Correction: Alcohol restrictions and suicide rates in South Africa during the COVID-19 pandemic: results of a natural experiment
Anthony Fish Hodgson, Jason Bantjes, Jane Pirkis, et al. Alcohol restrictions and suicide rates in South Africa during the COVID-19 pandemic: results of a natural experiment. BMJ Glob Health 2025;10:e017171.
Some lockdown level 3 observations were miscategorised in relation to alcohol restrictions, resulting in two inaccurate estimates in tables 3 and 4. These estimates are corrected in the relevant tables, as well as within the text of the paper. These corrections do not impact the study’s overall findings and conclusions, but do affect two effect sizes. A few additional corrections are also included to rectify typographical errors and improve clarity.
The final sentence of the “Findings” section has been changed from,
“However, the shift from partial to full restriction on the sale of alcohol was associated with an 18% (95% CI 10% to 25%) reduction in suicides for both sexes combined and a 22% (95% CI 13% to 30%) reduction in suicides among men, but no significant reduction among women.” to
“However, the shift from partial to full restriction on the sale of alcohol was associated with an 23% (95% CI 15% to 30%) reduction in suicides for both sexes combined and a 27% (95% CI 19% to 35%) reduction in suicides among men, but no significant reduction among women.”
In paragraph two of the section titled “Data Analysis”, "To assess the association between various lockdown levels and alcohol restrictions, we compared the dates of the lockdown periods in 2020/2021 with the corresponding periods in 2017 (table 3)." has been changed to: "To assess the impact of the alcohol bans, we compared suicide rates across the various lockdown periods and alcohol restrictions , stratified by sex (table 3)."
To:
To:
In paragraph one, “the shift from partial to full restrictions coincided with an 18% reduction in risk of suicide, equivalent to a change in crude suicide mortality rates from 12·0 per 100 000 to 9.90 per 100 000.” has been changed to “the shift from partial to full restrictions coincided with a 23% reduction in risk of suicide, equivalent to a change in crude suicide mortality rates from 11.81 per 100 000 to 9.26 per 100 000.”
References 25 and 47 are both replaced with the final published version of the paper:
Barron K, Parry C, Bradshaw D, Dorrington R, Groenewald P, Laubscher R, Matzopoulos R. 2022. Alcohol, Violence and Injury-induced Mortality: Evidence from a Modern-day Prohibition. Review of Economics and Statistics. https://doi.org/10.1162/rest_a_01228
Reference 27 is replaced by:
Roomaney RA, Mhlongo S, Dekel B, et al. The 3rd Injury Mortality Survey: A national study of injury mortality levels and causes in South Africa in 2020/21. Cape Town: South African Medical Research Council, 2023. https://www.samrc.ac.za/sites/default/files/attachments/2023-11/IMS3Report.pdf
We are grateful to the National and Provincial Departments of Health and the various Forensic Pathology Services for providing access to the mortuaries and to the SAMRC scientists in the Burden of Disease Research, Gender and Health Research for making these data available. We acknowledge the contribution of co-authors of the 2020/1 Injury Mortality Survey, as cited within this manuscript, and the fieldwork and management team from Geospace International (Pty) Ltd for data collection.
