Linking inpatients with alcohol use disorder to follow-up: what has worked, and how little is known

Design

Systematic review of trials and cohort studies (GRADE)

Population

17 studies of 21 hospital-based interventions for inpatients with alcohol use disorder

Primary outcome

Linkage to outpatient follow-up after discharge

Effect

8/21 interventions improved linkage; RCT effects −8% to +30% absolute

Admissions driven by alcohol use disorder are common, and what happens after discharge largely decides whether the next one follows. This systematic review, published in Annals of Internal Medicine on 8 September, looked for hospital-based interventions that improved outpatient follow-up. It included 17 studies of 21 interventions: 12 randomised trials (1,917 patients), 3 non-randomised trials and 2 cohort studies.

Eight of the 21 interventions (38%) improved linkage to outpatient care. Those that worked generally combined behavioural support and care coordination with either alcohol use disorder medication or social support. Across the randomised trials, effects ranged from an 8-point absolute fall to a 30-point absolute rise in follow-up. Certainty was low, because of risk of bias and imprecision.

So the evidence does not name a single model to copy. What it does suggest is that a referral letter alone is not enough, and that the elements most often present in successful programmes — a person who arranges the appointment, and medication started before discharge — are cheap to try on any medical ward.

• Document alcohol use with a structured tool such as AUDIT-C on admission, not only when withdrawal appears.
• Discuss naltrexone or acamprosate before discharge where there is no contraindication; starting in hospital avoids a missed outpatient step.
• Book the follow-up appointment before the patient leaves, rather than asking them to arrange it.
• Name a contact person — nurse, counsellor or social worker — in the discharge summary.
• Check liver function before naltrexone and renal function before acamprosate.

Why it matters

Discharge is where most alcohol-related admissions lose contact with care, and the ward team controls that moment.

Don't overread it

Low-certainty evidence: no single intervention model has been shown to work reliably.

The statistics, in plain English

An effect range from −8 to +30 percentage points means some interventions did nothing or worse, and some helped a lot; the studies were too different to pool into one average. Low certainty means the true effects could differ substantially.

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.