The Responsible Retailing Forum ("RRF") seeks to develop a new intervention, Stop Service to Obviously- Impaired Patrons ("S-STOP"), to reduce the incidence and harm associated with overservice of alcohol. Modeled after RRF's effective program to reduce alcohol sales to minors using Mystery Shopper feedback on staff ID-checking conduct, SSTOP would (1) conduct "Pseudo-Intoxicated" Mystery Shop" ("P-I/MS") inspections of serving establishments, employing actors who seek to purchase an alcohol beverage while showing obvious signs of intoxication, (2) provide licensees with confidential feedback on actual staff conduct and a video link to view the behavior of the P-I/MS that visited their establishments, (3) provide staff with brief online training in the recognition and skillful refusal of service to intoxicated patrons, and (4) provide communities with a measure of the prevalence of overservice. The proposed study will: (1) determine the effectiveness of S-STOP in improving recognition and refusal to serve an obviously- impaired customer. To do this, we will implement S-STOP in 10 pairs of demographically matched college and university communities, employing a cross-over design. After a 3-month baseline, we will implement S- STOP in one community in each pair (Cohort 1), while the second community serves as a control (Cohort 2). After 6 months, we will end S-STOP in Cohort 1 communities but continue inspections to measure the effects of decay; and we will begin S-STOP in Cohort 2. (2) examine how licensees utilize the S-STOP program and the extent to which utilization moderates the effectiveness of the program. To do this, we will measure the number and percentage of managers who visit the S-STOP website and register their staff for training, measure the number of staff that complete the training, and conduct analyses to investigate the dose-response relationship between utilization of the S-STOP program and likelihood of overservice. (3) investigate why some owner/managers did not participate in S-STOP. To achieve this, we will interview 20 owner-managers who did not access the S-STOP website.

The Responsible Retailing Forum (RRF) has developed a community-based program - S-STOP (Stop Service To Obviously-Impaired Patrons) - that assists communities and their on-premises alcohol establishments to address the problems of over-service of alcoholic beverages. S-STOP is a six-month program, administered by RRF, which:

• Conducts Pseudo-Intoxicated Mystery Shop (P-I/MS) visits at on-premises establishments employing actors who seek to be served an alcoholic beverage while showing obvious signs of intoxication;
• Provides on-premises owner-operators and managers with confidential feedback on staff conduct and an Internet link to a video that shows how the P-I/MS that visited their establishments typically would have behaved;
• Provides resources to licensees to assist them in complying with over-service laws;
• Provides on-premises staff with a brief online training module on how to recognize and skillfully refuse service to potentially intoxicated patrons;
• Provides community stakeholders with aggregated data on the prevalence of alcohol over-service in their communities.

In this project, we will conduct and evaluate S-STOP in selected college and university communities in California.

Specific Aim 1 of this project is to demonstrate the effectiveness of S-STOP in reducing over-service. To achieve this, RRF will conduct a randomized control trial using a cross-over design. We will:

• Identify ten pairs of demographically matched college/university communities for a total of 20 communities and, in each community, recruit a minimum of 12 on-premises serving establishments for participation in the project;
• Conduct 3 P-I/MS baseline inspections in all 20 communities to observe pre-intervention over-service conduct;
• Randomly assign one community from each pair to receive S-STOP immediately (Cohort 1) and the other (Cohort 2) to serve as a control group and then to receive S-STOP later;
• Implement S-STOP in Cohort 1 communities for 3 months and conduct P-I MS in Cohort 2 control communities with no feedback to the licensees or community. At the end of the 3 months, we will end S-STOP in Cohort 1 communities, but continue unreported P-I/MS and begin S-STOP in Cohort 2 communities for 3 months;

Specific Aim 2 is to examine how licensees utilize the S-STOP program within their establishments and the extent to which utilization moderates the effectiveness of the program. To achieve this aim, we will:

• Measure the number and percentage of owner-operators and managers who visit the S-STOP website to view the typical behavior of P-I/MS inspectors and register their staff for training;
• Measure the number of staff who complete the S-STOP online training;
• Investigate the dose-response relationship between utilization of the S-STOP program (number of owner-manager visits to S-STOP website; percentage of staff completing S-STOP training) and likelihood of over-service as assessed by the P-I/MS.

Specific Aim 3 is to investigate why some owner/managers did not participate in S-STOP in order to improve the program. We will:

• Interview 20 owner-managers who did not access the S-STOP website to find out why and determine how the program might be modified to encourage their use of it.

The S-STOP program has three key components:

• Pseudo-Intoxicated Mystery Shops (P-I/MS). S-STOP conducts P-I/MS inspections by actors who seek to be served an alcoholic beverage while showing obvious signs of impairment. In order to appear realistic, the P-I/MSs act out levels of intoxication that staff of bars, clubs, and restaurants are likely to encounter in the course of their jobs. Owner-operators and managers are notified by mail that the S-STOP inspections recently occurred and are provided with the results (refused service/did not refuse service) of inspections in their own establishment. In addition, the mailings include materials describing best practices for on-premises serving establishments, providing guidance on what steps staff should take if a customer is obviously impaired and describing the potential legal consequences for licensees and staff who serve alcohol to an obviously-impaired customer. The mailings also provide a link to an internet site with a video showing the typical behavior of actors during the P-I/MS visits and for registration of staff for online training.
• Online refresher training in over-service recognition and refusal. S-STOP provides a brief online training module for staff on how to recognize and skillfully refuse service to potentially intoxicated patrons. The online training presents staff with scenarios involving customers who have been drinking and models several ways in which staff can firmly, but respectfully decline to serve. The video also reinforces the importance of management supervision and reinforcement of responsible beverage service practices. The online training lasts under 4 minutes and can be accessed using a mobile device, tablet, or computer. Owners/managers will be notified when their staff members complete the training.
• Community engagement. S-STOP also reports aggregated community-level P-I/MS results to licensees, their industry partners, and public health, safety, and licensing agencies to provide these community stakeholders with a measure of over-service prevalence and changes in refusal rates in their communities over time.

Inclusion Criteria:

• Must be server/bartender in sampled on-premises alcohol serving establishment

Exclusion Criteria:

• Establishment must have at least 3 servers/bartenders on staff.