Merrill, Ray M. PhD, MPH, MS, FACE, FAAHB
Substantial evidence supports the effectiveness of worksite wellness programs on the improvement of employee health behaviors and personal health.1–10 The worksite increasingly is viewed as an appropriate setting in which to offer comprehensive employee wellness programs.11 Worksite wellness programs can promote better employee health and worker productivity by delivering health promotion initiatives in an environment conducive to healthy behaviors among employees.12,13
The success of worksite wellness efforts ultimately depends on supportive policies and worksite environments that facilitate the ability of employees to adopt and maintain healthy lifestyle choices.14,15 Nevertheless, small businesses often face barriers to developing and implementing effective wellness programs, some of which include lack of health-promotion staff, cost barriers, employees located across wide geographical areas, and insufficient time and expertise to design and deliver behavior-change programs. For small companies, the maximum personnel commitment often is limited to the designation of a company wellness representative, a wellness committee that can work with an external wellness program provider, or both. The purpose of this study was to evaluate the effectiveness of a worksite wellness program delivered by WellSteps, LLC, with coordination by employee committees at five small companies.
Analyses are based on employees representing five small businesses in the United States. The classification of small business is based on the size of the companies, each having fewer than 500 employees,16 namely 21, 48, 49, 343, and 483. These businesses represent a brokerage firm, a professional employer organization, an integrated health care service company, an engineering firm, and a food sales and service corporation, respectively. Average age and sex distributions differed significantly across the companies. The numbers of employees in the five companies that completed a baseline personal health assessment (PHA) sometime during 2008 to 2010 were 19 (90%), 20 (42%), 39 (80%), 229 (67%), and 311 (64%), respectively. The numbers of employees in the five companies completing the PHA twice were 5 (24%), 2 (4%), 33 (67%), 149 (43%), and 171 (35%), respectively.
All employees in the companies were encouraged to complete the PHA and to participate in specific behavior-change campaigns. A behavior-change campaign is an evidence-based set of simple behavioral tasks performed by employees over a 3- to 8-week period. Of the 618 total employees who completed the PHA, 484 (78.3%) also participated in one or more behavior-change campaigns. The average participation rate in these campaigns was 41.5% (SD, 30.7%). The campaigns were designed to promote healthier lifestyle behaviors, thereby reducing chronic disease risk and escalating health care costs.17,18
The wellness program was delivered by WellSteps, LLC. The WellSteps model has been designed to provide an evidence-based solution for small- to mid-sized companies that lack the resources to develop and manage a wellness program independently. The program is based on and includes seven best practices stemming from three decades of research and field experience with highly successful programs by the Wellness Councils of America.19 A more complete description of the WellSteps model is presented elsewhere.20
Employees who completed the PHA received individual feedback about their health risks. Data from the PHA also were aggregated and discussed by a WellSteps program manager with the designated wellness coordinator, the committee, or both at each company. The wellness coordinator, the committee, or both selected the campaigns to be delivered at their respective places of work on the basis of the population health priorities reflected in the PHA aggregate feedback. They also selected the incentives associated with each campaign.
Employee recruitment was conducted through e-mail, posters at the worksite, and postcards delivered to the homes of the employees. Messages tailored on the basis of the transtheoretical model also were sent to individual employees who reported readiness to change.21–25 All campaigns were based on the social cognitive theory26 and from evidence produced by the Coronary Health Improvement Project.27–31
Six behavior-change campaigns were offered at each company during a 1-year period. Employees were eligible to participate in any or all of the behavior-change campaigns. Completion of a campaign required the employee to finish at least half of the weekly tasks. The total behavioral dose of all the campaigns combined ranged from 28 to 34 weekly tasks, depending on campaigns selected by each company.
Small incentives (eg, a book or a pedometer) were used to reinforce participation in the campaigns. In addition, a chance to win a modest reward was available to those who completed each campaign. The reward recipient was selected using a statistical random selection process. Participation in future campaigns also was reinforced by public announcement of the winners. Each campaign was designed to build behavioral capability by providing immediately applicable skills and tools. An attempt to increase self-efficacy was made by segmenting the behavior-change process into weekly, manageable doses. The campaigns are described in Table 1.
