Anda di halaman 1dari 10

Annals of Oncology 13: 1414–1423, 2002

Original article DOI: 10.1093/annonc/mdf255

Informing breast cancer patients about clinical trials:

a randomized clinical trial of an educational booklet
P. M. Ellis1*, P. N. Butow2 & M. H. N. Tattersall3
Medical Oncologist Hamilton Regional Cancer Center, Hamilton, Ontario, Canada; 2Medical Psychology Unit, 3Department of Cancer Medicine,
University of Sydney, Sydney, Australia

Received 7 January 2002; revised 4 April 2002; accepted 12 April 2002

Background: To evaluate the impact of an educational booklet on women’s knowledge of and

willingness to participate in a randomized clinical trial of treatment for breast cancer.
Materials and methods: Women undergoing surgery for newly diagnosed early stage breast cancer
were randomized to receive, or not, an information booklet explaining the need for and manner in which
randomized trials are conducted.
Results: Eighty-three women with newly diagnosed early stage breast cancer completed a question-
naire assessing attitudes to random clinical trials (RCTs) and were randomized to receive usual
information treatment options provided from their oncologist, or the educational booklet in addition to
usual information from their oncologist (42 usual information, 41 booklet). Fewer women who received
the clinical trials booklet (40% versus 47%) would consider participating in the hypothetical clinical
trial (P = 0.6). Mean knowledge scores increased for both groups; moreover, women who did not
receive the booklet showed similar improvements to women who received the booklet [mean difference
0.09, 95% confidence interval (CI) –0.66 to 0.83]. In a multivariate analysis women who would con-
sider participating in the clinical trial were more anxious [odds ratio (OR) 5.9, P = 0.02] had involved
lymph nodes (OR 5.8, P = 0.02) and were less influenced by negative aspects of clinical trials (OR 7.7,
P = 0.0001). After adjustment for these variables women who received the educational booklet were
significantly less likely to consider trial participation (OR 0.22, P = 0.05).
Conclusions: Educating women about clinical trials in this manner appears ineffective in improving
recruitment to RCTs. Women appear to be more influenced by their perception of risk than understand-
ing. This finding has ethical implications for communication of information about RCTs.
Key words: attitude to health, breast neoplasm, patient compliance, patient education, randomized
controlled trials

Introduction RCT [4]. There is an ethical obligation that trials be of suffi-

cient quality to have a favorable impact on society in the future
Randomized clinical trials (RCTs) require the doctor to com-
[5, 6]. Additionally, requirements for informed consent for
bine the traditional role of healer with that of trialist or
clinical trials are often more stringent than those for treatment
researcher [1]. According to Pellegrino [2], society permits
outside of the setting of a clinical trial [7–11]. The goal is to
experimentation with human subjects because of the benefits
provide sufficient information about the treatments offered on
that the whole society, as well as the experimental subject,
the trial, the benefits, side effects and alternative treatment
may gain from the new medical knowledge. Therefore, a
approaches, plus procedural information about the clinical
fundamental ethical issue surrounding participation in RCTs
trial for patients to make an autonomous decision.
reflects a need to safeguard individual rights versus an obliga-
tion to society to facilitate research [3]. Review of the literature suggests that the goals of informed
In order to safeguard individual autonomy, additional ethical consent for clinical trials are not always met [12]. A number of
requirements are imposed upon treatment received as part of a researchers have found that patients fail to recall much of the
information presented to them during a consent interview
[13–16]. Patient information sheets for clinical trials are often
*Correspondence to: Dr P. M. Ellis, Hamilton Regional Cancer Center, too complex for many patients to read [17, 18]. Poor recall and
699 Concession St, Hamilton, Ontario, L8V 5C2, Canada. understanding may also arise due to variation in the amount of
Tel: +1-905-387-9495; Fax: +1-905-575-6326;
E-mail: information provided by doctors during the informed consent

