart, for example, the formal and informal relations among managers and clinicians, the frequency of different arrangements for intemal communication, types of departmentalization, and management systems. * The systematic design, testing, and study of explicit quality monitoring and assurance systems. Such studies should include the analysis of interactions between managers and clinicians, especially as they constrain clinical autonomy and decision making. * Studies to determine which social, personal, organizational, and educational factors account for managerial skill and success in managing a hospital - to get the information needed for the recruitment and training of successful hospital managers. * Studies of the structure and dynamics of medical labor markets, to improve understanding of why there is an oversupply of doctors in so mary different countries.

This paper is a product of the Health and Nutrition D.vision, Population and Human Resources Department. Copies are availablefree from the World Bank, 1818H Street NW, Washington DC 20433. Please contact Sonia Ainsworth, room S6-065, extension 31091 (37 pages with chans).

The PPR Working Paper Series disseminates the findings of work under way in the Bank's Policy, Plaruning,and Research Complex. An objectivc of the series is to get these findings out quickly, even if presentations are less than fully polished. The findings, interprctations, and conclusions in these papers do not necessarily represent official policy of the Bank. Produced at the PPR Dissemination Center

l E-

Table of Contents

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ANALYTICAL FRAMEWORK ..... BASICISSUES .

. .. . .. .. .. .. .. ....
.

1 4 11 11 12 25 29

............. .. .. ............ MeasurementIssues.. . . . . . . . . . . . . . . . . . . . . . . SubstantiveIssues ..... . . . . . . . . .. ...... . . .

. .

ISSUES RELATEDTO THE TRAININGOF HOSPITALMANAGERS . . ... TOWARDS A RESEARCHAGEMDA .....

. .

. .

. .. ... ... . .... . ..

ACKNOWLEDGEMENTS

We are grateful to Dr. Willy De Geyndt for his advice on the structure and content of this paper. We would also like to acknowledge the help of Beatriz Zurita, Michal Frejka, and Luis Miguel Vidal in completing the literature search and organizing the material. The shortcomings of this paper are solely the responsibility of the authors.

Throughoutthe world, hospitalshave come to epitomizemodern medical care. For many years, a major policy concern in the health arena was to provide communities with enough hospitals. More recently,however, the focus of concern has shifted to the overdominantrole of the hospitalwithin the health system. In developed countriesthere is an excess of beds. In most developingcountries the concern is that, even without having fully satisfied overall requirementsfor hospitals,they already absorb such a high proportionof resourcesthat they seriouslythreatenany effort to achieve full coverageof the population. Furthermore, is widely believed it that a health care system centeredaround hospitals is intrinsically incompatible with the geographic,economic,and culturalattributesof many populations. In addition,the mix of servicesoffered by hospitals emphasizingacute, episodic,and curativeactivities-- is believedto poorly match the prevailingepidemiologic profile and the populationneeds for preventiveand continuouscare. This inconsistency becomingeven is more marked as an increasing epidemiologic transition, number of countr,es whereby chronic ailments undergo a profound

are becoming more

important,with the ensuing requirementsfor long-termservices that most general hospitalshave traditionally had difficultysupplying(Omran 1971; Frederiksen1969). As with physician supply hospitalsseem to have moved from deficit to excess without ever having achievedsome kind of equilibrium (Starr 1982, pp. 421-427). Evidently,a health system dominatedby hospitals is not the only possible organizational model. In fact, for most of the history of health

2 care hospitalsrepresenteda rather marginal element. As Foucault (1978) points out, during a long period of time the hospitalwas a nonmedical institution, and medicinewas not a hospital-based profession. "The hospital as a therapeuticinstrumentis a relativelymodern concept,dating from the end of the eighteenthcentury" (Foucault1978). Since then, a number of social, economic,scientific,and technologicchanges,which have been summarizedelsewhere (in particularby Rosen 1972), have made the hospital the "fulcrumof care" (Berki 1972, p. 8). The dominance of hospitals is one of the most striking featuresof convergenceamong the health systems of countriesat all levels of economic developmentand with all forms of politicalrepresentation (Mechanic1975; Frenk and Donabedian 1986). Togetherwith the importantprogressthat they have produced,hospitalshave also given rise to the set of concerns mentioned above. As the ambitious goal of achievingHealth for All by the Year 2000 is universallyadopted, it becomes increasingly crucial to understandthe functioningof that segment of the health care system where most resourcesare spent. UNICEF has estimated that, while in many countries 85% of the national health budget is spent in hospitals,these serve less than 10% of the population. For example, in Mexico hospitals represent less than 1% of all the health care facilitiesof the Ministryof Health, but employ over 40% of the Ministry'sphysiciansand nurses (Secretariade Salud 1985, pp. 213-319). Many countriesface, therefore,a double concentration health care: of geographicconcentration large urban areas and technological in concentration large hospitals (Soberonet.al. 1986). The problem is in further compoundedby the effects of concentration the distribution on of resources. For instance,in many countriesefforts at regionalization have

3 been bedeviledby the tendency of hospitals to mix all three levels of care. This is.in part due to the weaknessof primary health care (PHC), which makes it necessary for the outpatientdepartmentsof many hospitals to become major providersof first-contact care. Thus, the concentration of resources in hospitals is both a cause and an effect of the weakness of PHC. Another reason for the mixture of levels of care is the tendency towards "tertiarization" many general hospitals. In either case, the end result of is the lack of clear patterns of patient referral,the difficultyof assigningdefined populationbases to differenttypes of health care facilities,the coexistencein the same facilityof cases with wide variationsof complexity,and the inefficient use of resources. Because their central position is likely to be maintainedin the foreseeablefuture, hospitalswill continueto be major determinants the of overall performanceof the health system. Any efforts to improve this performancemust therefore give the highest priorityto hospitalefficiency. This is the perspectivethat guides the present paper. The purpose of the paper is to discuss some fundamentalissues of hospitalmanagement,with special emphasison staffingand training. To achieve this, the paper is divided into three parts. First, an analyticalframeworkis presented that helps orient the discussion. Hospitalsare conceivedof as complex organizations, with goals, tasks, control systems,and relationships of authoritythat are articulatedin both formal and informalways (Scott 1966). The performanceof the hospital is conceptualized terms of three in different types of efficiency: managerial,clinical,and production efficiency. We also analyze the elementsof the internalorganization design and of the externalenvironmentthat influencethe level of performarnce analyzed. Second, some issues that refer to each of the are

4 elementsof the analyticalframeworkare identified. Finally,a research agenda that may help to better understandthe issues and thereby to improve the performance hospitals is presented. of

FRAMhWORK ANALYTICAL Figure 1 presents a schematicmodel for the study of hospital efficiency. This model begins by positing that there are two major groups

of actors in the hospital:managers and clinicians. Each of the two major types has many different subgroups. Among the managers, there are distinct to levels, ranging from members of the directorate, senior executives,to the middle and lower echelons. Clinicians,on the other hand, comprisea in variety of professions. Nevertheless, our discussionwe will focus on physicians,since they still constitutethe principalgroup of providers,in of terms of number, importance,autonomy,and economicconsequences the decisionsthat they make. Insofar as the same person can have both rather than managerialand clinical functions,we speak of ro'les occupationalgroups (Allisonet.al. 1983). This is particularly important for physicians,who often occupy importantadministrative positions in hospitals. For the purposes of this paper, when a physicianassumes the managerialrole, he or she will be considereda manager. As we shall discuss later on, one of the issues in health care organizationsis preciselythe convenience having physiciansperform administrative of functions. For the time being, however,the point is that the actors are conceivedof in terms of their roles and not of their professional origins.

...

