• Promoting a Healthier Workforce

    Health and Productivity Management

    landing-header-kc-hpm1_37748_17213
  • Health and Productivity Management Center

    Quantifiable Impact of the Contract for Health and Wellness: Health Behaviors, Health Care Costs, Disability, and Workers' Compensation

    Gregg M. Stave, MD, JD, MPH; Lamont Muchmore, MBA; Harold Gardner, MD

    Journal of Occupational and Environmental Medicine 2003; 45(2):109-117

    Learning Objectives

    • Distinguish between employers' and employees' prerogatives in setting up the Contract for Health and Wellness program.
    • Identify the degree of employer participation in the program, and the changes in health behaviors documented in the course of the 4-year study.
    • Comment on the cost savings realized by the company from the Contract program, and the chief underlying reason or reasons for these savings.

    Abstract

    Current literature about the long-term impacts of corporate health and wellness programs has brought to light new evidence about the cost savings associated with health-promotion interventions. A critical element in these initiatives is attracting the participation of employees at risk for high benefits use. This study presents evidence that suggests accomplishing this task has economic savings implications to large employers. A health and wellness intervention program offered at GlaxoSmithKline, entitled the Contract for Health and Wellness, is examined. Focusing on a group of 6049 employees, the study examines the impact on health behaviors and on integrated health benefits use of this continuously employed population from 1996 to 2000. Total benefits costs are examined for participants and nonparticipants, and the annual savings associated with the isolated impact of the program are, on average, $613 per participant. Reductions in disability costs accounted for the majority of these savings.


    From GlaxoSmithKline (Dr Stave); and Options and Choices, Inc., Cheyenne, WY (Mr Muchmore and Dr Gardner).

    Address correspondence to: Lamont Muchmore, MBA, Options and Choices, Inc., 2232 Dell Range Boulevard, Suite 300, Cheyenne, Wyoming 82009; E-mail: lamont-muchmore@oci.com.

    Lamont Muchmore serves as a consultant for Glaxo SmithKline.

    Journal of Occupational and Environmental Medicine 2003; 45(2):109-117
    Copyright © 2003 Lippincott Williams & Wilkins
    All rights reserved


    There is an increasing understanding that the health of a company's employees affects the success of the company. 1 A comprehensive corporate health effort directed at keeping employees healthy, productive, and working goes beyond treatment services to include prevention programs. Health-promotion programs are an important component of these prevention efforts. 2-8

    In 1996, GlaxoSmithKline developed a voluntary, incentive-based, umbrella program that unites and focuses all health promotion activities. This program, the Contract for Health and Wellness, was implemented in 1997. It incorporates the concept of contracting, that is, people will be more committed to completing a program if they sign a contract to do so.

    The program covers five focus areas: tobacco use, nutrition, activity, stress management/depression, and preventive health practices. These topics represent drivers of healthcare costs, areas of demonstrated interest for employees, and leading contributors to mortality. 9, 10 In fact, the leading contributor to mortality in the United States in 1990 was tobacco, followed by diet and activity patterns.2 Additionally, the contract differs from traditional health risk appraisal programs in that it focuses on the same health behaviors for all participants regardless of health status or health risks.

    The ongoing goals of the Contract for Health and Wellness are: 1) to promote a culture of health and productivity; 2) to encourage employees to adopt or maintain positive health behaviors; and 3) to provide economic benefit to the company through reduced healthcare expenditures and a higher level of productivity.

    In 1997, the first year of the program, employees were required to complete a contract for good health behaviors. Programs and services were offered to support these behaviors, however, there was no requirement to affirm that the agreed health behaviors were adopted or maintained. Beginning in 1998, the program was modified to incorporate an affirmation of health behaviors and program attendance at the end of the calendar year.

    The Contract for Health and Wellness covers health behaviors that are the major preventable cause of disability, disease, and death. The focus areas of the program are: 1) avoiding tobacco use; 2) encouraging physical activity; 3) encouraging appropriate nutrition; and 4) practicing stress management. The program also encourages preventive health screenings and immunizations.

