Support Care Cancer (2004) 12:626633

DOI 10.1007/s00520-004-0622-5

Phyllis M. Lau
Kay Stewart
Michael Dooley

Received: 19 October 2003

Accepted: 1 March 2004
Published online: 3 April 2004

Springer-Verlag 2004

P. M. Lau ()) K. Stewart M. Dooley

Department of Pharmacy Practice,
Monash Univeristy,
381 Royal Parade, Parkville, VIC, 3052,
Australia
e-mail: phyllis.lau@vcp.monash.edu.au
Tel.: +61-39-9039625
Fax: +61-39-9039629
M. Dooley
Pharmacy Department,
Peter MacCallum Cancer Centre,
St. Andrews Place, East Melbourne, VIC,
3002, Australia

ORIGINAL ARTICLE

The ten most common adverse drug reactions

(ADRs) in oncology patients:
do they matter to you?

Abstract Aims: To assess incidence,

predictability, preventability and
severity of adverse drug reactions
(ADRs) in hospitalised oncology patients. Patients and methods: Patients
hospitalised at Peter MacCallum
Cancer Centre from 28 February to
2 June 2000 were selected for interviews about symptoms related to
their drug therapy. Medical records
were also reviewed. Causality, predictability, preventability and severity were assessed for each ADR.
Results: One hundred and sixty-seven
patients associated with 171 admissions were interviewed. Four hundred
and fifty-four ADRs were identified
in 127 (74.3%) separate admissions
(mean ADRs per admission 2.7;
range 018). Eighty-eight percent of
ADRs were predictable. Of these,
1.6% was classified as definitely
preventable and 46.1% probably
preventable. The ten most common
ADRs were constipation, nausea
vomiting, fatigue, alopecia, drowsiness, myelosuppression, skin reactions, anorexia, mucositis and diarrhoea. These ADRs have high-docu-

Introduction
In the practice of oncology, adverse drug reactions
(ADRs) of cancer treatments have become almost synonymous with the treatments themselves. The low therapeutic index of chemotherapeutic agents and the predictable and common adverse events of cancer treatments
mean that these events are seen as an unavoidable com-

mented incidence rates and were also

the ten most predictable ADRs in this
study. Common reasons for ADRs to
be assessed as definitely or probably
preventable were omission or inadequate/inappropriate use of preventative measures. The results also
showed a discrepancy between clinical severity and patients perception
of the impact of ADRs on well being.
Conclusions: ADRs are common in
hospitalised oncology patients and
are predictable and at least probably
preventable in many instances. Improved use of preventative measures
has the potential to contribute to reducing the incidence and severity of
ADRs. Recognition and understanding of the discrepancy that exists
between clinical severity and patientperceived severity of ADRs will enable specific areas to be identified
and targeted for vigourous intervention.
Keywords Adverse drug reactions
Cancer patients Causality
Predictability Preventability
Severity

ponent of treatment. These events are often accepted not

only by the patients but also by health-care providers. In
many instances, ADRs are even used as end-points for
treatment protocols. With this mindset, it is not surprising
that it is not known how many of the ADRs experienced
by oncology patients could be prevented.
Research on broader population groups in Sweden, the
United Kingdom, Germany and the United States has

