Attitudes towards clinical research amongst participants and nonparticipants


Dr Soren Madsen MD Research Associate, Department of Medical Gastroenterology C, Copenhagen University Hospital in Herlev, Herlev Ringvej 75, DK-2730 Herlev, Denmark (fax: +45 44 94 40 56; e-mail:


Abstract. Madsen SM, Mirza MR, Holm S, Hilsted KL, Kampmann K, Riis P (Copenhagen University Hospital in Herlev, Herlev, Denmark; Odense University Hospital in Odense, Odense C, Denmark; University of Manchester, Manchester, UK; Copenhagen University Hospital in Herlev, Herlev, Denmark; and Odense University Hospital in Odense, Odense C, Denmark). Attitudes towards clinical research amongst participants and nonparticipants. J Intern Med 2002; 251: 156–168.

Objectives. To investigate attitudes to clinical research amongst cancer trial participants and nonparticipants, and to compare results with those from previous studies amongst participants in noncancer trials.

Design. Trial participating respondents were given three questionnaires during the clinical trials. Respondents amongst patients declining randomization answered a single questionnaire.

Setting. Participants and nonparticipants in randomized clinical cancer trials.

Subjects. Forty-one participants and 47 nonparticipants in cancer trials.

Results. Altruistic motives of physicians to conduct medical research were highly rated. Attitudes towards clinical research were positive in all groups, with nonparticipant respondents being the least positive. Eight to nine tenths found scientific testing necessary before general health service implementation. Trial participants were, as compared with nonparticipating respondents, more positive towards both participation of self and others. Both personal and altruistic motives for participation were highly rated. Primary reasons for nonparticipation were fear of ‘the unknown’ and/or unease with randomization. Only a minority felt a moral problem created by declining trial participation. Respondents amongst noncancer participants were more satisfied with the information given than both cancer participants and cancer nonparticipants. Negative experiences in cancer participants generally dealt with frustration related to seeing too many physicians at check-up appointments.

Conclusion. Attitudes towards clinical research are generally positive even in cancer nonparticipants. Both personal and altruistic motives for participation were highly rated. A fear of ‘the unknown’ and resentments towards randomization were primary reasons to renounce participation. Seeing too many physicians at check-up appointments seems to be an important factor for negative experiences in cancer trial participants.


In recent years the fraction of eligible patients recruited to clinical trials have decreased world wide [1–3], postponing trial completion and making the projective universality of trial results unpredictable [4–10]. The declining willingness to participate can only be partly explained by physician factors [2, 3, 11–16], whereas patient factors probably have gained increasing importance [3, 13, 14, 17–20]. There is, however, a lack of knowledge about general attitudes towards clinical research, how patients actually experience trial participation, and why patients accept or decline participation.

To investigate these issues we performed a questionnaire-survey amongst cancer patients accepting and patients declining randomization to trials including chemotherapy. We further compared the results with those from previous questionnaire-surveys amongst participants in two noncancer trials [21].

Materials and methods

All patients found eligible to participate in the clinical cancer trials were also eligible to participate in the questionnaire study. Questionnaires were given at the first visit after the patient’s decision whether to be randomized or not to 41 patients accepting and 47 patients declining randomization. Only six of the approached patients declined to participate in the questionnaire study (7%). Further, questionnaires were handed out to cancer trial participants during the course and after completion of the trial. Distributions of cancer disease, sex and age are shown in Table 1. In the breast cancer group only premenopausal patients were included, as the protocol for postmenopausal patients did not include chemotherapy but endocrine therapy. The study was performed at the oncological departments in Herlev University Hospital and Odense University Hospital during 1997–1999.

Table 1.   Number of respondents, sex and age distribution amongst respondents in questionnaire survey Thumbnail image of

The initial questionnaire probed attitudes towards aspects of clinical trials including reasons for participation and nonparticipation. In addition, the respondents were asked to evaluate the information they had received about the clinical cancer trial earlier. The questionnaires given during, and after the completion of the trial included questions about perceived satisfaction with participation. Most questions were on a multiple-choice form, with an additional space for free text. Comparisons between groups were only made, when questions were phrased identically.

All patients gave both verbal and written informed consent. The survey fulfilled the demands of Danish law and the Helsinki Declaration, and was approved by the Research Ethics Committee of Copenhagen County and the Scientific Ethical Committee for Vejle and Funen Counties (reg. no. KA 95059).

