First, do no harm. Second, do no direct, qualitative research with psychotherapy clients…
In Part 1 of this article (Volume 10, No. 2, Winter 2018), Peter Jenkins set out the current context in which psychotherapy research is taking place and argued that university ethics committees are becoming increasingly focused on rules and risks. He continues the theme here, exploring particular risks that concern ethics committees, excluded research topics, and the impact of university ethics committees on psychotherapy research.
Risk of causing respondent distress?
University ethics committees seem to be highly sensitive to research which may trigger distress in participants. This might seem to be a reasonable position to adopt. Yet psychotherapy research, particularly qualitative research, often seeks to engage with topics which are, by their very nature, distressing, or even traumatic, such as loss, conflict, separation, abuse and sexual assault. This is not surprising, as psychotherapists work with client distress as a core part of their professional activities. In one sense, this might equip them particularly well for undertaking research into sensitive topics, hopefully in an appropriate manner as researchers. The risk of generating a degree of client distress is perhaps seen by ethics committees as a detriment simply to be avoided, rather than as one potential outcome, which can be justified on the basis of the support to be made available to participants and the beneficial research outcomes which might be ultimately forthcoming.
Takesawa et al (2004) discuss this issue in the context of research into the highly sensitive topic of palliative care, considering the likely balance of risk and benefit in generating a degree of distress when carrying out palliative care research. This study found that “only a minority of respondents reported any distress, and those who did generally reported distress that was mild or moderate. This suggests that these studies may be much less distressing to study subjects than researchers and IRBs [Institutional Review Boards] might expect” (Takesawa et al, 2004, p.596). It concludes that there is a utilitarian case for carrying out this research, despite the possibility of causing some, albeit a limited, degree of respondent distress: “it seems likely that these small risks are outweighed by the benefits that family members may experience when they have the opportunity to discuss these issues” (Takesawa et al, 2004, p.597). University ethics committees would do well to follow the example set here, in adopting a reality-informed utilitarian approach towards the risk of generating respondent distress, rather than one which avoids all risk, by vetoing research into sensitive, but socially necessary, topics, such as palliative care, transgenderism, and eating disorders, to name several such topics.
Risk of the university being sued?
The second apparent driver for the increased regulation of psychotherapy research by university research ethics committees has been “the increase in litigation seeking legal redress and financial compensation” (Danchev and Ross 2014, p.107). However, this is, arguably, based on a misapprehension of the nature of the risks posed to the institution by carrying out research, other than that involving primarily medical or physiological interventions, such as the caffeine test described in Part 1. Even here, the substantial fine levied on the university was for breach of health and safety law, rather than for professional negligence (Weaver, 2017). The guidance from the Department of Health requires that ethics committees and parent institutions “ensure that they are able to compensate anyone harmed as a result of negligence” i.e. where physical damage has been caused by the research (DoH, 2005: 3). This might be taken as strong support for the university’s need to take robust action to defend against the risk of litigation for negligent research. However, this argument is unconvincing, given that universities rarely experience litigation of this kind, unlike the NHS. It seems much more likely that medical and nursing academic staff, who work on university ethics committees, have imported a set of assumptions about the high risk of medical negligence claims, based on their experience within the NHS. The appreciably high risk of litigation by patients, for alleged medical negligence within the NHS, simply do not apply to the very different world of university-based, non-medical, psychotherapeutic research (Jenkins, 2016).
The other relevant aspect of the supposed risk of universities being sued for damages arises from data protection law. The key reference here was the Data Protection Act (DPA) 1998, now replaced with regard to research data responsibilities by the Data Protection Act 2018 and the General Data Protection Regulation. The DPA defined ‘damage and distress’ which was eligible for compensation as a result of data breaches in very limited and specific terms:
“ – substantial damage would be financial loss or physical harm;
– substantial distress would be a level of upset, or emotional or mental pain, that goes beyond annoyance or irritation, strong dislike, or a feeling that the processing is morally abhorrent” (ICO, 2009, p.140).
The definition of distress employed here by data protection law sets a very high threshold of eligibility, similar to the standard required for compensation via professional negligence law, namely that of a ‘psychiatric injury’. This would need to be capable of diagnosis via psychiatric textbooks, such as DSM-5. It seems hard to envisage the kind of data breach in a psychotherapy research study which could realistically cause distress of this severity. Again, university ethics committees could take a reality-based position in considering the risk of distress caused by possible data breach, rather than taking refuge in a standard risk-avoidant stance.
