Skip Navigation LinksHome > April 2012 - Volume 54 - Issue 4 > Impact of a Web-Based Worksite Health Promotion Program on A...
Journal of Occupational & Environmental Medicine:
doi: 10.1097/JOM.0b013e31824d2e43
Original Articles

Impact of a Web-Based Worksite Health Promotion Program on Absenteeism

Niessen, Maurice A. J. MA; Kraaijenhagen, Roderik A. PhD; Dijkgraaf, Marcel G. W. PhD; Van Pelt, Danielle MSc; Van Kalken, Coen K. PhD; Peek, Niels PhD

Free Access
Continued Medical Education
Article Outline
Collapse Box

Author Information

From the NDDO Institute for Prevention and Early Diagnostics (NIPED) (Mr Niessen, and Drs Kraaijenhagen and Van Kalken), Amsterdam, the Netherlands; Clinical Research Unit (Dr Dijkgraaf), Academic Medical Center, Amsterdam, the Netherlands; BETER (Ms Van Pelt), Tilburg, the Netherlands; and Department of Medical Informatics (Dr Peek), Academic Medical Center, Amsterdam, the Netherlands.

Address correspondence to: Maurice A.J. Niessen, MA, NDDO Institute for Prevention and Early Diagnostics (NIPED), Amsteldijk 194, 1079 LK Amsterdam, the Netherlands (m.a.j.niessen@niped.nl).

Drs Van Kalken and Kraaijenhagen are directors and co-owners of NIPED. This institute developed the studied program and currently markets it in the Netherlands.

Mr Niessen is full-time employed as researcher by NIPED. Ms Van Pelt is employed at the safety, health, and welfare service of the financial company (BETER) evaluated in this study. All other authors are employed by the Academic Medical Center, University of Amsterdam. They received no additional funding for this study and report no competing interests.

Authors Niessen, Kraaijenhagen, Dijkgraaf, van Pelt, van Kalken, and Peek have no relationships/conditions/circumstances that present potential conflict of interest.

The JOEM Editorial Board and planners have no financial interest related to this research.

Collapse Box

Abstract

Objective: To evaluate the effect of participation in a comprehensive, Web-based worksite health promotion program on absenteeism.

Methods: Study population consists of Dutch workers employed at a large financial services company. Linear regression was used to assess the impact of program attendance on the difference between baseline and follow-up absenteeism rates, controlling for gender, age, job level, years of employment, and noncompletion of the program.

Results: Data from 20,797 individuals were analyzed; 3826 individuals enrolled in the program during the study period. A 20.3% reduction in absenteeism was shown among program attendees compared with nonparticipants during a median follow-up period of 23.3 months.

Conclusions: Participating in the worksite health promotion program led to an immediate reduction in absenteeism. Improved psychological well-being, increased exercise, and weight reduction are possible pathways toward this reduction.

Learning Objectives:

* Outline the design and elements of the PreventionCompass worksite health promotion program, including the authors' reasons for describing it as “comprehensive.”

* Summarize the new findings on reductions in absenteeism among employees participating in the PreventionCompass program, compared to nonparticipants.

* Discuss the study implications, including possible mechanisms of the program's effectiveness.

In Europe, 86% of deaths and 77% of the disease burden are caused by noncommunicable diseases that are linked by common risk factors with shared, underlying determinants and common opportunities for intervention and prevention. Almost 60% of the disease burden is accounted for by seven leading risk factors: high blood pressure (12.8%); tobacco (12.3%); alcohol (10.1%); cholesterol (8.7%); overweight (7.8%); low fruit and vegetable intake (4.4%); and physical inactivity (3.5%).1 The workplace is considered to be an excellent setting to target these risk factors because a large proportion of the population can be reached and workers spend about half their waking hours at work.2 Health risk factors have been associated with a loss of on-the-job productivity.38 The latter makes worksite health promotion programs especially interesting for employers. However, these programs should be effective enough to persuade employers to invest in them.9

Studies that evaluate the effectiveness of worksite health promotion programs have recently been reviewed by Soler and colleagues.10 These programs include three elements: (1) collection of information about personal health behaviors or measurable health indicators, or both; (2) translation of the collected information into individual risk scores; and (3) feedback to participants regarding their risk status. Health promotion programs that have been described in the literature vary considerably for these elements. The effectiveness of worksite health promotion programs is usually evaluated for behavioral aspects and physiologic indicators.4,11,12 For employers, changes related to productivity are particularly interesting. Thus far, no consistent reduction of absenteeism after participation in a worksite health promotion program has been reported, although moderate reductions in absenteeism are shown in studies evaluating worksite health promotion programs with additional interventions.13,14