Previous questionnaire studies amongst noncancer trial participants [21] were performed amongst patients with inflammatory bowel disease (IBD) participating in one of two clinical studies (IBD-participants). One study was an open-labelled, randomized study comparing treatment of active left-sided ulcerative colitis with either self-administered subcutaneous injections of interferon-α-2a (Roceron®, Roche Inc., Hvidovre, Denmark) or prednisolone enemas (Predniment®, Ferring Inc, Hørsholm, Denmark). Thirty-two patients were randomized [22], and all completed the questionnaires given (Table 1). The other clinical study evaluated low-field magnetic resonance imaging (MRI) in active inflammatory bowel disease. Forty-seven patients participated [23], and all completed the questionnaires given (Table 1).

Statistical methods

Nonparametric statistical tests (Kruskall–Wallis tests, Mann–Whitney tests) with a significance level of 0.05 were performed using the GraphPad Prism program version 3.00 for Windows (GraphPad Software, San Diego, CA, USA). All P-values are given as two-tailed. To examine a possible influence of sex and age of respondents on results a multinominal logistic regression analysis was performed using the SPSS statistical package version 10.1 for Windows (SPSS Inc., Chicago, IL, USA). For each dependent variable the analysis was performed with group, sex and age as independent variables. In performing the regression analysis age was categorized into three age-groups, age 0–35 years, 36–50 years, and older than 50 (Table 1).


All patients answered the initial questionnaire. Amongst cancer trial participants 20 patients (49%) answered the second questionnaire, and 15 patients (37%) the third questionnaire. A few questions were not answered by all respondents.

Cancer trial participating respondents were significantly older than trial nonparticipants. Amongst patients with cancer of the breast a borderline significant lower age in nonparticipants was seen (P=0.06). There was a preponderance of men in the groups with gastrointestinal cancer, whilst the group with cancer of the breast only involved women (Table 1).

Patients’ apprehension of physicians’ motives

About one-third of respondents amongst cancer trial participants and a quarter of nonparticipants either fully or partly agreed with the statement, that medical research is primarily performed to promote doctors’ careers (participants 34%; nonparticipants 23%), whilst little less than a third of participants and about half of the nonparticipating respondents either partly or fully disagreed (participants 31%; nonparticipants 47%). The remaining respondents neither agreed nor disagreed. There were no significant differences between groups and no significant influence of age or sex.

The respondents graded the importance of five possible motives of doctors to plan and conduct medical research. No significant differences between groups were seen, and no significant influence on results of age or sex were shown (Table 2).

Table 2.   Respondents’ comprehension of doctors’ motives to plan and conduct medical research Thumbnail image of

Former trial participation – impact on attitudes towards clinical trials

Nine respondents amongst trial participants and five nonparticipants were former trial participants. Eight cancer trial participating respondents and four nonparticipants had changed their attitudes negatively by this experience, significantly different from the results amongst IBD-participants, where four-fifths reported a positive change in attitude (cancer trial participants versus IBD-participants: P=0.01; cancer trial nonparticipants versus IBD-participants: P=0.02) [21].

Attitudes towards clinical trials at study entry

There was no significant difference in the general attitude towards clinical trials amongst trial participants (cancer trial participants, IBD-participants), but not unexpectedly the participants were significantly more positive than cancer trial nonparticipants. Still, a majority of cancer trial nonparticipating respondents were relatively positive. No significant influence on results of age or sex was seen (Table 3).

Table 3.   Attitudes towards medical research Thumbnail image of

Almost all respondents amongst both cancer trial participants and nonparticipants considered it necessary to test both new drugs and diagnostic methods scientifically before implementation in clinical practice. This was also seen amongst IBD-participants, where respondents only were asked about the necessity of testing new drugs. No significant influence on results of age or sex was seen (Table 3). In free text a majority stated the primary reason to be a concern with unknown, and unwanted effects of a new intervention.

