Mental health and perceptions of biomarker research – possible effects on participation

On 20 November, 2007, the UK chancellor Alistair Darling admitted that two computer discs containing the details of over seven million families claiming child benefit had been lost. Missing information included the names, addresses and dates-of-birth of the children, and national insurance numbers and – in some cases – bank details of their parents; 25 million individuals were affected. In the public furore and panic that followed Darling's disclosure, the media were full of commentary reflecting on the grave risks that databases – whether those held by government, by the NHS, or by those conducting research – posed vis-à-vis confidentiality and security. This sceptical view of databases was typified by Simon Jenkins, ex-editor of The Times, who argued that electronic databases are ‘inappropriate for confidential material of any sort’:

A secure online computer is a contradiction in terms. What needs to be private must stay in our heads or be put on paper under lock and key. We should get out of this cul-de-sac and invest in Manila envelopes. (Jenkins, 2007)

If Jenkins' desire for a return to Manila envelopes were to be found representative of what the general public also wants, it would severely cramp much current research in mental health. That research, like medical research more broadly, is increasingly dependent on the use of electronic patient records and the creation of databases that can integrate information from individuals such as fMRI scans, blood and tissue samples, as well as their health and treatment history. This is particularly the case with research involving biomarkers – those characteristics that can be ‘objectively measured and evaluated as … indicator[s] of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention’ (Biomarker Definitions Working Group, 2001). Indeed, biomarker research cannot proceed without an infrastructure comprising complex databases that contain confidential medical information.

Researchers and clinicians are therefore strongly committed to the development and use of such databases because of the many potential benefits that they can yield – for example, within epidemiological research and in aiding the recruitment of research participants. There is therefore widespread apprehension on the part of researchers that perceived risks in relation to privacy, confidentiality and consent could result in overly restrictive access to and use of personal medical data (see for example Academy of Medical Sciences, 2006; Walley, 2006). But of course the practice and validity of such research are ultimately dependent on participants in mental health research agreeing to endorse and be involved in this process of acquiring, bringing together and interpreting data. Currently, we know little about whether mental health service users – and, indeed, carers and other family members – do endorse this process of acquiring and interpreting data, or whether they share Simon Jenkins' wish for a return to Manila envelopes.

Such lack of knowledge is of particular import in relation to the newly-formed National Institute for Health Research (NIHR) Biomedical Research Centres (BRCs) in the United Kingdom, one of which specializes in mental health. The BRCs are aiming to find new ways of preventing, diagnosing and treating ill-health by translating advances in biomedical research into benefits for patients. Much of the focus of the BRC for Mental Health will be on brain imaging scans and biomarkers to understand neurobiological pathways, to ascertain the risk of developing particular conditions, and to predict responses to pharmacological as well as psychological treatments. Research using genetic and genomic models through which to understand the aetiology of mental health conditions and the likelihood of treatment response raises numerous ethical questions about the practice, content and societal ramifications of psychiatric genetics and genomics research (Biesecker & Peay, 2003; Hedgecoe, 2001, 2004; Nuffield Council on Bioethics, 1998; Soskolne, 1997). The BRC for Mental Health is explicitly committed to ensuring that service users and their carers are fully engaged in the research process at all stages. What makes the translational focus of the BRC for Mental Health particularly challenging and fascinating in relation to the involvement and engagement of service users is the complexity that biomarkers pose for aetiology, susceptibility and treatment response. One of the central roles of the ‘Stakeholder Participation’ theme within the BRC for Mental Health – to which both of us belong – is therefore to understand the particular concerns or challenges that biomarker research poses in relation to service user engagement and participation.

Recent studies conducted in the United Kingdom indicate that the public is, in principle, enthusiastic about biomedical research, though keen to know more about the research process, why certain data are required and how they are stored and interpreted (Armstrong et al., 2007; Ipsos MORI, 2007). A volunteer sample of working adults from two employers in the United States was strongly in favour of mental as well as physical illness genetic research, and endorsed many positives and benefits associated with participating in genetic research (Roberts, Warner, Geppert, Rogers, & Green Hammond, 2005). We cannot assume, however, that this enthusiasm vis-à-vis biomedical/genetic research is necessarily shared by those with neurodegenerative and/or psychiatric diagnoses – and to their carers and families – given the prevalence of stigma and complex debates over risk and transmission. As it stands, however, relatively little research has focused specifically on these groups (for exceptions, see Austin, Smith, & Honer, 2006; Jones, Scourfield, McCandless, & Craddock, 2002; Lock, Freeman, Sharples, & Lloyd, 2006; Meiser et al., 2007; Meiser, Mitchell, McGirr, Van Herten, & Schofield, 2005; Steinbart, Smith, Poorkaj, & Bird, 2001; Trippitelli, Jamison, Folstein, Bartko, & DePaulo, 1998).