In this study, the effect on absenteeism of participation in a comprehensive, Web-based worksite health promotion program is evaluated among workers in a Dutch multinational in the financial services industry. The health promotion program includes biometric measurements, laboratory testing, and assessment of lifestyle behavior, mental health disorders (depression, anxiety), and psychological strain (stress, burnout). The latter is particularly important because in the Netherlands, about one in every three new recipients of work disability benefit is disabled for work because of mental health problems.15 A central goal of the health promotion program is to change how people think and behave: it aims to raise awareness, educate, motivate, and empower individual health care users to take steps to promote their own health. The transtheoretical model of Prochaska and DiClemente16 plays an important role as participants are guided through the stages of precontemplation, contemplation, preparation, action, and maintenance. The program offers individuals a personal action plan, which may include advice to visit a (mental) health professional for further diagnostic testing or evaluation, to change health-related lifestyle behavior(s), or simply to maintain present healthful behaviors.

Back to Top | Article Outline

METHOD

Worksite Health Promotion Program

The health promotion program is called the Prevention Compass. Its core is a computerized knowledge-based reasoning system. In the system, risk assessment algorithms, as well as test and treatment thresholds regarding disease processes with a major impact on quality of life, productivity, vitality, and disability-adjusted life-years17 are kept up-to-date according to prevailing guidelines and best practices. To be included in the system, the “medical benefit versus risk” balance of risk factors and disease processes must be positive in accordance with the criteria of Wilson and Jungner.18 As a result, it integrates multiple evidence-based disease risk algorithms focusing on (the risk profiles of) disease processes with proven prevention or early diagnostic options (including cardiometabolic diseases, common mental disorders, musculoskeletal problems, and [colorectal] cancer), and (risk profiles for) decreased workability and work engagement. One of the main communication principles of the health promotion program is individual tailoring. Compared with generic information, tailored feedback is more effective in creating awareness and intention to change unhealthy behavior.19 Another general principle concerns the importance of supportive and nonjudgmental communication, regardless of the client's risk level or motivation for behavior change. For example, smokers who express low or no motivation to quit are not encouraged to feel “guilty” or otherwise “pressured” to quit. It is recognized that intrinsic motivation is a necessary ingredient for lasting behavioral change. Instead the unmotivated smoker's freedom of choice is affirmed; he or she is respectfully informed of the health benefits of smoking less or quitting, and offered resources for bolstering resolve and self-confidence to become smoke free. Furthermore, consistent with stepped care principles, the personal health management plan emphasizes low-intensity self-management and lifestyle change wherever possible. When lifestyle change is indicated, individual preferences—regarding for instance independent versus professionally supervised interventions—are taken into account.

Back to Top | Article Outline
Participants

The research sample consisted of individuals employed at a large Dutch financial services company during the period January 2007 to July 2009. Individuals whose employment ended during the baseline period, or who became employed during the follow-up period were excluded from analyses. Data from persons employed on freelance basis as well as data from persons with missing sociodemographic variables were also excluded. Because the capacity for onsite collection of biometric measurements was limited, employees were invited gradually to ensure contained influx into the program. From August 1, 2007, to June 30, 2009, anonymous e-mail invitations to participate in the worksite health promotion program were sent by the human resources department, based on a random selection within employee month of birth. For example, the first batch of invitees were an at random selection within all employees born in the month of January. A single reminder was sent after 2 weeks. The invitation e-mail included a description of the worksite health promotion program and informed employees that participation was voluntary and at no cost, that all personal data would be treated confidentially, and that no individual results would be shared with their employer or any other party. All participated on a voluntary and informed basis.

Back to Top | Article Outline
Intervention

Employees were classified as program attendees when they activated their online account after which containers for the collection of laboratory (urine and feces) samples were sent to participants' home address. Each attendee completed a Web-based electronic health questionnaire, followed by biometric measurements (length, weight, waist circumference, blood pressure) conducted at the worksite by trained and certified staff. Participants handed the collected laboratory samples to the staff during this visit. At the same visit, blood samples were collected for laboratory testing of total cholesterol, high-density lipoprotein, low-density lipoprotein, triglycerides, glucose, and glycated hemoglobin A1C. A personalized Web-based health report and health plan was automatically generated only after all health data were collected. At this point, the health promotion program was completed. Each health plan comprised (1) explanation of the assessed risk for each of the targeted preventable conditions, using a three-color “traffic light” system (green, normal risk profile; orange, moderately elevated risk profile; red, seriously elevated risk profile), (2) explanation of the threats associated with elevated risk and potential gains of taking preventive action, and (3) opportunities for taking preventive action on the basis of the participant's stated motivation for health-behavior change (physical activity, smoking cessation, alcohol intake, dietary habits), self-efficacy, and preferences with respect to interventions (eg, guided vs nonguided interventions). When seriously elevated cardiovascular risks were detected, the health plan included referral for further medical evaluation and treatment. A 30-minute health counseling session with the program physicians was also available upon request for all attendees.

