A new study published in Psychotherapy and Psychosomatics examines previous findings on the tendency for therapists to overestimate their progress with clients, and presents the efficacy of alternatively using client self-report of feedback throughout therapy.
“Clinicians appear to be overly optimistic about patient benefits compared to measured outcome. Failure to recognize that a patient is not responding to treatment is a serious problem in routine care and one that appears to be made worse by the clinicians’ confidence in their clinical judgment and unique healing gifts.”
Michael Lambert, a professor at Brigham Young University, summarized the results of 12 clinical trials that measured and monitored patient progress and feedback throughout the course of therapy and observed the effect of feedback on treatment outcomes.
Previous literature has found that approximately 20-40% of patients fail to respond to treatment in clinical trials, and prospects are even worse in child or adolescent samples. This is despite deliberate attempts to provide evidence-based treatment from demonstrably successful therapists after patients with the same “disorder” have been carefully screened.
Furthermore, clinicians tend to overestimate their abilities and the progress of their clients. One in four therapists rate themselves within the top 10% of therapists, and none had the impression that they were below average, one study found. It may be this false perception that one is exceptionally qualified to provide effective services that also prevents therapists from being able to assess client progress accurately.
“Unfortunately, the clinicians’ view of their own patients’ outcome is much more positive than the measured outcome using self-report scales. In their survey of clinicians, Walfish et al. suggest that they estimate about 85% of their patients improve or recover, an estimate that far exceeds estimates based on measured outcomes in clinical trials and routine care. The discrepancy between clinician estimates of success and measured success suggests the need for formally measuring and monitoring treatment response.”
Lambert suggests that this discrepancy between clinicians’ ability to discern client progress and their realistic state of progress may be underlying issues in psychotherapy treatment including patient deterioration, treatment failure, and client dropout. It is not as though clients at-risk of deterioration are necessarily difficult to detect, either. Rather, studies demonstrate that patients at risk of deterioration can often be identified after just the first few sessions.
Lambert summarizes additional research suggesting the ways in which feedback can improve therapist performance. The greater the discrepancy between therapist perception of progress and measured progress, the more likely it is that feedback will be beneficial. Other researchers have conceptualized a theory to determine how feedback is most useful, and found it is most effective when therapists are committed to improving, if they are made aware of the discrepancy between their views and measured progress, the source of the feedback is deemed credible, and when feedback is timely, simple, and offers concrete suggestions to improve.
“Clinicians could benefit from employing formal mental health vital sign tracking systems because of their proven accuracy in identifying treatment failure and thereby overcoming a clinician’s overly optimistic estimates of their patients’ treatment responses and because of their inability to predict treatment failure, specifically negative change.”
In this study, Lambert observed the usefulness of feedback through two measures, the Outcome Questionnaire (OQ) and the Youth Outcome Questionnaire (Y-OQ), which can be recorded with a computer tracking system. As the founder and partial owner of OQMeasures, the owner and distributor of the OQ, Lambert reports the potential for conflict of interest within this study and acknowledges limitations of the measures within the limitations section of this report.
The primary research question was whether or not clients have better outcomes when therapists are receiving feedback and monitoring of client self-reported progress. A secondary question is if so, then to what degree is it more useful relative to outcomes in clients receiving care from therapists who did not receive feedback?
Lambert’s summary indicates that feedback was helpful for some clients more than others. Most clients who receive psychotherapy tend to progress at a relatively steady rate. For those clients, feedback did not make a significant difference in their rate of progress.
For the other 20-40% of clients at risk of deterioration or dropout, feedback significantly and positively altered the course of their treatment such that deterioration rates were reduced from 21% to 13%, and recovery rates increased from 20% to 35%. When concrete strategies for improvement accompanied this feedback, deterioration rates were further reduced to 6%, and recovery rates further increased to approximately 50%.
While therapist qualities and skills are not solely responsible for barriers in treatment progress, monitoring treatment and identifying improvement strategies may serve to direct therapists’ attention to other therapeutic factors, prompting them to carefully attend to the client-therapist relationship, consider social supports, develop greater attunement toward client readiness to change, and other components.
Yet, sometimes, Lambert notes, monitoring the course of psychotherapy is unfortunately imposed on therapists by systems of care in a manner which might elicit resistance of what feels like external control and management. “Nevertheless,” he writes, “it appears that this research-based innovation (formal monitoring and problem-solving) has little downside for clinicians (it is cheap and effective) and large upsides for patients.”
