The independent and joint risks of alcohol consumption, smoking, and excess weight on morbidity and mortality: a systematic review and meta-analysis exploring synergistic associations

Citation

Burton R, Fryers PT, Sharpe C, Clarke Z, Henn C, Hydes T, Marsden J, Pearce-Smith N & Sheron N (2024) The independent and joint risks of alcohol consumption, smoking, and excess weight on morbidity and mortality: a systematic review and meta-analysis exploring synergistic associations. Public Health, 226, pp. 39-52. https://doi.org/10.1016/j.puhe.2023.10.035

Abstract
Objective Alcohol consumption, smoking, and excess weight independently increase the risk of morbidity/mortality. Less is known about how they interact. This research aims to quantify the independent and joint associations of these exposures across health outcomes and identify whether these associations are synergistic. Study design The protocol for this systematic review and meta-analysis was pre-registered (PROSPERO CRD42021231443). Methods Medline and Embase were searched between 1 January 2010 and 9 February 2022. Eligible peer-reviewed observational studies had to include adult participants from Organisation for Co-Operation and Development countries and report independent and joint associations between at least two eligible exposures (alcohol, smoking, and excess weight) and an ICD-10 outcome (or equivalent). For all estimates, we calculated the synergy index (SI) to identify whether joint associations were synergistic. Meta-analyses were conducted for outcomes with sufficiently homogenous data. Results The search returned 26,290 studies, of which 98 were included. Based on 138,130 participants, the combined effect (SI) of alcohol and smoking on head and neck cancer death/disease was 3.78 times greater than the additive effect of each exposure (95% confidence interval [CI] = 2.61, 5.48). Based on 2,603,939 participants, the combined effect of alcohol and excess weight on liver disease/death was 1.55 times greater than the additive effect of each exposure (95% CI = 1.33, 1.82). Conclusion Synergistic associations suggest the true population-level risk may be underestimated. In the absence of bias, individuals with multiple risks would experience a greater absolute risk reduction from an intervention that targets a single exposure than individuals with a single risk.

Keywords
Alcohol; Smoking; Excess weight; Synergistic interaction; Meta-analysis

Journal
Public Health: Volume 226