Factors influencing the effectiveness of server-intervention education.

The purpose of the study was to assess the effect of server-intervention education and the effect of various situational variables upon program effectiveness. A 6-hour server education program was administered to 876 servers and 203 managers of 100 licensed establishments in eight states. Participants in the program were administered knowledge, attitude and behavior self-report measures prior to and following the program. Each establishment was also visited three to four times, before and after the program, by staff observers exhibiting visible signs of intoxication. Equal numbers of visits were also made to a comparison group of 138 establishments in the same communities as the establishments participating in the program. Significant improvement in knowledge, attitude and self-reported behavior was found among program participants. A significant overall increase in observed intervention was found in visits to participating establishments and no change was found within the comparison group. However, there was a significant intervention-by-state interaction with positive changes occurring in only five of the eight states. Differences among states in observed intervention were not related to dram shop law, but did appear to be related to prior level of intervention, type of establishment and business volume. Unfortunately, even after training, intervention was observed in only 20% of the 1,590 visits, and only 7% of the visits resulted in terminating service. The lack of incentives for discontinuing service to intoxicated patrons is discussed.