It is important to consider that there was a limited range of studies reviewed by Lambert which featured a small variety of researchers using client self-report measures of progress and outcome. Finally, Lambert warns against the reliance on feedback tools and processes, rather than interpersonal attunement and skill, to improve outcomes:
“Mental health self-report data cannot capture the full range of psychological functioning any more than a thermometer can detect cancer, diabetes, or heart disease. Furthermore, collecting such data cannot cure mental illness any more than sticking a thermometer in a patient’s mouth can cure the flu. Feedback data themselves are not helpful unless clinicians know how to use the data to improve treatment.”
****
Lambert, M. J. (2017). Maximizing Psychotherapy Outcome beyond Evidence-Based Medicine. Psychotherapy and psychosomatics, 86(2), 80-89. doi: 10.1159/000455170 (Abstract)
Read: “The Legacy of Saul Rosenzweig: The Profundity of the Dodo Bird” by Barry Duncan. It’s available on Researchgate. Duncan’s article explores the history of what he calls “common factors” in psychotherapy, a reconceptualizing of what Lambert identified in 1992 as “therapeutic factors.” The idea that factors common to different psychotherapies could account for therapeutic change dates back to Saul Rosenzweig’s 1936 paper. Quoted from the Duncan article:
“Lambert (1992) identified four therapeutic factors (extratherapeutic, common factors, expectancy or placebo, and techniques) as the principal elements accounting for improvement in psychotherapy. Inspired by Lambert’s proposal, Miller et al. (1997) expanded the use of the term common factors from its traditional meaning of nonspecific or relational factors to include four specific factors: client, relationship, placebo, and technique.”
“The Heart and Soul of Change,” by Mark Hubble, Barry Duncan and Scott Miller (now in its 2nd edition) showed how these common factors demonstrate that improvement during psychotherapy can be attributed as follows: client (40%), relationship (30%), placebo (15%), and technique (15%). The relationship factor is seen as fifty-fifty client and therapist. So improvement in psychotherapy from this perspective is 70% due to the client and 30% due to a therapist and technique. So feedback from clients is an important part of making therapy work.
Without feedback form the client, how would you ever know if you were making progress? This seems to obvious as to be trivial, but it’s clear from the data that the majority of therapists don’t practice this way. It’s a head scratcher, for sure!
I would think if therapists had many effective methods, and just needed to find the right one, they would want feedback. But since they usually don’t, I question that they really have that many tools. Unless the client can tell them how to do their job, they wouldn’t know what to do with the feedback.
It is not surprising that if you have psychologists being trained with the medical model as backdrop, then your product will be someone who believes that they use “evidence based treatments” (usually CBT) for particular conditions (MDD, agoraphobia, GAD, etc.). Just like doctors they don’t treat people, but their “illness” and thus there is no need to consider context, how a person’s life has unfolded given his/her circumstances or what might have caused or contributed to their distress – all you have to focus on is symptom reduction and that is also the only feedback you need to elicit from them. When they “relapse” after 3 to 6 months (which is usually the case), you can blame them for not using the skills that you have taught them or that they were not overly “psychologically minded” to start off with in the first place. To these psychologists technique is everything and the relationship a mere practical coincidence. I see this particularly here in Australia and remain hopeful that one day the profession as a whole will look back and recognise that, first and foremost, people in distress need connectedness with others and that there needs to be a far greater emphasis in psychologists’ training on selecting people who have natural facilitation skills, empathy, kindness and respect as opposed to the current trend of selecting the most academically successful ones and turning them into psycho-technicians (I can’t think of another way of describing them)
Sadly, therapists are not very good at appraising client progress. Lambert and the work of many others has brought these limitations to light. Hubble, Duncan, and Miller have developed several really good process oriented assessments (outcome rating scale, session rating scale) that are not expensive or cumbersome to administer, many of which take less than a minute to complete. But, it gives the therapist immediate feedback on their performance from the perspective of the client, a perspective that is often ignored or at least under appreciated. The OQ and YOQ instruments are excellent but the downside is that they are proprietary and costly.
I just ran across this blog that offers some good insights: http://www.betterevaluation.org/en/blog/best_practices_arent
A very good article! I’ve always been very suspicious of “EBT” for the very reasons mentioned in the article, namely that it is both politically and financially motivated. The one issue not addressed in the article is that in order to be an “EBT”, someone has to spend the money to research something in a systematic way. Guess who has the most money to fund such research? Drug companies, of course. Drug companies can afford to run multiple research studies (and of course can afford to buy “results”) in a way that other entities can’t. So of course, most “EBTs” are drugs. Pretty sneaky!
I also like the emphasis on humility and individualized approaches, even in the purely medical world. That thinking is 50 times more applicable to the “mental health” world, where there is no concrete way to even say who is “ill” or if anyone is “ill” at all.
Thanks for the link!
— Steve
That was helpful. Thankyou.