With regard to attitudes towards personal participation there was no difference between cancer trial participants and IBD-participants, both being more positive than noncancer trial participating patients. No significant influence on results of age or sex was seen (Table 3). In the free text responses several cancer trial participating patients stated hopes for a personal benefit, and a wish of helping future patients to be the primary reasons for their positive attitude. In the group of cancer trial nonparticipating patients several stated that the reasons for their hesitation was anxiety towards ‘the unknown’, and a wish to maintain a personal influence on decisions. As an example one cancer trial nonparticipant wrote:

I’m scared of something, of which consequences are unknown. Naturally, I wish to have the best treatment – and let others be guinea-pigs, but if the trial treatment could save my life, and there were no other treatments available, I would not hesitate.

When comparing the patient groups’ attitudes towards family or friends participating in trials the same order of positiveness between groups as seen in relation to personal participation was found. A tendency towards cancer trial nonparticipating respondents being more positive towards participation of family or friends than towards personal participation (P=0.1) was seen. No significant influence on results of age or sex was seen (Table 3).

Research ethics committees’ impact on decisions to participate

The majority stated that the presence of public, and now legally based, research ethics committees in Denmark had had an impact on their decision to participate (cancer trial participants 63%, IBD-participants 60%). Approximately a quarter stated that the existence of the committees had had no impact on their decision (cancer trial participants 24%, IBD-participants 28%), or they had no opinion on this issue (cancer trial participants 10%, IBD-participants 12%). No significant influence on results of age or sex was seen.

Declining participation in the trial – a moral problem?

Only a minority felt, that declining participation constituted a moral problem (cancer trial participants 24%, IBD-participants 15%, cancer trial nonparticipants 12%), whilst a majority did not (cancer trial participants 56%, IBD-participants 65%, cancer trial nonparticipants 80%), or had no opinion on the issue (cancer trial participants 15%, IBD-participants 17%, cancer trial nonparticipants 8%). No difference between the groups was detected, and no significant influence on results of age or sex was shown.

Randomization – impact on attitudes towards clinical trials

Approximately half the cancer trial participants and IBD-participants had a positive attitude to randomization. A third was hesitating, whilst only a minority were negative. There were no differences between cancer trial participants and IBD-participants, whereas comparisons against attitudes in cancer trial nonparticipants revealed significant differences in both cases, with cancer trial nonparticipating respondents being more negative. The multinominal logistic regression analysis showed no sex related influence on results, whereas age significantly influenced results with younger respondents being most positive. Odds for a positive opinion versus a hesitating/negative opinion was 2.65 (95% CI: 0.95–7.42) in respondents younger than 36 years (P=0.02) (Table 3). In free text several respondents amongst cancer trial participants stated as the primary reason for their attitude, that drawing lots was the best method to ensure a random distribution between groups, although most would have liked to choose their treatment themselves. As a typical example one respondent wrote:

It is very unpleasant, that you in such a serious matter seems subjected to a matter of chance, which can have decisive consequences. I would have liked to choose myself.

Some respondents were on the other hand pleased with randomization taking the responsibility of choosing away from the patient. In several free text statements from cancer trial nonparticipating respondents a strong resentment towards drawing lots in the case of trials involving life-threatening diseases was stated, and the majority of the free text statements mentioned a wish to choose their own treatment, or a wish for the doctor to choose for them. Two respondents wrote:

You shouldn’t use the drawing of lots to decide in life-threatening diseases. If the doctor had said to me that treatment A was the best, I would have chosen that. It is totally incomprehensible for me that you impose a principle of drawing lots when you – as a patient – feel you are dying.

Emotionally it is very difficult to handle in serious diseases. In a way you need someone to tell you what is best for you.

Reasons to participate

Both cancer trial participants and IBD-participants were asked to grade the importance of a number of possible reasons for their choice to participate in the actual trial. There were no significant differences in gradings between groups.

Both wish to get access to the new drug or new diagnostic tool and a wish to be more closely monitored were rated very important or important by a majority. In contrast, the majority in both groups rated a wish to obtain a good relation to the department performing the study either of minor or no importance for their choice. A majority of respondents felt a wish to help future patients to be either very important or important, whereas the importance of positive experiences from former participation in trials in the majority was found of only minor or no importance. In general no significant influence on results of age or sex was seen. However, concerning a wish to get access to the new drug or new diagnostic tool a small, but significant influence of age was seen with this wish being less important amongst respondents 36–50 years of age (P=0.02; odds for a very important/important response versus a minor importance/unimportant response was 1.07 (95% CI: 0.17–6.81) (Table 4).

