How do we know whether a patient is likely to do well in the psychologically-based treatment we offer them? The truth is, at least for the moment, we don’t. At least not in any way that is evidence-based or precise.
If you work clinically and are anything like me, this might sit rather uncomfortably. I work with patients with chronic pain and quite regularly feel that I have a “sense” early on about who is likely to engage well and make changes in treatment. Surely there must be some way of quantifying this? What about how long they have been in pain? How many previous treatment attempts have failed? How intense or debilitating their pain is? What about how anxious they are? How depressed they are? How old they are? Surely there must be something that will give us a clue to how they might respond?
Psychological treatments are increasingly recommended for chronic pain, but they don’t seem to be effective for everyone. In fact, in one recent study of Acceptance and Commitment Therapy (ACT) and Cognitive Behavioural Therapy (CBT) only 27% of participants were classified as treatment responders [1], although various other studies of ACT or mindfulness-based treatments have found much higher treatment response rates ranging between 51% and 75% [2-5]. Outcome domains and measures tend to differ between studies, which likely contributes to these differences. It has also been argued that using “responder” analysis in itself can be misleading as any observed effects are not necessarily specific to treatment but may reflect more general prognostic factors [6]. Across treatments for chronic pain more generally (including both pharmacological and non-pharmacological approaches), around 40% of patients are not satisfied with the treatment they receive [7].
Let’s consider these figures for the moment, along with the fact that in the UK the NHS spends an estimated £12 billion a year on back pain alone. That is a lot of money being spent on a lot of patients receiving a treatment that did not help them. Arguably, funds that could be better spent on developing more targeted treatments for those patients who are not currently benefiting.
It seems clear that in order to develop better treatments we need to understand how and why one particular treatment might work well for one patient yet not for another. Predictors are general factors that are associated with outcome independent of any particular treatment approach, whilst moderators are factors where their association with outcome differs depending on the specific treatment approach [8]. In a recent systematic review [9], we aimed to establish patient baseline characteristics associated with outcome in Contextual forms of Cognitive Behavioural Therapy, such as Acceptance and Commitment Therapy (ACT) and mindfulness-based approaches. Studies were only included in the review if they investigated adults with chronic pain, if they were longitudinal in design, and if the results allowed conclusions to be drawn about moderators or predictors (even if this was not actually an aim of the study). Data extraction was performed by two independent reviewers and included information on study design, sample size, setting, treatment content and format, outcome assessment measures and details of potential moderators or predictors. An adapted version of the Hayden criteria (a tool designed specifically for studies of prognostic factors) [10] was used to assess the methodological quality of the studies. Twenty studies were included in the review but across these studies no firm conclusions could be drawn about predictors or moderators. One characteristic that appeared to be potentially important was baseline emotional functioning, although the direction of this effect was inconsistent in that higher distress predicted better outcome in some trials and worse outcome in others.
Various methodological limitations were identified across the studies included in the review. These included differences in the treatment protocols and outcome assessment measures used, which made it difficult to compare or combine results across studies, as well as ad hoc rather than pre-planned analyses of predictors. Perhaps most interestingly, there appeared to be little theoretical guidance to selecting potential predictors for investigation. This last point seems particularly relevant as ACT is a model that is inherently grounded in psychological theory. ACT draws upon the “psychological flexibility” model, a model of a person’s ability to respond sensitively and flexibly in a given situation to select a behaviour that leads them in the direction of their valued goals [11-13]. The key processes of the psychological flexibility model are explicitly targeted in treatment. Yet none of the 20 studies had investigated the key processes of the psychological flexibility model as possible predictors of outcome.
If we want to better understand if a particular patient is likely to respond to treatment, we need to understand how our treatments are working in the first place. Once we better understand the processes of change in a particular treatment, perhaps we can start to predict which patient characteristics might interact with these processes to make someone more or less likely to respond to treatment. We suggest that using a theoretical approach (for example, the psychological flexibility model) to investigate predictors or moderators may help us better understand our patients, better formulate their needs, and better design more targeted and effective treatments. Watch this space.