.: :

ENVIRONMENT ( EPIDEMIOLOGIC, ECONOMIC, POLITICAL)

. . . .... .

..:...

:':~ ~ ~ ~~~~~~~~... ~ ~ ~ ~~~~~ ..........

.. .

.. . . .. . . . . .

ACTOR.S

I..TERVEI..G

. .:: :

OR

VARIABLES

. . N D S G

::'---

.....

. . . . . . . .

............

OCTIVES

PRODUCTS

....

........

Managerial

Decision Skills

Making

POLICIES-

MANAGFIERS

Managerial

J~MANAGERtIAL
EfICIENCY

SUPPORT SERVICESS

Managerial

Autonomy

HEALTH
Design of Production :' Process
.

SERVICE EFEICIENCY

CARE-: SERVICES

1ciA

Clinical p< Clinical

Decision Skills Autonomy

Making CLiiTICAL
EFFICIENCY

HEALTH

_________________Clinical
.~~

~ ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. ...

. .. ..

Figure I CONCEPTUAL FRAMEWORKFOR THE STUDY OF HOSPITAL 6FFIC[ENCY

6 Figure 1 shows the interaction the two basic groups of actors in the of frameworkof a complex organization, the hospital. This interactionis affected by the specificdesign that the organization adopts. Furthermore, the organizationitself is surroundedby an environment, where it interacts with other organizations and with formal and informalgroups of clients. Through its environment,the organizationis shaped, as we shall see later on, by complex epidemiologic, economic,and politicalprocesses. Within the context of specificenvironmentsand organizational designs, the core of Figure 1 portraysa dynamic conceptionof the interaction betweenmanagers and clinicians. Through the operationof certain intervening variables,the interaction generates a set of products. The quantity and quality of these products is determinedby the efficiencyof the organization. In this respect,we propose that there are three types of efficiency, which ought to be kept analytically distinct. We call these clinicalefficiency(CE), service productionefficiency(PE), and managerial efficiency(ME). The distinctionbetween clinicalefficiencyand productionefficiency has been proposedby Donabedianet.al. (1982). Basically,CE refers to the productionof health, however defined,whereas PE has to do with the productionof health services. Thus, CE is the extent to which a physician "combines,times, and sequencesservices...to produce the greatest increment of health, given a specifiedavailableor permissibleexpenditure" (Donabedian et.al., 1982). The combination, timirg, and sequencingof health services in the managementof a case is called a "strategyof care" by these authors. Hence, CE is the efficiencyof the strategiesof care. The clinicallyefficientstrategywill be the one that produces the largest improvement health for a given amount of expenditureor, alternatively, of

7 the one that produces a certain level of health with the least costly utilizationof resources. Needlessto say, the improvementin health status must be attributableto the strategyof care. It is clear that CE is a componentof the quality of care. The concept has the merit of combining health outcomeswith resourceconstraintsin the definitionof quality. As a componentof quality,CE is determinedby the appropriateness the of clinical decisionsto select a certain strategyof care, by the skill with which the strategy is carried out, and by the degree of clinicalautonomy, i.e., the extent to which the cliniciancan control the content of his/her own work (Freidson 1970, pp. 71-84). Even when a physicianhas selectedthe optimal strategyof care, there may be inefficiencies the process ox producingthe servicesthat form in this strategy, leading to a waste of resources. For example,there may be delays in processingor reportinglaboratorytests, or there may be a low occupancy rate, or the hospitalmay be using more costly personnel than warranted by the complexityof tasks. Donabedianet.al. (1982) suggest that such inefficiencies the productionof servicesshould not be considereda in part of the definitionof quality, althoughthey certainlyinfluencethe level of quality that is achievedper dollar of expenditure. As can be seen, PE is dependent,not on clinicaljudgement,but on the proper design of the service productionprocess, so that the amount of servicesspecified by a certain strategyof care can be produt at the lowest cost.

The concepts of clinicalefficiencyand s-rvice productionefficiency introducea useful distinctionin the analysis of the substantivefunction of a hospital, i.e., the productionof services that will generatean improvement health. In a parallel fashion to the notion of clinical of efficiency,Figure 1 proposesthe concept of managerialefficiency.

FIGURE 2

DIMENSION OF ANALYSIS Political

TYPE OF RELATIONSHIP Ezternal the State

.slatsions vith

Internal
Povr ad authority rolationx betvoe

managers clinicians

and

Economic

- Ovrall economic codit factor markets .

Characteristics t production

of

TYPOLOGY VARIABLES THAT AFFECTHOSPITAL PERFORMANCE OF

9 Dependingon the level of the manager, the product of ME are policies or support services: As in the case of CE, ME depends on the appropriateness of managerialdecisions,the skill in managing the organization (as evidenced,for example, in styles of leadership, capacityto solve conflict, handling of time, financialability,etc.), and managerialautonomy,either from the cliniciansor from officialsat higher levels of decisionmaking. The two main actors in our model -- cliniciansand managers -- interact in complexways. We have already seen that each group can interferewith the autonomyof the other. In addition,because of the characteristics of medical work, which is dominatedby professionals, both groups participate in the design of the productionprocess and thereforedeterminePE. The relationshipbetween managers and cliniciansdoes not occur in a vacuum. It takes place in the context of an internalorganization design, which in turn is surroundedby an externalenvironment. Followingthe work of Zald (1970),we can classifythe variablesthat operate inside or outside the organizationinto economicand political. Figure 2 presentsa framework for analyzingthese relationships. There are many potentialvariables for each cell in Figure 2. However,we have includedonly those that are most pertinent for the analysisof the contextualfactors that affect hospital efficiency. Let us first briefly identifythe variablesthat define the exchanges of a given hospitalwith its externalenvironment. On the political dimension of analysis,the main set of relationships refers, in most countries,to those that the hospitalmust establishwith the State, either because the hospital is part of a larger network of public organizations, and hence is owned by the State, or because it derivesmost of its income from social insurancefunds, or, at the very least, because the hospital is

10

subject to the regulatoryauthorityof the State (Frenk and Donabedian 1986). The hospital also faces a complex externaleconomicenvironment. At its highest level of aggregation,this environmentis formed by the overall economicsituationof a country. For example, economiccrises impose several constraintsthat require creativeresponseson the part of both private and public hospitals. At a more immediatelevel, the hospital interactswith various product and factor markets. Because this paper focuses mainly on issues of staffing,the variable that we considermost importantin this respect is the structureand dynamicsof the labor market, particularlythe professionallabor markets from which the hospitalmust recruit its managersand clinicians. In additionto the politicaland economicvariables shown in Figure 2, the external environmentof the hospital is defined by the epidemiological context of the area. As will be discussed later on, when this context is in rapid transitionit can severely strain hospitalresources. Moving to the internalsituationof the hospital, the most important political aspects of organizationdesign are those that specify the legitimatepower and authority relationships between physiciansand managers. In turn, the economicdimension centers around the design of the productionprocess. There are several economicmodels attemptingto understandthe hospitalas a firm (Jacobs 1974). For example,Harris (1977) has presenteda model based on internalsupply and demand functions. Regardlessof which model is adopted,some of the basic variables that need to be understoodin the internaleconomicorganization the hospital of includethe definitionof tasks (e.g., the mix of routine and nonroutine tasks), the divisionof labor, the service productionfunctions,and the systems for assuringthe quality of the product.

11

Figures 1 and 2 should not be seen as rigid depictionsof what are really very complex processes. They are not the only possible representation these processeseither.1 Instead,our conceptual of framework is meant simply as a guide to the identification and analysisof more specificresearch issues.