Excluded psychotherapy research studies
It is hard to scope the full extent of the rejection of proposed psychotherapy research topics. It may also be the case that students avoid contentious topics, on the advice of their supervisor, to ensure acceptance of their preferred topic and avoid delays to starting their research. Some excluded research proposals include the following:
- the value of counselling for the fathers of children who died at or near birth;
- clients on the autism spectrum;
- clients’ experience of traumatic disability and the role of counselling in rehabilitation;
- young people under 18, whose parents were drug users; and
- young people’s experiences of having a parent in prison.
It is notoriously hard to research an absence, as presumably university ethics committees do not necessarily keep records of research projects which are rejected, any more than they research their own outcomes, or the effectiveness of the criteria which they apply in their decision-making (Goldacre, 2012). Of this small sample of rejected research projects, the first three derive from Gladwell’s paper (2015). The fourth is interesting, because it relates to research participants under the age of 18. The authors simply note that the rejection “fits with standard expectations”, which, from an ethical point of view, might sit as a normative judgment, but avoids unpicking the more complex ethical issues involved in this rejection (Danchev and Ross, 2014, p.107). The fifth example, from the same authors, again notes that this research project was perceived to be “an ethical minefield”, relying on a prominent metaphor of death and destruction to power the argument. However, it could be argued that ethics committees are precisely the bodies best equipped to engage with this kind of argument, rather than simply retreat from the complexity it might present (Danchev and Ross, 2014, p.85).
Risks involved in researching young people?
Research committees seem to be particularly reluctant to authorise research with young people under 16 years old, presumably out of a fear of falling foul of the law, or of incurring parental disapproval or complaint. However, this attitude has led to some strange inconsistencies in relation to perceptions of the need to obtain parental consent for young people in research. In the field of school-based counselling, testing the possible effects of counselling on pupils’ ability to study and learn has become something of a Holy Grail issue. However, there are puzzling differences between protocols for research participation of pupils in Scotland and those in England. One study included 21 pupils, aged from 12-17 years, with pupils in Scotland presumed to be able to consent to participation in research if also competent to receive counselling without parental consent, under the Age of Legal Capacity (Scotland) Act 1991. For pupils in England in the same study, however, “parental consent was obtained in addition to consent from the participants themselves” (Rupani et al, 2012, p.503). Presumably, this is based on legal advice on the need for parental consent for young people to participate in research in England, as compared with those in Scotland.
However, this view on young people requiring parental consent for participation in research in England seems to be mistaken. After all, if a young person in England under 16 can consent to a termination without requiring parental knowledge or consent, then presumably they can complete a questionnaire, or take part in a research interview. The reluctance to authorise research with young people may therefore reflect a more risk-averse approach within the university ethics committee than is justifiable according to the law. Rather than taking the view that research with young people is an ‘ethical minefield’, and therefore to be avoided, it is actually the case that such research can be authorised, if there is a willingness to apply the existing law. For this sharp distinction to be drawn between young people in Scotland, and those in England, reflects not so much a radical difference in the respective laws about adolescent autonomy, albeit, in statute as opposed to case law, but a shying away from acknowledging the legal rights of young people in England. One of the clearest statements about the legal position regarding the competence of children and young people to consent to participation in research, without requiring additional parental consent, can be found in the Code of Practice for the Mental Health Act 1983, applying to England (DoH, 2015). This provides a clear statement of the Gillick principle with regard to children and young people under the age of 16, in the context of mental health law, but the essential principles of Gillick are not restricted to this specific context.
“…in the case of Gillick, the court held that children who have sufficient understanding and intelligence to enable them to understand fully what is involved in a proposed intervention will also have the competence to consent to that intervention. A child may be Gillick competent to consent to admission to hospital, medical treatment, research, or any other activity that requires their consent” (DoH, 2015, p.176, emphasis added).
This is just one example, but there does appear to be an exaggerated fear of incurring liability, litigation and compensation claims in research involving young people. However, according to Masson (2004, p.45), a leading authority on the law in this area, “researchers are not at risk of legal proceedings merely through involving children or young people in research”.
Charting the impact of ethics committees
In order to try to assess the impact of university ethics committees on the development of research into psychotherapy, I carried out a small survey on research articles published over the period between 2001 and 2010 in the BACP journal, Counselling and Psychotherapy Research. From my own anecdotal experience of working in universities during this period, psychotherapy research at postgraduate level (i.e. MA or MSc) did not routinely require approval from university ethics committees during the early part of this period; this was reserved for projects which were deemed to be exceptionally problematic in ethical terms, particularly if relating to research on sexual topics.
Diagram 3: Sample from CPR 2001.
