Abstract
Weight management programs are becoming increasingly common in workplace settings; however, few target middle-aged men. The purpose of this article is to describe the process evaluation of a worksite translation of the Diabetes Prevention Program in a predominantly middle-aged male population. The translated program, FUEL Your Life, was largely self-directed, with support from peer health coaches and occupational health nurses. The RE-AIM (Reach Effectiveness Adoption Implementation Maintenance) framework was used to examine the factors that influenced program implementation using data from an environmental assessment, participant surveys, peer health coach surveys, and occupational health nurse interviews. An overwhelming majority of the employees who enrolled in the study were overweight or obese (92%). Overall, the program was effective for weight maintenance; those with higher levels of participation and engagement had better weight loss outcomes. The peer health coach and family elements of the intervention were underused. The program was successful in reaching the intended population; however, the program had limited success in engaging this population. Not surprisingly, weight loss was a function of participant engagement and participation. Increasing participant engagement and participation is important to the success of weight management interventions translated to the worksite setting. Garnering buy-in and support from management can serve to increase the perceived importance of weight management in worksites. With management support, weight management protocols could be integrated as a component of the mandatory safety and health assessments already in place, fostering promotion of healthy weight in the workforce.
Keywords
Introduction
The increasing rates of overweight and obesity among employees (Hertz, Unger, McDonald, Lustik, & Biddulph-Krentar, 2004) and the effects of excess weight on absenteeism, presenteeism, and medical costs (Finkelstein, DiBonaventura, Burgess, & Hale, 2010) challenge employers to find ways to manage these conditions and contain costs. With the majority of adults working and spending a significant portion of their waking hours at the worksite, worksites offer a unique opportunity to translate weight loss programs that are proven effective in other settings. The Diabetes Prevention Program (DPP) was a highly effective weight loss program in a motivated, prediabetic population. DPP participants lost 6.8 kg, resulting in a 16% reduction in diabetes (Knowler et al., 2002). DPP has been translated to several community settings with moderate success, and reported weight loss has been between 2 and 4 kg. These translations typically had small sample sizes and restricted participation to obese or prediabetic individuals (similar to the original DPP; Ackermann, Finch, Brizendine, Zhou, & Marrero, 2008; Cramer, Sibley, Bartlett, Kahn, & Loffredo, 2007; Davis-Smith et al., 2007; Jackson, 2009; Mayer-Davis et al., 2004; Pagoto, Kantor, Bodenlos, Gitkind, & Ma, 2008; Seidel, Powell, Zgibor, Siminerio, & Piatt, 2008). Similarly, small-scale translations focusing on at-risk or prediabetic employees have reported encouraging weight loss of more than 2 kg (Aldana et al., 2005; Barham et al., 2011).
This article describes the process evaluation of Fuel Your Life (FYL), a worksite translation of the DPP suitable for a wide range of work settings and all employees, regardless of weight and risk status. Glasgow’s RE-AIM framework (Reach Effectiveness Adoption Implementation Maintenance; Glasgow, Vogt, & Boles, 1999) was used to examine factors that influenced program adoption and implementation in the FYL study. The RE-AIM framework has been used as a process evaluation tool in worksite weight management studies previously (Estabrook, Zapka, & Lemon, 2012).
Theoretical Basis
FYL was designed to be suitable for a wide range of work settings and for all employees regardless of weight or diabetes risk status. The translated program is described in detail below.
FUEL Your Life
The FYL intervention was based on the DPP Lifestyle Intervention (The DPP Research Group, 2002). The key features of the original DPP were the following: (a) weight loss and physical activity goal-based behavioral intervention, (b) case manager/lifestyle coach intervention delivery, (c) intensive and ongoing intervention, (d) individual tailoring, (e) materials/strategies for an ethnically diverse population, and (f) interventionist network support. The FYL translation was designed to be used by worksites with varying levels of resources. The translated program used a participant manual that was modified to facilitate self-study. The FYL participant manual retained the key concepts of the DPP Lifestyle Balance participant manual, including 16 lessons completed over 24 weeks. Research staff enrolled participants on-site at each study location and measured height and weight. Participants met with a dietitian or health educator to review their baseline weight and body mass index (BMI) and discussed the goals and expectations of the FYL program. Participants were given all study materials during this session.