BASIC ISSUES In a first approximation, is possible to identifythree major groups it of issues that can orient the formulationof a researchagenda on hospital management. One group refers to issues of measurement. Indeed, it is necessary to develop and test specificand sensitive indicatorsof the various elementsthat are shown in Figures 1 and 2, especiallythe three types of efficiencythat we have proposed. The second and largest group of issues are substantive. In accordancewith our general frameworkof analysis, these include three subsets: (a) those that refer to the relationships the hospitalwith its external context;(b) issues about of the internalorganizationdesign; and (c) those that have to do with the core of organizational performance. Finally,the third large group of issues are related to the trainingof hospitalmanagers for efficiency. We will next examine each group of issues, so tnat we can then proceed, in the last section of this paper, to outline a researchagenda.

Neasurin

nt Issues

Because of the nature of this paper, we will not go into great detail in the analysis of the issues that deal with the operationalization and For a differentthough related approach,see Kovner and Neuhauser (1983).

12 measurementof the conceptsproposed in our analyticalframework. It should be pointed out, however, that a great amount of methodological work is required in order to answer such basic questionsas the following:
--

What are sensitiveand specific indicatorsof managerial,

clinical,and service productionefficiency?

- -

How can one assess such attributesas managerialor clinical

judgement,skills, and autonomy?

- -

What is the appropriatemeasure of hospitaloutput? If services

are considered,how should one account for the groups of activitiesthat go into a hospital day? Should certain by-productsof the hospital, such as information, professionaleducation,research,and referral,be included? If output is conceived in terms of health status,what measures are there availableto solve the problem of attribution, that a change in health so status is validly related to hospitalcare?
--

Given the multidisciplinary nature of hospital care, how can one

relate each output to the contribution distinct inputs? Conversely,how of can one assign specificportions of an input (e.g., time equivalentsof physicians)to the productionof multiple outputs? These are just a few of the methodological issues that would need to be solved in any specificstudy of hospitalperformance.

Substantive External

Issues Environinnt. economic, Whole disciplines and political are devoted to the study of conditions focuses that prevail in a

the epidemiologic,

society. On the other

hand, our main interest

on the performance

core of hospitals,especiallyas it is affectedby training and staffing. Hence, our analysisof the externalenvironment the hospitalwill of

13 necessarilybe limited. Nevertheless,it is fundamentalto keep in mind that no research agenda on managerial,clinical,or productionefficiency can be completewithout at least some consideration the environmental to conditionsthat shape the organization. A first problem arises in the precise definitionof what is external and what is internal. Indeed, definingthe boundariesof any organization, and especiallyof a human serviceorganizationsuch as a hospital is not a straightforward matter (Hasenfeld1983). For instance, it could be stated that one of the guiding principlesof the primary health care approachis to deliberatelyblur the organizational limits of health care facilities that they outreach into the communitywith active programsof health promotion,disease prevention,and early detectionof cases. As Miles et.al. (1982) point out, "the definitionof the organization's boundary should be consistentwith the problem under investigation." In our case, the purpose is to operationally distinguishbetween those processesthat take place within given hospitalsand those that are externalto any individualhospital. Bearing the foregoingcaveats in mind, we can proceed to considersome issues that derive from the epidemiologic, economic, and political environmentof hospitals. Epidemiolotic Environment. The fundamental capacity mortality of hospitals to adapt issue here refers to the so

to changingpatterns of morbidity and

in the community. This issue is particularly salient in some

developingcountries that are experiencinga complex epidemiologic transition(Soberonet.al. 1986). It is beyond the scope of this paper to make a detailed analysisof the present characteristics and likely evolution

14 of this transition. Suffice it to point out the followingcritical problems: -- What information identify
--

systems

can hospitals

devise

to opportunely

new trends in basic epidemiologic and demographicvariables? What economicallyfeasibleschemes are there to convert current

hospital capanityso that it respondsbetter to the aging of the population and the emergent:s chronic ailments? What new linkagesmust hospitals of develop with other health care facilitiesso that they can provide the necessarycontinuityfor the long-termmanagementof chronic diseases? -- How must the staffingof hospitals adapt to new epidemiologic and demographiccontexts? Is it possibleto retrain specialistsso that they can take care of differentconditionsor age groups? What is to be done with specialtiesthat become epidemiologically obsolete (witness,for example, the case of phthisiology and of tuberculosis hospitals)? PoliticalEnvironment. Out of the whole gamut of politicalvariables that confronta hospital,we will concentrateon those that have to do with its relationshipto the States 2 In a long process that began approximately in the eighteenthcentury (Foucault1977; Rosen 1972), the State has become the largest owner, payer, or regulator in the health industryof practically every country, so much so that Donnangelospeaks of the "universality" of State interventionin medical care (1975, p. 4). In fact, it would be impossibleto understandthe dominant role of hospitalswithout referenceto the fact that, especiallysince the 1950s, a growing number of governments itshas become increasingly customary in the literature, adopt the we narrow definitionof the State as the institutions governmentproviding of the administrative, legislative, and judicial vehiclesfor the actual exerciseof public authorityand power, instead of the broad definitionof the State as the total politicalorganization a society, includingits of citizens. >-2

15 adopted and stimulateda paradigm of medical care based on specialtycare of high technological complexityin hospitals (Frenk 1983). Likewise,the current concern with the high cost and low coverageof hospitalshas been largely promptedby governmentsthat begin to shift towards a new paradigm based on the tenets of primary health care. Even the search for formulas to stimulate private sector participationin the financingand provision of health care have many times been conductedby governmentsthat seek to reduce their financialrisk in this area. In fact, those countrieshave adopted explicit formulasto reduce State intervention have found that the public vctor still remains as the principal actor in the health field (for example see Klein 1984). There are two major spheres in which the relationship between the State and the hospitals has direct consequences over the performanceof the latter. The first one refers to the reimbursement formulas,which have been shown to affect the internalpower equilibriumbetween managers and clinicians(Young and Saltman 1983; Spivey 1984). The second deals with the limitationsthat governmentimposes on managerialautonomy,especiallyin public hospitals that form part of larger bureaucracies such as ministries of health. 'AAnumber of importantissues derive from these two spheres: -What reimbursement mechanismsexist that will generate incentives

for managerialand service productionefficiency,without reducingclinical efficiency?

3 Actually, such limitationson managerialautonomyalso appear to take place in private multihospital systems (Weil and Stam 1986). Thus, an importantquestionfor researchwould be to find out whether the critical variable is the type of ownershipof the hospital -- public versus private - or the existenceper se of an additionallayer of managersthat control several hospitals.

