Clinical Ethics >
Volume 3, Number 4 >
Pp. 180-184
Empirical Ethics |
* Public Health Sciences, University of Edinburgh, Edinburgh
Princess Anne Hospital, Southampton and University of Southampton, Southampton
The Ethox Centre, University of Oxford, Oxford, UK
E-mail: nina.hallowell{at}ed.ac.uk
The aim of this study is to assess patients' recall of their previous research participation. Recall was established during interviews and compared with entries from clinical notes. Participants were 49 patients who had previously participated in different types of research. Of the 49 patients, 45 (92%) interviewees recalled 69 of 109 (63%) study participations. Level of recall varied according to the type of research, some participants clearly recalled the details of research aims, giving consent and research procedures. Others recalled procedures (e.g. DNA testing) but were unclear about their purpose. There was no significant effect of time on recall. Some types of research participation (e.g. DNA testing) may be recalled as clinical care. We argue that such misunderstandings may have the potential to undermine participants' ongoing consent, particularly in ongoing/longitudinal studies. Valid consent may be best achieved by re-assessing the scope of consent and relating it to the nature of the interventions themselves rather than the reasons for undertaking them.
 | Introduction |
|---|
 Top Introduction MethodsResultsDiscussionReferences |
|---|
The validity of consent is generally seen to be dependent on participants' understanding at the point during which their consent is given (i.e. the study invitation is accepted). However, it must be noted that participants' understanding may change over time for: (1) participant-related reasons (e.g. competence diminishes and/or participants forget that they are involved in research);10 and/or (2) research-related reasons (e.g. the study design is altered). These changes in understanding can be seen as compromising the initial consent and may require participants to be re-consented.
To support the rapidly evolving field of cancer genetics, many cancer genetics centres in the UK undertake their own or collaborate with national and international research projects. Cancer genetics services have evolved in ways that potentially make it difficult for patients to distinguish clinical management from research, because the ‘procedures’ offered to patients are sometimes funded by the research rather than the clinical/service budgets. Patients may be approached to take part in cancer genetics research (clinical and screening research, DNA research, epidemiological research or social science) by health-care professionals (clinical geneticist/genetic nurse counsellor) during a clinical consultation, sent recruitment information by post, visited at home by a member of the research team or invited to attend a stand-alone recruitment appointment at the genetics clinic.
The ROCC project, which generated the data presented in this paper, sought to establish lay and expert perceptions of the activities that take place within cancer genetics clinics in the UK. One of the project's aims was to determine the extent to which patients differentiate procedures undertaken for research purposes from those undertaken as clinical care. One of the ways that patients' understanding of clinic activities can be investigated is to explore the ways in which they recall their previous research participation, and these data are presented below.
| Methods |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
Participants were purposively sampled to include those who had been invited to take part in a range of different types of research studies.
Research studies involving physical interventions that have the potential for therapeutic benefits
Non-therapeutic research that is unlikely to have individual benefit
Ethical (MREC) approval was obtained. A total of 119 patients attending the Wessex Genetics Service were sent an invitation pack containing an information sheet and an expression of interest form to be returned to the research team. Fifty-two (44%) patients agreed to be interviewed.
In-depth interviews were carried out by SC and GC between March 2006 and 2007. The interviews were carried out at the participant's home, work place or the genetics clinic, lasted between 40 and 90 minutes, and were tape-recorded with consent.
Interviews began with a number of demographic questions (Table 1). Interviewees were then asked to provide a narrative account of their involvement with the genetics clinic and any research projects they had previously participated in. Discussion about a range of clinical and research activities (such as screening, blood samples, etc.) was prompted and interviewees' responses were explored (Box 1) to establish whether they were perceived as part of a research protocol or clinical service.
|
Participants' clinical genetics records were thoroughly examined to collect information recorded about research participation. This included data about invitations to research studies, the response (i.e. declined/consented) and the year consent/refusal was received. Recalled and recorded research participation was compared.
| Results |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
Two participants had received an inconclusive result in service DNA testing for a rare condition and had been subsequently approached for their consent for storage of their blood samples and research testing, if suitable research projects should arise. Neither of these participants had been invited to participate in other research projects. We made the decision to exclude these participants from the remaining analyses because although they had given consent to use their blood in future studies, technically speaking they had not been involved in any research at the time of the interviews.