© 2002 European Society for Medical Oncology


consultation. Williams and Zwitter [19] found considerable Patients and methods
variation in the information that investigators state they would
The women in this study were identified from a larger study assessing
routinely give to patients considering entry into phase III clin- knowledge and attitudes to RCTs, which has been reported elsewhere [25].
ical trials. All women undergoing a definitive surgical operation (lumpectomy and
Several studies have examined ways to improve the provi- axillary dissection or mastectomy) for early stage invasive breast cancer
sion of information about randomized trials. Davis et al. [20] at the Sydney Breast Cancer Institute (SBCI) during 1998 were eligible
for inclusion in this study. Women with locally advanced breast cancer
randomized cancer patients to receive or not, the National
treated initially with chemotherapy and/or radiotherapy were not eligible
Cancer Institute (NCI) booklet on clinical trials. Patients for this study. In addition, women were ineligible for this study if they had
receiving the NCI booklet were more knowledgeable about metastatic disease at presentation, were unable to read English or were
clinical trials in general. Simes et al. [21] and Aaronson et al. unable to complete a questionnaire.
[22] randomized patients considering entry into a specific Women were identified via a multidisciplinary breast cancer meeting
clinical trial to receive more detailed information about the in which all new breast cancer cases from the preceding week were
discussed. Women were approached after they had received news of their
trial that was being discussed versus their standard practice.
own histopathology results, but prior to seeing a medical oncologist to dis-
They found that patients who received more detailed informa- cuss further treatment options. The purpose of the research was explained
tion were more knowledgeable about the trial, but both authors and women were given an information sheet to read. Written informed
observed a small (statistically insignificant) reduction in consent for this study was obtained from women who agreed to partici-
willingness to participate in the trial. pate. Institutional ethics committee approval was obtained from Central
Sydney Area Health Service.
Earlier research by our group suggested that cancer patients
Women were asked to complete a baseline questionnaire (see below),
and members of the general public have a poor understanding
then randomized centrally to a usual discussion about treatment options
about the need for RCTs to establish the worth of new and from their doctor (including clinical trials if appropriate), or an informa-
existing treatments, or the manner in which this would happen tion booklet about clinical trials in addition to usual discussion about
[23, 24]. Many patients did not appear to understand the treatment options from their doctor. No attempt was made to try to stan-
rationale for randomization. A number of women attending dardize the usual information provided by the doctor. The information
booklet explained the need for and manner in which RCTs are conducted.
focus group interviews indicated that they would be more
This was a short booklet which discussed what clinical trials are, how
willing to consider participating in a clinical trial once they treatment is decided on a RCT, potential advantages and disadvantages of
were better informed. The findings of a more recent larger participating in a clinical trial and common misunderstandings about
survey suggested that women who were more knowledgeable randomized trials identified from earlier research [23, 24]. It also pro-
about RCTs, in general, were more willing to consider partici- vided a list of questions to help women find out more information from
pating in a trial themselves [25]. their medical oncologist about specific clinical trials appropriate to their
own situation. The booklets were printed in black and white on neutral
Many cancer patients are approached to participate in clini-
colored paper. Women receiving the booklet who wanted additional
cal trials soon after their diagnosis when they are feeling general information about clinical trials could also receive a more detailed
stressed and vulnerable [26]. Discussions about RCTs at this booklet produced by the New South Wales Cancer Council. Four to six
point may add to feelings of anxiety and uncertainty. In order weeks later women were mailed a follow-up questionnaire to complete
to overcome this Baum [27, 28] has suggested that there is a and return in a reply paid envelope. Information was collected from the
women’s medical oncologist concerning disease data (tumor size, nodal
need to educate the public about the need for RCTs and the
involvement, histological grade, hormone receptor status), eligibility for
manner in which they are conducted. inclusion into a randomized trial of adjuvant therapy, whether clinical
Patient education booklets are one type of intervention that trials were discussed with the patient, the outcome of such discussions and
has been evaluated in a number of different settings to the actual treatment received.
improve information exchange. Educational booklets have
been shown to increase knowledge in patients with arthritis Survey instruments
and hypertension [29, 30]. Other authors report additional The baseline questionnaire measured:
• demographic details including age, marital status, education, occupa-
benefits including improved satisfaction [31], fewer GP visits,
tion, ethnicity and medical/allied health training;
less referrals to physiotherapy and fewer hospital admissions • Hospital Anxiety and Depression Scale (HADS) [34, 35];
among patients with back pain [32]; and reduced anxiety, • women’s preferences for the amount of information they wish to
greater physiologic recovery and improved mood among receive from their doctor (three item scale previously described by
patients undergoing coronary artery bypass graft surgery [33]. Cassileth et al.) [36] and their level of involvement in clinical deci-
This study reports the results of a RCT evaluating the impact sion-making (five item scale previously described by Degner and
Sloan) [37];
of an informational booklet, explaining the need for and
• knowledge about the need for and manner in which RCTs are
manner in which RCTs are conducted, on women’s know- conducted using a seven item scale developed for this questionnaire;
ledge of and willingness to participate in a RCT of treatment • attitudes to RCTs. This was a 36 item scale assessing the impact of
for breast cancer. individual items on women’s willingness to participate in randomized