The PHA questions were modeled on the basis of the 2006 Behavioral Risk Factor Surveillance System Questionnaire.32 The PHA questions and feedback were written at the sixth- to seventh-grade reading level. The PHA included questions about health behaviors (exercise, food intake, and sleep patterns), emotional health (depression, anxiety, and stress), self-reported health risks (body mass index [BMI], blood pressure, total cholesterol, and blood glucose), health-related work outcomes (presenteeism and absenteeism), health care utilization (doctor visits and hospital use), and demographics. The average time to complete the PHA was about 8 minutes. Employee participants completed the PHA online. Immediate electronic feedback was provided to each participant with an option to print a feedback report. The feedback report consisted of visual risk stratification for each risk factor (green, yellow, or red) and recommendations for making small changes in behaviors to reduce these risks, if any. A small number of the employees who did not have Internet access mailed their completed PHA to WellSteps headquarters, where the data were entered. Feedback reports then were provided through the mail. Approval to assess these data was granted by the institutional review board at Brigham Young University on September 15, 2011.
Frequencies, means, and standard deviations were used to describe the data. A repeated measures design using a mixed (random effects) model was used to assess change in health behaviors through 12 and 24 months of follow-up. Mean change scores were evaluated using the t statistic and the F statistic. The McNemar test was used to compare paired dichotomous responses over time. The Spearman correlation coefficient was used to measure the association between health perceptions/life perceptions/job satisfaction and selected behaviors. Two-sided tests of significance were based on the 0.05 level against a null hypothesis of no association. Analyses were performed using SAS version 9.2 (SAS Institute Inc, Cary, NC).
Employees completing the PHA at baseline ranged in age from 24 to 71 years (mean, 44.5 years; SD, 8.7 years); 326 (52.8%) were women and 292 (47.2%) were men. Of these individuals, 360 (58.2%) completed the PHA a second time, generally more than 1 year later (mean, 404 days; SD, 71 days). There was no significant difference between employees who completed the follow-up PHA and those who did not with respect to age, sex, BMI, blood pressure, cholesterol, blood glucose, absenteeism, or health care visits. Nevertheless, those who completed the PHA at baseline and follow-up were significantly more likely to participate in one or more campaigns (87.8% vs 65.1%; χ2 = 45.45; P < 0.001).
Participation in one or more of the wellness initiatives was more common in women than in men (86.8% vs 68.8%; χ2 = 29.30; P < 0.001), in those with lower BMI (27.2 vs 28.3; t = −2.21; P = 0.028), and in those with lower blood pressure (16.5% borderline/high vs 25.0% borderline/high; χ2 = 4.77; P = 0.029). Participation in the wellness initiatives was not significantly associated with age, cholesterol level, or blood glucose level.
Change scores for selected health behavior and health outcome variables are shown in Table 2. Because change scores tended to be similarly favorable across the five companies, the results represent combined data. Most of the change scores were significantly different from zero; that is, each of the health behaviors significantly improved, with the exception of smoking and seat belt use. The rate of current smokers was 10.7 per 100 and did not significantly change on the basis of the McNemar test. Health perception and life satisfaction improved, but job satisfaction declined, based on a 10-point self-reported scale, with 10 being the best. Body mass index decreased slightly, but not significantly so.
At baseline, after adjusting for age and sex, health perception was significantly positively associated with days exercised (Spearman ρ = 0.36), minutes exercised (0.26), whole-grain intake (0.17), fruit intake (0.13), vegetable intake (0.18), and sleep (0.29). Nevertheless, life satisfaction was significantly positively associated with only days exercised (0.19), minutes exercised (0.13), and sleep (0.35). Job satisfaction was significantly positively associated with only sleep (0.25). After adjusting for age, sex, and the baseline levels of these variables, improvement in health perception was significantly positively associated with increases in days exercised (0.27), minutes exercised (0.17), whole-grain intake (0.18), fruit intake (0.13), vegetable intake (0.11), and sleep (0.21); increase in life satisfaction was significantly positively associated with increases in minutes exercised (0.11), fruit intake (0.17), and sleep (0.17); and increase in job satisfaction was significantly positively associated with an increase in fruit intake (0.12).
Blood pressure, cholesterol, and blood glucose did not change significantly from baseline to follow-up. The change in absenteeism from baseline to follow-up was not significantly different from zero and did not vary across the levels of age and sex after adjusting for the baseline level of absenteeism. Change in absenteeism was also not significantly associated with participation in the wellness initiatives. Similar results were observed for change in health care services.
Selected health behaviors at baseline and follow-up among PHA participants are presented in Table 3. Significant improvements occurred for exercise; consumption of whole grains, fruit, and vegetables; and nights of restful sleep. The percentage of employees wearing a seat belt at least 90% of the time increased slightly, but not significantly so.