16
- -

Should the State attempt to control hospital performance mostly

through incentivesystems based on reimbursement, should it attempt more or direct supervisionand control? What is the role of consumergroups in this process? How can accountability the public be maintainedin governmentto run hospitals?
--

In the case of public hospitals,should goals be set by each

hospital,or should this be a functionof the larger public organizationto which the hospital belongs? Should ministriesof health actually run hospitals,or should their role be limited to setting,enforcing,and supervisingstandardsof care? What mechanismsare there to increase managerialautonomy in public hospitals? What are the consequences of decentralizing goal-settingand operating authorityto hospitals in a previouslycentralizedsystem? What formulasare there to monitor performancein a decentralized public system? Rcono ic Environment. Issues dealing with the economicenvironmentof the hospitalwill be approachedat two different levels. The first one refers to the overall economicsituation of a country. The second one has to do with the immediateenvironmentrepresentedby the markets in which the hospital must act. The fundamentalissue at the higher level of analysis is the adoptive responseof hospitalsto situationsof economiccrises such as the ones faced by many developingnations. Economic crises seem to have a dual effect on hospitals. On the one hand, health conditionstend to deteriorate so that the need for hospitalservices increases. At the same time, however, the standardpolicy response to such crises has been to cut budgets for social programs, includinghealth care (Brenner1979; Brenner and Mooney 1983; Soberon et.al. 1986). Public hospitals face an additionalburden,

17 since they have to absorb part of the demand previouslysatisfied by private facilitiesthat a growing number of clients can no longer afford. As hospitals in many countriesattempt to deal with this complex set of strains,several importantresearchquestionsemerge: -What are the cost savings and effectiveness alternativemodes of

of providingservicesthat have traditionally been the domain of general hospltals,such as normal deliveriesor minor surgery? Is it economically and clinicallyfeasible in developingcountriesto shift to alternative settingsfor care that may satisfy a larger volume of demand at lower costs (e.g. "birth centers"or ambulatorysurgery centers)? -What are the effects of new methods of financing,such as

communityprepaymentschemes,which can be implemented deal with some of to the consequencesof economiccrisis on the utilizationand financingof hospitals? -What mechanismscan be designedto improve the flow and control of

material resourceswithin hospitals so that waste can be prevented? -More generally,what is the repertoireof survival strategiesthat

hospitalsmust employ under conditionsof economic strain? Intimatelylinked to this last questionis the whole issue of the ways in which hospitals participatein the product and factor markets that form their immediateeconomic environment. As we pointed out earlier,our current focus on issues of staffingmakes it necessaryto restrictthe discussionspecificallyto labor markets. The entire world has witnessed a dramatic increase in the supply of physicians. As Kindig and Taylor (1985) demonstrate, this increasehas occurred in countriesat all levels of economicdevelopment. From 1950 to 1979 the number of physiciansper 10,000 people grew by 96X in

18 industrialized countries,by 223% in centrallyplanned economies,by 164% in middle income nations,and even by 29% in low income countries. The growing supply of physicianschanges the operatingenvironmentof the hospitals in two fundamental ways. First, it gives the hospital,as an employer, greater leverageto impose working conditionsthat are more favorableto its interests. Second, as the competitionfor profitableclinicalpositions increases,it is likely that more doctors will shift from patient care to management (Tarlov 1983). Indeed, it has been shown that physicians'career preferencesare significantly affectedby their perceptionsof the medical labor market (Frenk 1985). As the conditionsin this market become more difficultfor doctors,they will increasingly seek stable employmentthrough salariedpositions,with less clinicalautonomy,larger managerial responsibility, and greater stratification within the medical profession (Freidson1985). Furthermore,to the extent that in many developing countriesthe increasingsupply of physicianshas not been accompaniedby a similar growth of paramedicaland technicaloccupations,it is not unrealisticto expect that some doctors will fill less skilled positions in the hospital, giving way to a new kind of medical underemployment. In sum, the main issues that derive from the foregoingconsideraticns can be synthesizedas follows: -What are the implicationsof an increasingsupply of physicians

for the hiring and staffingpractices of hospitals? Should the substitution of physiciansfor less skilled positionsbe allowed and even encouraged? Should hospitalsexpand their staffs of residents to accommodatethe growing demand for graduatemedical education,or should they strictlymaintainthe number that they require to fulfill their medical care functions?

19
--

In order to contain competition, practicingphysiciansare likely

to impose barriersto the attainmentof hospital privilegesby their younger colleagues. Should managementinterveneto reduce such barriers? Should it press for an increasein salariedpositions at the hospital?
--

Faced with a choice between physiciansand administrators the as

senior managers of the hospital,what criteriashould guide the higher authoritiesin their hiring policy? Should physiciansbe preferred,as they are in many countries,simply because they have the knowledgeabout the substantivefunctionsof the hospital? Or should managerialefficiencybe the guiding criterion? As can be seen, some of these issues begin to have a direct bearing on the design of the hospital, a topic to which we turn next. OrganizationDesign. Organization design has been defined as "the way authority,responsibility and information are combinedwithin a particular organization"(Kimberlyet.al. 1983). A design allows "to tailor the organizationso that it can monitor its environmentand respond to the constraintsand opportunities presentedby the environment..." (Kimberly et.al. 1983) and to achieve coordination and integrationof tasks across parts of the organization(Lawrenceand Lorsch 1967). There are two main issues that determinedifferent types of designs: how activitiesshould be grouped within the organization and how decisions will be made. In fact, these issues illustratetwo differentanalytical dimensionsof the same concept. On the one hand, the organization design is representedby the structure,i.e., the type, number, and size of units, spans of control,and the arrangementof units along the lines of authority. On the other, one can identifythe more subtle and dynamic elementsof a standardization, design, such as degree of centralization-decentralization,

20 formalization, mechanismsfor coordination, communication and control, as well as rewards systems. This section will be focused mainly on the structuralissues. Since there is a more evident relationship between the more dynamic elements of the organizationdesign and organizational performance,these will be analyzed in the followingsection. Three types of structureshave been traditionally identified: functional,divisional,and matrix (Daft 1983). Functionalstructuremeans a divisionof labor into departmentsspecializedby functionalareas, i.e., departmentsof surgery,medicine,nursing,medical records,and so on. Kimberly et.al. (1983) mention that this type of structure is more common in relativelysmall (100-200bed) communitygeneral hospitals. On the other hand, divisionalstructuresare organized around serviceshaving, in many cases, their own clinicaland administrative support services. This type of structuremay be seen more often in large teachinghospitals (Howe 1969). Finally, matrix structuresare the most infrequentones in hospitals. They are characterized a dual authoritysystem designed to improvelateral by coordinationand informationflow across the organization'1euhauser1972, Gray 1974). All of these possible configurations might respond to traditional arrangementsof the structurerather than to actual environmental demands or to the need to improve organizational performance(Mintzberg1981). If one considersthe possible role of hospitals in primary care provided through outreachprograms,one could ask which of these structuralalternatives is the most appropriate(Shortell1984; Aday 1984). In addition, it is very importantto consider the particular characteristic hospital structureswhere two chains of command coexist. of

21 For Mintzberg (1981),hospitalsare a "professional bureaucracy," since who must be their structuralconfiguration relies on trained professionals given considerablecontrol over their own work. In this case, one can identifyparallelhierarchies,one for the professionals and another for the support staff. The existence of two main chains of command in hospitals -- medical -staff and administration has been well documentedby several authors in developed countries (Perrow 1961; Georgopoulos1962; Bucher and Stelling 1969; Engel 1969; Scott 1973; Robb 1975; Longest 1980; Shortelland Evashwick 1981; Scott 1982; Leatt et.al. 1983; Kinston 1983). However, in of developingcountriesthe high predominance cliniciansover professional managers in hospital administration might blur the limits between the two hierarchies. This is even more so when one considersthat in most government-owned hospitalsphysiciansare salaried;therefore,they are accountLblenot to the medical staff organization but to the administration. From all these aspects of the structuraldimensionof organization design in hospitals,several issues can be identified:
--

Since physicianspredominatein top administrative positions,can

two chains of command still be clearly identified? Is the scope of their expert power clinical,managerial,or both? How is this situation influencing the professionalautonomyof cliniciansand professional managers?
- -

What structuralarrangements are necessary to improve the balance

between these two groups so that technicalexpertise in medicine and administration can be better allocated?
--

Within this particulartype of structure,how and by whom are the

goals of the hospital defined?

22
--

Communication betwoenmembers of differentprofessional groups in

hospitals has always beer a difficulttask, not only because of their different backgroundsbut also because of deficienciesin organization design (Robb 1975). This is also true with regard to communication between providers and clients. How can hospitalorganization better designed to be improve the flow of information between departments, providers,and clients? (Hasenfeld1983). -What are the differentimplications the organizational of design

of private versus public hospitals for clinical,managerial,and production efficienc,?

