- Design
- Randomised non-inferiority trial
- Population
- 172 adult primary care patients with mild to moderate stress-related mental ill health
- Primary outcome
- Perceived Stress Scale-10 over 12 weeks
- Effect
- d = 0.22 favouring guided CBT; CI upper bound 0.52 exceeded the 0.3 margin
A Swedish non-inferiority trial in BMJ Mental Health (7 October) randomised 172 adult primary care patients with mild to moderate stress-related mental ill health to 12 weeks of nurse-led collaborative care or psychologist-guided self-help CBT. The primary outcome was the 10-item Perceived Stress Scale.
Both groups improved substantially on every measure. The between-group difference favoured guided CBT (d = 0.22), and the upper bound of its confidence interval (0.52) crossed the prespecified margin of 0.3, so collaborative care was not shown to be non-inferior. Guided CBT also reduced exhaustion slightly more (d = 0.28). No other secondary outcome differed.
Where a psychologist is available, guided CBT remains the stronger option for stress-related problems. Where one is not, a structured nurse-led approach still produced large within-group improvement, which makes it a reasonable fallback rather than a reason to offer nothing.
- Offer guided CBT first for stress-related mental ill health when a psychologist-supported programme is available.
- Where psychology access is limited, consider structured nurse-led collaborative care rather than watchful waiting alone.
- Measure perceived stress and exhaustion at baseline and during treatment to track response.
- Screen for depression and anxiety, which often accompany stress-related presentations in primary care.
Why it matters
Many services are tempted to substitute nurse-led care for psychology; this trial says the gap is small but present.
Don't overread it
Without a no-treatment arm, the large improvement in both groups may partly reflect natural recovery.
The statistics, in plain English
Non-inferiority means proving a new option is not worse than the standard by more than a set margin. Here the margin was an effect size of 0.3 and the worst plausible difference was 0.52, so the trial could not rule out a meaningful disadvantage, even though the best estimate (0.22) was small.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for psychiatry, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free