Table 4.   Different reasons for consent to participation in a clinical trial Thumbnail image of

We further asked respondents to mention other factors influencing their decision to participate. Several stated a health benefit as the primary reason, several expressed a wish to help others, and some had felt their freedom of choice limited considering their ‘desperate’ situation.

Reasons not to participate

A majority of cancer trial nonparticipants stated a fear of adverse effects and/or emphasized a need for avoiding ‘the unknown’. Several stated an unease by letting chance decide and/or mentioned a lack of personal resources. Some mentioned a feeling of lacking information, some would save ‘the new’ treatment for a possible recurrence of their disease, and some explained their choice being purely emotional.

Evaluation of information given prior to the choice whether to participate in a trial

There were overall significant differences with IBD-participants being the most satisfied, followed by respondents amongst cancer trial participants, and with cancer trial nonparticipants being the least satisfied. There were, however, only significant differences in two-group comparisons between IBD-participants and cancer trial nonparticipants. No significant influence on results of age or sex was seen (Table 5).

Table 5.   Patient evaluation of oral and written information given prior to potential consent to participate in a trial Thumbnail image of

Experiences from trial participation

More than half of the cancer trial participating patients failed to answer the second and third questionnaires. Therefore, only a brief summary of results are given, as the potential for bias is substantial. However, the majority of the remaining respondents stated full satisfaction in the ‘during study’ questionnaire, whilst a large minority stated some areas of doubt remaining in the ‘after study’ questionnaire. The time needed in the trial was in agreement with the expectations for most and the general impression of participating in the trial was either very positive or positive in a majority of respondents.

The most frequent theme in positive free text was a feeling of receiving more extensive monitoring than expected outside of a trial setting. Other recurrent themes were the feeling of helping future patients, the feeling of being treated as a human being, and being content with not having renounced the possibility of receiving something ‘better’. Statements from two patients largely cover the typical content:

It is positive that I possibly have helped others in the same situation as mine. I haven’t felt much being in a trial except for the fact that I was examined more thoroughly at treatment sessions than I think I would have been outside the trial.

I was a little afraid of renouncing a ‘better’ treatment. It was therefore a relief to let fate or the computer make the choice for me. It is positive for me that I haven’t said ‘no’ to any treatment.

A common theme in statements concerning negative experiences in the trial was a frustrating experience of seeing too many physicians at check-up appointments.

Participation in the future?

One cancer trial participating respondent did not want to participate again, whereas the remaining respondents were willing to participate in future trials. Whether patients would advise family members or close friends to participate followed the same pattern. The same tendency was seen amongst IBD-participants. Several respondents gave free text answers giving reasons for their willingness to participate and to advise close kindred to do the same. The reasons were, as amongst IBD-participants, a wish to help future patients, the respondents themselves, and/or science in general.

General comments about trial participation

Several cancer trial participating patients provided more comments. To have seen a row of different physicians at sequential check-up appointments was the most frequent theme. A typical example:

It was frustrating always seeing a new doctor at check-up appointments. You should see the same or at least only two doctors, as they wouldn’t need to start all over each time. It would feel much more reassuring, and it is not satisfactory, that you all the time meet new, young, and ‘green’ doctors, who are just passing the department for educational reasons.


The response rate of the initial questionnaire was very high, whereas the second and third questionnaires in the group of cancer trial participants were only answered by less than half of the respondents. The results concerning the latter two must therefore be interpreted with caution, as a considerable risk of bias is present.

In a previous questionnaire study in the healthy Danish background population exploring the same issues [24] we found women younger than 46 years having a less positive general attitude towards trials [95% CI’s of scores (same scale as in Table 3): age ≤ 46 years, 2.67–2.96; age > 46 years, 2.45–2.65; P=0.008). The younger women rated both a wish to get access to the new drug (95% CI’s of scores: age ≤ 46 years, 1.88–2.67; age > 46 years, 1.72–1.91; P=0.03) and a wish to obtain a good relation to the department performing the study (95% CI’s of scores: age ≤ 46 years, 2.81–3.24; age > 46 years, 2.34–2.64; P=0.0004) less important than the remaining respondents as motives to trial participation. Because of the inclusion bias introduced by only including premenopausal breast cancer patients, the present study in particular solicited the opinions of younger women, who thus could be expected to have a relatively critical approach towards clinical research. However, the multinominal regression analysis performed showed only very few items being influenced by age and none by sex, when one controls for which group the respondent belongs to. These results must be interpreted cautiously as our samples are relatively small. They, however, seem to indicate that other factors than age and sex are the main determinants for a person’s attitude towards research.

