Using Crowdsourcing to Test Treatments for Alcohol Use Disorder

Newswise — Recruiting participants to test treatments for alcohol use disorder (AUD) is time consuming and expensive, limiting the development of new interventions for the disorder.  Crowdsourcing is a sampling method that can effectively and efficiently recruit large numbers of participants who have different health histories. Researchers tested the feasibility and acceptability of delivering cognitive training interventions, a commonly used treatment approach for AUD, via crowdsourcing. They recruited 476 participants from the crowdsourcing website Amazon Mechanical Turk (mTurk) who completed a baseline survey.  Participants were then randomly assigned to receive an inhibitory control, working memory, or control training treatment intervention and asked to complete training tasks daily over a two-week period. Follow-up assessments evaluating the acceptability of the intervention and effects on alcohol and soda consumption were completed immediately following the training and then again two weeks later.

The researchers found that participants completed nearly two-thirds of the training tasks over the two-week intervention period and their performance on the tasks was consistent with the expected effects. Almost 95% of participants said that they were satisfied with the study procedures, more than 95% said that they would participate again, and more than 80% said that they would consider incorporating the training task in their daily life. They reported modest reductions in alcohol consumption, primarily in the inhibitory control group, and these reductions that did not extend to soda consumption.

The authors concluded that using crowdsourcing methods to develop interventions for AUD is both feasible and acceptable to potential study participants. These findings help to establish crowdsourcing as a method to recruit large samples of participants for future studies testing novel interventions for AUD and other substance use disorders.