A randomized trial to evaluate a management training program to prevent illegal alcohol sales.

Abstract

To evaluate effects of a training program for owners/managers of alcohol establishments-Alcohol Risk Management (ARM)-on: (i) propensity to sell alcohol to obviously intoxicated patrons; and (ii) changing establishment-level policies/practices. We assigned alcohol establishments randomly to intervention (full-ARM) and delayed-intervention/control (ARM Express) conditions. One large metropolitan area in Midwestern United States. Owners and managers at 231 on-premise alcohol establishments (i.e. bars, restaurants). Training programs consisted of one-to-one sessions with the owner/manager at each establishment. The goal of training was to help owners/managers to select and implement alcohol control policies in their establishments. The full-ARM training consisted of four one-to-one sessions and the ARM Express was a single session. We measured intervention effects through baseline and follow-up pseudo-intoxicated alcohol purchase attempts (i.e. feigning intoxication while attempting to purchase alcohol) and telephone surveys of owners/managers at alcohol establishments. Sales rates to pseudo-intoxicated patrons reduced 23% (relative to delayed-intervention/control condition) at the first follow-up purchase attempt (P = 0.06) but returned to baseline levels 3 months later. On average, establishments selected 13 of 18 recommended policies, but in multivariate models we observed no significant differences at follow-up in reported policies/practices across establishments. Reliance on manager training to promote responsible establishment alcohol policies is not sufficient to prevent illegal alcohol sales to obviously intoxicated patrons and to reduce alcohol-related problems.