In grading importance of physician’s motives to conduct medical research altruistic motives were highly rated by a large proportion in all groups, with cancer trial nonparticipants rating these motives the highest. This could seem inconsistent with the large proportion of respondents that regarded the motive of promoting doctors’ own careers as either important or very important. This apparently self-contradictory high importance gradings of both altruism and ‘self-interest’ as motives for doctors merely reflect the area’s complexity. In the light of large-scale public health system retrenchment programmes in Denmark in recent decades it is interesting, that the wish to minimize total expenses via research was regarded to have only minor or no importance by approximately two-thirds in all groups. The finding that the group of respondents with ages between 35 and 50 years rated the motive of getting access to a new drug or new diagnostic tool less important than both younger and older respondents is difficult to interpret, but the difference is small.

It is not surprising, that the chronically diseased IBD-participants with a regular contact to research active university departments had a higher proportion of former trial participants than the cancer patients. It is however, surprising, that the majority of previous trial participants amongst both groups of cancer patients had had their attitudes changed in a negative direction, whilst the majority of IBD-participants had changed their attitudes in a positive direction. The difference is statistically significant but must be interpreted cautiously as the number of respondents amongst cancer trial participants and cancer trial nonparticipants is small. That IBD-participants, whose former trial participation in a majority had been related to their bowel disease, had had their attitudes changed positively is maybe explained by our department having a long tradition for a special concern for ethical problems in research and patient care. This probably have had an impact on the experiences of the patients attending the department.

It is an important finding, that large majorities of respondents a priori had a positive attitude towards clinical trials in general. This is in agreement with previous findings [25–28]. It was expected that respondents in the group of cancer trial nonparticipants would have a less positive a priori attitude. It is, however, surprising that even these respondents stated such high degrees of positive attitudes towards clinical trials.

As expected answers revealed a significantly more positive attitude towards both personal participation and participation of others amongst both groups of trial participants, as compared with cancer trial nonparticipating respondents. At the same time considering trial participation as a moral obligation was rejected by the majority amongst both cancer trial participants, cancer trial nonparticipants, IBD-participants, outpatients, and the public [24]. As argued by Rossel this way of thinking implies ethical problems, as a person willing to accept a certain good (i.e. the new treatment) procured by a social activity (clinical trial), but who is not willing to contribute to the social activity being a prerequisite for the availability of the good can be characterized as a ‘free rider’ [29]. The results are probably another expression of an apparent conflict between rights and duties in modern societies with an increasing tendency to focus on individual benefits.

A majority amongst both cancer trial participants and IBD-participants stated that the presence of public research ethics committees had an impact on their decisions. That a clinical study has been reviewed and approved by an ethics committee apparently gives a sense of security and probably increases willingness to participate, which emphasizes the importance of such information. It must, however, be noted that at least some respondents probably were not a priori aware of the existence of this official control system [30].

In contrast to trial participants, a majority of cancer trial nonparticipants had a negative or a hesitating opinion of the principle of randomization, and several expressed a resentment towards this principle in free text responses. The importance of these aspects has also been found by others [4, 20, 31–34]. It is significant, that a major reason for a critical attitude towards ‘the drawing of lots’ was the fact, that the disease was a potential life-threatening condition. However, as found in the study by Fallowfield et al. further information can increase willingness to participate in trials, although a certain proportion of patients probably do not wish to participate, whatever information is provided [20].

Trial participants amongst both cancer patients and patients with inflammatory bowel disease gave almost identical answers concerning a priori attitudes towards clinical trials, the necessity of testing, attitudes towards both participation of self and others, importance of research ethics committees, and finally the issue of a possible moral obligation to participate. These similarities in answering indicates that the attitudes – at least concerning these issues – are to a large extent independent of age, gender, disease and potential burdens that are part of the trials.