    At the beginning of each year, employees complete a self-assessment for each focus area based on a stage of a readiness-for-change model.1 The five stages of this model are precontemplation, contemplation, planning, action, and maintenance. After determining where their behaviors fit along the readiness-for-change continuum, employees select the health behaviors they will commit to for the year. On-site seminars and programs are designed to meet the needs of employees in each focus area based on their stage of readiness-for-change. Feedback collected after every seminar is analyzed and used in designing subsequent programs.

    The Contract for Health and Wellness is promoted using a variety of marketing strategies. This includes making participation in the program a "vital sign" in on-site clinics, just as tobacco use is a vital sign. This means that information on health habits and program participation is elicited during clinical encounters whenever the traditional vital signs of blood pressure, pulse rate, respiratory rate, and weight are measured and recorded.

    Employees earn points for program attendance and for the health behaviors they have engaged in over the course of a year. At the end of the year, employees receive rewards based on the number of points earned. The average cost of these rewards is less than $50.

    This analysis evaluates the impact of the Contract for Health and Wellness on health behaviors and total health benefits cost for program participants.

    Methods

    Sample

    Data are drawn from the GlaxoSmithKline/OCI Integrated Database (GlaxoSmithKline - IDb), which includes integrated, person-centric data for almost 20,000 employees. The GlaxoSmithKline - IDb includes employment, healthcare, disability, absence, workers' compensation, and survey data from various sources.

    The study population contains 6049 GlaxoSmithKline employees who were continuously active and employed during a 5-year period from 1996 to 2000. The population is diverse with respect to job type, age, gender, and location. Descriptive statistics associated with the population are shown in Table 1.

    TABLE 1 Descriptive Statistics of the Population

    Table 1 Feb 03

     

    Participants

    Data

    Health behaviors in the contract focus areas were assessed by means of a self-report questionnaire at the time of contracting and at the end of each calendar year during program completion.

    Healthcare costs are obtained from claims transaction data linked to the individual by a unique identifier. Individuals in the study sample come from multiple health care plans offered by GlaxoSmithKline, including indemnity plans and managed care plans. Mental health costs are included as part of healthcare costs. Prescription drug costs are also available and are distinguished from other healthcare costs. Healthcare and prescription drug costs represent only employer-paid claims for services and exclude deductibles and copays. This means that only a portion of the actual charged fees are included in the cost, and that portion is only the amount paid by the employer. Additionally, only payments for services rendered to employees are included and all dependent costs are excluded.

    Both short-term (STD) and long-term disability (LTD) claims data are obtained from the employers' disability insurance carriers. Costs are determined according to the dollar amount of actual indemnity checks paid to the employee during the disability period. LTD costs include reserves that are established in order to estimate additional costs that will be paid in the future. All employees in the sample are covered by the same disability policy.

    Costs for workers' compensation include all costs associated with the claim, including healthcare costs, indemnity costs, legal fees, and reserves. As with LTD, all claim costs associated with a given claim are recorded in conjunction with the beginning date of the claim.

    The study examines the impact of Contract for Health and Wellness participation on health behaviors and total health benefits cost. Total health benefits include costs from the previously mentioned benefit categories: healthcare, mental health, drug, STD, LTD, and workers' compensation. Lost time from the STD, LTD, and workers' compensation benefit types has been cleaned to remove any potential overlap.

    Statistical Adjustments for Existing
    Differences Between Groups

    From a large set of available datafields, variables having a plausible causal relationship with total health benefits cost were chosen as covariates. The set of covariates includes age, tenure, gender, race, marital status, full-time/part-time status, employment status, and salary. Inclusion of these covariates in the regression models is to control for observed differences between participants and nonparticipants.

    Adjustments for Cost

    All costs are adjusted to represent 2000 dollars using inflation factors based on the Consumer Price Index (ftp://ftp.bls.gov/pub/special.requests/cpi/cpiai.txt). All costs are determined by using the service incurred date because the method of comparison involves participation in a program that has finite time frames. Claims and cost related to LTD and workers' compensation can extend far beyond this 5-year study period. Therefore, the severity of such cases must be accounted for by including reserve amounts.

