Obesity, a major risk factor for diabetes, affects more than one-third of adults in the United States and is associated with several demographic and socioeconomic factors, including low income (1). Several studies have found that obesity rates are generally higher among working class occupations than professional occupations, even after controlling for demographic factors (2,3).

From a sociological perspective, the environments in which people live and work are strong influences on obesity and diabetes (4,5). The work environment is especially important because many adults spend a significant amount of time at work and because obesity affects employers through reduced productivity and absenteeism as well as increased health care costs and disability (6). Numerous studies acknowledge the negative health consequences of workplace factors such as stress, low autonomy, poor coworker and managerial support, and unhealthy physical work environments (2,7). These workplace risk factors may be more common in low-wage and working-class jobs and may explain some occupational differences in obesity prevalence (2,8).

Promoting health through worksite wellness programs is a national priority. The Affordable Care Act creates new incentives to promote employer wellness programs and encourage opportunities to support healthier workplaces (9). The National Institutes of Health and the Centers for Disease Control and Prevention have targeted worksites as a priority location for health interventions because they offer an efficient means of delivering and evaluating programs and provide opportunities to reach socially disadvantaged populations (10,11). However, data for the effectiveness of workplace health programs are limited and may not be generalizable to all types of workers (6,11–13). National data show that blue-collar and service workers are less likely to work for an employer who offers health promotion activities and are less likely to participate in such programs when offered (14).

This study focused on a little-studied health disparity — workplace health promotion among low-wage workers. The objective of the study was to examine through interviews and focus groups 1) worksite culture, environment, and policies that influence healthy eating and physical activity; and 2) barriers that reduce worker participation in workplace health promotion programs. An understanding of how the workplace affects health behaviors is can inform design of effective interventions to reduce and prevent obesity.

We partnered with a large health care system and a national labor union representing retail workers to recruit study participants. Qualitative data collection included interviews with key informants (eg, employer representatives, union leaders, benefits administrators) and worker focus groups with both partner organizations. The workforce in the union was relatively homogenous with regard to income and included workers in jobs such as cashier and merchandise stocker. Within the health care system, we targeted hospital work departments and locations that employed a large proportion of low-wage workers, including housekeepers, patient care technicians, and food service workers. This study was approved by the Washington University institutional review board.

We interviewed 10 individuals from the union partner: 4 local union leaders, 5 store representatives, and 1 health benefits administrator. Key informants were recruited in person or through email, and interviews were conducted in person or over the telephone. We asked about current and previous wellness initiatives offered to employees, employee participation in these initiatives, and potential barriers to participation. Informants were also asked about workplace factors that influenced health behaviors (ie, physical activity and healthy eating) and employee attitudes about health and wellness.

We conducted a total of 9 focus groups involving 61 workers. Twenty hospital employees (4 men and 16 women) participated in 4 groups. Forty-one unionized retail workers including 12 men and 29 women participated in 5 focus groups. Focus group participants were recruited through their work department, store, or local union hall. The research team attended union meetings to recruit members in person and posted flyers in break rooms at selected stores and hospital departments. We used a semistructured script to guide focus group discussions. The scripts covered 11 broad domains with follow-up questions and prompts for each domain (Table 1). All group discussions were audio recorded and transcribed. Transcriptions were entered into QSR International’s NVivo 10 software (QSR International Pty Ltd), and all were coded by 2 independent raters using a predefined code book based on the domains in the focus group script. After initial coding and consensus of all transcripts, we applied a phenomenological approach for data analysis to find the “essence” or common themes across individual experiences (15). The purpose of the thematic analysis was to answer 2 questions: “what impacts healthy eating and physical activity” and “what can be modified at the workplace?” Through systematic review and discussion, codes were merged and grouped under main themes. Each transcript was re-read and re-coded for consistency.

This study highlights factors related to obesity as described by 2 low-wage work groups; our findings are consistent with results from a similar study among low-wage workers in various industries (8). The workplace was often viewed as a barrier to healthy eating and physical activity; however, workers supported the concept of workplace health promotion and offered suggestions for overcoming many of the identified barriers. As demonstrated in this study, the workplace may be effective in engaging populations at risk for obesity and related illnesses, though it may be necessary to go beyond traditional workplace wellness approaches. Using more innovative methods may increase program reach, effectiveness, and sustainability.

Policy changes have increasingly been recognized as essential components of worksite health promotion (16) and are more sustainable than individual-level behavior interventions (17). Policies promoting a culture and environment conducive to reducing obesity can be a strong catalyst to behavior change. These can include top-level policies, such as offering a health care plan that has wellness options or implementing organizational policies that provide for access to low-cost healthy foods at the worksite, encourage active transportation to and from work, or allow for flexible work schedules to encourage lunch or break-time physical activity. The work environment (both indoor and outdoor) is also an important component of behavior change and can have a significant impact on behavior choice (18). An environment that encourages less sedentary work and more physical activity could include well-placed and maintained stairwells for stair use versus elevators or distant parking.