trials. This scale has high internal reliability (Cronbach α 0.96) [25] Results
and measures four factors: perception of the positive aspects of clin-
ical trials; perception of the negative aspects of clinical trials; percep- One hundred and one women underwent definitive surgical
tion of loss of control/inconvenience on a clinical trial and views treatment for invasive breast cancer during the study period
about altruism. Higher scores (range 0–7) reflect greater willingness (Figure 1). Ninety-five women were approached about the
to participate in a RCT; study and 83 (87%) agreed to participate. Four women with-
• general willingness to participate in RCTs. drew after randomization and follow-up questionnaires were
The follow-up questionnaire measured: returned by 67 (71%) of women. Demographic information on
• information and involvement in decision-making preferences; the 83 women randomized is shown in Table 1. Women ran-
• knowledge about clinical trials; domized to receive the clinical trial booklet were on average
• attitudes to RCTs; 3 years older [95% confidence interval (CI) –1.5 to 7.5 years]
• willingness to join a hypothetical trial of adjuvant therapy for breast than women in the control group, but this difference was not
cancer (see Appendix 1); statistically significant. A greater proportion of women who
• satisfaction with the amount of information in the consultation, received the booklet had received tertiary education, but again
communication skills of the doctor and the level of participation in the this difference was not significant (P = 0.44). Significantly
consultation. This scale has been extensively used by the Medical more women who received the clinical trials booklet were
Psychology Unit to assess satisfaction in doctor–patient communica- classified as possible or definite cases of anxiety (63 versus
tion research [38–40]; 33%, P = 0.0008) and depression (21 versus 5%, P = 0.03)
• women’s use of the clinical trials booklet (four item scale—‘not at according to HADS. Disease characteristics of the 83 women
all’, ‘somewhat helpful’, ‘moderately helpful’, ‘very helpful’); are summarized in Table 2. The two groups were well
• recollections of discussions about clinical trials (did your doctor matched according to tumor size, lymph node status, histo-
discuss clinical trials with you?); logical grade and hormone receptor status.
• other sources of information.
Over 95% of women wanted to receive all information
Data analysis was undertaken using SPSS, version 6.1 (SPSS, Chicago,
whether good or bad from their doctor. These preferences
IL, USA) [41]. Analysis was by intention-to-treat. At the time of this study,
remained stable over the 4–6 week study period. Short-term
few women were invited to participate in an actual clinical trial, therefore
changes in women’s preferences for involvement in decision-
the main outcome of the randomized trial was willingness to join the
making were evident. Twenty-two per cent of women would
clinical trial described in a hypothetical scenario. However, women were
prefer to adopt a passive role in decision-making, 44% a col-
asked to make this decision at a salient time point (soon after making
decisions about their own breast cancer treatment). Secondary outcomes
laborative role and 34% an active role. Preferences remained
included changes over time in women’s knowledge and attitudes towards unchanged for 45 of 67 (68%) women in the follow-up
randomized trials, plus women’s satisfaction. Descriptive information questionnaire 4–6 weeks later. However, 11 (16%) women
was collected on women’s reasons for agreeing, or not agreeing, to wanted a greater role in decision-making and 11 (16%)
participate in the hypothetical trial. Differences in baseline variables wanted the doctor to play a greater role in decision-making.
between the randomized groups were assessed using χ2-tests or Student’s Knowledge scores were calculated by summing the number
t-test. Differences in willingness to participate in the hypothetical breast of correct responses to the seven items (range 0–7). The mean
cancer trial were assessed using χ2-test. Change scores for knowledge and knowledge score was 4.1 (SD 2.0). Baseline knowledge was
attitudes to clinical trials were assessed using paired Student’s t-tests. A slightly higher among women randomized to receive the
stepwise backward multivariate logistic regression was conducted to clinical trials booklet (difference 0.9, 95% CI –0.05 to 1.7,
examine the impact of receiving the information booklet on willingness to P = 0.06). There was a small increase in knowledge between
participate in the hypothetical RCT after adjustment for the effects of baseline and reassessment 4–6 weeks later (mean 0.6 95% CI
confounding variables. Information on women’s use of the clinical trials 0.2 to 0.9, P = 0.003). However, women who received the
booklet, other sources of information and their recollection of discussions clinical trials booklet showed no greater improvement in
about clinical trials will be presented in a separate paper.
knowledge than women in the control group (difference 0.09,
Earlier research suggested that 15–20% of women with early stage
95% CI –0.66 to 0.83, P = 0.82).
breast cancer treated in the Medical Oncology Department at Royal
Comparison was made of the four factor scores assessing
Prince Alfred Hospital were recruited onto clinical trials [42]. A sample of
attitudes to RCTs between women randomized to receive the
122 would be able to detect an increase from 20 to 45% with a power of
clinical trials booklet and the control group. Higher scores
0.80 and a significance level of 0.05. An interim analysis was planned
when approximately two thirds of women had been recruited. At this point
equate to greater willingness to participate in randomized trials
83 women had been recruited. A decision was made to close the trial at (range 0–7). There were no differences in baseline attitudes to
this time because of both a lack of efficacy and low salience of the inter- randomized trials. Women’s attitudes to randomized trials
vention. Assuming a best case scenario (i.e. all the remaining women in changed over time. At follow-up, women were less influenced
the intervention group decided to join the trial) the difference between the by positive aspects of clinical trials (–0.22, 95% CI –0.04 to
groups would not have achieved statistical significance. The final study –0.4, P = 0.02) and altruistic motivations (–0.23, 95%
had a power of 0.60 to detect a difference of 25% and a power of 0.80 to CI –0.04 to –0.43, P = 0.02). Their was no evidence of any
detect a 35% difference in willingness to join the hypothetical trial. change in women’s perceptions of the negative aspects of