Occupational health nurses (OHNs) who were employed by the organization received a manual and attended training with the research team to discuss their program responsibilities. OHN responsibilities included delivering six presentations on FYL topics at staff meetings, sending FYL health messages to supervisors to be read at the beginning of each shift, and displaying FYL posters at the worksites. OHNs had regular calls with the research project director during the intervention period for implementation support.
Peer health coaches (PHC) were identified by the OHN(s) at each site with the intent of representing each shift/department. The PHCs received a manual and had an individual meeting with the research team to discuss their responsibilities. PHCs were expected to have ongoing informal contact with participants and use talking points to reinforce the FYL lessons, providing social support. FYL posters served as environmental prompts at the worksite, reinforcing the FYL lessons. Additionally, sites were encouraged to apply the company vending policy, which was guided by the “Fit Pick Standards” (“NAMA Extends Healthy Vending Program,” 2008). The guidelines require that 25% of items meet healthy criteria (35% of calories or less from fat, less than 10% of calories from saturated fat, and less than 35% of total weight from sugar), be labeled, and placed at eye level. Additionally, vending guidelines promoted displaying nutritional information for each product. A family packet was given to each participant, describing ways the family could support the employee’s FYL participation and with instructions on how to access the FYL website. A website was created to provide additional resources to family members, including shopping lists, recipes, and cooking strategies to reduce fat. The research team measured height and weight at the conclusion of the core intervention period (6 months) and reviewed the participants’ progress in the program. A 6-month maintenance phase followed the core intervention period. Participants were encouraged to maintain any weight loss or keep working toward their original weight loss goal. During the maintenance phase, OHNs delivered three additional presentations on FYL topics and hung posters. There was no participant manual for the maintenance phase. The maintenance phase concluded with a final height and weight measurement. Table 1 summarizes the adaptations of DPP translated to FYL.
Key Features of DPP Translated in FYL
NOTE: DPP = Diabetes Prevention Program; FYL = FUEL Your Life; PHC = peer health coach; OHN = occupational health nurse.
Method
Study Setting and Employee Population
The study was conducted in six Union Pacific Railroad locomotive maintenance facilities. Sites were matched based on the number of employees and randomly assigned to treatment or control group. All employees were eligible to participate, participation was entirely voluntary, and all data collection was conducted during work time at the worksites. Paper-based surveys were completed independently as outlined below. Each participant received a $10 incentive for enrolling in the program and participating in the initial data collection procedures. Participants received additional $10 incentives for participating in data collections at 6 months and 12 months. The average age of employees was 46 years, 94% were male, approximately 75% were Caucasian, and 95% were union members. At baseline, there were 2,819 employees at the study sites, with 1,301 in control sites and 1,518 in treatment sites. The site size ranged from 232 to 933 employees. The process evaluation includes data from the three treatment sites.
Data Collection Methods and Protocol
Multiple sources of quantitative and qualitative data were used for the process evaluation, and included the following: (a) an environmental audit tool designed to measure environmental and organizational support for weight management for each worksite, (b) interviews with the OHNs to gather perceptions of program implementation, and (c) questionnaires to assess participant engagement and perceptions of FYL. Most process evaluation data were collected using anonymous surveys developed specifically to capture process data. The surveys were anonymous to reduce participant burden and improve response rate. For purposes of this article, we are focusing on the process evaluation. Table 2 describes the RE-AIM dimensions, the data collection method, and parameters assessed.
RE-AIM Dimensions Applied to FYL
NOTE: RE-AIM = Reach Effectiveness Adoption Implementation Maintenance; FYL = FUEL Your Life; PHC = peer health coach; PA = physical activity; EAT = Environmental Assessment Tool; OHN = occupational health nurse.
Environmental Assessment Tool (EAT)
The EAT is an audit tool administered at each worksite to document environmental and organizational supports for weight management. Trained research staff completed the EAT at the beginning and end of the program at each site. Additional information about the EAT can be found elsewhere (DeJoy et al., 2008; Parker, DeJoy, Wilson, Bowen, & Goetzel, 2010).
