Myndgard

Blog · · 4 min read

Supporting clients between sessions: what the evidence says

Routine outcome monitoring, between-session work and brief questionnaires: what the research supports, where it is modest, and what to plan for.

By Myndgard Limited · Updated

Banner reading: Supporting clients between sessions: what the evidence says

Updated October 2026: this article has been rewritten as a general, evidence-based piece. An earlier version made product claims that no longer reflect Myndgard.

A client might spend one hour a week with their counsellor and the other 167 without them. Most of what shapes how therapy goes, including mood, sleep, stress and whether a new skill gets tried, happens in those other hours. The clinician usually learns about them through the client's recollection at the start of the next session.

There is reasonable evidence for two ways of making that time more visible and more useful: measuring progress regularly, and giving structured work to do between sessions. Both have limits worth being clear about.

Measuring progress as you go

Routine outcome monitoring means asking clients to complete a short measure at regular points, often every session, and showing the results to the clinician. The idea is simple: clinicians are not always good at spotting when a client is quietly getting worse, and a regular measure gives an early signal.

The evidence is encouraging but modest. A 2018 meta-analysis of 24 studies of the two most-studied feedback systems, the Partners for Change Outcome Management System and the Outcome Questionnaire System, found that two-thirds of the studies showed feedback-assisted therapy doing better than treatment as usual offered by the same practitioners. Effects ranged from small to moderate. The clearest benefit was for clients predicted to have a poor outcome, where feedback reduced deterioration.

A larger 2021 multilevel meta-analysis, covering 58 randomised and non-randomised studies and 21,699 patients, found a small effect on symptom reduction (d = 0.15) and a small favourable effect on dropout. The authors noted that results varied with the feedback instrument, how often it was used, treatment intensity and the country in which the study was conducted, and called for more consistent reporting of those details.

Read together, the two reviews suggest that measurement helps most when it is frequent, when clinicians actually look at the results, and when there is a plan for what to do with a client who is not on track. Measurement on its own changes little.

Choosing short, familiar measures

Brief, validated questionnaires make regular measurement practical. Two of the most widely used are the PHQ-9, for low mood, and the GAD-7, for anxiety.

In its original validation study of 6,000 patients in US primary care and obstetrics-gynaecology clinics, a PHQ-9 score of 10 or more had a sensitivity and specificity of 88% for major depression, and scores of 5, 10, 15 and 20 marked mild, moderate, moderately severe and severe levels (Kroenke and colleagues, 2001). The GAD-7 was validated with 2,740 adults in 15 US primary care clinics, with a cut point giving 89% sensitivity and 82% specificity for generalised anxiety disorder (Spitzer and colleagues, 2006).

Both are screening and severity measures. A score is a prompt for a conversation, not a diagnosis.

Work between sessions

Many therapies, cognitive behavioural therapy in particular, set tasks for clients to try between sessions. A 2016 meta-analysis of 17 CBT studies with 2,312 clients found that clients who completed more of their homework tended to have better outcomes at the end of treatment (g = 0.79 for quantity of homework compliance). The few studies that also measured the quality of the work, meaning whether the client had actually acquired the skill, found a similar relationship, though there were only three of them and the uncertainty was wide.

These are associations, not proof that homework causes improvement: clients who are already doing better may also find it easier to do the work. But they support the common clinical view that what happens between sessions matters, and that the quality of practice deserves as much attention as the quantity.

Practical cautions

  • Consent and data protection. Questionnaire answers and mood records are health data. In the EU this is special category data, and the Data Protection Commission notes that its processing is prohibited except in limited circumstances set out in Article 9 of the GDPR. Clients should understand exactly who sees what, and be able to change their mind.
  • Risk. If a measure includes a question about self-harm, decide in advance who sees a concerning answer, how quickly, and what happens outside working hours. Make sure clients know a digital tool is not an emergency service.
  • Burden. Short, regular measures are more sustainable than long, occasional ones. Ask only what you will use.
  • The relationship comes first. Scores are most useful when they are discussed with the client, not just recorded.

In practice

If you want to support clients between sessions, start with one or two brief, validated measures at a regular rhythm, agree with each client what will be shared and why, and build a clear routine for reviewing results and acting on concerning ones. Set between-session tasks collaboratively and check in on how they went, including whether the skill made sense, not just whether it was done.

Myndgard Share, which we build for university counselling services, is designed around this approach. The studies above concern feedback and homework in therapy generally; they are not evidence about Myndgard, and we make no claim that it changes outcomes.

If you're struggling: Samaritans 116 123, text 50808, or 112/999 in an emergency.

Sources