Clinical notes indicate that 121 research invitations were issued to the 49 remaining interviewees. The interviewees received between 1 and 9 invitations (mode = 2) and 109 of 121 (89%) invitations were accepted. Four participants did not recall taking part in any research, their notes indicated that all had joined a research register, plus one was also involved in a screening study and another in a DNA testing study. The remaining 45 interviewees recalled accepting 69 of 109 (63%) research invitations and their study participation to some degree (completely, partially, procedurally or they recalled research as a clinical service) (Table 2). Analysis of the clinical notes revealed that eight of the 49 interviewees failed to respond or had declined at least one research invitation in the past, although they had taken part in other research studies. Only three of these eight interviewees recalled declining any research invitations.
|
|
Thirteen participants recalled participating in 15 studies (10 psychosocial, two chemoprevention, two epidemiological projects and one research database) that were not documented in their clinical notes. However, we were unable to confirm that these invitations had been issued.
With one exception, all recorded study consents/declines occurred between 1995 and 2006. The largest number (18) of invitations issued in a year was in both 2002 and 2005, and the lowest level of recall (50%) were for consents recorded in 2002 (9 of 18 invitations) and 2006 (4 of 8 invitations). A Kruskal-Wallis one-way analysis of variance was performed to see if time elapsed since recorded consent affected the total (
2 = 10.4, P = 0.495) or partial (2 = 15.8, P = 0.148) recall, there was no significant effect.
| Discussion |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
Previous research which focuses upon recall of consent procedures3 for research has recorded a lower rate of recall than that observed here for certain types of research. This may be owing to the fact that this study recorded recall of other aspects of research in addition to consent procedures. Moreover, there was evidence that in some cases participants misinterpreted the status of some of the procedures they had undergone, particularly screening and DNA testing research. For example, although every interviewee who had participated in screening research recalled undergoing screening, nearly half mistakenly thought that it was a clinical procedure or were uncertain of its aims. It is possible that participants may take part in screening studies to gain reassurance,11 however, their inability to recall the research objectives suggests that research participation may provide false reassurance in some cases. Similarly, while participants could recall having a DNA test, only one-third of these were clearly identified as research tests. These observations are unsurprising, for there is a longstanding ambiguity about the status of DNA tests in general,8 and in cancer genetics in particular.9 Indeed, it can be hypothesized that the lower level of recall of participation in DNA-based research may occur because the participants had originally attended genetic clinics with the aim of accessing genetic tests and they are less concerned about motivations for testing.12 Uncertainty about the nature of different activities may also explain the lower level of recall of participation in epidemiological, psychosocial and database research, because as with DNA testing and screening, the information (e.g. family history and risk evaluation) collected during these types of research studies may be similar to that collected during clinical consultations.
There are a number of alternative explanations for interviewees' failure to recall their research participation. First, it is possible that research participation may be perceived as a less important aspect of participants' cancer experience. In other words, research participation may just be a more forgettable event when compared with the diagnosis and treatment of cancer or confirming and managing genetic risks. Secondly, it must be noted that some of our interviewees had received a number of research invitations, in one case nine, so it is possible that those who had received more than two invitations were more liable to fail to recall them or mistakenly recall procedures as clinical care. However, research invitations were issued over a 12-year period and as approximately 75% of the interviewees had only been issued with two invitations during this period according to their notes, this seems an unlikely explanation. Nevertheless, we want to draw attention to the fact that approaching patients to be involved in multiple studies has ethical implications, not least because they may confuse the procedures, risks and benefits associated with the different studies. Arguably, these observations suggest that further systematic investigation of the external and personal factors which effect recall is required.