Changes solely in the workplace environment may not be enough to encourage healthy behaviors (19). Health behavior decisions are affected by the social context in which they are made, such that the social support and social norms surrounding a health issue have a substantial effect on how that health behavior is perceived. Changing social norms and fostering a supportive work environment for the desired behavior is a necessary complement to the other levels of intervention. Social norms have been studied as a way to promote nutrition (20) and physical activity (21).

Workplace participatory approaches may foster social support and help to overcome organizational and employee barriers to program success. Most worksite weight-loss programs have relied on a top-down approach, rather than a participatory approach based on employee involvement in the design of interventions (22). In workplaces where employees generally have little influence on their work environment, similar to those sampled in this study, participatory approaches can result in better program implementation and subsequent health improvement (22). The recently described Healthy Workplace Participatory Program (HWPP) includes work environment changes, as well as healthy eating and physical activity interventions (23). A small study based on HWPP found promising changes in behaviors and weight loss in a pre–post evaluation of a participatory worksite intervention (24). To our knowledge, this HWPP-based study is the only controlled study to date using a worker health participatory program to attain weight loss. Future research should implement and evaluate workplace participatory interventions for weight loss.

Workplace wellness programs should also use effective communication strategies to engage workers from diverse work groups and backgrounds. As demonstrated with the health care system in this study, many low-wage workers were not aware of the wellness programs that were available to them. The same programs, however, have good participation from other work groups in the health care organization, primarily because of the method of communication. Rapid changes in information technology have enabled new interventions that use mobile telephones and other mobile devices (mHealth). These techniques show great promise for weight reduction in low-income populations (25), and such interventions are readily scalable to larger populations (13).

Although we did not directly ask about incentives, several participants discussed monetary incentives as a possible motivator to eat healthy and exercise. The use of incentives is common in workplace wellness programs; employers could maximize the benefits of incentives by incorporating lessons from behavioral economics. For example, the increasingly popular approach of delivering incentives through health insurance premium adjustments is unlikely to be as effective as more frequent and immediate rewards for behavior. This is because people tend to discount the future, meaning that they respond more readily to immediate than delayed costs and benefits (26). The participants in our study commonly discussed cost as a barrier to eating healthy and exercising. As suggested by others (27), low-income workers may be more likely to change and sustain healthy behaviors if provided with financial support for healthy food and participation in other weight-loss activities. Employers should also be aware of the limitations of incentives for behavior change. Recent reviews have shown behavioral effects to be relatively short-lived after incentives are removed (27), and considerable attrition is found in workplace programs for weight loss (28). More research is needed to determine the optimal timing, magnitude, and structure of incentives, but results to date suggest that incentives may need to be an ongoing feature of the workplace to have maximum impact.

Finally, employers may consider integrating traditional occupational safety and health programs (ie, those that focus on health hazards unique to the workplace) with health promotion and wellness programs (ie, those that focus exclusively on lifestyle factors off the job). The Total Worker Health program was launched by the National Institute of Occupational Safety and Health (NIOSH) to support the development and adoption of research and best practices to integrate these approaches and address health and safety risks at multiple levels, including the work environment (physical and organizational) and individual behaviors. This integrative approach may lead to greater adoption of interventions by management and workers and hence to improvements in the health of workers (11), but more research is needed to evaluate both the development process and the effectiveness of integrated programs (29).

The results of this study can help inform future worksite interventions for low-wage workers; however, our study has several limitations. First, we collected data from key informants who could be contacted or agreed to be interviewed. Second, although the participants in the focus groups represented a range of positions and worker groups, they were limited to those available during the implementation of the focus group discussions. Although using a convenience sample may be a limitation, those who elected to participate in the interviews or focus groups were able to provide helpful insights on the topic. Future intervention planning would need to be preceded by additional input from a broader participant base. Third, the information we collected may not be generalizable to other health conditions or work settings. Despite these limitations, the key informants and focus group participants provided rich and potentially actionable information on addressing obesity at the worksites of these worker populations.

Workplaces can provide an effective venue for engaging low-income populations at risk for obesity and related illnesses. Results of this study suggest that future worksite interventions for low-wage workers can improve reach, effectiveness, and sustainability if they embrace more innovative methods than those used in current workplace wellness programs. Future interventions should address workplace policies and environment and social norms that affect health behavior decisions. Communication strategies and financial incentives should be better aligned with the needs of low-wage workers. Workplace participatory programs are a promising approach to engage workers in health improvement.