627

found that the incidence of ADRs is high and is dominated by reactions that are predictable and preventable [5,
10, 17, 24, 26]. From their review of Australian studies,
Roughead and colleagues found that between 32 and 69%
of drug-related admissions reported were classified as
definitely or possibly preventable [22].
The magnitude of ADRs endured by oncology patients
is significant. The 1992 Australian National Hospital
Morbidity Data (excluding NT) reported that 11% of
ADRs in Australian hospitals were associated with antineoplastic and immuno-suppressive drugs [19]. The 1995
Quality in Australian Health Care Study, which reviewed
medical records of over 14,000 admissions to 28 hospitals
in New South Wales and South Australia, also implicated
anti-neoplastics amongst the most common agents responsible for medication-related hospitalisations [27].
Advances in the development of supportive care in oncology, for example, 5HT3 antagonists for the control of
chemotherapy-induced nausea, have led to reductions in
drug-related toxicity. However, there is a need to quantify
the frequency and severity of ADRs experienced by oncology patients to enable the development and implementation of intervention strategies. Appropriate and adequate use of preventative measures to combat ADRs and
the optimal management of these reactions in cancer patients are important issues in the quality use of medicines.
The impact of ADRs on the well-being of cancer patients has been a much discussed topic in both public and
medical arenas. There is evidence to show that medical
perspectives do not necessarily coincide with those of
patients who are being treated[8, 9]. Patients are more
likely to report bothersome but not necessarily clinically
important symptoms [12, 16]. It is important that patients
perspectives and psychosocial issues be considered when
attempting to identify, prioritise and develop strategies to
combat the problem [15]. Discrepancies between patients
and clinicians perception of what constitutes a significant
adverse reaction is worthy of further investigation.
Strategies to reduce the burden of ADRs and to improve the quality use of medicines in health care are
clearly needed [14]. The British Audit Commissions
2001 report Spoonful of Sugar highlighted the important proactive role that pharmacists play in medicine
management [2]. The 2002 report by the Clinical Oncological Society of Australia, The Cancer Council Australia and the National Cancer Control Initiative similarly
highlighted the contribution of pharmacists to the quality
of cancer treatment and patient care [7]. An increased
understanding of the prevalence and nature of ADRs,
as well as an appreciation of patients perspectives on
quality of life, would enable us as a profession to identify and target specific areas for vigourous pharmaceutical
care.
The aim of this study was to assess the incidence,
predictability, preventability and severity of ADRs in hos-

pitalised oncology patients and to determine patients

perception of how ADRs impact their well-being.

Patients and methods

Ethics approval
Ethics approvals were obtained from the Clinical Research and
Ethics Committee at Peter MacCallum Cancer Centre and the
Monash University Standing Committee on Ethics in Research
Involving Humans.
Definition of ADR
This study used the World Health Organisation (WHO) definition
of an ADR: a response to a drug which is noxious and unintended,
and which occurs at doses normally used or tested in man for
prophylaxis, diagnosis, or therapy of disease, or for modification of
physiological function [3].
Recruitment
In-patients 18 years old or older at the time of admission to Peter
MacCallum Cancer Centre from 28 February to 2 June 2000 were
systematically selected into an ADR study. Study admissions were
then randomly selected to participate in interviews.
Exclusion criteria
1. Patients who were unable to give informed consent or were
deemed inappropriate by the attending medical practitioner for
interviews
2. Admissions to day ward with durations of shorter than 4 h
3. Patients treated in HIH (hospital in the home), celltrifuge unit or
PET (positron emission tomography) unit.
Selection process
Each weekday morning, daily admission lists that included planned
admissions to the centres day ward and five inpatient wards for
that day, and unplanned admissions for the day before or the previous weekend were obtained from the admission office.
Every third admission was selected for consideration for inclusion into the study. Selection on day 1 started with the first
admission on the lists. Selection on day 2 started with the second
admission on the lists, and this process was repeated thereafter. An
admission found to be ineligible was replaced by the next admission on the lists on the same day.
Five study admissions were then randomly selected each
weekday from the study admission list for that day using random
numbers generated by a computer software program. On days when
there were five or fewer study admissions, all patients were invited
for interview. A study admission omitted because of physical inaccessibility or patient inability to participate (e.g. unconscious, too
ill, too emotional) was replaced from the remaining study admissions for that day.

