Shifting the Paradigm for Health Care Research
The undoubted highlight of my week was the first Annual Research Meeting between colleagues at the Tavistock and Portman and those at our academic partner at the University of Essex. It produced some very rich discussions no more so than a “question time” debate on the nature and purpose of research.
For good reasons research is a major voice in setting the direction of policy and action in the NHS. The aspiration for evidence based policy and practice is an accepted paradigm of how we do business although its application in the real world of health care delivery and commissioning is still patchy and inconsistent. Nonetheless at a time of ever more limited resources the mantra of doing “what works” is undoubtedly compelling. I would not want to question this view but I do think that we need to take a more sophisticated approach to formulating the questions on which we base the judgements on “what works” so that it better reflects the complexities of patient choices and the psychological and social factors which drive those choices.
Research has been one of the key domains in which the Cinderella status of mental health has been visible. Despite representing 23% of the demand on the NHS mental health accounts for only 9% of reach spending. Research in mental health has also been a fraught area at times, illustrating deep differences of view in the understanding of the causes of mental ill health of a scale unlikely to be seen in many areas of physical medicine. However mental health’s historic weaknesses in the area of research can also be turned to advantage in pointing to the need to change the paradigm of how we undertake health care research. Some of the lessons learnt in doing so can equally be applied to areas of physical health care, especially as the focus of the NHS’s work shifts from the treatment of single diagnosable conditions to the management of multi-morbidity.
Despite the historical claims of some of its practitioners, mental health has always been in a relatively poorer position than physical health in being able to categorically demonstrate what works. This has often been put at the door of a lack of scientific rigour and at times there has been some truth in this claim. However there is another factor. That is the much more visible and significant role which individual agency and social circumstances play in contributing both to the development of mental health problems and to the pathways of recovery from them.
Traditional healthcare research, epitomised by the gold standard RCT, has been historically good at demonstrating the population impact of various treatments. This model works well in situations, for instance pharmaceutical interventions, where the objectives and impact of an intervention can be clearly isolated and measured objectively, ideally over a relatively short period of time. It is also an approach which has its limitations.
There are two in particular which I would like to focus on. First is the alignment of a patient’s preferences and wishes with the purpose and potential of treatment. The best example of this is the role of anti-psychotic medication. If research on the impact of medication is purely focused on the alleviation of voice hearing it will tell a very partial story about the efficacy of those interventions. It was little surprise that the excellent work done by the James Lind Alliance in working with people with lived experience to identify priorities for research into schizophrenia and psychosis gave a significant level of prominence to work to understand the impact of treatment on sexual functioning.
Rebalancing the focus of research (as well as service delivery) by taking full account of the perspectives of people who use services is important not just to tick a box but to ensure that the questions we ask in research are driven in utilitarian terms by the interests of those who are affected by illness and not just by those who have an interest in the provision of services. Some researchers are beginning to focus on this and develop tools and techniques to take ensure patient preference can be properly taken account of in healthcare research and development. It has been very exciting, as part of our work to put together new models for CAMHS services, to work with colleagues from the Dartmouth Center for Health Delivery Science who have produced the CollaboRATE as one approach to systematically measuring shared decision making.
A more sophisticated model of asking what works also needs to look not just at preferences but also needs to be ready to take a more holistic view of the interventions which impact on health and wellbeing , including crucially the role of individual agency. Consciously or subconsciously most therapeutic processes require a level of concordance. In mental health services we know that there are major issues with dropout rates from services such as IAPT and in CAMHS the modal number of sessions is only one. These are crucial issues in a resource constrained system. We need much more research which focuses on evaluating and better understanding the qualitative impact of clinical interactions with the aim of identifying the factors in both clinical and patient behaviour which best promote positive engagement with treatment and the achievement of successful outcomes.
None of this is to decry the appropriate place of RCTs where they can best and most definitively answer key questions about effectiveness. Mental health, in particular, needs more well-constructed research of this kind and some treatments such as CBT can clearly show the benefits of building up a strong evidence base.
However we need far more than just RCTs and, as I have tried to argue in this blog, we need to put a much greater emphasis in incorporating patient preferences and a deeper understanding of individual agency as a key factor in therapeutic processes. In doing so we are more likely to research and by implication invest in the health and other care services which are most likely to increase human life and happiness.