Figure 1. Flow diagram of the study population.

clinical trials (0.03, 95% CI –0.16 to 0.22, P = 0.73) or about described along with a hypothetical RCT in which women
practical issues/loss of control (0.06, 95% CI –0.17 to 0.28, may receive tamoxifen alone, chemotherapy alone or chemo-
P = 0.63). therapy followed by tamoxifen. Twenty-six women (43%)
In response to the initial questionnaire, 20 women (25%) indicated they would be prepared to join the clinical trial,
would consider participating in clinical trials, 24 women (30%) while 34 women (57%) were not prepared to. Data were
would not and 35 women (45%) were unsure. In the follow-up missing for the remaining seven women. Fourteen women
questionnaire a scenario describing a hypothetical woman (47%) randomized to the standard information arm were pre-
with a node-positive breast cancer was presented (see Appen- pared to join the clinical trial, compared with 12 women
dix 1). Treatment options of tamoxifen or chemotherapy were (40%) randomized to receive the clinical trials booklet. There-

Table 1. Comparison of demographic characteristics of patients randomized to receive the clinical trials booklet
or standard information

Standard information Booklet P value

Intervention 42 (51%) 41 (49%)
Age (95%CI) 52.8 years 49.8 years 0.19
(49.5–56.1) (46.5–53.1)
Marital status 0.62
Single 3 (8%) 5 (13%)
Married/defacto 26 (65%) 25 (66%)
Separated/divorced/widowed 11 (27%) 8 (21%)
Education 0.44
Prior to HSC 16 (40%) 9 (24%)
HSC 2 (5%) 2 (5%)
Tertiary, non-university 9 (23%) 9 (24%)
Tertiary, university 13 (32%) 18 (47%)
Occupation 0.85
Managerial/professional 18 (43%) 20 (49%)
Para-professional 1 (2%) 2 (5%)
Sales/clerical/trades 10 (24%) 8 (19%)
Home duties 13 (31%) 11 (27%)
Country of birth 0.34
Australia 29 (71%) 24 (63%)
Other Anglo–Saxon 7 (17%) 7 (18%)
European 2 (5%) 6 (16%)
Other 3 (7%) 1 (3%)
Previous breast cancer 0.34
Yes 2 (5%) 4 (10%)
Medical/allied health training 0.86
Yes 8 (20%) 8 (21%)
Anxiety 0.008
Non-case 28 (67%) 14 (37%)
Possible/definite case 14 (33%) 24 (63%)
Depression 0.03
Non-case 40 (95%) 30 (79%)
Possible/definite case 2 (55) 8 (21%)

CI, confidence interval; HSC, Higher School Certificate, 12 years of formal education.

fore slightly fewer women (7%) receiving the clinical trials medical/allied health training), disease variables (tumor size,
booklet than women in the standard information arm were nodal involvement) preferences for involvement in clinical
willing to join the clinical trial, but this difference was not decision-making, anxiety, depression, along with the four
statistically significant (95% CI –32% to 18%, P = 0.60). factor scores measuring attitudes to clinical trials were entered
There was no evidence of any difference between the groups into the model. The results are summarized in Table 3. After
in women’s satisfaction with the consultation (P = 0.65). adjustment for other variables, women who received the
A multivariate logistic regression analysis was conducted to clinical trials booklet were significantly less likely to join
examine the impact of the clinical trials booklet on women’s the hypothetical clinical trial (OR 0.22, 95% CI 0.04–1.0,
decision to join the hypothetical clinical trial of adjuvant P = 0.05). Women with possible or definite cases of anxiety
therapy for breast cancer, after adjustment for potential con- were more likely to join the clinical trial (OR 5.9, 95% CI
founding factors. Demographic variables (age, education, 1.2–30.1, P = 0.02), as were women with involved lymph

Table 2. Comparison of disease characteristics of women included in the evaluation of the clinical trials booklet

Variable Standard information Booklet P value

Tumor size (mm) 20.4 (95% CI 15.8–25) 19.1 (95% CI 14.5–23.7) 0.68
T status 0.55
T1 (≤20 mm) 25 (63%) 30 (73%)
T2 (21–50 mm) 13 (32%) 9 (22%)
T3 (≥51 mm) 2 (5%) 2 (5%)
Lymph node status 0.28
N0 (not involved) 29 (69%) 23 (58%)
N1 (involved) 13 (31%) 17 (42%)
Stage 0.58
I 20 (50%) 23 (58%)
II 20 (50%) 18 (42%)
Oestrogen receptor 0.22
Positive 31 (78%) 36 (88%)
Negative 9 (22%) 5 (12%)
Progesterone receptor 0.56
Positive 29 (73%) 32 (78%)
Negative 11 (27%) 9 (22%)
Histological grade 0.44
Grade I 14 (34%) 10 (24%)
Grade II 11 (27%) 16 (39%)
Grade III 16 (39%) 15 (37%)

CI, confidence interval.