Among those who completed the PHA at both baseline and follow-up, selected self-reported health outcome measures are presented in Table 3. The percentage of employees rating their life satisfaction as 9 to 10 and self-perceived health as 9 to 10 significantly increased. However, the percentage of employees rating their job satisfaction as 9 to 10 significantly decreased. The percentage of obese employees remained constant over the study period.
Finally, the aforementioned analyses reflect data combined across five companies. Of interest is whether the results apply to the separate companies, given their different demographic makeup. Note that we combined three of the very small companies because of small numbers of employees. Mean scores at baseline and mean change scores for the selected health behaviors and health outcomes are presented by company group in Table 4. After adjusting for age, sex, and baseline health behavior/outcome level, a significant difference in change in mean scores occurred for fruit intake, nights of restful sleep, smoking, and job satisfaction. The three combined companies had greater improvement in fruit consumption, nights of restful sleep, smoking, and job satisfaction. For the other variables, there was no difference in mean change scores between company groups. In general, the change scores are favorable across the company groups.
Improvements in exercise, diet, and nights of restful sleep were observed from baseline to follow-up. Health perception and life satisfaction also improved, but job satisfaction declined. Although layoffs or salary reductions did not occur during the study period, a poor economy and anxieties regarding employment might have contributed to the small decline in job satisfaction.
Improved health perception was directly related to increased exercise, improved dietary practices, and sleep; improved life satisfaction was directly related to increased exercise, fruit intake, and sleep; and improved job satisfaction was directly related to increased fruit intake. The positive association between health perception and these health behaviors is consistent with previous research.27,33–38
Body mass index remained constant. Although there was not a significant decrease in BMI, maintenance of BMI is preferred to the increasing trend that is being seen nationally.39 This result is consistent with the results of previous studies evaluating the effects of PHA and wellness participation on health behaviors.1,28 Smoking and seat belt use did not significantly change, but smoking started low and remained low, and seat belt use started high and remained high.
Blood pressure, total cholesterol, and blood glucose did not significantly change over the study period. This is in contrast to previous research showing that physical activity and dietary behavior could improve these health outcomes in as few as 6 weeks.29–31 Nevertheless, it is possible that short-term beneficial changes occurred in these measures but did not persist throughout the study period, yet our results are consistent with another recent study that found no protective association between fruit and vegetable intake and biomarkers for the risk of chronic disease (ie, C-reactive protein, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, total cholesterol, fasting plasma glucose, and glycosylated hemoglobin).40
The insignificant change in absenteeism from baseline to follow-up may be because absenteeism began low, with only 20% of employees missing three or more days of work and approximately 10% missing four or more days of work in the past year because of illness or injury. Previous research has shown a decrease in absenteeism attributable to sickness, albeit statistically not significant (Relative Risk = 0.78; 95% CI: 0.10 to 1.57), after a worksite health promotion intervention.41 Another study showed a significant decrease in absenteeism among employees who participated in a worksite wellness program, but mean absenteeism was higher at baseline than in this study.10
Health care visits did not decrease over the study period. This finding is consistent with other studies that have found that people who are more health oriented tend to use health services more often.2,9,10 Although the number of health care visits remained constant, it is possible that the nature of the visits changed. For example, perhaps more visits were made to consult a physician about the employee's PHA results or to receive screening tests. Data were not available to determine the nature of health care visits or to assess the cost per visit.
Completion of the PHA involved more than 50% of employees in all but one of the companies, where the participation level was 41.7%. A review of several studies indicated that participation in worksite wellness programs is generally less than 50%.42 Greater participation levels in four of the companies represented in this study may be explained by program incentives, small incremental changes in behaviors, and multifaceted interventions that aimed to reach a high percentage of workers. In one study, financial incentives were reported as the primary reason for participating in worksite wellness programs, followed by a concern for personal health.9 Other reasons that contribute to participation may include convenience, motivational activities, a fun social environment, and supportive management.43–46 On the contrary, factors that contribute to nonparticipation may include a general lack of interest, insufficient time, or employees are already healthy and do not perceive an immediate need to change their lifestyle.43–46
Participation in the wellness program was significantly greater among women than men. A similar result has been reported in other worksite wellness programs.5,9,10,42 This may be because women are more concerned about their health and more likely to comply with the requirements of a worksite wellness program. In addition, women are more likely to have sedentary jobs, wherein participation in the wellness program is more attractive to them, and men have greater access to fitness centers than women beyond the workplace. Age had no influence on participation, possibly because the wellness initiatives were designed for a range of ages. In addition, those who completed a second PHA were, on average, 1 year older, but this difference in age was not statistically significant. Nevertheless, the older age is consistent with older individuals being more interested in monitoring their health.