What do we know about mental health service users' attitudes to biomedical research? There is some indication from small research studies that those with a diagnosis of bipolar disorder are in favour. In one study, participants responded positively to research on susceptibility genes and pre-symptomatic testing for adults (though were far more ambivalent about prenatal testing) (Jones et al., 2002). In two other studies, the majority of participants (who comprised those with a diagnosis of bipolar disorder and unaffected spouses/family members) expressed relatively positive attitudes toward predictive genetic testing (Meiser et al., 2005; Trippitelli et al., 1998). In addition, service users with depression have been found to be significantly more likely to use a biological model to account for their condition than lay people (Kuyken, Brewin, Power, & Furnham, 1992), and there are some indications that individuals with a family history of alcoholism are more likely to believe in a genetic cause of alcoholism (Creeden, Manowitz, Hamer, & Ballou, 2000). These studies suggest that investment in biological causation models and endorsement of genetic testing might also indicate a willingness to participate in biomarker research – though this would certainly need to be verified through additional research.

But other research has revealed more equivocal feelings towards genetic research and genetic models of causality. A study in Germany, for example, that assessed the attitudes of the public in general and those of service users and their relatives indicated that, though the majority of people were in favour of psychiatric genetic research, they were ambivalent about the moral implications of such research. In addition, there were significant variations in how the groups spoke about pre-symptomatic testing (Illes et al., 2003). Williams and Healy's study of individuals who had recently been referred to a community mental health team indicated how some people can move from one strongly held view about the causation of mental health problems to another ‘relatively quickly and unproblematically’ (Williams & Healy, 2001, p. 474). This suggests that more research is required to explore why and when causal beliefs about mental health problems are held on to strongly and when they are abandoned – and what the implications might be in relation to individuals' willingness to participate in mental health research involving biomarkers. It should also be emphasized that we still know very little about the views of those with certain diagnoses (e.g. schizophrenia) towards genetic/biomedical research, and how views might vary according to diagnosis, gender, age, ethnicity, and so on.

Because biomarkers can indicate aetiology, susceptibility and treatment response, they can be used for multiple purposes: stratifying study groups in clinical studies, establishing diagnosis, monitoring and predicting treatment response, measuring toxicity, investigating pathophysiological mechanisms, exploring concentration – response relationships, and screening potential new therapies in vitro and in vivo. Biomarkers might point to environmental exposure (e.g. to stressors or toxins) as well as to individual susceptibility, to an early disease stage, or to indications of response to psychological as well as biomedical treatments. Biomarkers therefore point to intricate relationships between biological and psychosocial influences, and pose difficult questions for attempts to establish causality. They raise far greater interpretative challenges than do Mendelian accounts of genetic transmission.

When ascertaining whether and how potential mental health research participants wish to engage with biomarker research, it is crucial not to assume that instances of reluctance ought necessarily to be interpreted as a lack of knowledge. The ‘deficit in knowledge’ model has been very powerful in attempting to explain why the public does not necessarily embrace bioscientific technologies, frameworks and innovations. According to this model, the public's ignorance can be addressed by conveying accurate scientific knowledge that will supersede mistaken lay accounts and trump the misrepresentations and inaccuracies circulated within the mass media. But the deficit model has come under sustained critique within sociology and studies of science and technology since the early 1990s. The sociologist Bryan Wynne, for example, has argued that the production and circulation of scientific knowledge is socially negotiated – both by the so-called ‘public’ and by scientists themselves (Wynne, 1993). Many social scientists have therefore moved away from the deficit model in approaching the public understanding of science and have instead proposed a much more complex two-way relationship between ‘science’ and the ‘public’, neither of which can be seen as simple unified entities. In this model, there is an ongoing negotiation of what both ‘science’ and ‘knowledge’ mean; a negotiation in which views and frameworks are contested, new understandings are generated, and shifts occur in what count as scientific facts. The responses of service users, carers and other family members to biomarker research are therefore likely to be influenced by their global beliefs about databases, and by the frameworks they use to understand the benefits (or lack of benefits) of research, to explain mental health problems, and to respond to biomedical and genetic accounts of the individual. (We should also not underestimate the effects that health professionals and the media might have on people's willingness or refusal to participate in such research.)