Table 3. Results of the multivariate logistic regression analysis examining the impact of the clinical
trial booklet on women’s decision to join a clinical trial after adjustment for potential confounders

Variable Odds ratio 95% CI P value

Intervention 0.05
Standard information 1.0
Clinical trials booklet 0.22 0.04–1.0
Anxiety 0.02
Non-case 1.0
Possible/definite case 5.9 1.2–30.1
Negative aspects of clinical trials 7.7 2.3–25.2 <0.0001
Nodal status 0.02
Not involved 1.0
Involved 5.8 1.1–28.8

CI, confidence interval.

nodes (OR 5.8, 95% CI 1.1–28.8, P = 0.02). Women who were (n = 10), there was nothing to lose by joining the clinical trial
less likely to be influenced by negative aspects of clinical (n = 8) and altruistic motivations—helping to advance med-
trials were also significantly more likely to join the hypo- ical knowledge (n = 8) and helping other people (n = 8). Other
thetical clinical trial (OR 7.7, 95% CI 2.3–25.2, P <0.0001). reasons included the possibility of receiving better treatment
Women were asked why they would choose to join or not to (n = 6) and the stage of the disease (n = 5).
join the clinical trial. The responses are summarized in Table 4. There were a greater number of reasons why women would
The most common reasons given for choosing to join the choose not to join the clinical trial. There were no major
clinical trial were that any of the treatments appear to be helpful differences between the randomized groups. The reasons

Table 4. Summary of women’s reasons for joining, or not joining, the clinical trial in the hypothetical

Frequency Proportion (%)

Reasons why women would choose to join the clinical trial (n = 24, more than one response allowed)
Any of the treatments appear to be helpful 10 42
Nothing to lose by joining the trial 8 33
Help advance medical knowledge 8 33
To help other people 8 33
Might receive better treatment 6 25

Reasons why women would choose not to join the clinical trial (n = 29, more than one response allowed)
Want to be able to pick the treatment myself 11 38
More confident about chemotherapy 6 21
Don’t like the idea of randomization 5 17
Don’t want to have 5 years of tamoxifen 4 14
Want to avoid the side effects of chemotherapy 4 14
Want my doctor to pick my treatment 3 10
Don’t want to be experimented on 3 10
Would worry whether I received the right treatment 2 7
Want to feel in control 2 7
Prefer the quickest treatment 1 3

predominantly related to dislike about treatment allocation on breast cancer. Women with involved lymph nodes would be
a clinical trial and concerns about receiving one of the treat- aware that this fact carries a greater risk of future relapse and
ments offered on the clinical trial. death from breast cancer. These findings are supported by
observations from Siminoff and others that women with early
stage breast cancer who overestimate their prognosis with
Discussion standard therapies are less likely to consider participating in a
Earlier research by our group suggested that willingness to clinical trial [43–45]. By chance, significantly more women
participate in randomized trials was associated with greater randomized to the information booklet were classified as
understanding of the issues involved [25]. These findings sug- cases of anxiety and depression. Following adjustment for this
gested that patient education about the need for and manner in imbalance, anxiety was still associated with increased willing-
which RCTs are conducted might improve patients’ under- ness to join a clinical trial. Therefore women’s perception of
standing about randomized trials and lead to improved recruit- the implications of their illness appears influential in their
ment. However, the results of this study suggest that general decision to join a randomized trial. The findings highlight the
education of patients about RCTs in this manner was an potential for subtle coercion when discussing both standard
ineffective strategy in promoting recruitment to clinical trials. treatment options and clinical trials.
Previous studies have demonstrated that patients receiving The results of this study suggest that women’s perceptions
more detailed information about specific clinical trials have of the negative aspects of clinical trials (randomization, experi-
improved knowledge about those trials [21, 22]. However, mental aspects, greater uncertainty) are far more influential in
women in this trial who received the clinical trial booklet decision-making than their perceptions about the advantages.
showed no greater improvement in knowledge than women Previous research has suggested that the main reasons patients
who received usual information about treatment options from give for joining a clinical trial are the potential to receive more
their doctor. These findings suggest that factors other than effective treatments and helping future patients [46, 47]. In the
knowledge or understanding about a trial may be more import- follow-up questionnaire, women in this study appear less
ant in the decision to enter a clinical trial. likely to be influenced by potential advantages of clinical
Characteristics associated with the women’s cancer were trials. This finding may reflect a satisfaction bias with the
influential in the decision to join the clinical trial. We did not treatment decision women made as only two women received
directly measure women’s estimate of their prognosis. How- adjuvant therapy as part of a clinical trial. However, an addi-
ever, nodal status is one of the major prognostic factors for tional explanation is that women, having reflected on the

