Whole-of-community interventions aim to reduce alcohol and other drug (AOD) use and harms by mobilizing community leaders, organizations and policy-makers to respond effectively to AOD use. The aim of this review is to estimate the effectiveness of whole-of-community interventions in reducing population-level harms arising from AOD use. A systematic review of electronic databases CENTRAL, Embase, Medline, Medline in Process and PsycINFO was conducted from database inception to August 2017. Eligible trials had a parallel comparison group, implemented interventions in two or more community settings, and reported data on AOD use or harms. Intervention settings included schools, sporting clubs, police and law enforcement agencies, community centres, local media and retail premises. Twenty-four trials from 63 publications were included (n=249 125 participants). Outcomes from AOD consumption (quantity and frequency), AOD-related crime and AOD-related accidents, injuries and hospital admissions. Data were pooled using random-effects inverse variance meta-analysis in Review Manager version 5.3. Risk of bias was mostly high, due to lack of random allocation, selective reporting and significant attrition. Meta-analyses indicated significant reductions in risky drinking [Alcohol Use Disorders Identification Scale (AUDIT) >8; three trials (7 data points), relative risk (RR)=0.78, 95% confidence interval (CI)=0.62-0.99)], but found no impact on past-month alcohol use (five trials, RR=0.95, 95% CI=0.89-1.02), binge drinking (five trials, RR=0.97, 95% CI=0.89-1.06) or 12-month marijuana use (two trials, RR=0.98, 95% CI=0.86-1.11). Narrative synthesis indicated some reductions in AOD-related assault rates and arrests, but were equivocal for quantity of alcohol consumed, 12-month illicit drug use, assault or abuse, motor vehicle accidents and hospital admissions. Interventions to reduce alcohol and other drug use and harms applied to whole communities have resulted so far in small reductions in risky alcohol consumption, but have had little impact upon past month alcohol use, binge drinking or 12-month marijuana use and the studies have been subject to high risk of bias.