    Data Analysis

    To ascertain the quantified impact of the Contract for Health and Wellness on total health-related benefits cost, regression modeling is used to isolate the impact of the Contract for Health and Wellness in each year (1997 through 2000). Regression models are used to evaluate each year separately. A detailed explanation of the methodology follows, beginning with the most recent year (2000).

    The Contract for Health and Wellness was first implemented in 1997. Therefore, this analysis uses 1996 as the base year for comparison purposes. The analysis quantifies the difference in each employee's cost from the base year (1996) to the current year (2000) and then compares this cost difference between employees who participated in the Contract for Health and Wellness in 2000 and employees who did not participate in the Contract in 2000. The comparison is made through the use of linear regression analysis. The dependent variable in the regression model is the change in cost from one time period to another. Therefore, the dependent variable does not suffer from any type of censoring or non-normality typically seen with cost variables in this type of analysis.

    Through the use of regression analysis, the estimated savings for a "typical" employee can be obtained. To obtain the total estimated savings for 2000, this result is multiplied by the total number of employees (in the sample) completing the Contract in 2000.

    The same methodology is used to estimate savings for years prior to 2000. The estimated savings for each year are summed to obtain the total estimated savings for the entire length of the program.

    Results

    Participation

    In each of the last 4 years, over 2000 employees in the study population fully participated in the Contract for Health and Wellness, representing almost 35% of eligible employees. For the purpose of this evaluation, only employees completing both an initial registration form and a final completion form at the end of the calendar year are considered participants, with the exception of 1997 when only an initial registration form was required. Employees completing the Contract for Health and Wellness represent about two-thirds of the employees that initially register for the Contract.

    Health Behavior Change

    A major goal of the Contract for Health and Wellness is to assist and motivate employees in adopting and maintaining healthy behaviors. Within each year of the program, there was substantial progression along the stages of readiness-for-change, with large increases in the percentage of participants in the maintenance phase in the nutrition, physical activity, and stress management categories. There was consistency between responses provided during the reconciliation of the contracts in December and responses provided the following January for employees who re-enrolled in the program. Similarly, there was a substantial increase in the proportion of employees in the maintenance phase for those who participated in the program during all three years of 1998-2000 (Fig. 1). For example, the percentage of employees who report eating at least five servings of fruits and vegetables each day increased from 35 to 67% among those who participated in the program for three years. Among other findings, the percentage of employees regularly practicing stress reduction techniques increased from 51 to 81%. Changes in tobacco use were small, however, the prevalence of tobacco use was relatively low at baseline (<10%).

    Fig. 1.
    States of change for multiple year participants. (This graphic represents responses from 1,275 employees in the study that participated in the Contract for Health & Wellness during 1998, 1999, and 2000.

    Fig 1 Feb 03Fig. 1.

    In support of employees seeking to enhance and maintain their health, on-site health enrichment and education programs are offered during lunchtime and after work. These programs are centered on the focus areas of the Contract. From 1997 through 2000, total attendance at these programs has increased more than 5-fold, from 3057 visits to 15,765 visits per year (Fig. 2). The number of employees potentially able to attend these on-site programs was 5727 in 1997 and 6398 in 2000. These programs are open to all employees-participation in the Contract was not required.

    Fig 2.
    Health enrichment attendance. (This graphic represents the total number of attendances in GlaxoSmithKline Health Enrichment Activities. All GlaxoSmithKline employees are eligible to attend regardless of participation in the Contract for Health & Wellness.

    Fig 2 Feb 03Fig. 2.

