My name is South Africa and I have a drinking a problem: a multicentre quasi-experimental analysis on alcohol regulation and injury presentations to emergency centres.

Thesis / Dissertation


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
Background The South African COVID-19 experience included several national regulatory changes to manage the additional demand placed on the healthcare platform. Since alcohol-related injury contributes a significant acute healthcare burden in South Africa, regulations to limit alcohol availability were also instituted. In this study we aimed to determine the impact of changing alcohol regulations (full availability, partial availability, and a complete ban on alcohol sales), across three time periods, on injury presentations to emergency centres in the Western Cape, South Africa using a quasi-experimental interrupted time series (ITS) design. Methods The study population included all patients who presented to a public sector emergency centre in the Western Cape with injury identified by a trained nurse on triage and identified by an emergency centre clinician by final ICD-10 code in facilities using routine real time electronic capture of emergency centre visits. Since the study design was a quasi-experimental ITS, we used an autoregressive integrated moving average (ARIMA) model with the level and slope of the model in the pre-intervention period being the counterfactual against the observed actual post-intervention level and slope. The primary outcome was the relative percent increase or decrease in the level and slope of injury presentations. Findings A total of 31,151 injury patients across the three periods were included in the analyses. A shift from full availability to partial (Monday to Thursday) retail alcohol availability resulted in an overall step reduction in daily injury presentations of 29·0% (absolute reduction [95% CI]: -71·7 [-102·9, -40·4]). A shift from partial to a complete ban resulted in a further step reduction of 26·2% in daily injury presentations (absolute reduction [95% CI]: -4·5 [20·6, -8·4]). This impact was consistent in terms of direction but ranged in magnitude across various sub-populations. Interpretation Our findings reflect the considerable impact of alcohol regulation on injury presentations to emergency centres. Partial retail alcohol restrictions, particularly in the South African context, could be considered for longer term, sustainable alcohol regulation policies to reduce the considerable burden of injuries on health services and society.