The potential effects of an extended alcohol withdrawal treatment programme on morbidity and mortality among inpatients in the German city of Bremen: a simulation study

Abstract Background According to the German guidelines, people with severe alcohol use disorders (AUDs) should receive withdrawal treatment. Compared to somatic withdrawal treatment (SWT), extended duration and psychosocial elements of so-called “qualified withdrawal treatment” (QWT) aim to reduce relapse rates. Despite promising results of prospective studies on QWT, only few German inpatients seeking withdrawal treatment receive QWT. We estimated the potential effects on mortality and morbidity for higher proportions of treatment-seeking patients receiving QWT rather than SWT in the German city of Bremen. Methods In 2016 and 2017, 2051 inpatients were admitted to two specialised hospitals for withdrawal treatment. The potential beneficial effects of QWT over SWT were estimated by simulating treatment outcomes taken from two prospective studies. Outcomes comprised number and length of all-cause hospitalisations within 5 years, as well as abstinence and all-cause mortality rates within 28 months. Outcomes were estimated for actual and increased rates of QWT (25, 50%) among inpatients seeking alcohol treatment. Results In the selected hospitals, 170 patients (8%) received QWT. If 25% of AUD inpatients were to receive QWT, benefits in abstinence rates (+ 18%), the total number of hospitalisations (− 9%) and hospital days (− 10%) could be expected. If 50% of AUD inpatients were to receive QWT, benefits in abstinence rates (+ 45%), the total number of hospitalisations (− 23%) and hospital days (− 26%) were more pronounced, in addition to reductions in mortality (− 20%). Conclusion Increasing the proportion of people with severe AUD enrolled in extended withdrawal treatment programs (such as QWT) may contribute to reduce overall alcohol-attributable burden of disease. Randomised controlled trials or other prospective studies controlling for confounding factors are needed to determine the potential at the population level.