    Cost Comparison

    Cost savings associated with completing the Contract for Health and Wellness were substantial. Over the four-year program period (1997-2000), the isolated impact of the Contract for Health and Wellness represents a total cost savings to GlaxoSmithKline of roughly $5.5 million. Savings in 1997 are estimated to be over $500,000 for almost 2200 employee participants. Subsequent results show increases in the cost savings per employee over time, indicating the program has gained additional impact during the time frame of the study. The savings associated with study participants during the year 2000 were equivalent to $950 per employee or just over $2 million in total savings, the highest savings per employee over the four years. Annual results are presented in Fig. 3. The weighted average of these annual cost savings per employee is equivalent to $613 per employee.

    Fig. 3.
    Total estimated savings.

    Fig 3 Feb 03

     

    Given the magnitude of savings associated with the contract in each individual year, the next logical step is to measure the effectiveness of repeated participation in the program. To do this, savings associated with employees who completed the Contract for Health and Wellness every year for the 3-year period, 1998-2000, were examined. This subset of employees was compared to GlaxoSmithKline employees that did not complete the Contract in 1998, 1999, or 2000. (Participation in 1997 was not considered, as the Contract for Health and Wellness was slightly different during that year.) Demographic differences between the two populations are shown in Table 2.

    Table 2: Comparison of Demographics Between Groups

    Table 2 Feb 03

    Regression modeling was used to isolate the impact of the Contract for Health and Wellness on benefits cost over a 3-year period. The difference in average annual cost from 1998 to 2000 and the base year (1996) is compared between employees who participated in the Contract during 3 years and employees who did not participate in the contract during the same time frame. Employees who completed the Contract for Health and Wellness had significantly lower total costs (P < 0.02). This was equivalent to a $991,204 reduction in total cost on an annual basis or $777 per participant.

    Table 2also reports annualized cost and lost day values for both participants and nonparticipants. Although comparing these values between groups may be of some value, this only tells part of the story. It is important to remember that the analytic methodology employed in this analysis compares the cost difference (from a base year to the year of evaluation) between each group. It is with this more sophisticated methodology that the above estimate ($777 savings per participant) is calculated.

    A major challenge for any health promotion program is to attract employees that are high-risk in terms of health benefits use. A comparison of total costs prior to program implementation (in a base year), shows that participants actually had higher total health benefits costs, on average, than the nonparticipant group. This is true for all years (1997-2000) that the Contract for Health and Wellness has been offered. Further analysis using an independent samples t test, shows that the lower cost associated with the nonparticipant group in 1998 is statistically significant (P < 0.04).

    Discussion

    The Contract for Health and Wellness was successful at attracting a large percentage of employees to participate and in assisting participants in adopting and maintaining healthy behaviors. There was an estimated $5.5 million in total health benefit cost savings for participants.

    Participants in the program showed not only improvements in health behaviors but also progression through the stages of readiness-for-change. Emmons et al 8 have described similar positive behavior change results for a worksite program targeting physical activity, diet, and smoking.

    A major concern in the analysis of health and wellness programs is selection bias, especially when analyses compare participants to nonparticipants. It is reasonable to be concerned that employees that choose to participate in a voluntary program could be healthier than nonparticipants and consequently show lower benefits costs than nonparticipants. If that is the case, the participant group may have lower costs even in the absence of a wellness program.

    Employees participating in the Contract for Health and Wellness had higher costs, on average, before program implementation than did nonparticipants. Additionally, the methodology employed within this analysis compares the cost difference for each employee from the year of program participation to a base year before program implementation. Since the unit of analysis is the difference in cost, each individual acts as his/her own control and the effect of program participation can be measured independently of factors affecting the likelihood of participation. In this model, employees with the most opportunity to save are those employees with historically high health benefits use costs. This group of employees also represents more risk to the employer in terms of lost time and productivity. The magnitude of the savings found in this analysis indicates that many high-risk employees participated in the program.

    Additionally, the regression models control for demographic differences between participants and nonparticipants. In all years (except 1997, since the program required completion of only one survey that year) the study classifies employees who started but did not complete the Contract as nonparticipants. However, this group of employees likely receives some benefit from initial participation in the Contract. Subsequently, these employees may have lower overall costs than they would have had in the absence of the Contract for Health and Wellness. Including these employees in the comparison group may produce somewhat conservative estimates of cost savings associated with the Contract.