--

What are the advantagesand disadvantages functional, of

divisional,or matrix structuresfor hospitals in developingcountries? -What might be the best alternativesfor structuringthe hospital

organizationaccordingto their external context,size, and types of servicesprovided? -- In light of the goal of "Health for All by the Year 2000," what are the best alternativesfor designingthe hospital organization, as to so provide better access and utilizationof hospital resourcesby the population? -Which environmental variableshave major effects on hospital

design? What is their impact? How are these variablesoperatingto influencehospitaldesign in developingcountries? Ortanizational Performance. Improvingorganizational performanceis perhaps the most importantchallengeto any hospitaladministrator. Shultz and Johnson (1976) have proposed some selectedmanagerialpracticesfor improvingperformance. These practiceswere grouped within three main

23 areas: managementof quality,managementof costs, and managementof conflict. Managementof quality involves,among other things,the implementation of assessmentand monitoringsystems and quality assurancemechanismsbased on a sound organizationdesign. The latter includesmanagerialdecisions regardingthe degree of standRrdization and formalization clinicaland of non-clinicaltasks, the degree of decentralization, and the implementation of adequate coordination and communication mechanismsthrough the developmentof quality assurance programs. Furthermore, managerial decisionshave to be made regarding the types of incentivesand specific control mechanismsfor clinicalperformance. Another very importantaspect of the managementof quality is the issue of staffing. Several authors have studied the relationships between hospitalmedical staff organizationand the quality of care (Shortelland Lo Gerfo 1981; Flood and Scott 1978; Roemer and Friedman 1971). On the other hand, staffing is also a relevantaspect of the management

of costs. Pauly (1978),Garg et.al. (1979), and Sloan and Becker (1981) have analyzed differentaspects of the relationship betweenmedical staff and costs. The ratio of managementto productionpersonnelas it affects the efficiencyof hospitalshas been studied by Rushing (1974). Scott and Shortell (1983) have made an extensivereview of the literatureon these topics under two major areas: effectiveness (qualityof care) and efficiency. These include the managementof quality and the managementof costs. It is very importantto mention that both managerial practices require a well designed information system that allows managers to obtain a real image of hospitalperformanceso that decisions are made on a more solid basis.

24 Managementof conflict is of paramount importancein hospitalsgiven the different professionalgroups involvedin patient care. Organization design, along with goal setting and negotiatingskills, are the best elements for managing conflict. Again, a neat organization design tends to improvecommunication and coordination and to prevent conflictby defining authorityand responsibility among hospital staff. Finally,organizational performanceseems to be associatedwith a linkage to the organizational environment, appropriateorganization an design, and the existenceof information systems that provide awareness of organizational functioningand the opportunityto take correctiveaction (Scott and Shortell 1983). Many issues could be raised around organizational performance. Some of them have already been mentioned in other sectionsof this paper, particularlywith regard to the relationships between the organizationand its externalenvironmentand some aspects of the organizational design. Nevertheless, there are still other relevant issues that deserve some consideration: -Which are the most common mechanisms in developingcountriesto

link hospitalswith their external environment? -What is the role of communitymembers in the administration of

hospitals?
-- -

What should be the compositionof hospital boards? Since quality assuranceexperiencesare only beginning in many

developingcountries,what might be the strategiesfor implementing quality assuranceprograms? What might be the characteristics an information of system in order to run an efficientand effectivequality assuranceprogram?

25
- -

Three types of quality assurance systems can be identified

accordingto the degree of decentralization and involvement hospital of staff: centralized,decentralized nonparticipative, and decentralized (Ruelas 1986). What should be the degree if decentralization participative for quality assuranceactivities? What are the best mechanismsfor involvinghospital staff in quality assuranceprograms?
- -

How much standardization and formalization professional of

activitiesis necessaryto assure quality of care?

- -

What might be the incentivefor cliniciansto increasetheir

compliancewith standardsof care?

--

Who should supervisethe differentprofessional activitieswithin

the hospital?
- -

What should be the adequateratios of general practitioners/

specialists,doctors/nurses, clinicalpersonnel/support personnel,according to case mix in developingcountries,in order to maintainan efficient level of hospitalperformance?
- -

What should be the criteria for establishing medical staff/ a

residents ratio that assures adequatesupervisionand quality of care?

--

How can the participation cliniciansin hospital-wide of decision

making be improved?
--

What kind of coordination and communication mechanismsmight be

implementated among hospitaldepartmentsin order to prevent ccnflictsand improve continuityof care?

ISSUES RELATED TO THE TRAININGOF HOSPITALKANAGE In accordancewith the frameworkproposedin this paper, managerial efficiencyis a result of three main components: managerialdecisionmaking,

26 skills, and autonomy. We have alreadymentioned several aspects of managerialdecisionmaking directedat improvinghospitalperformance,as well as some issues regardingthe relativeprofessional autonomyof managers within the hospital structure. According to Katz (1974), there are three kinds of skills necessary for an effectiveadministrator adequatelyperform his or her role: to conceptual,technical,and human skills. On the other hand, there are several studies that attempt to elucidatethe differenttypes of roles that administrators perform (Mintzberg1975; Kuhl 1977; Allison et.al. 1983). The developmentof managerialskills to adequatelyperform different roles depends on two importantaspects:experienceand training. Given the complexityof hospital administration, learningthrough the day to day experiencemight be a trial-and-error process that is very costly for the organization. On the other hand, even though formal training cannot substitutefield experience,it provides a broader frame of reference for decisionmaking and facilitates the learningprocess from field experiences. Ruelas and Leatt (1985) have proposed that trainingprograms should be designedconsideringthree aspects:the level of the executivewithin the structure,and the kinds of roles to be performed to deal with these problems. At the same time, the developmentof conceptual,technical,and human skills should also be consideredaccordingto the hierarchicallevel of the hospital executive. Specificprogramsand contentscan then be established. It is interestingto mention that hospitaladministration a is relativelynew discipline. Hospitals in North America have been under the dominationof differentgroups (Perrow 1961). At some point in time trusteesdominated. The basis for their control was primarilyfinancial.

27 Then, major decisions had to be based upon a medical competencethat trustees did not posses, so physiciansbecame the dominantgroup. When and needed more coordination, hospitals became more complex organizations acquired increasingpower. hospital administrators This evolutionmight not be the same in developingcountries,where is physiciansstill tend to dominateand where hospital administration not of well establishedyet. The implications this situationare twofold: of first, there is a need to provide clinicianswith a better understanding so hospitaladministration that they can improve their managerial health care performance;second, it is necessary to professionalize managementby developingformal trainingprograms in this field, which by groups. necessity will include physiciansas well as other occupational Different alternativesfor providingadequate trainingin hospital have to be better explored in developingcountries,namely, administration master's, doctoral, continuingeducation,and even undergraduateprograms. Sending students to developedcountriesrepresentsa differentkind of alternativethat must also be considered. The following issues illustratejust some of the major questionsthat need to be answered: -How are managerialproblemsperceived by hospitalexecutivesat

different levels of the hierarchy and different types of hospitals in developingcountries? How can trainingprograms be designed to take account of such variation? What should be the main contents? -face As trainingprogramsfor health servicesadministrators

one response growing competitionfrom programs in businessadministration, has been to emphasizethe strictlymanagerialaspects in the curriculum,at the expense of health contentssuch as epidemiology. If, however,hospitals

