Do alcohol control policies work? An umbrella review and quality assessment of systematic reviews of alcohol control interventions (2006 – 2017)

Abstract

Background The 2010 World Health Organization Global Strategy to Reduce the Harmful Use of Alcohol recommends countries adopt evidence-based interventions. Aim To update, summarize, and appraise the methodological rigour of systematic reviews of selected alcohol control interventions in the Strategy. Methods We searched for systematic reviews across PUBMED, EMBase and The Cochrane Library in 2016 and updated in 2017 with no language limits. Two investigators independently in duplicate conducted screening, eligibility, data extraction, and quality assessment using the ROBIS tool. We categorised interventions according to the WHO recommendations, and rated reviews as at high, low or unclear risk of bias. We applied a hierarchical approach to summarising review results. Where overlap existed we report results of high quality reviews and if none existed, by most recent date of publication. We integrated the ROBIS rating with the results to produce a benefit indication. Results We identified 42 systematic reviews from 5,282 records. Almost all eligible reviews were published in English, one in German and one in Portuguese. Most reviews identified only observational studies (74%; 31/42) with no studies from low or lower-middle income (LMIC) countries. Ten reviews were rated as low risk of bias. Methodological deficiencies included publication and language limits, no duplicate assessment, no assessment of study quality, and no integration of quality into result interpretation. We evaluated the following control measures as possibly beneficial: 1) community mobilization; 2) multi-component interventions in the drinking environment; 3) restricting alcohol advertising; 4) restricting on- and off-premise outlet density; 5) police patrols and ignition locks to reduce drink driving; and 6) increased price and taxation including minimum unit pricing. Conclusions Robust and well-reported research synthesis is deficient in the alcohol control field despite the availability of clear methodological guidance. The lack of primary and synthesis research arising from LMIC should be prioritised globally.