We conducted a preassessment and postassessment (no comparison group) of employers' implementation of evidence-based WHP best practices and their attitudes toward WHP after they received the HealthLinks intervention. We conducted the study from January 2009 through September 2009 in Washington State. The University of Washington Institutional Review Board exempted the study from further review after receiving a summary of the procedures and a copy of measures.

We targeted Mason County, a largely rural community that has elevated health risk behaviors. The county reports a 29% obesity rate (2% higher than the state) and 29% current smoking rate (14% higher than the state) (12). The average income of residents in the county is 20% less than state levels, $45,417 versus $56,317 (13).

We recruited 23 small workplaces (defined as a workplace with fewer than 250 workers) in Mason County, Washington. We identified workplaces that met the inclusion criteria by using several approaches: 1) accessing a public database of businesses in the region (14); 2) obtaining referrals from the Washington State Department of Health; and 3) identifying workplaces that had a prior relationship with ACS-GWD through participation in fundraising events or other activities.

HealthLinks consists of 5 steps: 1) recruitment of workplaces, 2) assessment of baseline implementation of best practices, 3) recommendation of best practices, 4) implementation of recommended best practices, and 5) assessment of employer's implementation of best practices at 6-months postintervention and their attitude about WHP. To recruit workplaces (step 1), an ACS-GWD interventionist telephoned the upper-level manager at each workplace and briefed the manager on HealthLinks. If the manager showed interest in participating in HealthLinks, the interventionist described the program in more depth. Information offered included 1) an outline of the relationships among missed work days, work productivity, and lost revenue, and 2) an outline of the relationship between WHP and return on investment. To assess baseline implementation of best practices (step 2), an hour-long, in-person assessment was conducted with the workplace manager. The ACS interventionist determined which among the WHP best practices (policy, program, and communication) were present or absent at the workplace (the best-practice instrument is in Appendix A). Two weeks after the assessment, the interventionist prepared a tailored recommendation report (step 3) that listed each best practice and noted whether the practice was fully or partially implemented. The interventionist then recommended actions to fully implement the best practices (the list of resources and services that could be recommended are in Appendix B). The ACS interventionist presented the recommendation report to the manager, who selected 2 to 4 recommended actions to implement. Working with the manager, the interventionist implemented the selected recommendations (step 4). For example, employers interested in a physical activity program might select the 10-week Active for Life (AFL) program (15,16). The interventionist and personnel from the Mason County Department of Health offered resources and training sessions for the workplace contact responsible for overseeing the AFL program. They also assisted in enrolling participants and tracking physical activity goals by using an electronic tracking system. Participants received incentives when they achieved their physical activity goals and completed AFL. Incentives included boxed lunches and gift cards to local grocery stores (eg, Top Foods, Trader Joe's, Subway). Six months after the recommendation report, HPRC staff re-administered the assessment instrument to evaluate changes in best-practice implementation from baseline to follow-up. We also assessed employers' attitudes about WHP.

We assessed employers' implementation of evidence-based best practices before and 6 months after HealthLinks by using the Employer Practices Survey, a 50-item instrument consisting of closed-ended, nonscaled questions. The survey included 12 questions on tobacco use, healthy eating, and physical activity policies; 7 items on physical activity and tobacco use cessation programs; and 19 items on communication of health information. Primary outcomes included percentage of implementation of policy, program, and communication best practices and overall best practices.

We assessed employers' perception of and satisfaction with HealthLinks at 6 months after the program by using an 8-item Employer Attitude and Satisfaction Survey comprising open- and closed-ended questions. The outcomes were 1) perceived barriers, 2) HealthLinks components employers liked most, 3) HealthLinks components most likely to affect future wellness activities, and 4) HealthLinks communication materials that were most helpful.

We scored most of the Employer Practices Survey questions dichotomously, using a score of 1 for the practices in place and a score of 0 for those not in place. We evaluated employers' implementation of tobacco policy by using 3 values, 0, 0.75, and 1. We assigned a score of 0 if the employer had no tobacco policy. We assigned a score of 1 if the employer had a complete tobacco ban policy (eg, tobacco use was not allowed anywhere on workplace grounds or in vehicles). We assigned a score of 0.75 if the employer did not allow using tobacco in the building(s). We assigned a score of 0.75 (rather than 0.50) because, by forbidding smoking indoors, employers were restricting most workers' tobacco use for most of their working hours. For each best practice, we created a summary score by summing the values, dividing by the number of possible points, and reporting the result as a percent. We calculated an overall best-practice score for each employer by summing each best practice score and taking the mean.

We used Wilcoxon matched pairs tests to analyze significant differences in best-practice implementation from baseline to follow-up. We analyzed the data by using SPSS 14.0 for Windows (SPSS, Inc, Chicago, Illinois), and we calculated all reported significant differences at the 95% confidence level.

To assess employers' attitudes and perceptions, we calculated frequency counts for employers' responses to closed-ended questions on the Employer Attitude and Satisfaction Survey. For responses to open-ended questions, we looked for responses with similar themes and reported the most common themes.

We contacted 69 eligible workplaces in Mason County, Washington, and intervened with 23 (33% participation rate); the workplaces had an average of 42 workers. Most (n = 20) workplaces had 200 workers or fewer. The top 5 industries were tribal centers, lumber and forestry, financial institutions, academic institutions, and public service agencies.

Several factors affected workplaces' capacity to participate in HealthLinks. More than half (n = 14) of participating workplaces had a previous relationship with ACS. Two factors most likely to influence workplaces' decision to participate in HealthLinks were upper management support (n = 8) and concern about the health needs of workers (n = 7). The HealthLinks characteristics that drove employers' participation included the reputation of ACS (n = 8) and the fact that HealthLinks was easy to implement, broad in scope, and free (n = 8).

workplaces' capacity to participate in HealthLinks depended on resources received. Our intervention tracking system documented which resources and educational presentations we delivered to the workplaces (Table 1). The 3 resources that we delivered to more than half of workplaces were access to fightcancer.org, e-newsletters, and the Quit Line promotional posters. In addition, 10 workplaces and 173 workers from these sites participated in AFL; the Lunch and Learn topic that workplaces most frequently requested was physical activity (n = 9), followed by healthy eating and stress management.

Overall, implementation of best practices increased significantly for all 3 practice types — policy, program, and communication (Table 2). On average, workplaces implemented 36% of the best practices at baseline and 59% at follow-up (P < .001).

At follow-up, 21 employers reported high satisfaction with HealthLinks (Table 3). The most popular HealthLinks components were the e-newsletter and the Lunch and Learn presentations. The most helpful communication material was the e-newsletter. Employers favorably rated the Lunch and Learn topics for ease of promotion, relevance, and appeal. Employers rated sessions on physical activity, healthy eating, and stress management as easiest to promote; ratings for relevance and appeal were similar (data not shown). The HealthLinks component most likely to influence future WHP decisions was the Employer Practices Survey assessment of best practices.

Of the 23 participating workplaces, 12 identified at least 1 barrier to HealthLinks implementation. The most common barrier was workers' not having the time to participate (n = 7) (Table 3).