628

Table 1 Naranjo algorithm or adverse drug reaction probability scale. The total score calculated from this table defines the category as:
Possibly (total score 14), Probably (total score 58), Definitely (total score >9)

1. Are there previous conclusive reports on this reaction?

2. Did the adverse event appear after the suspected drug was administered?
3. Did the adverse reaction improve when the drug was discontinued or a specific antagonist was
administered?
4. Did the adverse reaction reappear when the drug was re-administered?
5. Are there alternative causes (other than the drug) that could solely have caused the reaction?
6. Did the reaction reappear when a placebo was given?
7. Was the drug detected in the blood (or other fluids) in a concentration known to be toxic?
8. Was the reaction more severe when the dose was increased, or less severe when the dose was
decreased?
9. Did the patient have a similar reaction to the same or similar drugs in any previous exposure?
10. Was the adverse event confirmed by objective evidence?

Data collection
The study involved both ongoing patient interviews and review of
patient medical records.
Patient interview
Patients were interviewed on alternate days during their hospital
stay and asked questions about adverse symptoms as a result of
drug treatment. Symptoms included those present on admission,
regardless of whether they related to reasons for admission, and
those that occurred during hospitalisation. Patients were also asked
to rate on a seven-point scale how they thought each symptom
impacted on their well-being. Labels were used to define only the
extreme categories with 0 being no or hardly any impact and 6
being totally changed my life [13].
Review of patient medical records
Patient medical records were reviewed daily to identify possible
ADRs. All symptoms reported during interviews and in medical
records were scrutinised for possible relationship to drugs by extensive review of the patients medical records, which included past
and present medical history, past medication histories, results of
any investigative or assessment procedures, notes recorded by
members of the attending medical team, medication administration
charts, and prescription records. Patients health status, diseases and
disease progression were all noted during the assessment of each
symptom.
For each ADR, a complete search on the ADR and the drug(s)
involved was accessed through the on-line MICROMEDEX
Health-care Series [25]. Information found or not found was verified with other resources, namely product information (PI), Australian Adverse Drug Reactions bulletins prepared quarterly by the
Australian Adverse Drug Reaction Advisory Committee (ADRAC)
and published by the Therapeutic Goods Administration (TGA),
PREMEDLINE and MEDLINE databases [21]; American Hospital
Formulary Service Drug Information (AHFS DI) [1]; Australian
Prescription Products Guide (APPG) [4]; and Meylers Side Effects
of Drugs [11].
Assessment algorithm and criteria
ADRs identified either from patient interview or from patient
medical records review alone, were assessed for causality, predictability, preventability and severity.

Yes

No

Do not know

+1
+2
+1

0
1
0

0
0
0

+2
1
1
+1
+1

1
+2
+1
0
0

0
0
0
0
0

+1
+1

0
0

0
0

Table 2 Criteria for determining predictability of an adverse drug

reaction (ADR). Patients who have had the drug on previous occasion(s): If the drug was previously well-tolerated at the same dose
and route of administration, the ADR is NOT PREDICTABLE; if
there was a history of allergy or previous reactions to the drug, the
ADR is PREDICTABLE. Patients who have never had the drug
previously: Incidence of the ADR reported in product information
or other literature determines its predictability
Incidence rate

Incidence description

Predictability

1/10
1/100 and <1/10
1/1000 and <1/100
1/10,000 and <1/1000
<1/10,000

Very common
Common
Uncommon
Rare
Very rare

Predictable
Predictable
Not predictable
Not predictable
Not predictable

Causality criteria
The Naranjo Algorithm or Adverse Drug Reaction Probability
Scale (Table 1) was used to evaluate the causality relationship
between a likely ADR and a drug [18]. Only symptoms with a
Naranjo score of 1, that is, at least possibly related to one or more
drugs, were recorded as ADRs, which were classified into three
categories: possibly (Naranjos scores 14), probably (Naranjos
score 58) or definitely (Naranjos score >9) related to a drug. They
were then assessed for predictability and preventability.
Predictability criteria
The Council for International Organizations of Medical Sciences
guidelines for preparing core clinical-safety information on drugs
were adapted for assessment of predictability (Table 2) [6]. Product
Information and references including the on-line MICROMEDEX
Health-care Series, ADRAC On-Line, MEDLINE, AHFS drug information, APPG and Meylers Side Effects of Drugs were used to
determine the documented incidence rate of an ADR.
Preventability criteria
The criteria of Schumock and Thornton were modified to assess
and categorise ADRs into definitely, probably or not preventable
(Table 3) [23]. The factor toxic serum drug level was omitted
from the original criteria, and two questions (4 and 5, section B) on
the use of preventative measures were added.