    The sample used in this analysis includes only GlaxoSmithKline employees that were continuously employed from 1996 to 2000. Other GlaxoSmithKline employees also participated but did not meet the criteria to be included in the sample. For example, in the year 2000, the total number of participants in the Contract was 3559 employees (38% of all eligible employees). Of the 3559 participants, 2122 (60%) met the criteria to be included in the sample. All estimates of cost savings represent only savings associated with the group of participants that are included in the sample.

    The impact of the Contract for Health and Wellness has increased over time (Fig. 3). This result is consistent with findings in other literature, Pelletier comments that "[E]xtending the length of follow-up does seem to be associated with a positive, measurable impact on the intervention.",12 It is also possible that enhancements made to the program between 1997 and 2000 increased the program's effectiveness. As mentioned above, the major change between 1997 and 1998 was the implementation of the follow-up questionnaire documenting employee goals and program completion. Participation in on-site health enrichment programs also increased dramatically over these years (Fig. 2).

    The financial impact associated with employees who completed the Contract for Health and Wellness during 1998, 1999, and 2000 represents annual estimated savings of $777 per employee. However, this annual estimate is smaller than the expected savings associated with employees who completed the Contract for Health and Wellness in 1999 ($944 savings per employee) or in 2000 ($950 savings per employee). This leads to a question about the residual or carryover effect of participation in the Contract. The results seem to indicate that the residual impact of the program into subsequent years may be limited in terms of short term cost savings. If the residual impact were significant, we would expect the annual estimated cost savings associated with the 3-year participant group to be higher than groups participating in any single year. Additional research in this area is necessary to fully understand the dynamics of a possible residual effect.

    Because benefits use among employees is shown to be maldistributed in previous literature, 13 the idea of cutting cost may not always be in direct correlation to decreasing Human Capital output. Previous research by Gardner et al13 has found that many employees with high benefits use consumption capital remain in a high-cost/high-risk group for many years. This group generally incurs a majority of benefits cost and is typically less productive (as measured in terms of lost time). This result should empower employers to think more about how they might work to decrease benefits cost while simultaneously increasing productivity.

    Results indicate that the Contract for Health and Wellness has simultaneously reduced total benefits cost for the participation group and increased productivity by reducing these costs, specifically disability use. The reduction in the use of disability is also related to less time away from work and subsequently increased productivity within the workgroup.

    It is also important to judge the savings associated with the contract relative to the cost inputs required to support the program. The estimated cost per participant is around $100. The weighted average of the yearly savings estimates per employee (Fig. 3) is equivalent to $613 per participant. Given a $100 investment per participant, this represents a return on investment of over 600%.

    The expected savings associated with employees who completed the Contract for Health and Wellness in 2000 was $950 per employee. Some of these savings come from reduced health care cost ($185). This finding is similar to savings estimates found in previous literature. In a study of the Live for Life program, Ozminkowski et al 14 found a $225 annual health care savings per employee.

    However, a majority of the savings associated with the Contract come from reduced use of the nonoccupational disability benefits STD ($217) and LTD ($545). Savings associated with workers' compensation were nominal and prescription drug costs increased slightly, offsetting the savings associated with workers' compensation. It is important to note that the majority of LTD costs are reserved, or projected, costs rather than actual, incurred costs. So, the magnitude of savings associated with LTD may vary somewhat from the above estimate, depending on the accuracy of the reserving practices in this particular instance. However, it is important to realize that reserving practices have been established to insure that institutions will plan appropriately for future costs associated with these claims. Although we don't know exactly what these costs are currently, it is in the best interest of the institution to estimate these future costs as accurately as possible. Therefore, it is appropriate to include these costs in this analysis.

    The finding that health promotion program participation is associated with reduced use of disability benefits is consistent with several recently published studies. Schultz et al. have shown that participants in a health promotion program in a manufacturing setting showed lower absenteeism and use of disability days compared with nonparticipants. 5 Serxner et al 7 studied a worksite health promotion program in a large telecommunications company and reported a reduction of disability days among self-selected participants compared with an increase in disability days for nonparticipants.