This study was limited because we did not have health information about employees who did not complete the PHA, making it impossible to determine whether those who completed the PHA at baseline differed in health from those who did not complete the PHA. The overall participation rate was 65.5%. It is likely that those who participated in the PHA were more likely women and were healthier. In addition, without a comparison group, we were unable to determine whether the results were influenced by factors beyond the PHA and behavior-change campaigns. Furthermore, employees self-selected to take the PHA and to participate in the wellness program. These individuals were likely motivated to make health behavior changes and, as such, it is possible that some selection bias may have occurred. Finally, the behavior and health outcome data were self-reported. Nevertheless, we can only assume that because people were completing the PHA to obtain useful information about their own health, they would be as accurate as possible in their responses.
A relatively high percentage of workers completed the PHA and participated in the wellness program, perhaps, in part, because of incentives that reinforced participation and because the behavioral tools provided in the behavior-change campaigns improved self-efficacy. Participation in the wellness program was lower among men and those who had poorer health. These groups should receive additional attention during the recruitment phase of wellness programs. Health perception and life satisfaction were directly associated with health. In addition, improvements in health perception and life satisfaction were related to improvements in health behaviors. It is likely that no significant improvements were observed in smoking, seat belt use, absenteeism, or some biometric measures because baseline levels of these variables were already favorable.
1. Merrill RM, Anderson A, Thygerson SM. Effectiveness of a worksite wellness program on health behaviors and personal health. J Occup Environ Med. 2011;53:1008–1012.
2. Hyatt Neville B, Merrill RM, Kumpfer KL. Longitudinal outcomes of a comprehensive, incentivized worksite wellness program. Eval Health Prof. 2011;34:103–123.
3. Merrill RM, Aldana SG, Bowden DE. Employee weight management through health coaching. Eat Weight Disord. 2010;15:52–59.
4. Koffman DMM, Goetzel RZ, Anwuri VV, et al. Heart healthy and stroke free: successful business strategies to prevent cardiovascular disease. Am J Prev Med. 2005;29:113–121.
5. Merrill RM, Aldana SG, Ellrodt G, et al. Efficacy of the Berkshire Health System Cardiovascular Health Risk Reduction Program. J Occup Environ Med. 2009;51:1024–1031.
6. Racette SB, Deusinger SS, Inman CL, et al. Worksite opportunities for wellness (WOW): effects on cardiovascular disease risk factors after 1 year. Prev Med. 2009;49:108–114.
7. Schultz AB, Lu C, Barnett TE, et al. Influence of participation in a worksite health-promotion program on disability days. J Occup Environ Med. 2002;44:776–780.
8. Serxner SA, Gold DB, Grossmeier JJ, Anderson DR. The relationship between health promotion program participation and medical costs: a dose response. J Occup Environ Med. 2003;45:1196–1200.
9. Merrill RM, Hyatt B, Aldana SG, Kinnersley D. Lowering employee healthcare costs through the Health Lifestyle Incentive Program. J Public Health Manag Pract. 2011;17:225–232.
10. Aldana SG, Merrill RM, Price K, et al. Financial impact of a comprehensive multisite workplace health promotion program. Prev Med. 2005;40:131–137.
11. US Department of Health and Human Services. Healthy People 2010: Understanding and Improving Health. 2nd ed. Washington, DC: Government Printing Office; 2000.
13. Lack DM. Presenteeism revisited. A complete review. AAOHN J. 2011;59:77–89.
14. Bandura A. Organizational application of social cognitive theory. Aust J Manage. 1988;13:275–302.
15. Prochaska JO, Velicer WF. The transtheoretical model of health behavior change. Am J Health Promot. 1997;12:38–48
17. Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity-related health risk factors. JAMA. 2003;289:76–79.
18. Boston Municipal Research Bureau. Municipal Health Reform: Seizing the Moment [A joint report by the Boston Municipal Research Bureau and Massachusetts Taxpayers Foundation]. Boston, MA: Boston Municipal Research Bureau; 2007. Available at: http://www.bmrb.org/content/upload/BMRBMTF.pdf
. Accessed September 5, 2010.