629

Table 3 Criteria for determining preventability of an adverse drug

reaction (ADR)
Section A
Answering yes to one or more of the following implies that an
ADR is DEFINITELY preventable
1. Was there a history of allergy or previous reactions to the drug?
2. Was the drug involved inappropriate for the patients clinical
condition?
3. Was the dose, route, or frequency of administration inappropriate
for the patients age, weight, or disease state?
If answers are all negative to the above, then proceed to Section B
Section B
Answering yes to one or more of the following implies that an
ADR is PROBABLY preventable
1. Was required therapeutic drug monitoring or other necessary
laboratory tests not performed?
2. Was a documented drug interaction involved in the ADR?
3. Was poor compliance involved in the ADR?
4. Was a preventative measure not administered to the patient?
5. If a preventative measure was administered, was it inadequate
and/or inappropriate? Answer NO if this question is nonapplicable
If answers are all negative to the above, then proceed to Section C.
Section C
The ADR is NOT preventable

Table 5 Adverse drug reactions (ADRs)/admission distribution

ADRs/admission

Proportion %

0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

24.4
16.7
16.1
11.9
9.5
8.3
3.6
4.8
0.6
1.8
0.6
0.6
0.6
0
0
0
0
0
0.6

Table 4 Diagnosis distribution at time of admission

Diagnosis

Proportion %

Breast
Haematological
Gastrointestinal
Lung
Dermatological
Head/neck
Urology
Sarcoma
Gynaecological
Brain
Others
Unknown primaries

22.3
14.0
13.4
12.5
11.5
8.5
6.2
4.9
4.0
1.3
1.1
0.4

Results
Interview sample
Patients from 276 admissions were selected to be approached for interviews. Of these, 194 were available to
be informed about the study and 183 consented to participate. In total, 167 patients from 171 (93.4%) admissions were interviewed. The interview sample comprised
admissions of 87 women and 80 men whose ages ranged
from 20 to 92 years with an average of 59.014.8 years.
The proportions and types of cancer at the time of admission are illustrated in Table 4.

Fig. 1 The ten most common adverse drug reactions (ADRs)

Number of ADRs identified

Four hundred and fifty-six ADRs were identified from
127 of the 171 admissions (74.3%). ADRs per admission
ranged from 0 to 18, with a mode of 0, a median of 2 and
a mean of 2.7 (Table 5).
Types of ADRs identified
One hundred and twenty different reaction types were
described from 32 drug categories. The ten most common
reaction types (n=289) comprised 63.4% of ADRs identified (Fig. 1).

630

Preventability of ADRs
Over half (53.4%) of ADRs were classified as not preventable, 45% probably preventable and 2% definitely
preventable. For the ten most common ADRs, only three
cases were assessed as definitely preventabletwo cases
of drowsiness, which were caused by inappropriate opiate
doses, and one case of diarrhoea, which was caused by
simultaneous use of three laxatives. High proportions of
constipation, nausea vomiting, diarrhoea and mucositis
were assessed as probably preventable. Alopecia and
anorexia were always assessed as not preventable (Fig. 3).
Patients impact rating of the ten most common ADRs

Fig. 2 Predictability of the ten most common adverse drug reactions (ADRs)

Predictability of ADRs
Eighty-eight percent of the ADRs were assessed as predictable. The predictability of the ten most common
ADRs closely matched the frequency (Fig. 2).