In contrast to the overall positive attitudes found amongst actual trial participants, the attitudes amongst cancer trial nonparticipating respondents towards the same issues in general were considerably more critical, although a surprisingly high proportion stated positive a priori attitudes towards trials, and almost all recognized the necessity of testing. It is furthermore interesting that only some cancer trial nonparticipants were frankly negative towards potential self-participation and/or participation of others.

In grading the importance of different motives for participation both self-centred and altruistic motives were highly rated by a large proportion in both groups of trial participants. These findings are in accordance with findings in other studies [13, 24, 25, 27, 28, 32, 35–37]. It is significant, that a majority in both groups of trial participants rated the possibility of closer monitoring either important or very important. In addition the majority of IBD-participants expressed a high grade of satisfaction by participating in the trials, primarily because of the feeling of being ‘specially handpicked patients’ [21]. The importance of this phenomenon was also shown in a questionnaire study by Slevin and coworkers, who examined the aspects of research trials which were considered appealing or unappealing by patients. The most appealing aspects were expectations of being more closely monitored, receiving more extensive information and being treated by a doctor with a special interest in their disease [38].

High satisfaction scores were recorded by trial participants concerning the initial oral and written information, although a rather large minority of cancer trial participants were not fully satisfied. In contrast, the information was only rated fully satisfactory by little more than half of cancer trial nonparticipating respondents. An important prerequisite for patients giving meaningful consent to participate is sufficient information. This is probably particularly important in patients with serious diseases, where consequences of choices can seem decisive. In fact, some cancer trial nonparticipants stated that the primary reason for not giving consent was the lack of sufficient information.

Although the data concerning experiences during the trials must be interpreted with caution – due to the failure of answering from more than half the respondents – it is reassuring, that a majority of respondents were relatively content with the total information given, and that the general impression of participating in the trial was either very positive or positive in a majority.

In the group of IBD-participants the most important factor in achieving high satisfaction scores and willingness of almost all respondents to participate in a future trial and to advise family or friends to do the same was the fact that these patients only saw one physician at control visits [21]. In contrast, the most frequent theme related to negative experiences amongst cancer trial participants was a strong discontent about seeing too many physicians at check-up appointments who were not always perceived competent. This experience of a shifting panorama of passing doctors in the out-patient clinic on days of check-up appointments, leaves an impression amongst patients of being ‘just a number in a row’. Although more than half of the respondents did not answer the second and third questionnaires, the statements from the remaining respondents were almost uniform.

Because a large majority of the remaining respondents in fact stated their concerns about this, it is a message of paramount importance not only in research but also with a perspective for all clinical work, that patients have a right to be taken seriously as human fellow-beings. The importance of a confident partnership between patient and physician/researcher should not be underestimated. The possibility of recruiting larger proportions of eligible patients in the future than today is maybe realistic, if researchers remember and acknowledge this, as high levels of positive attitudes towards many different aspects of clinical research are found in the public, amongst out-patients [24], amongst participants in noncancer trials [21], amongst participants in cancer trials, and even amongst patients renouncing participation in cancer trials.


The authors wish to acknowledge Poul Schlichting, DMSc, Medical Gastroenterological Department C, Copenhagen University Hospital at Herlev, Pia Munkholm, DMSc, Medical Gastroenterological Department, Copenhagen University Hospital at Hvidovre, and Bogi Davidsen, DMSc, Department of Medicine, Slagelse County Hospital for valuable comments to the manuscript and for general guidance. We are grateful to both physicians and nurses in the Oncological Departments at Copenhagen University Hospital at Herlev, where especially the research nurses Helle Rudbæk (Herlev) and Karen Friis (Odense) are thanked for help by the inclusion of patients and collection of questionnaires.

The implementation of the studies was made possible by grants from (alphabetically): Aage Louis-Hansen’s Foundation; Anna og Jakob Jakobsen’s Foundation; Astra Denmark; Beckett Foundation; Danish Medical Research Council; Danish Research Council for the Humanities; Hede Nielsen’s Foundation; Jacob Madsen’s and Olga Madsen’s Foundation; Danish Hospital Foundation for Medical Research. Region of Copenhagen, the Faroe Islands and Greenland; Ove Villiam Buhl Olesen og Edith Buhl Olesen’s Foundation; Roche Inc.; Sigrid Rigmor Moran’s Foundation, and Vibeke Binder and Povl Riis’ Foundation.