19. Hunnicut D, Leffelman B. WELCOA's 7 benchmarks of success: developing results oriented programs one company at a time. Absolute Advantage. 2006;6:1–28.
20. WellSteps: Effective employee wellness solutions. Available at: www.wellsteps.com
. Accessed September 15, 2011.
21. Spencer L, Wharton C, Moyle S, Adams T. The transtheoretical model as applied to dietary behavior and outcomes. Nutr Res Rev. 2007;20:46–73.
22. Spencer L, Adams T, Malone S, et al. Applying the transtheoretical model to exercise: a systematic and comprehensive review of literature. Health Promot Pract. 2006;7:428–443.
23. Migneault J, Adams T, Read J. Application of the transtheoretical model to substance abuse: historical development and future directions. Drug Alcohol Rev. 2005;24:437–448.
24. Spencer L, Pagell F, Adams T. Applying the transtheoretical model to cancer screening behavior. Am J Health Behav. 2005;29:36–56.
25. Spencer L, Pagell F, Hallion M, Adams T. Applying the transtheoretical model to tobacco cessation and prevention: a review of literature. Am J Health Promot. 2002;17:7–71.
26. Bandura A. Social cognitive theory: an agentic perspective. Annu Rev Psychol. 2001;52:1–26.
27. Thieszen CL, Merrill RM, Aldana SG, et al. The coronary health improvement project (CHIP) for lowering weight and improving psychosocial health. Psychol Rep. 2011;109:338–352.
28. Merrill RM, Aldana SG, Pope JE, et al. Evaluation of a best-practice worksite wellness program in a small-employer setting using selected well-being indices. J Occup Environ Med. 2011;53:448–454.
29. Aldana SG, Greenlaw RL, Diehl HA, et al. The behavioral and clinical effects of therapeutic lifestyle change on middle-aged adults. Prev Chronic Dis. 2006;3:A05.
30. Aldana SG, Greenlaw RL, Diehl HA, et al. The effects of a worksite chronic disease prevention program. J Occup Environ Med. 2005;47:558–564.
31. Aldana SG, Greenlaw RL, Diehl HA, et al. Effects of an intensive diet and physical activity modification program on the health risks of adults. J Am Diet Assoc. 2005;105:371–381.
35. Anderson JW, Hanna TJ, Peng X, Kryscio RJ. Whole grain foods and heart disease risk. J Am Coll Nutr. 2000;19(suppl):291S–299S.
36. Cade JE, Burley VJ, Greenwood DC. Dietary fibre and risk of breast cancer in the UK Women's Cohort Study. Int J Epidemiol. 2007;36:431–438.
37. Jensen MK, Koh-Banerjee P, Hu FB, et al. Intakes of whole grains, bran, and germ and the risk of coronary heart disease in men. Am J Clin Nutr. 2004;80:1492–1429.
38. Merrill RM, Taylor P, Aldana GS. Coronary Health Improvement Project (CHIP) is associated with improved nutrient intake and decreased depression. Nutrition. 2008;24:314–321.
40. Fisk PS II, Middaugh AL, Rhee YS, Brunt AR. Few favorable associations between fruit and vegetable intake and biomarkers for chronic disease risk in American adults. Nutr Res. 2011;31:616–624.
41. Kuoppala J, Lamminpää A, Husman P. Work health promotion, job well-being, and sickness absences—a systematic review and meta-analysis. J Occup Environ Med. 2008;50:1216–1227.
42. Robroek SJW, van Lenthe FJ, van Empelen P, Burdorf A. Determinants of participation in worksite health promotion programmes: a systematic review. Int J Behav Nutr Phys Act. 2009;6:26.
43. O'Donnell MP. Definition of health promotion 2.0: embracing passion, enhancing motivation, recognizing dynamic balance, and creating opportunities. Am J Health Promot. 2009;24:iv.
44. Spittaels H, De Bourdeaudhuij I. Who participates in a computer-tailored physical activity program delivered through the Internet? A comparison of participants' and non-participants' characteristics. Int J Behav Nutr Phys Act. 2007;4:39.
45. Tavares LS, Plotnikoff RC. Not enough time? Individual and environmental implications for workplace physical activity programming among women with and without young children. Health Care Women Int. 2008;29:244–281.
46. Emmons KM, Linnan L, Abrams D. Women who work in manufacturing settings: factors influencing participation in worksite health promotion programs. Womens Health Issues. 1996;6:74–81.
Copyright © 2013 by the American College of Occupational and Environmental Medicine