Characterizing the Variation of Alcohol Cessation Pharmacotherapy in Primary Care

Alcohol use disorder (AUD) imposes a high mortality and economic burden. Effective treatment is available, though underutilized.

Describe trends in AUD pharmacotherapy, variation in prescribing, and associated patient factors.

Retrospective cohort using electronic health records from 2010 to 2019.

Primary care patients from 39 clinics in Ohio and Florida with diagnostic codes for alcohol dependence or abuse plus social history indicating alcohol use. PCPs in family or internal medicine with at least 20 AUD patients.

Pharmacotherapy for AUD (naltrexone, acamprosate, and disulfiram), abstinence from alcohol, patient demographics, and comorbidities. Generalized linear mixed models were used to identify patient factors associated with prescriptions and the association of pharmacotherapy with abstinence.

We identified 13,250 patients; average age was 54 years, 66.9% were male, 75.0% were White, and median household income was $51,776 per year. Over 10 years, the prescription rate rose from 4.4 to 5.6%. Patients who were Black (aOR 0.74; 95% CI 0.58, 0.94) and insured by Medicare versus commercial insurance (aOR 0.61; 95% CI 0.48, 0.78) were less likely to be treated. Higher median household income ($10,000 increment, aOR 1.06; 95% CI 1.03, 1.10) and Medicaid versus commercial insurance (aOR 1.52; 95% CI 1.24, 1.87) were associated with treatment. Receiving pharmacotherapy was associated with subsequent documented abstinence from alcohol (aOR 1.60; 95% CI 1.33, 1.92). We identified 236 PCPs. The average prescription rate was 3.6% (range 0 to 24%). The top decile prescribed to 14.6% of their patients. The bottom 4 deciles had no prescriptions. Family physicians had higher rates of pharmacotherapy than internists (OR 1.50; 95% CI 1.21, 1.85).

Medications for AUD are infrequently prescribed, but there is considerable variation among PCPs. Increasing the use of pharmacotherapy by non-prescribers may increase abstinence from alcohol.