About Helen Gilpin
Helen worked and completed her MSc in Cognitive Neuroscience and Neuroimaging at the University of Nottingham, before continuing her research as part of the Body In Mind team. She went on to complete her Doctorate in Clinical Psychology back in the UK. Her doctoral research focused on trying to better understand the factors that might make someone more or less likely to respond to a psychologically-based treatment for chronic pain. She also has a specific interest in the relationship between body representation disturbances and chronic pain. Helen now works as a Clinical Psychologist in a specialist pain management service in London, delivering interdisciplinary pain management programmes based on Acceptance and Commitment Therapy. In her spare time she enjoys cheese, wine, trail running, and travelling as much of the world as she possibly can.
References
[1] Wetherell, J.L., Petkus, A.J., Alonso-Fernandez, M., Bower, E.S., Steiner, A.R., Afari, N. (2016). Age moderates response to acceptance and commitment therapy vs. cognitive behavioural therapy for chronic pain. Int J Geriatr Psychiatry, 31, 302-8.
[2] Kaplan, K.H., Goldenberg, D.L., Galvin-Nadeau, M. (1993). The impact of a meditation-based stress reduction program on fibromyalgia. Gen Hosp Psychiatry 15, 284-9.
[3] McCracken, L. M., Sato, A., & Taylor, G. J. (2013). A Trial of a Brief Group-Based Form of Acceptance and Commitment Therapy (ACT) for Chronic Pain in General Practice: Pilot Outcome and Process Results. The Journal of Pain, 14(11), 1398–1406.
[4] Simister, H. D., Tkachuk, G. A., Shay, B. L., Vincent, N., Pear, J. J., & Skrabek, R. Q. (2018). Randomized Controlled Trial of Online Acceptance and Commitment Therapy for Fibromyalgia. J Pain. Advanced online publication doi: 10.1016/j.jpain.2018.02.004.
[5] Vowles, K.E & McCracken, L.M. (2008). Acceptance and values-based action in chronic pain: A study of treatment effectiveness and process. J Consult Clin Psychol, 76, 397-407.
[6] O’Connell, N.E., Moseley, G.L., McAuley, J.H., Wand, B.M., Herbert, R.D. (2015). Interpreting Effectiveness Evidence in Pain: Short Tour of Contemporary Issues. Phys Ther, 95(8), 1087–94.
[7] Breivik, H., Collett, B., Ventafridda, V., Cohen, R., & Gallacher, D. (2006). Survey of chronic pain in Europe: prevalence, impact on daily life, and treatment. Eur J Pain, 10(4), 287-333.
[8] Hancock, M., Herbert, R., Maher, C.G. (2009) A guide to interpretation of studies investigating subgroups of responders to physical therapy interventions. Phys Ther, 10, 698–704.
[9] Gilpin, H.R., Keyes, A., Stahl, D.R., Greig, R., McCracken, L.M. (2017). Predictors of treatment outcome in contextual cognitive and behavioural therapies for chronic pain: a systematic review. J Pain, 18(10), 1153-64.
[10] Hayden, J.A., Cote, P., Bombardier, C. (2006). Evaluation of the quality of prognosis studies in systematic reviews. Ann Intern Med. 144, 427-37.
[11] Hayes, S.C. (2004). Acceptance and commitment therapy, relational frame theory, and the third wave of behavioral and cognitive therapies. Behavior Therapy. 35, 639-65.
[12] Hayes, S.C., Luoma, J.B., Bond, F.W., Masuda, A., Lillis, J. (2006). Acceptance and commitment therapy: model, processes and outcomes. Behav Res Ther. 44, 1-25.
[13] Hayes, S.C., Pistorello, J., Levin, M.E. (2012). Acceptance and Commitment Therapy as a Unified Model of Behavior Change. The Counseling Psychologist. 40, 976-1002.