On average, all measurements were collected within 8 weeks after enrollment into the worksite health promotion program. With the exception of attendees registering in the final 4 months, data of attendees who had not completed all measurements by the end of the study period were analyzed as a separate subgroup (enrolled but not completed participation).

Back to Top | Article Outline
Outcome Measure

For all attendees of the worksite health promotion program, baseline and follow-up periods were determined by the program enrollment date. Study participants who did not enroll in the worksite health promotion program during the study period were classified as nonparticipants. For them, the baseline period ended at August 1, 2007, the day invitation e-mails to participate in the worksite health promotion program were first sent.

For each study participant, the absenteeism rates during baseline and follow-up periods were determined as follows: first, the total number of workable days was calculated for both periods. If employment started during baseline or ended during follow-up, workable days were adjusted proportionally. For part-time employees, workable days and absence episodes of 3 days or more were multiplied by the fraction of employment. In accordance with the Dutch practice, maternity leave was not recorded as absenteeism. Dividing the total number of absence days by the workable days resulted in the absenteeism rate. The difference between baseline and follow-up absenteeism rates was used as the primary outcome measure and dependent variable in the regression equation.

Equation (Uncited)
Equation (Uncited)
Image Tools
Back to Top | Article Outline
Statistical Analysis

Linear regression was used to assess the impact of worksite health promotion program attendance on the difference between baseline and follow-up absenteeism rates, controlling for gender, age, job level, years of employment, and program completion. Job level explains at what education level an employee is functioning (source: Hay Group International). Subsequently, for all study participants, the observed absenteeism rate during baseline and the regression equation that resulted from the preceding step were used to estimate the expected absenteeism rates during follow-up under presumed participation and presumed nonparticipation in the program. The average difference in expected absenteeism rates during follow-up under presumed participation and presumed nonparticipation was used as an estimate of the overall effect on absenteeism of the program. Data were analyzed using SPSS for Windows, version 19.

Back to Top | Article Outline

RESULTS

A total of 23,258 individuals were employed during the study period; 1297 individuals were excluded because their employment ended during the baseline period (n = 550), or started during the follow-up period (n = 747). Data of 613 persons employed on freelance basis were excluded from analysis as were data from 551 persons with an unknown job level. Data from 20,797 individuals were analyzed. Approximately 11,252 employees were invited to participate in the health promotion program during the study period. From this cohort, a total of 3826 individuals enrolled in the worksite health promotion program.

Table 1 lists the baseline characteristics of the study participants. Program attendees were slightly older and were functioning at a higher job level. They worked more hours per week and more often longer than 10 years employed at the company. Also, more males were present among the program attendees. The fixed versus variable baseline/follow-up periods for the nonparticipants and the worksite health promotion program attendees are reflected in the observed group differences in the duration of these periods.

Table 1
Table 1
Image Tools

Table 2 presents the results of the linear regression analysis. Compared with nonparticipants, there is an absolute decline in absenteeism rate of 1.0% (95% confidence interval [CI], 0.6 to 1.4) in the worksite health promotion program attendee group during follow-up. There was no statistically significant effect of attrition. The analysis also indicates that, in absolute terms, the absenteeism rate of females increased with 1.1% (95% CI, 0.7 to 1.4) during follow-up. An increase in absenteeism is observed for older employees (50 to 59 years). Although the effect was not statistically significant, results point toward an increase in absenteeism among employees functioning at lower and secondary vocational levels and a decrease in absenteeism among employees functioning at university level. Also, a decrease in absenteeism is observed for employees with more than 10 years of employment. The absenteeism rate for employees working at the company for more than 30 years showed a decline of 1.7% (95% CI, 1.0 to 2.4) during follow-up. On the basis of the regression equation, the absenteeism rate at follow-up estimated for program attendance is 3.93%. For nonparticipation, the estimated absenteeism rate at follow-up is 4.93%. The relative percentage gain of worksite health promotion program attendance, therefore, is calculated as follows:

Table 2
Table 2
Image Tools

Equation (Uncited)
Equation (Uncited)
Image Tools
Back to Top | Article Outline

DISCUSSION

In this study, a 20.3% reduction in absenteeism was shown among voluntary participants in a Web-based worksite health promotion program, when comparing their absenteeism rates to those of nonparticipants. Absenteeism was also lower for males and employees working more than 10 years at the company. A trend toward a decrease in absenteeism was found for employees functioning at university level. Absenteeism was higher among employees aged 50 to 59 years and a trend toward an increase in absenteeism was found for employees functioning at lower and secondary vocational levels.