issues over a few weeks, believed there were fewer advantages in a clinical trial for patients with advanced or metastatic
to participating in clinical trials than they originally considered. cancer may be quite different from those of patients with early
There were a variety of reasons given by women for not stage disease.
agreeing to join the clinical trial of adjuvant therapy. The Nevertheless, the findings of this study suggest that the
major reasons reflected specific concerns about the treatments decision to enter clinical trials in oncology is a complex inter-
offered on the trial (concern about side effects or wanting to action of patients’ pre-existing beliefs, the perception of the
avoid one of the treatments), or dislike about the allocation of seriousness of their disease and endorsement of the trial by the
treatment (want themselves or doctor to select treatment or physician. Strategies designed to educate patients about trials
dislike of random allocation). A desire by patients for a in general, or about specific clinical trials, have not proven
specific treatment recommendation from the doctor appears to effective in improving recruitment. Further research should
be an important obstacle to recruitment to clinical trials [48], examine more closely the exchange of information that takes
although a recent study by Fallowfield et al. [49] demonstrated place during the informed consent interview(s). In the interim
that clearer explanations about randomization increased it is important that patients’ concerns about the negative aspects
patients’ willingness to participate in trials in general. of randomized trials be aired in consultation to ensure patient
Common reasons given by women for agreeing to join the understanding and avoid undue coercion.
clinical trial included ‘any of the treatments appear to be help-
ful’ and ‘there did not appear to be anything to lose by joining
the trial’. These responses suggest an understanding of issues
relating to equipoise. Equipoise exists when either indivi- Dr Peter Ellis was supported by a postgraduate scholarship
dually or collectively, doctors are uncertain about the com- from the National Health & Medical Research Council of
parative merits of different treatment approaches. It is a Australia from 1997 to 1999.
crucial issue for doctors when deciding whether to participate
in RCTs [3, 5, 50]. These findings support those of Fallow-
field et al. [49, 51], that greater understanding of this issue by References
patients may also be an important factor in their decision to 1. Scoles L, Silagy C. Are clinical trials in general practice ethical? Aust
enter a randomized trial. Fam Physician 1993; 22: 182–183.
Unlike previous interventions [21, 22], we were unable to 2. Pellegrino ED. Beneficence, scientific autonomy, and self-interest:
document any effect from the clinical trial booklet in this ethical dilemmas in clinical research. Camb Q Healthc Ethics 1992;
study in improving knowledge about clinical trials among 4: 361–369.
women with breast cancer. Despite the lack of effectiveness 3. Lilford RJ, Jackson J. Equipoise and the ethics of randomization. J R
Soc Med 1995; 88: 552–559.
on women’s knowledge, receiving the clinical trial booklet
4. Beauchamp TL, Childress JF. Principles of biomedical ethics. New
was associated with reduced willingness to participate in a
York: Oxford University Press 1994.
clinical trial of adjuvant therapy for breast cancer. Patient
5. Lilford RJ. The substantive ethics of clinical trials. Clin Obstet
education booklets have been shown to have a number of
Gynecol 1992; 35: 837–845.
improved outcomes in a number of different settings [29, 32,
6. Palter SF. Ethics of clinical trials. Semin Reprod Endocrinol 1996;
33, 52, 53]. In general, the information contained in these 14: 85–92.
educational interventions was immediately relevant to the 7. Miller C, Searight HR, Grable D et al. Comprehension and recall of
problem with which the patient presented. Few women in this the informational content of the informed consent document: an
trial were presented the option of participating in an actual evaluation of 168 patients in a controlled clinical trial. J Clin Res
clinical trial. The clinical trial booklet may have been ineffect- Drug Dev 1994; 8: 237–248.
ive because the information contained in it was not particu- 8. Rimer B, Jones WL, Keintz MK et al. Informed consent: a crucial
larly salient to decision-making for the majority of women in step in cancer patient education. Health Educ Q 1984; 10 (Suppl):
this study. If so, the results highlight the importance of 30–42.
physician reinforcement of information about clinical trials. 9. Dal Re R. Elements of informed consent in clinical research with
There are some limitations to this study. The number of drugs: a survey of Spanish clinical investigators. J Intern Med 1992;
231: 375–379.
women included in the study was relatively small and approx-
10. Tattersall MHN, Simes RJ. Issues in informed consent. In Williams
imately 25% of women did not complete the follow-up assess-
CJ (ed.): Introducing New Treatments for Cancer: Practical, Ethical
ment. However, there were no major differences in baseline
and Legal Problems. Chichester, UK: John Wiley & Sons 1992; 79–90.
characteristics of those women not completing the follow-up
11. Feldman-Stewart D, Chammas S, Hayter C et al. An empirical
assessment and those included in the analysis (data not approach to informed consent in ovarian cancer. J Clin Epidemiol
shown). This study was conducted among women with early 1996; 49: 1259–1269.
stage breast cancer. It is not clear whether these findings can 12. Cassileth BR, Zupkis RB, Sutton-Smith K et al. Informed consent:
be generalised to patients with other cancers considering entry why are its goals imperfectly realized. N Engl J Med 1980; 302:
into a clinical trial. Additionally, the motivation to participate 896–900.