Participant Outcome Questionnaire
Participants completed paper surveys at three time points: baseline, posttest (6 months after baseline), and follow-up (12 months after baseline). This was primarily an outcome measure, but seven process evaluation questions were added to the 6-month posttest survey, and nine questions were added to the 12-month follow-up survey. These questions assessed the frequency and intensity of program services, number of lessons completed, and the frequency of talking with a PHC or a family member about FYL. Participants completed the survey in approximately 30 minutes. Example questions included the following: “How often did you review the lessons in the FYL manual?” “Out of the 16 lessons in the FYL manual, how many lessons did you complete?” and “How often did you talk with a peer health coach about the FYL program?”
Participation and Satisfaction Survey
This anonymous survey was completed at posttest (6 months) to gauge participant perceptions of FYL. Participants completed the survey in less than 10 minutes. Five program dimensions were assessed: (a) participant manual, (b) PHC and OHN, (c) posters, (d) healthy vending items, and (e) FYL website. Sample questions included the following: “During the program, how often did you review the FYL manual?” (5-point scale, never to daily), “Were the PHCs helpful to you during the program?” and “Did the nurse provide regular information about the FYL program?” (dichotomous scale, Yes/No). The supportiveness of shop management and the participants’ families was assessed. Finally, participants were asked open-ended questions regarding their likes and dislikes about the program, and any suggestions for improving FYL. These anonymous responses could not be linked to programmatic outcomes.
Health Coach Questionnaire
The PHCs completed an anonymous survey at posttest (6 months). It took participants less than 10 minutes to complete the survey. This survey gauged PHC perceptions of program components, including usefulness of the PHC manual and management support for FYL. Using a 5-point scale, PHCs responded with their level of agreement (strongly disagree to strongly agree) to statements such as “The posters were helpful for reminding participants about the program,” and “The PHC manual was helpful for me as a health coach.” The PHCs were also asked about the following: (a) their frequency of contact with participants, (b) their perceptions of how helpful they were to participants, (c) their familiarity with and the perceived helpfulness of FYL posters, (d) healthy vending items, and (e) the website.
Occupational Health Nurse Interviews
Two members of the research team conducted 45-minute telephone interviews with the OHNs at the completion of the project (12 months). The interviews were structured around the following themes: (a) employee program participation (e.g., “How would you rate employee participation in FYL survey and biometric measurements?”), (b) site characteristics (i.e., “Was there anything unique to this site that affected participation [at each time point]?”), (c) contact with participants (i.e., “How frequently did you talk with employees about nutrition, physical activity, and/or weight management during the FYL program?”), and (d) perceptions of the PHCs (i.e., “Were the PHCs effective in encouraging participation?”).
Intervention Dose
Intervention dose was calculated using the following scoring rubric. Four questions assessed use of the FYL intervention tools, and one question assessed the number of lessons completed. Participants responded with their level of engagement (5-point scale, daily to never), to four statements: “How often did you review the lessons in the FYL manual?” “How often did you check the FYL website?” “How often did you talk with a PHC about the FYL program?” and “How often did you talk with your family about FYL?” Points were assigned based on their answer and summed. Participants earned 2 points for each lesson completed. Dose was calculated by summing engagement points and points earned from completing lessons. Dose scores could range from 0 to 48.
Data Analysis
Survey data were analyzed using SPSS 20. Descriptive statistics were generated for all variables. Continuous variables were summarized by mean and standard deviation and categorical variables by frequency and percentage. Qualitative data were collected from open-ended survey questions, the EAT, and interviews conducted by trained research staff. Qualitative data were analyzed to identify commonly occurring themes.
Results
Reach: Participation and Representativeness
Overall, 479 employees enrolled in FYL at the treatment sites, which represents 30% of the employee population at those worksites. The average age of participants was 44 years. The majority of participants were either overweight (BMI = 25-29.9) or obese (BMI ≥ 30; 32.9% overweight, 59.4% obese). At the end of the intervention (posttest), there were 236 participants, representing a 49% attrition rate. The sample was representative of the overall employee population.