Our study stands out by the large sample size and the availability of employee records of study participants (recorded absenteeism, start/end and fraction of employment, job level, years of employment) as well as health promotion program enrollment dates of attendees. This resulted in an accurate calculation of absenteeism rates and baseline periods. Furthermore, potential confounding by age, sex, job level, years of employment and health program completion was corrected for in the statistical analysis.

An important limitation of our study is the fact that we were unable to distinguish between employees who did not enroll in the program because they were not invited, and employees who did not enroll because they were not interested. Both types of employee were classified as nonparticipants. On the basis of a participation rate of 34% that was found in another evaluation study of the same worksite health promotion program,20 we estimate that approximately 11,252 of the 20,797 study participants were invited during the study period, from which 3826 chose to enroll in the program. Another limitation is the fact that the study period was artificially split up into baseline and follow-up periods for nonparticipants at the fixed date of August 1, 2007. This way, no invitation bias could have influenced baseline absenteeism of the nonparticipants, because August 1, 2007, is the date invitation e-mails to participate in the worksite health promotion program were first send out to employees. On the downside, seasonal fluctuations in absenteeism may have introduced some bias in the results. With regard to the participation rate it is important to note that the expected company-wide absenteeism decline as a result of implementing this program is of course greatly affected by the participation of employees in the program.

Thus far, the only Dutch effectiveness study reporting about worksite health promotion in relation with absenteeism is the so called “Brabantia Project.”21 After the program, a decrease in absenteeism of 51% was shown in the intervention group compared with a decrease of 34% in the control group. However, in that particular study, the 14.3% to 15.8% baseline absenteeism rates were more then double the national average for workers in the light metal industry at that time, leaving much room for these rates to decline. Also, differences in absenteeism rates at follow-up were not adjusted for confounding variables and baseline absenteeism. Feedback to the participants regarding their risk status was limited to the biomedical measures of the assessment and interventions were not tailored at the individual level. For example, all participants from the experimental group received health education about alcohol use, regardless of individual alcohol intake. The worksite health promotion program evaluated in this study has a more sophisticated approach. Collected health data are translated into a personalized health report and advice, which is then used to drive behavioral change. The premise that lasting behavioral change can only be achieved when an individual is intrinsically motivated to change his or her behavior is a key underlying principle in this program. Therefore, the advice is tailored even further on the basis of the participant's readiness to change health related behaviors. Recently, Mills and colleagues8 reported about the effectiveness of a similar worksite health promotion program. In addition to a reduction in the cumulative count of health risk factors and an increase in on-the-job productivity, a significant reduction in absenteeism was found in the intervention group. An interesting future study would be to examine differences in effectiveness between “relatively simple” and “more sophisticated” health promotion programs.

It is postulated that changes toward a healthier lifestyle will reduce the risk for future chronic disease. In this study, however, an immediate impact of attending a worksite health promotion program on absenteeism was shown. The question is which pathway or underlying mechanism is responsible for this decline. Burton and colleagues4 reported that both alcohol use and smoking are relatively stable over a short period of time (2 years), whereas physical activity, weight, life dissatisfaction, and stress have the greatest amount of churn in a working population that is not participating in any particular health promotion program. For the worksite health promotion program in this study, 58% of the attendees initiate health behavior change after participation.20 It is plausible to assume that the initiated health behavior change has positively affected psychological well-being, either in isolation or combination with increased exercise and weight reduction. Obviously, these health risk factors are related, since physical activity not only contributes to weight loss but also is inversely related to depressive symptoms.22 Overweight has been associated with increased absenteeism23 and recently it was reported that for overweight individuals, a lack of physical activity increases absenteeism even further.24 Furthermore, by engaging the psychological self-help modules that are available in the worksite health promotion program, or by seeking counseling, a number of attendees probably have successfully addressed their mental health problems, whereas others could have diminished their stress levels, all of which resulted in an immediate effect on the absenteeism rate.

This study showed that participating in a worksite health promotion program can lead to an immediate reduction in absenteeism. Future research is necessary to identify the mechanisms responsible for this short-term effect.

Back to Top | Article Outline

REFERENCES

1. Gaining Health. The European Strategy for the Prevention and Control of Noncommunicable Diseases. Geneva, Switzerland: WHO Press; 2006.

2. Engbers L. Monitoring and Evaluation of Worksite Health Promotion Programs – Current state of knowledge and implications for practice. Geneva, Switzerland: WHO Press; 2007.