Fig. 3 Preventability of the ten

most common adverse drug reactions (ADRs)

Patients perception of how ADRs impacted their wellbeing ranged from no impact at all to totally changed
my life, with an average of 3.5 on a seven-point scale.
The proportions of patients who responded in each category are shown in Table 6. Not all patients were able to
rate their ADRs. This question was not pursued further in
interviews when patients expressed difficulty. Except for
myelosuppression and alopecia, high proportions of the
common ADRs were rated above mid-point on the scale.
Patients who responded to the question in relation to

631

Fig. 4 Patient impact rating of

the ten most common adverse
drug reactions (ADRs)

Table 6 Impact rating distribution

Impact rating

Proportion %

0 (no impact at all)

1
2
3
4
5
6 (totally changed my life)

7.0
12.7
11.0
15.9
17.2
16.8
19.3

anorexia all rated it as having totally changed their lives

(Fig. 4).

Discussion
It is clear from the results that ADRs are common in
hospitalised oncology patients. Extrapolating the results
from the study, about three quarters of admissions to Peter
MacCallum Cancer Centre have at least one ADR either
on admission or during hospital stay, and over 40% have
three or more ADRs. Comparatively, figures from this
study are higher than those reported in the 1992 Australian National Hospital Morbidity Data because of the
difference in methods of event identification [19]. The
national database is a collection of confidential summary
records for admitted patients separated from public and

private hospitals in Australia, while the data from this

study were obtained through personal interviews with
patients and comprehensive review of patients medical
records. This method clearly resulted in the identification
of more ADRs.
The range of ADRs experienced by cancer patients is
varied; however, the ten most common ADRs made up
almost two thirds of the incidents (Table 6). The point to
note is that these ten ADRs have high-documented incidence rates, that is, many of the ADRs that arose from
anti-cancer treatment were expected, confirming the findings of other researchers in this area [5].
When a predictable ADR occurs, questions should be
raised about whether it is preventable and if so, whether
known preventative measures have been used in an appropriate, adequate and timely manner. Using the modified Schumock and Thornton criteria, very few ADR incidents in this study were definitely preventable, and reactions such as fatigue, alopecia and anorexia were mostly assessed as not preventable. On the other hand, almost
half of the ADRs were probably preventable, for example,
constipation, nausea vomiting, dermatological reactions, diarrhoea and mucositis.
Opiates were the apparent cause of the majority of the
constipation incidents (63.8%), all of which were predictable and almost 90% classified as probably preventable. It was very clear from the study data that constipation was common as a result of opiates being initi-

632

ated or administered without a concurrent prophylactic

laxative, or, in some cases, prophylactic laxatives were
being administered at too low a dose. The situation was
similar with nausea vomiting. Anti-neoplastics were the
apparent cause of the majority of the nausea vomiting
incidents (69.4%), all of which were predictable and almost 60% of which were probably preventable. In contrast, although the majority of the fatigue incidents
(81.1%) caused by the administration of antineoplastics
were predictable, over 90% were not preventable.
Of significance in terms of improving patient care is
the finding that common reasons for ADRs to be assessed
as probably preventable were omission of or inappropriate
or inadequate use of known preventative measures. Possible explanations for this sub-optimal use of preventative
measures are manifold; for example, poor communication, inadequate ADR prevention guidelines and poor
implementation of guidelines. However, there seems to be
a more fundamental difference in perspectives and focus
between health-care providers and patients. The common
occurrence of many ADRs associated with cancer treatments and the non-life-threatening nature of these reactions in the context of the disease often means that what is
non-serious to a physician may be of considerable importance to a patient. The management of ADRs is currently more a reactive rather than a proactive exercise.
Further research is required to investigate how attitudes of
health-care providers in oncology can be influenced and
how ADRs could be prospectively identified so that
available measures and actions could be more reliably
undertaken to prevent and reduce the severity of ADRs.
Patients own evaluation of the impact of ADRs on
their lives adds a valuable perspective to the process of
medical decision making. In this study, patients rated
reactions that affected their normal daily functions, such
as constipation, nausea vomiting, fatigue and drowsiness, much higher than a reaction such as myelosuppression, the clinical significance of which may not be
apparent to a patient.
Our data identified reactions that patients regarded as
impacting significantly on their well-being that were
predictable and could have been better prevented. Constipation caused by the use of opiates and nausea
vomiting from chemotherapy are two obvious examples.
While these may seem trivial in the context of the disease
in question, they are obviously important to patients
themselves. The fact that these ADRs were still the most
commonly occurring ADRs in this study despite preventative measure being available indicates that current ADR
prevention and management practices require attention.
An increased understanding of patients perspectives
would enable specific areas to be identified and targeted
for vigourous and vigilant intervention [8, 15].
It must be noted that this study found fewer than two in
every 100 (1.6%) ADRs were definitely preventable. This
reflects the complexity of cancer therapy and the fact that

