The association between involuntary alcohol treatment and subsequent emergency department visits and hospitalizations: a Bayesian analysis of treated patients and matched controls.

Abstract

Many nations have provisions for involuntary treatment of alcohol dependence where the person is at serious risk of harm to themselves. To date, there has been little thorough evaluation of its effectiveness. This study aimed to determine if there were differences between involuntary and voluntary treatment for alcohol dependence on subsequent emergency and hospital care. A retrospective cohort design using linked routinely collected administrative data on health-care utilization. Hospital and community-based alcohol treatment, New South Wales, Australia. A total of 231 patients who were involuntarily treated for alcohol dependence and 231 matched controls who received treatment as usual within the period May 2012 to April 2018. Involuntary treatment comprised a 28-day mandated hospital admission which included supervised withdrawal, comprehensive assessment, rehabilitation and support followed by voluntary aftercare support for up to 6months. Treatment as usual comprised three not mutually exclusive forms of intensive voluntary alcohol treatment: withdrawal management, rehabilitation and pharmacotherapies for alcohol dependence. Outcome measures: changes in the number of emergency department (ED) visits and number of unplanned hospital admissions 12 months before and 12 months after completion of index treatment. Both groups showed a reduction in ED visits (incidence rate ratio (IRR) = 0.56, 95% credible intervals (CrI) = 0.39-0.78) and unplanned hospital admissions (IRR = 0.49, 95% CrI = 0.37-0.65). There was no statistically significant difference between the two groups (IRR = 0.77, 95% CrI = 0.58-1.03 for ED visits and IRR = 0.79, 95% CrI = 0.62-1.01 for hospital admissions). The Bayes factors were 0.925 and 0.936 for ED visits and unplanned hospital admissions, respectively, interpreted as weak evidence in support of the null hypothesis of no difference between the interventions. Involuntary treatment of alcohol dependence was associated with reduced health service utilization in the year following treatment, and the outcomes did not differ from those of a matched control group.