28 must respond to their changingepidemiologic environment, this trend could have very negative consequences. What new trainingapproachescan be devised so that future health care managersdo receive the complex contents of managerialscience,while at the same time preservingthe fundamental conceptsand methods of epidemiology? If such an integrative approach is not feasible,would it then be necessary to have an epidemiologist the in senior managementgroup of a hospital? -How should existingtrainingprograms in health care

administration respond to the increasein the number of physician administrators? Should new programs,different from the traditional master's degrees,be designedto meet the special backgrounds and needs of physicians? -Regarding the level of training,would undergraduate programs in

hospital administration useful? Should professionally be orientedor academicallyoriented postgraduate programs be predominantin developing countries? Should there be a sharp distinctionbetween both types? What should be the role of master's,doctoral, and continuingeducationprograms in order to meet the need of traininghospitaladministrators developing in countries? -How convenientare residencyperiods,under what circumstances,

and for how long? -Is there enough faculty in developingcountries to supporthigh

quality education in hospitaladministration? What might be the strategies for faculty development? -How useful is the trainingof professionalsin foreign countries,

as opposed to concentrating their nationalexperiences? What strategies on should be consideredto assure that experiencesobtained abroad will have an

29 impact in the country of the traineeswhen they return? How useful are exchangeprograms between developedand developingcountries? What should be done in order to take advantageof such programsso as to achieve a balance between academicquality,on the one hand, and relevanceto the context of the trainee,on the other?

TOWARDS A RLAR

AGUD

Most of the issues that ve have discussed throughoutthis paper represent importanttopics for research. The fact that we posed them as questionswas intended,precisely,to emphasizetheir researchability and to convey the sense that it is necessaryto seek answers through sound studies. The problem,of course, is that the number of issues is too large to constitutea workable researchagenda. It is necessary,therefore,to establish priorities. In this last section of the paper we will briefly sketch what such prioritiesmight be. A first consideration designinga researchagenda on a topic such as in hospitalmanagement is to strive for a balance between relevanceto decision making and excellencein the strict adherence to the norms of scientific research (Frenk et.al. 1986). Within this broad guideline prioritiesmust

be defined on two aspects: the type of research and the topics to be researched. With respect to the former,we believe that the order of priorities should begin with observational studies that document levels of hospital performanceand correlatethem with organizationdesign and environmental variables. Apart from offeringbasic descriptiorns that are much needed, especiallyin developingcountries,such studies would make it possible to operationalize and measure the constructsthat we have proposed in our

30 analyticalframework. As indicatedin the section on measurementissues, it is particularlyimportantto devise and test sensitiveand specific indicatorsof managerial,clinical, and service productionefficiency. In addition, it is necessary to determine the internaland external correlates of these dimensionsof performance. Observational studies would make it possibleto diagnosethe most criticalareas for the second type of research,namely, intervention studies that would introduceplanned change in hospitalsand would assess its consequences. It is fundamentalthat intervention studies be based on comparativedesigns. Indeed,a problem with evaluationsof the effectiveness specific interventions the frequent lack of control of is groups, which makes it impossibleto attributeany observedchange to the interventionitself, rather than to another source of variation. Thie externalvalidity Ofthese typesof studies is also often threatenedby the choice of highly specificsites that make it very difficultto generalize the findingsand to truly build a body of knowledge. If the ideal randomizedtrials cannot be achieved,then quasiexperimental designs with clear control groups should be used. These kind of studies should be analysesof the complementedby cost/benefitand cost/effectiveness interventions (Wortman1983). Turning to the prioritieson the topics for research, it must be stated, at the outset, that any ranking of topics is doomed to seem arbitrary,unless it is based on some explicitmethod to poll the perceptionsof large numbers of experts and consumersof research. Nevertheless, will attempt to offer what we believe is a preliminarylist we of the most urgent areas for inquiry,particularlyin developingcountries.

31 The first need is for good descriptivestudies of the hospital system design. In many developing and of the main aspects of organization on countrieswe are lacking the most basic information the compositionand of characteristics hospitals. Critical items that are often not known include the exact magnitudeof the private sector, the proportionof total health care resourcesthat is absorbed by hospitals,and the unit costs for there is a specifichospital services,to name only a few. Furthermore, lack of data on the structure of hospital organization. Whereas in developedcountriesextensive empiricalstudies have been conductedto define, for example,the two lines of authority,in many developingnations we are often ignorantof the ways in which formal and informalrelations among managers and cliniciansare structured. Likewise, it is necessaryto for internalcommunication, know the frequencyof differentarrangements types of departmentalization, and management systems.

Beyond broad descriptionsof the structureof hospitals in developing countries,the second priorityrefers to the systematicstudy of quality monitoringand assurance systems. In the final analysis,hospitalsshould be producing improvementsin health, however we define it. The design and testing of explicitsystems to assure the quality of care would therefore seem to be of the utmost importance if we are to gain some understanding of

to what exactly are hospitals contributing society and at what cost. Such between managers and studies should includethe analysisof the interactions clinicians,especiallyas they constrainclinicalautonomyand decision making. As pointed out earlier in this paper, there are several variantsof quality assurancesystems for hospitals. Assessing their relative effectiveness and costs should be a high-priorityitem on a research agenda.

32 The third area for research centers around the social, personal, organizational, and educationaldeterminants managerialskill. Indeed, of we need to know what are the factors that account for different degrees of success in managinga hospital. These studies should not be limited to psychological variables,although they should certainly include them. The challenge,however, is to ascertainthe relativecontributions managerial to skill of personalvariablesversus educationalbackgroundand organizational structure. Clearly,this kind of study would have major policy implications for the recruitmentand trainingof hospitalmanagers,which in turn might help to alleviate the critical shortageof skilledmanagementin underdeveloped countries. Finally, the magnitude,repercussions, visibility,and universalityof physicianoversupplymake this a high priorityfor research. In this respect,we are in need of studies about the structureand dynamicsof medical labor markets,which would allow us to understandthe origins of the oversupplyof doctors and the reasons why it has occurred in such a wide variety of countries. The coexistence,in many nations, of medical underemployment with lack of universalaccess to medical care is probably the most eloquent indicatorof the shortcomings current ways of of organizinghealth systems. Hospitalsare undoubtedlya major part of this picture. We should thereforeunderstandthe consequences that the oversupplyof physicianshas for the operationand staffingof hospitals, and for the design of innovativetrainingprograms. While still incomplete, this initial researchagenda might begin to illuminatesome of the basic issues that concern policy makers, managers, clinicians,and clients in the common search for higher levels of efficiency and equity in health care.