even though adequate preventative measures may have

been implemented, it does not necessarily mean than an
ADR will not occur. For example, patients having chemotherapy who also receive state-of the-art prophylactic,
multi-drug anti-emetic therapy often still got nausea and
vomiting.
While it could be argued that inviting patients to rank
their own perception of severity introduces a relatively
high degree of subjectivity, their perceptions are valid and
vital to an understanding of appropriate patient care. To
achieve optimal patient outcomes, there needs to be a
paradigm shift from the exclusive application of concrete
and limited categorisations such as the WHOs common
toxicity criteria [28]. The WHOs grading is based on
clinical criteria and is therefore considered very objective;
it does not take into consideration the patients perspectives.
In the last decade, the pharmacists traditional role of
compounding, dispensing and supplying drugs has been
expanded to include clinical participation. There has been
increasing involvement by pharmacists in clinical ward
rounds, in active monitoring of patients health status and
in clinical interventions in patients pharmaceutical care.
Many opportunities and platforms exist for pharmacists to
be involved in treatment decision making, ADR prevention and improved patient care [20] as well as in educating cancer patients about cancer treatments and ADRs.
Pharmacists should be proactive in counselling patients to
arm them with knowledge to make informed choices and
decisions about drug treatments. Pharmacists should
consider how effective they are in the community as
health educators; for example, in health promotion, in
providing advice on medications, in providing advice on
self-care and in ensuring quality of medicines. They
should be vigilant in preventing and reducing the severity
of ADRs that the treatments they dispense are likely to
cause. Pharmacists must continue to persist in establishing themselves as integral partners in interdisciplinary
primary health-care teams. The education and training of
pharmacists in effective communication and collaboration
with other health-care stakeholders should begin at the
undergraduate level and continue with on-going support
and education at the graduate level to maintain high levels
of professional confidence and skills.
A pertinent question for health-care professionals today is: How much does the impact ADRs have on our
patients well-being matter to us? In oncology practice
where a patients general health status is heavily compromised, many factors influence patient management.
More often than not, there is little choice between giving
a life-preserving drug that would be likely to elicit an
undesirable reaction and not giving it. The challenge in
these circumstances becomes how well we can prevent
predictable ADRs. The vigilant prevention of ADRs is
therefore an important issue in the quality use of medicine.

633

Conclusions
ADRs are common in hospitalised oncology patients,
predictable in many instances but definitely preventable
in only a few. There are, however, many occasions where
improved use of preventative measures has the potential

to contribute to reducing the incidence and severity of

ADRs. An understanding and appreciation of patients
perceptions of these events will help health-care professionals prioritise targeted interventions and prevention
strategies.