One problem with this study is that checking clinical notes to validate recall is not foolproof. Indeed, there was evidence that the interviewees recalled research participations that were not documented in their clinical notes; this may have been because invitations were issued by other departments or because the clinical genetics staff may have failed to record them. Whatever the reason, it is clear that clinical notes may offer an inaccurate comparison. In response, it can be argued that, even if the notes are inaccurate they do provide a record of some, if not the majority of, study participations against which we can compare recall.
In addition, there may be a sampling bias in this study. Participants were recruited from one clinic and thus the data may reflect this clinic's approach to recruitment and taking consent. However, it should be noted that this study focused upon research participation over a 12-year period, which involved over 30 health-care practitioners in a recruitment capacity. Moreover, those who had been responsible for recruitment were following the recommended national and international study/recruitment guidelines where appropriate.
Finally, as noted above, the level of recall recorded in this study may, in part, be owing to the method of recording recall that we developed. We became aware early on that we needed to introduce a number of follow-up questions/prompting to determine whether interviewees recalled various research procedures (e.g. ovarian screening) as relying on spontaneous mentions of research participation produced a low level of recall, primarily because interviewees frequently confused research procedures with clinical care.
By comparing participants' reports of their previous research participation with that recorded in their clinical notes, this study suggests that some types of research studies (e.g. screening and chemoprevention trials) in cancer genetics are more frequently recalled than others (e.g. psychosocial research). The findings indicate that the status of certain research activities, particularly DNA testing, may be misinterpreted or assumed to be clinical care and this may explain the low levels of recall for some types of research. In other words, some activities are just not understood to be research and hence, are not recalled as such.
Such observations raise questions about the validity of participants' ongoing consent in these cases. It can be argued that if patients do not recall/realize that certain procedures are undertaken under the auspices of a research protocol, then their consent may be invalid. Such observations are particularly important in studies where data are collected over a number of years (e.g. epidemiological or DNA mutation searching studies), as although the consent may have been valid at the outset, over time individuals may fail to recall their research involvement and come to view research procedures as part of their clinical care. One solution for this problem would be to re-consent the participants after a period of time has elapsed. However, it would be difficult to determine the appropriate time period in such cases. Moreover, re-consenting participants in such cases raises a range of pragmatic or bureaucratic issues, such as who should contact them and what should happen if they could not be contacted?
An alternative solution would be to re-define the scope of consent. Arguably, our interviewees' confusion about the status of various procedures only undermines their ongoing consent if we view consent as a contingent upon maintaining a distinction between research and clinical practice. If this distinction is ignored, and consent is related to the procedures (e.g. DNA testing or screening) and their associated risks and benefits rather than their function (increasing the knowledge base versus providing therapy) or the intellectual or epistemological motivations for undertaking them,13 then recall failures/confusions such as those witnessed in this study could be seen as much less problematic. However, this solution raises problems of its own, for example, how easy it is to explain the risks and benefits associated with medical procedures such as genetic testing, what risks and benefits would need to be explained and who would determine which risks and benefits should be explained? Finally, how could we ensure better understanding of procedures rather than purposes?
Box 1 Examples of general prompting of activities
|
| Acknowledgements |
|---|
| Footnotes |
|---|
Gillian Crawford is a genetic nurse counsellor who specializes in cancer genetics in the Wessex Genetics Service. She also works part-time as a Cancer Research UK-funded researcher and contributed to data collection on the ROCC project.
Michael Parker is Professor of Bioethics at the University of Oxford. He is also the Director of the Ethox Centre.
Anneke Lucassen is Professor of Clinical Genetics at the University of Southampton. She also works as a Consultant in Clinical Genetics in the Wessex Genetic Service, where she specializes in cancer genetics.
Nina Hallowell is Reader in Social Science and Public Health at the University of Edinburgh. She teaches a course on research ethics.
| References |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
