Education may help people stop certain sleep medicines, but evidence is uncertain
For adults using benzodiazepines or related sleep medicines, patient education may increase discontinuation compared with usual care. Evidence for stopping medication was low certainty, and better health outcomes were not established.
Based on the published abstract. The full paper may contain additional methods, results and limitations.
The 30-second takeaway
Patient education, medication review and pharmacist-led education may help more adults discontinue these sleep medicines compared with usual care, but the evidence is low certainty. Patient education probably makes little or no difference to physical function, mental health or insomnia symptoms. The review found no compelling evidence supporting discontinuation benefits from the other intervention categories.
Evidence for discontinuation was low certainty, so the estimated benefit could change with further research. The abstract does not report when discontinuation was assessed.
What researchers found
Stopping benzodiazepines or related sleep medicines
Versus usual care: 144 more discontinuations per 1000 patients, with a 95% confidence interval of 61 to 246 more. Evidence was low certainty; assessment timing was not reported.
Stopping medicine and health outcomes
The central outcome was whether patients stopped medication. For patient education, separate evidence of moderate certainty suggested little or no effect on physical function, mental health or insomnia symptoms. More discontinuation therefore should not be presented as better sleep.
Missing accompanying outcomes
For medication reviews and pharmacist-led education, the review found no evidence on those additional health outcomes. This leaves it unclear whether increased discontinuation accompanies improvements in sleep or functioning, or whether those outcomes change at all.
The original publication
Comparative effectiveness of interventions to facilitate deprescription of benzodiazepines and other sedative hypnotics: systematic review and meta-analysis.
Zeraatkar D, Nagraj SK, Ling M et al.
BMJ · 2025
- PubMed ID
- 40527546
- Record checked
AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.
One more question, understood.
Keep track of the research you’ve explored.