13. Harth SC, Thong YH. Parental perceptions and attitudes about 34. Zigmond AS, Snaith RP. The hospital anxiety and depression scale.
informed consent in clinical research involving children. Soc Sci Acta Psychiatr Scand 1983; 67: 361–370.
Med 1995; 41: 1647–1651. 35. Moorey S, Greer S, Watson M et al. The factor structure of the
14. Olver IN, Buchanan L, Laidlaw C et al. The adequacy of consent Hospital Anxiety and Depression Scale in patients with cancer. Br
forms for informing patients entering oncological clinical trials. Ann J Psych 1991; 158: 255–259.
Oncol 1995; 6: 867–870. 36. Cassileth BR, Zupkis RV, Sutton-Smith K et al. Information and
15. Sutherland HJ, Lockwood GA, Till JE. Are we getting informed participation preferences among cancer patients. Ann Intern Med
consent from patients with cancer? J R Soc Med 1990; 83: 439–443. 1980; 92: 832–836.
16. Dunn SM, Butow PN, Tattersall MHN et al. General information 37. Degner LF, Sloan JA. Decision making during serious illness: what
tapes inhibit recall of the cancer consultation. J Clin Oncol 1993; 11: role do patients really want to play? J Clin Epidemiol 1992; 45:
2279–2286. 941–950.
17. Grossman SA, Piantadosi S, Covahey C. Are informed consent forms 38. Butow PN, Dunn SM, Tattersall MHN et al. Patient participation in
that describe clinical oncology research protocols readable by most the cancer consultation: evaluation of a question prompt sheet. Ann
patients and their families? J Clin Oncol 1994; 12: 2211–2215. Oncol 1994; 5: 199–204.
18. Taylor KM, Bezjak A, Hunter R et al. Informed consent for clinical 39. Brown R, Butow PN, Boyer MJ et al. Prompting patient participation
trials: Is simpler better? J Natl Cancer Inst 1998; 90: 644–645. in the cancer consultation: evaluation of a question prompt sheet and
19. Williams CJ, Zwitter M. Informed consent in European multicentre coaching in question asking. Br J Cancer 1999; 80: 242–248.
randomised clinical trials—are patients really informed? Eur J Cancer 40. Dunn SM, Patterson PU, Butow PN et al. Cancer by another name: a
1994; 30A: 907–910. randomized trial of the effects of euphemism and uncertainty in com-
20. Davis SW, Nealon EO, Stone JC. Evaluation of the National Cancer municating with cancer patients. J Clin Oncol 1993; 11: 989–996.
Institute’s clinical trials booklet. J Natl Cancer Inst Monogr 1993; 14: 41. SPSS for Windows: Base system user’s guide, Release 6.1. Chicago,
139–145. IL: SPSS 1994; 828.
21. Simes RJ, Tattersall MHN, Coates AS et al. Randomised comparison 42. Ellis PM, Butow PN, Tattersall MHN et al. Accrual to clinical trials in
of procedures for obtaining informed consent in clinical trials of breast cancer. Ann Sci Meeting Clin Oncol Soc Aust 1996: A12
treatment of cancer. Br Med J 1986; 293: 1065–1068. (Abstr).
22. Aaronson NK, Visser-Pol E, Leenhouts GHMW et al. Telephone- 43. Siminoff LA, Fetting JH. Effects of outcome framing on treatment
based nursing intervention improves the effectiveness of the decisions in the real world: impact of framing on adjuvant breast
informed consent process in cancer clinical trials. J Clin Oncol 1996; cancer decisions. Med Decis Making 1989; 9: 262–271.
14: 984–996. 44. Sheldon JM, Fetting JH, Siminoff LA. Offering the option of
23. Ellis PM, Butow PN. Focus group interviews examining attitudes to randomized clinical trials to cancer patients who overestimate their
randomised trials among breast cancer patients and the general prognoses with standard therapies. Cancer Invest 1993; 11: 57–62.
community. Aust N Z J Public Health 1998; 22: 528–531. 45. Fetting JH, Siminoff LA, Piantadosi S et al. Effect of patients’
24. Ellis PM, Dowsett SM, Butow PN et al. Attitudes to randomised expectations of benefit with standard breast cancer adjuvant chemo-
clinical trials among outpatients attending a medical oncology clinic. therapy on participation in a randomized clinical trial: a clinical
Health Expectations 1999; 2: 33–43. vignette study. J Clin Oncol 1990; 8: 1476–1482.
25. Ellis PM, Butow PN, Tattersall MHN et al. Randomised clinical trials 46. Verheggen FW, Nieman F, Jonkers R. Determinants of patient
in oncology: understanding and attitudes predict willingness to participation in clinical studies requiring informed consent: Why
participate. J Clin Oncol 2001; 19: 3554–3561. patients enter a clinical trial. Patient Educ Couns 1998; 35: 111–125.
26. Thornton H. Clinical trials: a brave new partnership? J Med Ethics 47. Edwards SL, Lilford RJ, Hewison J. The ethics of randomised
1994; 20: 19–22. controlled trials from the perspectives of patients, the public and
27. Baum M. New approach for recruitment into randomised controlled healthcare professionals. Br Med J 1998; 317: 1209–1212.
trials. Lancet 1993; 341: 812–813. 48. Johnson JD, Roberts CS, Cox CE et al. Breast cancer patients’
28. Baum M. Clinical trials: a brave new partnership: a response to Mrs personality style, age, and treatment decision making. J Surg Oncol
Thornton. J Med Ethics 1994; 20: 23–25. 1996; 63: 183–186.
29. Maggs FM, Jubb RW, Kemm JR. Single-blind randomized control- 49. Fallowfield LJ, Jenkins V, Brennan C et al. Attitudes of patients to
led trial of an educational booklet for patients with chronic arthritis. randomised clinical trials of cancer therapy. Eur J Cancer 1998; 34:
Br J Rheumatol 1996; 35: 775–777. 1554–1559.
30. Watkins CJ, Papacosta AO, Chinn S et al. A randomized controlled 50. Freedman B. Equipoise and the ethics of clinical research. N Engl
trial of an information booklet for hypertensive patients in general J Med 1987; 317: 141–145.
practice. J R Coll Gen Pract 1987; 37: 548–550. 51. Fleissig A, Jenkins V, Fallowfield L. Results of an intervention to
31. O’Neill P, Humphries GM, Field EA. The use of an information improve communication about randomised clinical trials of cancer
leaflet for patients undergoing wisdom tooth removal. Br J Oral therapy. Eur J Cancer 2001; 37: 322–331.
Maxillofac Surg 1996; 34: 331–334. 52. Fender GRK, Prentice A, Gorst T et al. Randomised controlled trial
32. Roland M, Dixon M. Randomized controlled trial of an educational of educational package on management of menorrhagia in primary
booklet for patients presenting with back pain in general practice. J R care: the Anglia Menorrhagia Education Study. Br Med J 1999; 318:
Coll Gen Pract 1989; 39: 244–246. 1246–1250.
33. Cupples SA. Effects of timing and reinforcement of preoperative 53. Brandberg Y, Bergenmar M, Michelson H et al. Six-month follow-up
education on knowledge and recovery of patients having coronary of effects of an information programme for patients with malignant
artery bypass graft surgery. Heart Lung 1991; 20: 654–660. melanoma. Patient Educ Couns 1996; 28: 201–208.

