Myndgard

Blog · · 5 min read

Measurement-based care in counselling: what the evidence says

What routine outcome monitoring is, what the research on progress feedback actually shows, and how services use it without sidelining clinical judgement.

By Myndgard Limited

Banner reading: Measurement-based care in counselling: what the evidence says

Most counsellors already ask clients how things have been since the last session. Measurement-based care takes that question and makes part of it routine: a short, validated questionnaire completed regularly, with the scores looked at by the counsellor, and often by the client too, as the work goes on.

It is a simple idea that attracts strong opinions. Some see it as the most practical way to notice when therapy is not working. Others see form-filling that gets in the way of the relationship. The research supports neither the strongest claims nor the strongest objections, and it is worth reading carefully.

What it is

A 2019 review in JAMA Psychiatry by Lewis and colleagues defines measurement-based care as the "systematic evaluation of patient symptoms before or during each clinical encounter to inform behavioral health treatment". The last part matters. Collecting scores is not the point. Using them to inform the next conversation is.

In practice that usually means three things:

  • a brief measure completed at intake and then at regular points, ideally every session;
  • the results shown to the counsellor in a form that makes change, or lack of change, easy to see;
  • a habit of talking about the results with the client, especially when progress has stalled.

The same review found that fewer than 20% of behavioural health practitioners use measurement-based care, and as few as 5% use it at the frequency the evidence suggests. Those figures are drawn largely from US research, so they describe a general pattern rather than any particular Irish or UK service.

What the evidence shows

The best-known research comes from Michael Lambert's group and the Outcome Questionnaire (OQ) system, along with the Partners for Change Outcome Management System (PCOMS). A 2018 meta-analysis by Lambert, Whipple and Kleinstäuber brought together 24 studies of these two systems. Two-thirds found that therapy with routine outcome monitoring did better than usual treatment from the same practitioners. The effects ranged from small to moderate. The clearest finding was for clients predicted to have a poor outcome: feedback "reduced deterioration rates and nearly doubled clinically significant/reliable change rates" in that group.

A broader 2021 multilevel meta-analysis by de Jong and colleagues in Clinical Psychology Review cast a wider net: 58 studies, randomised and non-randomised, covering 21,699 patients. It found:

  • a small but significant effect on symptom reduction compared with control groups (d = 0.15);
  • a similar small effect for clients who were "not on track" (d = 0.17), which rose to d = 0.36 when the feedback came with clinical support tools, prompts that help the counsellor think through why progress has stalled;
  • a small favourable effect on dropout (odds ratio 1.19);
  • no significant effect on deterioration or on the number of sessions.

It also flagged sources of bias. Effects were larger when the same questionnaire was used to give feedback and to measure the outcome, and in the not-on-track group, larger when the system's developers were involved in the study.

Not every review is positive. A 2016 Cochrane review by Kendrick and colleagues looked at 17 trials (8,787 participants) of routine outcome measures in common mental health problems among adults. Pooling 12 of them, it found no evidence of a difference in symptoms between feedback and no-feedback groups, and judged the evidence low quality because of high risk of bias. Its conclusion was that there was "insufficient evidence" to support routine monitoring for improving outcomes, and that better research was needed.

The fair summary is that the overall effect is small, it is not guaranteed, and how feedback is delivered matters. The case is strongest for the clients who are quietly not improving.

Student counselling in the UK

UK university counselling has a long history of using outcome measures, mostly the CORE family. A 2021 study by Broglia and colleagues for the SCORE consortium pooled data on 5,568 students from four UK university counselling services using either CORE measures or the Counseling Center Assessment of Psychological Symptoms (CCAPS).

Two points from that paper are useful for any service thinking about measurement. First, it notes that an earlier and often-quoted figure, 70% of students showing reliable improvement on the CORE-OM, was "based only on those students completing counselling". Second, the services in the study used a short measure (CORE-10 or CCAPS-34) at the start of every session. That meant that when a student stopped attending without a planned ending, their last available score could still be used. Measuring only at the start and end leaves a service blind to exactly the students it most needs to understand.

Common objections, and how services answer them

The Lewis review lists the barriers that come up again and again.

"Measures are no better than clinical judgement." The same review notes that practitioners "typically overestimate how well patients are doing in treatment". That is not a criticism of counsellors. It is a known limit of anyone judging change from inside a relationship. A score does not replace judgement; it gives judgement something to check against.

"It takes time." Short measures exist for this reason; the CORE-10, for example, has ten items and can be completed before the session starts. Much of the time cost sits in looking at the results, which is where clear displays help.

"Clients will find it impersonal." Some will. Services that do this well explain why they ask, show clients their own scores, and treat a surprising result as a conversation starter rather than a verdict.

"The data will be used against us." The review lists concern about punitive use as a real practitioner barrier. Services need to be explicit that session-level measures exist to support the work with each client, and that any service-level reporting is about planning, not individual performance tables.

"What about confidentiality?" Clients raise this too. They should know who sees their scores and why, and should be able to ask questions about it.

In practice

For counselling services considering measurement-based care, the evidence points to a few modest, practical steps:

  • Measure often enough to be useful. Session-by-session measures catch unplanned endings and early stalls.
  • Look at the results before or during the session, not months later in a report.
  • Give extra attention to clients who are not on track. That is where feedback has the clearest effect.
  • Keep clinical judgement in charge. Scores inform the conversation; they do not decide it.
  • Be clear with clients and staff about who sees what, and why.

Done this way, measurement is less about proving that counselling works and more about noticing, sooner, when it isn't working for a particular student.

Sources