Abstract
BACKGROUND AND AIMS: The global growth of mobile phone use has led to new opportunities for health interventions, including through text messaging. We aimed to estimate the effects of text messaging interventions on alcohol consumption among risky drinkers. METHODS: Systematic review and meta-analysis of reports on randomized controlled trials (RCTs) published in English. Searches were conducted on 23 May 2019 in PubMed; PubMed Central; CENTRAL; CDSR; DARE; NHS-EED; Scopus; PsycINFO; PsycARTICLES; CINAHL; and Web of Science. Measurements included number of episodes of heavy drinking (HED) per month and weekly alcohol consumption (WAC) in grams. Trials among risky drinkers who were not receiving co-interventions were included in the review (n = 3481, mean age 29 years, 41% female). Data were extracted from reports and authors were contacted for additional data. RESULTS: Ten trials were included and all analyses were based on random-effects models. Primary analyses, including seven trials (n = 2528) for HED and five trials (n = 2236) for WAC, found that the interventions may reduce self-reported HED [-0.33 episodes per month; 95% confidence interval (CI) = -0.79, 0.12] and WAC (-18.62 g per week; 95% CI = -39.61, 2.38), although both estimates included the null. The Grading of Recommendations, Assessment, Development and Evaluations (GRADE) quality of evidence was judged to be low for both HED and WAC, primarily due to risk of attrition and performance bias, heterogeneity and influence of pilot trials on estimates. CONCLUSIONS: Text messaging alcohol interventions may reduce alcohol consumption compared with no or basic health information; however, there are doubts about the overall quality of the evidence.