I have already written a blog post ... "Warwick BABCP conference: 3rd day - even more evidence that therapists themselves are central to improving outcome (5th post)" ... about the great last morning symposium "The singer and not the song? Evidencing therapist effects across the IAPT stepped care model".  I have described in some detail the first two symposium presentations ... Nick Firth's "Therapist effects and moderators of effectiveness and efficiency in psychological wellbeing practitioners: a multilevel modelling analysis" and Dave Saxon's "Variability in practice: therapist effects in an IAPT service delivering CBT and counselling".  These talks were great and underlined the crucial importance of the, as yet, under-explored area of between-therapist outcome variability. However, in some ways their findings were needed but "me-too" replications of earlier work.  The third presentation of the morning broke newer ground.  It was by Jo-Ann Pereira on "Examining practitioners' personal aspects that contribute to effective practice".   

Jo-Ann asked the intriguing question "What are the personal aspects of practitioners who provide more effective practice relative to those providing less effective practice?".  Partly picking up from hints suggested by earlier research, she looked at the possible importance of therapist Resilience, Empathy, and Mindfulness.  She also explored how patient intake severity (PHQ-9) affected the contribution of these three qualities.  She included 37 practitioners (from the already described Sheffield IAPT group) for whom she had data on both personal qualities and patient outcomes.  These 37 therapists were made up of 8 Psychological Wellbeing Practitioners, 12 CBT therapists, and 17 counsellors.  Their mean scores on the CD-RISC Resilience measure were 69.6 (SD 9.5), on the BES-A Empathy questionnaire 75.7 (7.2), and the MAAS Mindfulness scale 65.2 (9.8).  Using mutilevel modelling to allow for variability in therapist patient numbers & patient severity, she showed intriguingly that the 7 above-average effective therapists had lower BES-A empathy scores than the 22 average effective therapists who in turn had lower BES-A empathy scores than the 8 below-average effective therapists.  Before we rush off and delete all our assumptions about the value of therapist empathy, it's important to clarify that the BES-A assesses three components of empathy ... emotional contagion, emotional disconnection, and cognitive empathy.  Early work on development of this scale showed clear correlations between emotional (but not cognitive) empathy and neuroticism.  Unfortunately I don't think Jo-Ann adequately discussed how the BES-A component scores related to therapist effectiveness.  I think this might well have provided much more interesting data, although looking at the BES-A with items like "I tend to feel scared when I am with friends who are afraid" and "I often get swept up in my friends' feelings" one can easily see how lower scores might well be associated with improved abilities as a therapist.  I do wonder why the BES-A was selected for use as the empathy measure in this research.  I have written at some depth in the past on these issues of empathy & resilience ... see the three blog posts "Do psychotherapists, doctors and leaders develop "emotional chainmail"?  Description of a possible problem", "Do psychotherapists, doctors and leaders develop "emotional chainmail"?  Two kinds of empathy" and "Do psychotherapists, doctors and leaders develop "emotional chainmail"?  Some ways of building both stability and empathy".

Although the findings from the empathy measure are interesting, as reported in their current form, they seem to me to produce more confusion than clarity.  Findings from the mindfulness and resilience measures appear more immediately useful.  So, using multilevel modelling again, more effective therapists scored higher than averagely effective therapists on the mindfulness and resilience measures.  They in turn scored higher than the less effective therapists.  Mindfulness was measured with the Mindful Attention Awareness Scale (MAAS).  This is a helpful questionnaire that reverse scores a series of "mindlessness" items like "I find it difficult to stay focused on what’s happening in the present" and "I could be experiencing some emotion and not be conscious of it until some time later".  It makes good sense that better scores on the MAAS might well be associated with improved ability as a therapist.  Here's a copy of the MAAS as a Word doc and here in PDF format ... and here's a daily MAAS diary in Word doc form and in PDF format.  For much more background and research on the MAAS, see the rather wonderful Self-determination theory website.  Of particular interest is that (especially if these MAAS/effectiveness findings are replicated) therapists could quite possibly train to improve their MAAS scores and potentially their effectiveness ... a good research question for the future.  And another good research question would be ... does the fairly narrowly focused MAAS measure link more or less strongly with therapist effectiveness than a broader measure of mindfulness like the (shortened) "Five facet questionnaire"? 