    Possible explanations for finding that the majority of cost savings are associated with reduced use of disability benefits were not explicitly evaluated in this study. One possibility is that people who participated in, and benefited from, the Contract felt a greater sense of company "loyalty," and were more likely to want to be at work when possible. As a result, they may have tried to avoid or minimize the use of disability benefits. There is some indirect evidence in support of this "cultural effect" hypothesis. When the company was recognized as the best place to work in North Carolina by the NC Business Journal in 1999, many of the programs offered in conjunction with the Contract were cited by employees as reasons they like working for the company. An assessment of a possible cultural effect would be a useful component of future research.

    The longitudinal nature of this program presents a robust set of data to analyze following implementation of the Contract for Health and Wellness. Many previous studies have focused on shorter time frames, such as 1 and 2 years postimplementation. 15 Savings may not be as well defined in these early years. This study examines a broad spectrum of data and the addition of detailed disability data and workers' compensation data reveal savings opportunities outside of the traditional healthcare insurance and prescription drug evaluations. Finally, all employees in the sample were eligible for all health-related benefits in all four years. Having enrollment data is insightful, as some employees filed no claims for healthcare, prescription drugs, workers' compensation, or nonoccupational disability during the time period. For example, in the year 2000, 598 employees had no health care, prescription drug, workers' compensation, or nonoccupational disability claims. This represents almost 10% of the entire sample. Inclusion of these data are informative, as many previous studies focused solely on claims data, possibly missing the roughly 10% of the population that does not incur any claims cost in a given year.

    As discussed, additional research may focus on estimating the residual impact of participation into subsequent years. Furthermore, some additional insight might be gained by altering the dependent variable, if possible, to include measures of sick leave and indirect costs such as lost time and decreased productivity.

    Although it is gratifying to be able to demonstrate short-term cost savings associated with some health promotion programs, it seems ironic that as a society we demand a short-term return on investment for primary prevention efforts. With increasing concerns over the cost of treatment and our willingness to pay for screening programs that cost more than $10,000 per quality adjusted life year, health promotion programs should be viewed by employers as a long-term investment with some potential for short-term savings.

    Limitations

    As with many studies including regression modeling, there are potentially other covariates that may affect the results, but these variables were not available for analysis. These variables may include other programmatic interventions at GlaxoSmithKline, merger and acquisition activities, and other operational/managerial changes that could not be accounted for in the regression model. The study did control for the impact of as many variables as were available.

    Although the time frame of this study is long in comparison to similar studies, it may take many years for the impacts of health promotion activities to fully develop. Diseases such as cirrhosis of the liver may be a function of alcohol abuse over a period of many years, and the serious cost implications of the disease may impact the individual for 10, 20, or even 40 years. This concept is especially true for the five program focus areas: tobacco use, nutrition, activity, stress management, and preventive health strategies. The most costly implications of these health habits are similar to the example above and may not be evident in the study time frame. Consequently, the results of this study may be conservative in light of longer-term effects.

    Inherent in the study are many factors specific to the GlaxoSmithKline employee population. Benefits structure and design, corporate culture, management philosophy, and workplace processes may all contribute to the effects of the program, but may not be applicable to other organizations. Therefore, when attempting to generalize these results to other populations, caution must be exercised to consider the scope of the study and factors that make the GlaxoSmithKline population unique.

    A valuable data element excluded from this analysis is sick leave data, as detailed sick leave data were available only on a portion of the sample. This precludes sick leave costs from being included as part of the dependent variable. In examining some elements around the cost of sick leave, our best estimate is that the inclusion of this variable would lead to an increase in total health benefits cost savings. Because there is a documented decrease in nonoccupational disability costs and evidence of a reduction in medical leave costs, the cost of lost productivity may also be a large factor that is not explicitly considered in this analysis.

    The Contract focuses on health behaviors. Biometric information, such as body mass index and lipid levels, was not collected. It was therefore not possible to determine whether the improvements in self reported health behaviors were associated with improvements in metabolic or physiologic parameters.