3. Burton WN, Chen CY, Conti DJ, Schultz AB, Pransky G, Edington DW. The association of Health risks and on-the-job productivity. J Occup Environ Med. 2005;47:769–777.

4. Burton WN, Chen CY, Conti DJ, Schultz AB, Edington DW. The association between health risk change and presenteeism change. J Occup Environ Med. 2006;48:252–263.

5. Ricci JA, Chee E. Lost productive time associated with excess weight in the U.S. workforce. J Occup Environ Med. 2005;47:1227–1234.

6. Bunn WB III, Stave GM, Downs KE, Alvir JM, Dirani R. Effect of smoking status on productivity loss. J Occup Environ Med. 2006;48:1099–1108.

7. Musich S, Hook D, Baaner S, Edington DW. The association of two productivity measures with health risks and medical conditions in an Australian employee population. Am J Health Promot. 2006;20:353–363.

8. Mills PR, Kessler RC, Cooper J, Sullivan S. Impact of a health promotion program on employee health risks and work productivity. Am J Health Promot. 2007;22:45–53.

9. Muto T, Yamauchi K. Evaluation of a multicomponent workplace health promotion program conducted in Japan for improving employees' cardiovascular disease risk factors. Prev Med. 2001;33:571–577.

10. Soler RE, Leeks KD, Razi S, et al. A systematic review of selected interventions for worksite health promotion. The assessment of health risks with feedback. Am J Prev Med. 2010;38:S237–S262.

11. Anderson J, Dusenbury L. Worksite cholesterol and nutrition: an intervention project in Colorado. AAOHN J. 1999;47:99–106.

12. Edye BV, Mandryk JA, Frommer MS, Healey S, Ferguson DA. Evaluation of a worksite programme for the modifıcation of cardiovascular risk factors. Med J Aust. 1989;150:574–581.

13. Bertera RL. Behavioral risk factor and illness day changes with workplace health promotion: two-year results. Am J Health Promot. 1993;7:365–373.

14. Goetzel RZ, Kahr TY, Aldana SG, Kenny GM. An evaluation of Duke University's Live for Life health promotion program and its impact on employee health. Am J Health Promot. 1996;10:340–342.

15. Arends I, Van der Klink JJL, Bültmann U. Prevention of recurrent sickness absence among employees with common mental disorders: design of a cluster-randomised controlled trial with cost-benefit and effectiveness evaluation. BMC Public Health. 2010;10:132.

16. Prochaska JO, DiClemente CC. Stages and processes of self-change in smoking: Toward an integrative model of change. J Consult Clin Psychol. 1983;51:390–395.

17. Murray C. Quantifying the burden of disease: the technical basis for disability-adjusted life years. Bull World Health Organ. 1994;72:429–445.

18. Wilson JMG, Jungner G. Principles and Practice of Screening for Disease. Geneva, Switzerland: WHO; 1968.

19. Kreuter MW, Bull FC, Clark EM, Oswald DL. Understanding how people process health information: a comparison of tailored and nontailored weight-loss materials. Health Psychol. 1999;18:487–494.

20. Colkesen EB, Niessen MA, Peek N, et al. Initiation of health-behaviour change among employees participating in a web-based health risk assessment with tailored feedback. J Occup Med Toxicol. 2011;9:5.

21. Maes S, Verhoeven C, Kittel F, Scholten H. Effects of a Dutch work-site wellness-health program: the Brabantia Project. Am J Public Health. 1998;88:1037–1041.

22. Sieverdes JC, Ray BM, Sui X, et al. Association between leisure-time physical activity and depressive symptoms in men. Med Sci Sports Exerc. 2012;44:260–265.

23. Neovius K, Johansson K, Kark M, Neovius M. Obesity status and sick leave: a systematic review. Obes Rev. 2008;10:17–27.

24. Van Strien T, Koenders P. How do physical activity, sports, and dietary restraint relate to overweight-associated absenteeism? J Occup Environ Med. 2010;52:858–864.

Cited By:

This article has been cited 1 time(s).

Journal of Medical Internet Research
Determinants of Participation in a Web-Based Health Risk Assessment and Consequences for Health Promotion Programs
Niessen, MAJ; Laan, EL; Robroek, SJW; Essink-Bot, ML; Peek, N; Kraaijenhagen, RA; Van Kalken, CK; Burdorf, A
Journal of Medical Internet Research, 15(8): -.
10.2196/jmir.2387
CrossRef
Back to Top | Article Outline

©2012The American College of Occupational and Environmental Medicine

Login

Article Tools

Images

Share