Looking at articles published in CPR for its first year of publication, 2001, there were 4 qualitative papers based on samples of therapists, 4 qualitative papers based on samples of clients, and 1 qualitative paper based on a joint sample of therapists and clients. None referred to the need for obtaining university ethics committee approval. So, in summary, 50% of the published papers in CPR in 2001 were based on qualitative research with clients.
Diagram 4: Sample from CPR 2010
By contrast, the picture looks very different in 2010. The volume of published articles doubles, compared with 2001. There are more quantitative studies of therapists, of students, and of clients. There is an increase in studies of students, perhaps due to their greater accessibility as research subjects, within university training programmes. Qualitative research with clients, however, has declined by more than half. In this context, it may almost be the case that student samples are beginning to replace client samples, and are serving as a kind of proxy for clients, in the world outside universities. All research projects (except one) make specific reference to approval by university ethics committees.
Correlation is not the same as causation, as any undergraduate course on research methodology will quickly establish. However, there is an interesting association here between the increasing use of university ethics procedures and a shift away from qualitative research towards a greater emphasis on quantitative methods. There is also a decline in publication of qualitative research with clients. This decline is partly masked by an increase in research using therapy students, possibly serving as a kind of proxy measure for clients in the world outside university training courses. A more detailed piece of research, perhaps looking at a wider range of psychotherapy research publications, could help to establish if this is a more widespread phenomenon, and, indeed, if there is a concrete link between the rise of university ethics committee procedures and the shift away from quantitative research with clients, which is perceived as being too risky for the parent institution.
Psychotherapy, as an emerging professional community, is in many ways subordinate to law and medicine, which are fully established professions, enjoying greater social status and hierarchical power. Medicine has been able to determine the research agenda for therapy via its preference for quantitative research, based on randomised controlled trials, which, in turn, provide the evidence base for the influential NICE guidelines within the NHS. In addition, the medical profession has exercised its power through university research ethics committees, in order to shape psychotherapy’s research agenda in significant ways. Topics which are central to psychotherapy, often based on using qualitative research, have been excluded as being overly risky for the university. This exaggerated assessment of risk refers to: the risk of causing litigation for negligence, or for breach of data protection law; of generating distress in research respondents when exploring sensitive topics; and finally, the risks caused by researching young people. A limited sample survey of psychotherapy research publications for 2001 and 2010 found that the proportion of qualitative research papers with therapy clients declined during this period. At the same time, there was evidence that compliance with university research ethics committee procedures became established as standard procedure within psychotherapy research. Whether, and to what extent, these two factors are directly related, would be a topic for other researchers to pursue further in the future.
This article is based on a paper entitled “Hegemony and risk in psychotherapy research” for the Centre for Psychoanalytic Studies at the University of Essex in February 2017. Thanks to Jochem Willemsen and Werner Kierski for their comments and support.
Peter Jenkins is a counsellor, supervisor, trainer and researcher. His most recent book is Professional Practice in Counselling and Psychotherapy: Ethics and the Law (Sage, 2017). Contact: [email protected]
Danchev, D. and Ross, A. 2014. Research Ethics for Counsellors, Nurses and Social Workers. London: Sage
Department of Health. 2005. Research Governance Framework for Health and Social Care. Second edition. London: DoH.
Department of Health. 2015. Mental Health Act 1983: Code of Practice. London: Stationery Office.
Gladwell, D. 2015. Do ethics committees facilitate research? [online] Contemporary Psychotherapy, 7(2) Winter Issue. Available at: https://www.contemporarypsychotherapy.org/volume-7-issue-2-winter-2015/do-ethics-committess-facilitate-research/
Goldacre, B. 2012. Bad Pharma: How Drug Companies Mislead Doctors and Harm Patients. London: Fourth Estate.
Information Commissioner’s Office. 2009. The Guide to Data Protection. Wilmslow: ICO.
Jenkins, P. 2016. Counselling in Higher Education settings: Working with risk, confidentiality and ‘duty of care’ issues, in D. Mair, ed, Short-term Counselling in Higher Education: Context, Theory and Practice. Routledge: London. pp. 45-66.
Masson, J. 2004. The legal context, in Fraser, S., Lewis, V., Ding, S., Kellett, M. and Robinson, C. eds, Doing Research with Children and Young People. London: Sage/Open University, pp. 43-58.
Rupani, P., Haughey, N. and Cooper, M. 2012. The impact of school-based counselling on young people’s capacity to study and learn, British Journal of Guidance and Counselling, 40(5) pp. 499-514.
Takesawa, J., Crowley, R. and Casarett, D. 2004. What is the risk of distress in palliative care survey research?, Journal of Pain and Symptom Management, 28(6) pp. 593-598.
Weaver, M. 2018. Cambridge panel rejected Facebook research project, The Guardian, 25 April.