Ongoing research is suggesting that expertise may be domain or disorder specific, rather than reflecting a core therapeutic attribute or skill. No matter how good we are, we can’t be all things to all clients. Even ‘supershrink’ has a weak spot, and it’s not kryptonite.

One of the stories I used to tell myself back in the day was that, as a therapist, I could work effectively with clients with depression, but not anxiety. I just didn’t ‘get’ anxiety and I felt out of my comfort zone working with it. Have you ever felt this way about working with particular issues?

Are there areas in which you consider yourself more or less strong or confident? Alternatively, maybe you consider yourself a good all-rounder? Whatever you may consider your strengths, it would appear that therapists who are effective in working across multiple problem domains are a rarity. A couple of studies, one from 2011 and the other from 2016, highlight that most therapists have suits in which they are particularly strong, but they are rarely strong in all.

The first study, by David Kraus and colleagues [i]  examined the outcomes of 6,960 clients seen by 696 therapists in naturalistic settings (described as traditional outpatient care) in the US. The measure used to determine their outcomes was the Treatment Outcome Package (TOP). The clinical scales consist of 58 items that assess 12 symptom and functional domains: work functioning, sexual functioning, social conflict, depression, panic (somatic anxiety), psychosis, suicidal ideation, violence, mania, sleep, substance abuse, and quality of life.

Kraus et al. 2011. Therapist effectiveness: Implications for accountability and patient care.

Client outcomes were analysed to determine the degree of change across the 12 domains. Change categories were: reliable improvement, no reliable change and reliable deterioration. The proportion of clients by domain that fell into each category are shown in the table above. Rates of reliable improvement across the domains ranged from a high of 55% on the depression domain, to 13% for mania. Reliable deterioration ranged from a high of 32% for social functioning to a low of 8% for mania.

Change at the therapist level

Data was then explored at the therapist level on the basis of whether their clients reliably improved, reliably worsened, or neither improved nor worsened (no reliable change). Therapists were classified (contentiously perhaps) as follows:

An ‘‘effective’’ therapist

one whose average patient reliably improves

An ‘‘unclassifiable/ineffective’’ therapist

one whose average patient neither reliably improves nor reliably deteriorates

A ‘‘harmful’’ therapist

One whose average patient reliable deteriorates

The table above shows the proportions of therapists, by domain, that fell into each category of effective, unclassifiable and harmful. As can be seen, only in four of the domains were 50% or more of therapists classed as effective. Therapists classed as harmful were generally in a significant minority. However, in eight of the twelve domains clients, 40% or more of therapists were unclassifiable (i.e. their clients neither reliably improved not deteriorated).

Therapists with multiple competencies

The researchers calculated the number of domains in which each therapist was effective, labelling this a competency. The table below shows the results. The average, and modal (most common), number of competencies was five. 96% of therapists were identified as competent in at least one domain. Just one therapist was competent in 11 domains, and none was competent in all.

Given the finding that no therapist was effective in every clinical domain, and the low correlations between domain rankings, the researchers were led to conclude that there appeared not to be a common or core competency or skill that would make a practitioner good in most or all domains. Rather, there are likely to be some domains, or symptom or functioning categories, in which practitioners are stronger than others.

A similar finding was made in the study from 2016 [ii] based on data for 3,540 clients seen in naturalistic settings by 59 therapists. Outcomes were pooled across therapists for each domain and a score calculated for the average therapist. Across the domains, some two-thirds of therapists had four or fewer above average domains. Excluding the domain of mania, just five (9%) of therapists were above average on the remaining 11 domains.

With this finding and the low correlations between domain rankings, there is some preliminary evidence to suggest that there does not seem to be a core competency or skill that renders a great clinician good in all or most domains. For example, the clinician who was ranked best at treating depression was also very good at treating social conflict and panic. On the other hand, he/she was one of the few with patients whose manic and violence symptoms reliably worsened.

Kraus et al. 2011

Do you know your strengths and weak spots?

In another blog I’ve written about the tendency across most professions, including therapy, to over-estimate our overall effectiveness. But how well are we able to judge our performance across a range of different presenting issues and domains? How well do we know our own strengths and weak spots? Answering that question is the subject of another study that is currently underway. [iii] Using the TOP, it aims to compare therapists’ predictions of their own effectiveness across a range of domains with their actual performance. You can be sure that as soon as it’s published I’ll report on it in a further blog. As for me, I’m no longer apprehensive about working with anxiety. I looked long and deep into it and developed a framework for approaching it that seems to work well enough with clients. Above average? I do hope so.


[i] Kraus, D. R., Castonguay, L. G., Boswell, J. F., Nordberg, S. S., & Hayes, J. A. (2011). Therapist effectiveness: Implications for accountability and patient care. Psychotherapy Research, 21, 267–276.

[ii] Kraus, D. R., Anderson, P., Bentley, J. H., Boswell, J. F., Constantino, M. J., Baxter, E. E., & Castonguay, L. G. (2015). Predicting therapist effectiveness from their own practice-based evidence. Journal of Consulting and Clinical Psychology, 84, (6)


Share with your networks
Posted by:Barry McInnes

2 replies on “Even ‘super-shrink’ has a weak spot. And you?

  1. I’m beginning to realise that CORE doesn’t have sufficient detail to rate myself according to the client’s conditions. Can you recommend the TOP system for routine use, and would it spot weak competencies?

    Incidentally, I’m not at all surprised you felt anxieties around your anxious clients. You are empathic, after all…..

    1. Hi David

      I guess my question would be how far down do you want to drill in the client’s conditions and how much data are you prepared to collect to get a definitive picture?

      While I like the idea that the TOP (Treatment Outcome Package) covers many domains, and certainly more than the 34 item CORE-OM, how ‘scientific’ do you want to get? The adult TOP measure has 58 items, and for me that’s way too many to use as anything more than pre and post, and maybe periodically. As a sessional measure for routine use it doesn’t feel feasible. If you want to use the TOP to try and establish areas of what you call ‘weak competencies’ I think you would need to collect a huge amount of client cases and have near perfect data for all clients so you could start to draw some conclusions. I have a lowish caseload and it would take me years! As an alternative you could use your own taxonomy of presenting problems and carefully monitor your outcomes and drop outs against those?

      Don’t get me wrong, the TOP looks like a good measure, and seems very scientifically robust, but for me it’s not feasible for tracking client progress sessionally. For me it’s CORE-OM pre, CORE 10 as sessional and CORE-OM post. Or for a really short measure the ORS Bear in mind the CORE-OM does tap into symptoms of depression, anxiety and trauma, ( the latter of which I don’t see in the TOP domains.

      I hope that helps!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.