What about the resilience findings?  The results don't seem surprising.  Spending day after day working at the coalface with suffering fellow human beings isn't easy.  The value of resilience makes good sense.  Jo-Ann seems to have used the full CD-RISC questionnaire ... see the Connor-Davidson website for more on this. There's a slightly geeky query I would raise here as to whether a shorter version of this measure might have been better ... see "Psychometric analysis and refinement of the Connor-Davidson Resilience Scale (CD-RISC): Validation of a 10-item measure of resilience".  Either way, it's easy to see that more effective therapists might well score better on items assessing ability to "Deal with whatever comes", "Can stay focused under pressure" and "Can handle unpleasant feelings".  The talk went on to describe some interesting themes that emerged when questioning the therapists more fully on their professional lives, personal approaches, and how they responded to challenging patients.  The answers threw doubt on the idea that empathy isn't relevant.  In fact, it seemed that the reverse pattern emerged with more effective therapists being quoted as giving empathic, mindful answers like "Being open to all experience other people bring and being open to all aspects (that I am aware of) in myself" or (with challenging patients) " ... respect, listening, realise my own limitations and be flexible with approaches".  Less effective therapists seemed to retreat into "technical concerns" e.g. emphasising becomng "More confident in my skills ... more flexible with ... models and treatment protocols."  

Unfortunately the symposium finished with a somewhat unrelated presentation by Glen Waller.  What a pity that we didn't close with a discussant looking at the issues raised by the three first talks and providing more time for general questions & discussion.  I found the symposium very interesting.  Sitting here now, two questions that occur to me are: 1.)  If this was a report on a cardiac surgery department, and differences like these between surgeons (and their surgical complication & death rates) were being discussed, it might well create something of a media storm if a response to the data was that we couldn't ... as an immediate & high priority ... identify & improve training for the less effective therapists.  The recent paper "Mortality in mental disorders and global disease burden implications: a systematic review and meta-analysis" notes "The median years of potential life lost was 10 years (n=24 studies).  We estimate that 14.3% of deaths worldwide, or approximately 8 million deaths each year, are attributable to mental disorders ... Efforts to quantify and address the global burden of illness need to consider the role of mental disorders in preventable mortality."  Another recent paper "Adherence to antidepressant therapy and mortality rates in ischaemic heart disease: cohort study" comments "The moderate and good adherence groups (of 63,437 patients followed for 4 years) had significantly reduced adjusted mortality hazard ratios of 0.83 (95% CI 0.78–0.88) and 0.86 (95% CI 0.82–0.90) respectively, compared with the non-adherence group. Conclusions Adherence to antidepressant pharmacotherapy is associated with reduced all-cause mortality in a population-based large sample cohort of patients with IHD. Physicians and health policy decision-makers should step up their efforts to sustain and enhance these patients’ adherence to their antidepressant regimen."  It's certainly possible to make a linked argument that making sure that patients see more effective psychotherapists rather than less effective may very well have mortality implications.  Maybe there's not such a difference between cardiac surgery and psychotherapy departments when it comes to death rate effects ... except that our psychotherapy patients' early deaths are less immediately evident to us.  Our key commitment is to our patients.  It's not good enough to simply know that about 15% of therapists are routinely linked with higher client dropout and significantly worse outcomes.  This has huge potential implications for suffering and early death.  We need to respond with more urgency to these findings.  

A second question that emerges for me is 2.)  We desperately want to improve our treatments of psychological disorders.  Four or more decades of effort trying to develop better psychological and pharmacological treatments has been largely crashingly unsuccessful in achieving these ends ... see, for example, the blog post "Psychotherapy (and psychotherapist) outcomes are good but largely stagnant" or the let's-not-bother-to-keep-on-trying (to develop more effective psychotherapies for depression) conclusion at Pim Cuijpers excellent talk on the evidence base a couple of days ago, reported in the post "Warwick BABCP conference ... a master presentation on four decades of outcome research".  But here, in this growing research literature on therapist outcome variability, we have an increasingly bright flashing beacon saying "If you want to improve therapy outcome, LOOK HERE!"  And extensions of the research reported in this symposium could lead the world on this hugely important issue.  Many of us record our therapy sessions with clients.  Just think about it.  If selected above-averagely, averagely, and below-averagely effective therapists recorded a whole series of client sessions (and the associated outcomes), what a wonderful research resource this could be for teasing out key differences between more & less effective therapist behaviours.  Constructive responses to these differences are very likely to be, at least partly, teachable ... so for example in the study "Using client feedback to improve couple therapy outcomes: a randomized clinical trial in a naturalistic setting" the researchers found that providing regular sessional feedback to counsellors & clients on both symptomatic & working alliance scores especially improved the outcomes of initially less effective therapists.  I think this issue of inter-therapist variability is immensely exciting.  We have been largely unsuccessfully hammering down the track of inter-therapy variability for four decades.  Come on guys, let's put more resources into exploring a different route.  So much good could emerge from this.

Note, since I wrote this blog post back in 2015, a paper by Jo-Ann Pereira & colleagues describing these findings has been published in free full text - "The role of practitioner resilience and mindfulness in effective practice: a practice-based feasibility study."