Aday, L.A., R. Andersen, S. S. Loevy, B. Kremer. 1984."fHbspital-sponsored Primary Care: II. Impact on Patient Care." American Journal of Public Health 74:792-98. Allison, R. F., W. L. Dowlig, F. C. Munson. 1983. "The Role of the Health ServicesAdministratorand Implications for Educators." In Kovner A. R., NeuhauserD. (eds). Health ServicesManatement: Readingsand Commentary(Second Edition). Ann Arbor, MI: Health Administration Press, pp. 32-67. Argote, L. 1982. "Input Uncertaintyand Organizational Coordinationin Hospital EmergencyUnits." Administrative Science Quarterly 27:420-34. Berki, S.E. 1972. Hospital Economics. Lexington,MA: LexingtonBooks. Brenner,M. H. 1979. "Mortalityand the National Economy:A Review and the Experienceof England and Wales, 1936-76." The Lancet 2:568-73. Brenner,M. H., A. Mooney. 1983. "Unemployment and Health in the Context of Economic Change." Social Science and i.ledicine 17:1125-38. Bucher, R., J. Stelling. 1969. "Characteristics Professional of Organizations." Journal of Health and Social Behavior 10:3-15. Daft, R. L. 1983. OrganizationTheory and Design. St. Paul: West PublishingCo. Donabedian,A., J. R. C. Wheeler,L. Wyszewianski. 1982. "Quality,Cost, and Health: An Integrative Model." Medical Care 20:975-92. Donnangelo,M. C. F. 1975. Medicine e Sociedade:0 Medico e seu Mercado de Trabalho. Sao Paulo: LivrariaPionera Editora. Engel, G. V. 1969. "The Effect of Bureaucracyon the Professional Autonomy of the Physician." Journal of Health and Social Behavior 10:30-41. Flood, A. B., W. R. Scott. 1978. "Professional Power and Professional Effectiveness: The Power of the SurgicalStaff and the Quality of SurgicalCare in Hospitals." Journal of Health and Social Behavior 19:240-54. Foucault,M. 1977."iistoria de la Medicalizacion. "EducacionMedica y Salud 11:3-25. Foucault,H. 1978. "Incorporacion del Hospitalen la TecnologiaModerna." EducacionMedica y Salud 12:20-35 Frederiksen,H. 1969. "Feedbacksin Economicand DemographicTransition." Science 166:837-47.

34 Freidson,E. 1970. Professionof Medicine: A Study in the Sociology of ARplied KnowledRe. New York: Harper & Row. Freidson,E. 1985. "The Reorganization the Medical Profession." of Medical Care Review 42:11-35. Frenk, J. 1983. Social OriRin, Professional Socialization, and Labor Market Dynamics: The Determinants of Career Preferences among Medical Interns in Mexico. Ann Arbor, MI: The University of Michigan, Ph.D. dissertation. Frenk, J. 1985. "Career Preferencesunder Conditionsof Medical Unemployment: The Case of Interns in Mexico." Medical Care 23:320332. Frenk, J., J. L. Bobadilla,J. Sepulveda,J. Rosenthal,E. Ruelas, M. A. Gonzalez-Block, Urrusti. 1986. "An InnovativeModel for J. Public Health Research: The Case of a New Center in Mexico." Journal of Health Administration Education 4: in press. Frenk, J., A. Donabedian. 1986. "State Intervention Medical Care: in Types, Trends, and Variables." Presentedat the 114th Annual Meeting of the American Public Health Association, Las Vegas, October 1. Garg, M. L., J. L. Mulligan,W. A. Gliebe. 1978. "Physicans'Specialty, Quality and Cost of In-patientCare." Social Science and Medicine 13c:187-90. Georgopoulos, S., F. C. Mann. 1962. The CommunityGeneral Hospital. B. Toronto:Macmillan. Gray, J. L. 1974. "MatrixOrganizational Design as a Vehicle for Effective Delivery of Public Health Care and Social Services." ManaRement International Review 14: 73-82. Harris, J. E. 1977. "The InternalOrganization Hospitals:Some Economic of Implications." Bell Journal of Economics8:467-82. Hasenfeld,Y. 1983. Human Service Organizations. Englewood Cliffs, NJ: Prentice-Hall. Howe, G. E. 1969. "Decentralization Aids Coordination Patient Care of Services." Hospitals 43:53-55. Jacobs, P. 1974. "A Survey of EconomicModels of Hospitals." Inquiry 11:83-97. Katz, D. 1974. "Skills of an EffectiveAdministrator." Harvard Business Review 52:90-120. Kimberly,3. R., P. Leatt, S. Shortell. 1983. "Organization Design." In Shortell,S., A. Kaluzny (eds). Health Care ManaRement: A Text in Organizational Theory and Behavior. New York: Wiley, pp. 291-332.

35 Physician Kindig, D. A., C. H. Taylor. 1985. "Growth in the International Supply. 1950 through 1979." Journal of the American Medical Association 253:3129-32. Kinston, W. 1983. "HospitalOrganizationand Structureand its Effect on Inter-professional Behaviorand the Delivery of Care." Social Science and Medicine 17:1159-70. Klein, R. 1984. "The Politicsof Ideologyvs. the Reality of Politics: The Case of Britain'sNationalHealth Service in the 1980s." Milbank Memorial Fund Quarterly/Health and Society 62:82-109. Kovner, A. R., D. Neuhauser. 1983. lntroduction. In Kovner, A. R., D. Neuhauser (eds). Health ServicesManaRement:Readingsand Commentary(SecondEdition). Ann Arbor, MI: Health Administration Press, pp. 3-18. Kuhl, I. K. 1977. The ExecutiveRole in Health Service Delivery Organization. Washington: Associationof UniversityProgramsin Health Administration. and Environment Managing Lawrence, P., J. Lorsch. 1967. Organization Differentiation and Integration. Boston: Harvard BusinessSchool Press. Leatt, P., V. Mickevicius,J. Barnsley,E. Vayda, J. I. Williams. 1983. An HospitalMedical Staff Organization: AnnotatedBibliography. Ottawa: CanadianHospitalAssociation. Longest,B. B. 1980. "A ConceptualFrameworkfor Understandingthe Multihospital ManagementStrategy." Health Care ManagementReview Winter:17-24. Longest,B. B. 1981. "An ExternalDependencePerspectiveof Organizational Strategyand Structure: The CommunityHospital Case." Hospitaland 26:50-69. Health ServicesAdministration Mechanic, D. 1975. "Ideology,Medical Technology,and Health Care Organizationin Modern Nations." American Journal of Public Health 65:241-47. Miles, R. E., C. C. Snow, J. Pfeffer. 1982. "Organization-environment: Conceptsand Issues. In Spirn, S., D. W. Benfer (eds). Issues in Health Care Management Rockville,MD: Aspen, pp. 421-38. Mintzberg,H. 1975. "The Manager'sJob: Folkloreand Fact. Harvard BusinessReview 53:49-61. Mintzberg,H. 1979. The Structuringof Organizations. New Jersey: Prentice-Hall, Inc. Design: Fashion or Fit." Harvard Mintzberg,H. 1981. "Organizational 1088-116. BusinessReview January-February:

36 Neuhauser,D. 1972. "The Hospitalas a Matrix Organization." Hospital Administration17:8-25. Omran, A. R. 1971. "The Epidemiologic Transition: A Theory of the Epidemiologyof PopulationChange." Milbank MemorialFund Quarterly 49:509-38. Pauly, N. 1978. "MedicalStaff Characteristics and HospitalCosts. Journal of Human Resources 13:78-111. Perrow, C. 1961. "The Analysis of Goals in Complex Organizations." American SociologicalReview 26:836-66. Ritvo, R. A. 1980. "Adaptationto Environmental Change: The Board's Role." Hospital and Health ServicesAdministration 25:23-37. Robb, J. H. 1975. "Power, Professionand Administration: An Aspect of Change in English Hospitals." Social Sciencesand Medicine 9:373-82. Roemer, M., J. Freidman. 1971. Doctors and Hospitals: Medical Staff Organizationand Hospital Performance. Baltimore,MD: Johns Hopkins UniversityPress. Rosen, G. 1972. "The Evolutionof Social Medicine. In Freeman,H.E., S. Levine, L. G. Reeder (eds). Handbookof Medical Sociology (Second Edition). EnglewoodCliffs, NJ: Prentice-Hall. Ruelas, E., P. Leatt. 1985. "The Roles of Physician-executives in Hospitals: A Framework for ManagementEducation." Journal of Health Administration Education 3:152-69. Ruelas, E. 1986. "Towardsa Typology of Quality Assurance Systems." Unpublishedpaper. Rushing, W. 1974. "Differencesin Profit and Non-profitOrganizations: A Study of Effectiveness and Efficiencyin General Short Stay Hospitals. Administrative Science Quarterly 19:473-84. Scott, W. R. 1966. "Some Implications Organization of Theory for Research on Health Services." Milbank Memorial Fund Quarterly 44:35-64. Scott, R. W. 1973, "The Medical Staff and the Hospital:An Organizational Perspective." The HospitalMedical Staff 2:33-8. Scott, R. W. 1982. -Managing Professional Work: Three Models of Control for Health Organizations." Health ServicesResearch 17:212-40. Scott, W. R., S. Shortell. 1983. "Organizational Performance: Managing for Efficiencyand Effectiveness. In Shortell,S., A. Kaluzny (eds). Health Care Management: A Text in Organizational Theory and Behavior New York: Wiley, pp. 418-55. Secretariade Salud. 1985. Anuario Estadistico1984. Mexico, D. F.: Direccion General de Informaciony Evaluacion.