    Conclusion

    Employees who participated in the Contract from 1998 through 2000 show substantial improvement in their awareness and performance of behaviors emphasized by the Contract's five focus areas. This indicates that the Contract is successfully influencing health behavior and leading participants toward the goal of a healthier lifestyle.

    Results also indicate that cost savings associated with the impact of the Contract for Health and Wellness are substantial. Over the 4-year period from 1997 to 2000, the total estimated savings associated with the Contract are $5.5 million. Additionally, the study finds the impact of the Contract for Health and Wellness has increased over time. The average estimated savings per employee have increased from $233 in 1997 to $950 in 2000. This represents an increase in savings of over 307%.

    The analysis also examines the relationship between employees who enrolled in the Contract for Health and Wellness in three consecutive years (1998 to 2000) and total health-related benefits cost. The findings show that the average annual estimated savings are equivalent to $777 per employee. The total savings associated with this group of 1275 employees are almost $1 million annually.

    Acknowledgments

    We wish to thank Justin Schaneman at Options & Choices, Inc. for help with data organization and definition.

    References

    1. Peterson K, Yarborough C, Ferguson E, Matthew S. The American College of Occupational and Environmental Medicine's Corporate Health Achievement Award. J Occup Environ Med. 1996; 3810: 969-972.

    2. Baun W, Bernacki E, Tsai S. A preliminary investigation: effect of a corporate fitness program on absenteeism and health care cost. J Occup Med. 1986; 281: 18-22.

    3. Jacobson B, Aldana S. Relationship between frequency of aerobic activity and illness-related absenteeism in a large employee sample. J Occup Environ Med. 2001; 4312: 1019-1025.

    4. Musich S, Napier D, Edington D. The association of health risks with workers' compensation costs. J Occup Environ Med. 2001; 436: 534-541.

    5. Schultz A, Lu C, Barnett T, et al. Influence of participation in a worksite health-promotion program on disability days. J Occup Environ Med. 2002; 448: 776-780.

    6. Serxner S, Gold D, Bultman K. The impact of behavioral health risks on worker absenteeism. J Occup Environ Med. 2001; 434: 347-354.

    7. Serxner S, Gold D, Anderson D, Williams D. The impact of a worksite health promotion program on short-term disability usage. J Occup Environ Med. 2001; 431: 25-29.

    8. Emmons K, Linnan L, Shadel W, Marcus B, Abrams D. The working healthy project: a worksite health-promotion trial targeting physical activity, diet, and smoking. J Occup Environ Med. 1999; 417: 545-555.

    9. US Department of Health and Human Services. Healthy people 2000. Washington, DC: USDHHS, Public Health Services. 1990. DHHS Publication No. (PHS) 91-50212.

    10. McGinnis J, Foege W. Actual causes of death in the United States. J Am Med Assoc. 1993; 27018: 2207-2212.

    11. Prochaska J, Velicer W. The transtheoretical model of health behavior change. Am J Health Promot. 1997; 121: 38-40.

    12. Pelletier K. A review and analysis of the clinical and cost-effectiveness studies of comprehensive health promotion and disease management programs at the worksite: 1995-1998 update. Am J Health Promot. 1999; 136: 333-345.

    13. Gardner H, Butler R, Gardner B, Bradshaw B. Pareto analysis of employee health benefits costs. Minneapolis, Minnesota: Minnesota Center for Insurance Research Publication; Winter 1999.

    14. Ozminkowski R, Ling D, Goetzel R, et al. Long-term impact of Johnson & Johnson's health & wellness program on health care utilization and expenditures. J Occup Environ Med. 2002; 441: 21-29.

    15. Aldana S. Financial Impact of health promotion programs: a comprehensive review of the literature. Am J Health Promot. 2001; 155: 296-320.

    Journal of Occupational and Environmental Medicine 2003; 45(2):109-117
    Copyright © 2003 Lippincott Williams & Wilkins
    All rights reserved

    Go top