References
1. American Hospital Formulary Service
drug information (1984) Board of Directors of the American Society of
Hospital Pharmacists, Bethesda
2. Audit Commission (2001) A spoonful
of sugarmedicines management in
NHS hospitals. Audit Commission,
London
3. Australian adverse drug reactions bulletin (1998) Australian Adverse Drug
Reaction Advisory Committee
(ADRAC)
4. Australian prescription products guide
(2000) 29th edn. Australian Pharmaceutical Publishing
5. Bates D, Leape L, Petrycki S (1993)
Incidence and preventability of adverse
drug events in hospitalized adults.
J Gen Intern Med 8:289294
6. Council for International Organisation
for Medical Science (CIOMS) (1994)
WGI, Guidelines for preparing core
clinical-safety information on drugs.
Council for International Organisation
for Medical Sciences, Geneva
7. Clinical Oncological Society of Australia (2002) Optimising cancer care in
Australia. The Cancer Council Australia
and the National Cancer Council Initiative, Melbourne, pp 1122
8. Coates AS, Abraham SB, Kaye T et al
(1983) On the receiving endpatient
perception of the side-effects of cancer
chemotherapy. Eur J Cancer Clin Oncol
19:203208
9. Dewland P, Dewland J (1999) At the
coalface, but on the receiving end.
J Med Ethics 25:541546

10. Dormann H, Muth-Selbach U, Krebs S

et al (2000) Incidence and costs of adverse drug reactions during hospitalisation: computerised monitoring versus
stimulated spontaneous reporting. Drug
safety 22:161168
11. Dukes J, Chalker M, Leuwer PKM et al
(2000) Meylers side effects of drugs,
14th edn. Elsevier, New York
12. Fisher S, Kent TA, Bryant SG (1995)
Postmarketing surveillance by patient
self-monitoring: preliminary data for
sertraline versus fluoxetine. J Clini
Psych 56:288296
13. Foddy W (1994) Constructing questions
for interviews and questionnaires. Theory and practice in social research.
Cambridge University Press
14. Green CF, Mottram DR, Rowe PH,
Pirmohamed M (2000) Adverse drug
reactions as a cause of admission to an
acute medical assessment unit: a pilot
study. J Clin Pharm Ther 25:355361
15. Griffin AM, Butow PN Coates AS et al
(1996) On the receiving end. V: Patient
perceptions of the side effects of cancer
chemotherapy in 1993. An Oncol
7:189195
16. Jarernsiripornkul N, Krska J, Capps PA
et al (2002) Patient reporting of potential adverse drug reactions: a methodological study. Br J Clin Pharmacol
53:318325
17. Mjorndal T, Boman MD, Hagg S et al
(2002) Adverse drug reactions as a
cause for admissions to a department of
internal medicine. Pharmacoepidemiol
Drug Saf 11:6572

18. Naranjo C, Busto U, Sellars E et

al(1981) A method for estimating the
probability of adverse drug reactions.
Clin Pharmacol Ther 30:239245
19. National hospital morbidity data collectionAustralia (excluding NT)
(19921993)
20. Pon D (1996) Service plans and clinical
interventions targeted by the oncology
pharmacist. Pharm Pract Manag 16:18
30
21. PREMEDLINE and MEDLINE version: rel 6.1.0, source ID 1.7672.1.63
22. Roughead EE, Gilbert AL, Primrose JG,
Sansom LN (1998) Drug-related hospital admissions: a review of Australian
studies published 19881996. Medi J
Aust 168:405408
23. Schumock G, Thornton J (1992) Focusing on the preventability of adverse
drug reactions. Hosp Pharm 27:538
24. Tegeder I, Levy M, Muth-Selbach U
et al (1999) Retrospective analysis of
the frequency and recognition of adverse drug reactions by means of automatically recorded laboratory signals.
Br J Clin Pharmacol 47:557564
25. Thomson MICROMEDEX. MICROMEDEX(R) Health-care Series Vol.
117. http://micromedex.hcn.net.au/
mdx-full/. Cited 19 Oct 2003
26. White TJ, Arakelian A, Rho JP (1999)
Counting the costs of drug-related adverse events. Pharmacoeconomics
15:445458
27. Wilson R, Runciman W, Gibberd R et
al (1995) Quality in Australian health
care study. Med J Aust 163:458471
28. World Health Organization (1999)
Common toxicity criteria (CTC).
1 August 1999