Effectiveness
FYL participants maintained their body weight when compared to employees at control sites who experienced a 2.6-pound weight gain at 6 months, and the findings remained consistent at 12-month follow-up. At 12 months, 55% of FYL participants lost weight compared with 35% in the control group. FYL participants experienced a significant decrease in minutes spent sitting and an increase in minutes spent walking. There were no observed differences in dietary intake.
FYL participant dose scores ranged from 0 to 44 (M = 10.91, SD = 10.66), and 75% had a dose score of 15 or less. A chi-square test examined the relationship between dose and total weight loss, and a significant relationship was found, χ2(3, N = 158) = 9.178, p = .027. Participants who received the highest intervention dose were more likely to lose weight than those who received lower intervention dose.
Adoption
Adoption was assessed by participant enrollment and evaluating management support. Fewer than half the participants reported they enrolled in the program to lose weight (40%). Other reported reasons for participating were to eat healthier (23%), the incentive (9%), exercise more (5%), or other reasons (21%).
Management support is critical to program adoption as management makes decisions about allocation of resources and business priorities. In this study, management support was measured using three data collection sources and the data were triangulated: participation and satisfaction survey, PHC questionnaire and the OHN interview. Overall, management was perceived as supportive from all sources. At the end of the intervention, 74.3% of participants indicated that management was supportive. At the 12-month follow-up, 38.5% of the participants responded that the site management was helpful. Half of the PHCs agreed that management was supportive, while the other half were neutral. The OHNs at one site used the shop management to get involved and make announcements to the employees. Additionally, several managers participated in the program. At another site, there was turnover in management at multiple levels; however, the OHN engaged the new management to promote FYL.
Implementation
Participant Manual
All FYL participants received a copy of the FYL manual during enrollment. Thirty-five percent of the participants reviewed the manual at least monthly (6.6% daily, 16% weekly, and 12% monthly). Sixty-eight percent of respondents completed at least one lesson in the manual; however, only 18% of respondents completed at least half of the lessons and 32% completed zero lessons.
Occupational Health Nurse
OHNs were the primary link between the worksites and the research team. Typically, one or two nurses were responsible for the health and safety of hundreds of employees. For example, there were 933 employees at the largest site with only two OHNs. The smallest site had 260 employees and one OHN. The OHNs had to balance FYL tasks with multiple job-related responsibilities, particularly mandated safety screenings for this population. Despite these limitations, the majority of participants (90%) indicated that the OHN was available to help them with the FYL program (range = 87% to 97%).
Peer Health Coach
Over half of the participants that responded (63%) indicated that a FYL PHC was available when they needed them. During the program, 62% of the participants never spoke with a PHC about the program. Half of the participants indicated that the PHCs were helpful during the program. Overall, 72 participants talked with the PHCs about FYL.
Family Support
Participants were given a packet at enrollment to share with family or friends. The packet provided recommendations for supporting the participant, and instructions on accessing the website. Twenty-nine percent of participants talked with their family about FYL at least monthly, and 13% had family members participating in FYL with them. The website was intended to be a resource to family members and included a calendar, a fat counter, message boards, resources, and recipes. The pages that were most frequently visited were the home page (40%) and resource page (22%). The majority of respondents (92%) indicated that their spouse/family did not use the website; however, 44% of the participants visited the FYL website. Approximately half of the PHCs found the FYL website helpful.
Worksite Environmental Support
OHNs placed FYL posters supporting the key lesson concepts in high-visibility areas during the intervention and maintenance phases of the program. The nurses felt the posters were an excellent component of FYL. Ninety-seven percent of participants indicated that they saw the posters around the shop, and 89% found the posters helpful. All of the PHCs felt the posters were helpful for reminding participants about FYL. In addition to the posters, weekly announcements were given at shift meetings.
