A community-wide education bundle cut Watch-group antibiotic use by two thirds

Design

Cluster-randomised controlled trial, 44 communities

Population

Community antibiotic providers and patients in Nanoro, Burkina Faso, and Kimpese, DR Congo

Primary outcome

Change in prevalence of Watch-group antibiotic use

Effect

Adjusted prevalence ratio 0.33 (95% CI 0.14–0.78)

CABU-EICO, published in June, randomised 44 villages and neighbourhoods in Burkina Faso and DR Congo to an intervention or control. Over nine months, the intervention combined community health education with training and feedback for every local antibiotic provider — health centres, private clinics, pharmacies and informal vendors — using the WHO AWaRe antibiotic guidance.

In the intervention communities, the prevalence of Watch-group antibiotic use fell from 26.8% to 17.1%, while it rose from 13.4% to 21.2% in control communities; the adjusted prevalence ratio was 0.33 (95% CI 0.14–0.78). Patient management, assessed by simulated patient visits, did not worsen.

The lesson for prescribers and stewardship teams is that most community antibiotic use happens outside clinics, and an intervention that reaches pharmacies and vendors as well as doctors can shift it. Similar mixed public-private-informal antibiotic supply exists in much of India.

• Use AWaRe categories in teaching: prefer Access antibiotics for common infections and reserve Watch drugs.
• Include pharmacies and informal sellers in stewardship efforts; they dispense much of community antibiotic use.
• Combine community education with provider feedback rather than relying on either alone.
• Reducing broad-spectrum use did not harm patient management in this trial.
• Measure Watch-group use locally to know whether stewardship is working.

Why it matters

Resistance is driven largely by community antibiotic use, which hospital stewardship never touches.

Don't overread it

The confidence intervals around each community's prevalence are wide, and the trial measured antibiotic use, not resistance.

The statistics, in plain English

An adjusted prevalence ratio of 0.33 means Watch-group use was about a third as common in intervention communities as expected compared with control; the interval 0.14–0.78 is wide but stays below 1.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.