Appendix 1. Hypothetical clinical trial nausea and vomiting, hair loss, sore mouth and eyes and
scenario diarrhea. There may also be an increased risk of infection. Not
every patient experiences side effects and many of these side
Now that you have made decisions about your own treatment
effects can be managed with medication.
we would like you to consider a hypothetical situation about a
Sheila’s doctor explained that either tamoxifen or chemo-
women called Sheila. Remember, while some of the details
therapy would be beneficial to her. She (or Sheila’s doctor)
may appear similar to yours, the details refer to Sheila.
could choose to have either one of these treatments. However,
Sheila is a 56-year-old woman who has recently had a
it is not known whether tamoxifen is better than chemo-
mastectomy for breast cancer. The cancer spread to several of
her lymph glands. Radiotherapy was not recommended for therapy, or whether combining tamoxifen and chemotherapy
Sheila. She has gone to talk to a medical oncologist who is better than either one on its own. Sheila’s doctor invites her
recommended that she have some additional treatment to to participate in an international clinical trial comparing these
reduce the chances of the cancer returning. In Sheila’s situ- treatments. If she were to agree, she would either receive
ation, both tamoxifen and chemotherapy, have been shown to tamoxifen on its own, chemotherapy on its own, or tamoxifen
be beneficial. plus chemotherapy. Her treatment would be chosen at random
Tamoxifen is a hormonal tablet taken once per day for (neither Sheila nor her doctor would pick her treatment).
5 years. The major side effects include hot flushes, vaginal Sheila has the option of choosing either chemotherapy or
dryness or itching, weight gain, mood changes and occasion- tamoxifen, or joining the clinical trial comparing these treat-
ally some nausea. Not all patients experience these side ments. Given the experience you have gained making deci-
effects. Chemotherapy is given as a combination of injections sions about breast cancer treatments, if you were Sheila,
and tablets over 6 months. The major side effects include would you agree to participate in the clinical trial?