There are some indications that the general public's understandings of and responses to genetic research are much more subtle and less one-dimensional (Condit, 1999) than had been assumed by those fearing that genes might be interpreted as the ‘blueprint of destiny’. Nevertheless, recent research on the stigma and prejudice experienced by those with psychiatric diagnoses gives cause for concern when assessing the societal implications of psychiatric genetic and genomic research. Optimists had assumed that genetic accounts would reduce stigma by virtue of rendering mental health problems more similar to physical illnesses, and of attributing the causes of mental disorders to factors outside of individuals' control. There are growing indications that this optimism is misplaced. A population survey in 2001 of over 5000 adults in Germany demonstrated that:

endorsing biogenetic causes, particularly seeing a brain disease as the cause, increases the likelihood that people with schizophrenia and major depression will be considered as individuals who are lacking in self-control, and who are unpredictable and dangerous, which is positively associated with fear. This, in turn, is linked with an increased desire for social distance. (Dietrich, Matschinger, & Angermeyer, 2006, p. 172)

In addition, a recent review of the ‘mental illness is an illness like any other’ approach to reducing the prejudice experienced by those with schizophrenia argues that, when ‘the disease model is applied to the brain, the assumption is that the person is incapable of judgments, reason, autonomy – that their personhood is negated’ (Read, Haslam, Sayce, & Davies, 2006, p. 311). This research points to the political and ethical importance of ascertaining whether biomarkers research will be interpreted – both by those with and without psychiatric diagnoses – as endorsing biogenetic causes. If this is indeed the case, then this raises serious questions for the practice of biomarkers research in general and for the engagement of participants in particular. One of the most sensitive studies to date that provides some assistance in addressing such questions is a vignette experiment involving individuals with mental illness that was embedded in a nationally representative survey in the US (Phelan, 2005). Phelan's study, as well as questioning the optimistic account of genetic explanations reducing stigma, showed that while a genetic attribution had no effect on desired social distance from the person with mental illness or on restricting his/her rights to reproduce, it did increase social distance from the mentally ill person's sibling – particularly vis-à-vis willingness to date and/or marry him/her. Overall, the study indicated that genetic attributions might be most beneficial to parents (through absolving them of responsibility and guilt vis-à-vis their offspring's mental health problems) and most damaging for younger relatives. Given the potentially very different ramifications of genetic accounts for service users, carers and other family members suggested by Phelan's research, it becomes even more pressing to ascertain whether or not biomarker research will be interpreted as falling squarely within ‘genetic accounts’ of mental illness. Phelan's study showed, it should be noted, that the three levels of genetic causation (genetic, partly genetic and not genetic) tended not to be linearly related to the outcome measures used for stigma. In other words, a ‘partly genetic’ attribution for mental health problems was not equivalent to a ‘weaker dose’ of a wholly genetic explanation. Phelan argues that this, alongside the fact that respondents tended to favour the ‘part genetic’ attribution, points to the pressing need for more research on the public's interpretation of gene-environment interactions.

While biomarkers research is certainly preoccupied with gene-environment interactions, it is not clear, then, whether it will be interpreted as such by the lay public. Current research involving biomarkers presumes complex interactions between genetic, environmental and stochastic phenomena in which the key terms are probability and susceptibility rather than determinism (Frazzetto & Gross, 2007). But a review of the genetics of Alzheimer's disease indicated that significant confusion surrounds the calculation and transmission of information about the risk associated with susceptibility genes to patients and their families. The authors argued that confusion, dispute, and uncertainty about molecular genetics characterized the views of many medical experts as well as those of lay individuals (Lock et al., 2006).

What do these various arguments and research studies imply about the possible effects on participation in mental health research that result from perceptions of biomarker research? Researchers and clinicians who are themselves involved in biomarker research in relation to psychiatric and neurodegenerative conditions and substance misuse disorders are likely already to be convinced by the benefits of such research. They are hence likely to feel frustrated when witnessing reluctance on the part of potential research participants to embrace these methods and models of psychiatric research. Any such frustration needs to be tempered through acknowledging the justifiable anxieties that individuals have about the potential transmission of confidential and personal medical data, and through considering that mental health conditions cannot automatically be regarded as analogous to physical health conditions.

Emerging research on the potential consequences of biogenetic models on levels of stigma and prejudice, and the well-known diversity of accounts that individuals hold, mean that those pursuing biomarkers research cannot assume that simply providing information on biomarkers is the way to engage service users, carers, other family members – and other members of the public – in such research. What is clear is that we do not need to go back to Manila envelopes: rather, we need a research programme that considers the active engagement of mental health service users at each stage of the research. This would include an understanding of the conceptual frameworks that influence the perceptions of biomarker research, and the complex and justifiable reasons that people might have for wishing not to engage in such research.

Acknowledgements

The authors acknowledge financial support from the Department of Health via the National Institute for Health Research (NIHR) Specialist Biomedical Research Centre for Mental Health award to South London and Maudsley NHS Foundation Trust (SLaM) and the Institute of Psychiatry at King's College London.