Healthy vending options provided additional environmental support. The intervention sites were located in settings with limited or no access to grocery stores and food establishments. Employees received multiple short breaks (10-20 minutes) and were restricted from leaving the worksite during breaks. There were no on-site cafeterias or food sources other than vending machines. Intervention sites were encouraged to adhere to the company policy on healthy vending. Fifty-seven percent of participants noticed healthy items in the vending machine; however, most participants (56%) reported that they did not purchase healthy items.
Maintenance
Fewer than half (43%) of the participants reported continuing FYL after the first 6 months. Even fewer participated in the final data collection activities at 12 months (31%). At the organization level, the program was not continued.
Discussion
The primary aim of the overall project was to translate the DPP to an intervention suitable for use in a variety of worksites. In this article, we use process evaluation data from this project to examine factors that influenced program adoption and implementation of the translated program, FYL. This translation was designed to reach a general work population and was tested in a predominantly blue-collar male population, a population that is often not reached with traditional weight management programs. The program was successful in reaching employees who were overweight or obese and was effective for preventing weight gain in this population. Participants who had the highest levels of participation and engagement lost the most weight. However, participation and engagement were relatively low. Despite these low levels, participants were generally satisfied with the program and FYL was well received.
The success of the FYL program in reaching a hard-to-reach, predominantly obese and overweight blue-collar population suggests that the intended reach of the program was achieved; however, fewer than half the participants joined the program to lose weight, suggesting that adoption of a worksite program intended to address obesity in this population may have had limited appeal among employees. Perhaps this sample did not perceive their weight as a problem or were not willing to do something about it at this point in time. With the high rates of overweight and obesity in the study sites, it is possible that some employees viewed their weight, which they frequently referred to as the “railroad physique,” as socially acceptable and without health consequence. These findings suggest there is a need to increase the awareness of the health consequences of overweight and obesity among blue-collar male employees, in particular. DeJoy and Southern (1993) have previously suggested that health promotion programs may be more acceptable in blue-collar employee populations when integrated with occupational health and safety initiatives. This may be particularly true in this population, which is required to participate in a number of mandatory safety and health assessments. Assessment of weight could be incorporated into other mandatory screening programs at the worksite. The assessment could be combined with programs like FYL to maximize their effectiveness. These efforts would not only increase awareness of the health consequences of overweight and obesity but also likely increase adoption of the program among employees.
The perception of management support was generally positive among participants, PHCs, and OHNs alike. However, process evaluation data reported here highlights challenges related to time and resources at the worksite. Management support is multidimensional. Generally, management must give permission to implement new programs within the worksite. Another level of support that makes the program a business priority and allocates needed staff time and resources for implementation is equally critical. The OHN played a key role in the implementation of FYL, but this intervention was one of many responsibilities. The OHNs were burdened with large caseloads and competing priorities. This was noted in the formative evaluation; therefore, PHCs were used to supplement the support of the OHNs. In this study PHCs were greatly underused. These are all factors that potentially had a negative effect on implementation. The lay health worker model was chosen for this project based on its successful use in community settings (Quinn & McNabb, 2001; Swider, 2002). Perhaps this population did not feel comfortable discussing weight and health concerns with peers, or they possibly were not aware of the PHC services. More research is needed to understand the role of the peer in promoting weight management in the worksite.
Several limitations of this study should be considered. The study population was largely male. Although this research is important to inform interventions targeting this population, the applicability in organizations with greater variability in gender and other participant characteristics is largely unknown. Self-report participant data used in the study could have limited accuracy. There may be recall issues regarding food intake, and participants could have provided inaccurate answers due to social desirability. The participants may have responded on the survey in a manner to make them look better. Another limitation is the minimal information about program maintenance at the institutional level. A better understanding of what effect is needed to warrant institutionalization is important.
FYL provided a low-intensity, low-cost, easy-to-implement translation of the DPP to an overweight and obese population of blue-collar male workers. FYL was effective for preventing weight gain, with weight loss positively correlated with intervention dose. Our findings suggest that barriers to program adoption and implementation at these worksites may have hindered program success. Weight management integrated with other worker health programs will likely be more effective in this population; however, integration will require another dimension of management support that makes total worker health a priority.
Footnotes
Funding from the Centers for Disease Control and Prevention (Grant/Award No. 5R18OH009396).