37 Shortell, S., C. Evashwick. 1981. "The StructuralConfiguration U.S. of Hospital Medical Staffs." Medical Care 19:419-30. Shortell, S., J. P. Lo Gerfo. 1981. "HospitalMedical Staff Organization and Quality of Care: Results for MyocardialInfarctionand Appendectomy.,, Medical Care 19:1041-53. Shortell, S., T. M. Wickzee,J. R. C. Wheeler. 1984. "Hospital-sponsored Primary Care: I. Organizational and FinancialEffects.'American Journal of Public Health 74:784-91. Shulz, R., A. C. Johnson. 1976. The Managementof Hospitals. New York: McGraw-Hill. Sloan, F., E. Becker. 1981. InternalOrganization Hospitalsand of Hospital Cost;' Inguiry 18:224-40. Soberon, G., J. Frenk, J. Sepulveda. 1986. "The Health Care Reform in Mexico: Before and After the 1985 Eart.,quakes." AmericanJournal of Public Health 76:673-80. Spivey, B. E. 1984. The RelationBetween HospitalManagementand Medical Staff under a Prospective-payment System.' New England Journal of Medicine 310:984-86. Starr, P. 1982. The Social Transformation AmericanMedicine New York: of Basic Books. Tarlov, A. R. 1983. "ShattuckLecture -- The IncreasingSupply of Physicians,the Changing Structureof the Health-services System, and the Future Practiceof Medicine." New England Journal of Medicine 308:1235-44. Tlhompson, D., W. J. McEwen. 1958. "Organizational,Roals J. and environment: Goal-settingas an interactionProcess." American SociologicalReview 23:23-31. Weil, P. A., L. Stam. 1986. "Transitionsin the Hierarchyof Authority in Hospitals: Implicationsfor the Role of the Chief ExecutiveOfficer." Journal of Health and Social Behavior 27:179-92. Wortman, P. M. 1983. "Evaluation Research: A Methodological Perspective." Annual Review of Psychology34:223-60. Young, D. W., R. B. Saltman. 1983. "Prospective Reimbursement and the Hospital Power Equilibrium. A Matrix-based ManagementControl System." Inquiry 20: 20-13. Zald, M. N. 1970. "PoliticalEconomy: A Frameworkfor Comparative Analysis. In Zald, M. N. (ed). Power in Organizations Nashville,TN: VanderbiltUniversityPress, pp. 221-69.

PPR WorkingPaper Series Contact

Title Author Date for paper

WPS146 ExchangeRate-Based Disinflation,Wage Rigidity,and Capital Inflows: Tradeoffs for Ch!le 1977-81

Timothy Condon Vittorio Corbo Jaime de Melo

February 1989

K Cabana 61539

WPS147 The Private Sector's Response to Financial Liberalizationin Turkey 1980.82 WPS148 Impact of the International Coffee Agreement'sExport Quota System on the World Coffee Market WPS149 Reflectionson Perestroyka and the Foreign Economic Ties of the USSR WPS150 Improvingthe Currency Composition of External Debt: Applicationsin Indonesia and Turkey WPS151 U.S. Trade PolicyTowards DevelopingCountries WPS152 Subsidies and CountervailingMeasures: Economic Considerations WPS153 An Analysisof Debt Reduction Schemes Initiated by Debtor Countries WPS154 Forecasting, Uncertainty and Public Project Appraisal WPS155 Measuring Adult Mortalityin DevelopingCountries: A Reviewand Assessment of Methods WPS156 Credit Cooperativesin Israeli Agiiculture

Izak Atiyrs

January 1989

W. Pitayatonakarn 60353

Takamasa Akiyama Panayotis N. Varangis Bela Balassa

February 1989

D. Gustafson 33714 N. Campbell 33769

January 19S9

Ken Kroner Stijn Claessens Bela Balassa

January 1989

L. Chavarria

33730 January 1989 N. Campbell 33769 N. Campbell 33769 L. Chavaria 33730 A. Kitson-Walters 33712

Bela Balassa

January 1989

Ishac Diwan Stijn Claessens Jock R. Anderson

March 1989

February 1989

Ian Timaeus Wendy Graham Yoav Kislev Zvi Lerman Pinhas Zusman Martha de Melo Marc Leduc Setareh Razmara Alice Galenson

April 1989

S. Ainsworth 31091 C. Spooner 37570

March 1989

WPS157 A PolicyModel for Tunisia with Real and Financial Flows

January 1989

A. Bhalla 60359

WPS158 Labor Redundancyin the Transport Sector

February 1989

W. Wright 33744

PPR Workinit Paper Series Contact for paper

Title WPS159 Current International Gas Trades and Prices

Author

Date

Kay McKeough

March 1989

M. Fernandez 33637

WPS160 Evaluating the Performance of Public Enterprises in Pakistan

Mary M. Shirley

March 1989

R. Malcolm 61708

WPS161 Commodity-indexed Debt in International Lending

Timothy Besley Andrew Powell

March 1989

J. Raulin 33715

WPS162 Ups and Downs in Inflation: Argentina Since the Austral Plan WPS163 The Impact of Infrastructure and Financial Institutions on Agricultural Output and Investment in Iidia

Miguel A. Kiguel

Hans P. Binswanger Shahidur R. Khandker Mark R. Rosenzweig

March 1989

J. Arevalo 30745

WPS164 Intersectoral Financial Flows in Developing Countries

Patrick Honohan izak Atiyas

March 1989

W. Pitayatonakarn 60353

WPS165 Developing Countries' Exports of Manufactures: Past and Future Implications of Shifting Patterns of Comparative Advantage WPSl66 Achieving and Sustaining Universal Primary Educatibn: International Experience Relevant to India

Alexander J. Yeats

Nat J. Colletta Margaret Sutton

March 1989

M. Philiph 75366

WPS167 Do Price Increases for Staple Food Help or Hurt the Rural Poor

Martin Ravallion

March 1989

M. Zee-Wu 37589

WPS168 Technological Change from Inside A Review of Breakthroughs!

Ashoka Mody

March 1989

W. Young 33618

WPS1-Q F:-ncial Sector Reforms in Adjustment Programs

Alan Gelb Patrick Honohan

WPS170 General Training Under Asymmetric Information WPS171 Cost-Effectiveness of National Training Systems in Developing Countries

Eliakim Katz Adrian Ziderman

April 1989

C Cristobal 33640

Christopher Dougherty

March 1989

C Cristobal 33640

WPS172 The Effects of Peru's Push to Improve Education

Elizabeth M. King Rosemary T. Bellew Julio Frenk Enrique Ruelas Avedis Donabedian

March 1989

C Cristobal 33640 S. Ainsworth 31091

WPS173 Hospital Management: Staffing and Training Issues

July 1989