Burton, Wayne N. MD; Chen, Chin-Yu PhD; Li, Xingquan MS; Schultz, Alyssa B. PhD; Edington, Dee W. PhD
* Discuss the potential role of workplace wellness programs in addressing racial/ethnic health disparities in employee population.
* Summarize the reported differences in health risks between ethnic groups at a large financial services company.
* Review changes in health risks in the study population, including racial/ethnic groups with larger or smaller changes in risk and specific risks within groups.
Health disparities among different racial/ethnic groups are pervasive, but there are relatively few employer-based health promotion programs that have measured their impact on health disparities among employees.1 The Agency for Healthcare Research and Quality reports annually on the differences in health factors associated with ethnicity in the United States.2 Although awareness of these disparities is growing on a national and international basis, employers rarely have access to data specifically related to health disparities for their employee populations. Similarly, no published studies, to our knowledge, have evaluated employee health promotion programs on how they may or may not have impacted diverse workforces through program participation and health risk change.
Currently, ethnic minorities, including Hispanics, African Americans, and Asians, represent about 36% of the total US population.3 The US Census Bureau projects that minorities will become the majority of the US population in the year 2042. In 2050, the working-age population in the United States is estimated to be 30% Hispanic, 15% African American, and 10% Asian compared with 15%, 13%, and 5%, respectively, in 2010.4
The Agency for Healthcare Research and Quality has shown that minority ethnic groups in the United States have worse access to care than whites in several core measures and that nonwhite groups also receive worse quality of care than whites.2 But these differences are likely confounded by income disparities because people below the federal poverty level receive worse care than high-income people (income greater than four times the poverty level) in 80% of the core measures.2 Studies have also shown that ethnic groups, even though they have medical insurance, may not receive appropriate preventive services.5–13 These differences in access to care and quality of care received should be better in an employed population than in the general population. Health risks, morbidity, and mortality are known to vary by ethnicity as well.14–16 Do these differences remain in a working population? The purpose of this study was to examine health disparities among the ethnic groups at a major corporation as well as the change in health risks over time at this corporation committed to helping employees maintain and improve their good health.
This is a retrospective observational analysis of a longitudinal cohort of employees from a Fortune 100 financial company based in the United States. This corporation has approximately 27,000 employees at multiple worksites in the United States, of whom 23,088 were enrolled in the company's medical plain in 2010. The firm has a diverse workforce with an active Corporate Diversity Council together with numerous employee diversity groups at the major US worksites. In 2010, the average age of the employee population was 42.3 years, and 65% of the workforce was female. The ethnicity distribution was as follows: 64% whites, 14% African American, 11% Hispanic, 8% Asian, and 2% other or multiracial. The demographics of this company's workforce were statistically similar in 2009 and 2010. A data warehouse was created, which comprised medical and pharmaceutical claims, health risk appraisal (HRA) responses, and personnel information including employee self-identified ethnicity designations. All data used in this study were linked and de-identified prior to export to the University of Michigan Health Management Research Center (Ann Arbor, MI) for analysis. This study was conducted in accordance with the University of Michigan's Institutional Review Board.
Starting in 2008, employees at this corporation were eligible to participate in a Consumer Directed Health Plan. These types of plans are typically characterized by relatively large deductibles before coverage begins. Some preventive care measures such as routine physical examinations, immunizations, screening blood tests for diabetes and lipid abnormalities, and cancer screening testing (eg, mammography and colonoscopy) are generally covered at 100%. A health savings account (HSA) is commonly used by employees to cover some or all of their noncovered medical costs. By providing patients with more information about the cost of different treatment options, Consumer Directed Health Plans are one proposed strategy for reducing the large increases in health care costs paid by employers.
The corporation first launched a comprehensive wellness program in the United States in 2009 in response to a comprehensive medical claims analysis and rising health care cost trends. The goals of the program included the following: (1) improve employee and dependent health, (2) create a culture of health, (3) manage medical cost trends, (4) improve employee performance, and (5) support the firm's focus on customer service. The wellness program included (1) free generic preventive drugs and vitamins for conditions such as diabetes, hyperlipidemia, and hypertension, (2) no cost in-network preventive care such as annual physical examinations and vaccinations, (3) on-site and telephonic health coaching and dietician consultations, (4) telephonic condition management for diseases such as diabetes, asthma, and cardiovascular disease, (5) HRA, and (6) biometric testing for total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglycerides, blood sugar, and blood pressure.
Nonsmoking employees pay $30 per pay period (24 pay periods annually) less for medical insurance than smokers. This nonsmoker premium also applies to covered spouses/domestic partners. Smoking cessation programs, including nicotine replacement therapy, are available at no cost.
Other programs and services available to employees include on-site clinics at major worksites staffed by health care professionals such as nurses, nurse practitioners, physicians, health coach and/or dietician, and other health care professions such as dermatologists who provide periodic screenings. All major worksites have an activity room for low-impact exercise (eg, yoga) or a fitness center equipped with various exercise machines such as treadmills, weights, bicycles, and other fitness equipment. A variety of worksite educational programs are offered tailored to the needs of the workforce such as weight management and exercise programs. Various incentives are offered including $100 for employees and $50 for spouses/domestic partners who are members of the firm's medical plan, which is deposited into the employee HSA for use for medical and other eligible expenses. All employees and their families have access, at no cost, to counseling services, both telephonic and face-to-face visits with a network of mental health counselors.
Assessing Health Risks and Status
An intranet link to an online HRA questionnaire was sent to all US employees in September 2009 and again in October 2010. A paper-based version of the HRA was also available; nevertheless, it was used by only a few employees. Employees who completed the HRA were provided with an individualized report regarding their health risks and suggestions for health improvement. A $100 incentive for completing the HRA was deposited into the employee's health spending account (HSA). Confidentiality of the HRA questionnaire responses was ensured by having the respondents reply directly to the vendor providing the HRA. Only aggregate HRA results were provided to the employer.
The HRA included questions regarding biologic and lifestyle health risk factors. Biometric testing was available at no cost to the employee at worksite screenings at the major worksite locations. The biometric testing included blood tests for blood glucose, total cholesterol, high-density lipoprotein cholesterol, calculated low-density lipoprotein cholesterol, triglyceride, and blood pressure. Height and weight were self-reported by the respondents. Biometric data were obtained from a blood specimen via a finger stick and analyzed on a Cholestech machine. Employees were asked to fast for the testing. Criteria for the 11 health risks can be found elsewhere.17 The cut point for obesity for most employees was 30 kg/m2. Research has suggested that Asian adults experience negative outcomes such as diabetes and cardiovascular disease at body mass indexes higher than 25 so that a different cut point of 25 kg/m2 is used to identify obese risk for Asians in this study.18–20 Each employee was classified as either having the particular health risk or not according to each specific risk criterion. To assess overall health risk status, the number of health risks was summed for each employee with low-, medium- and high-risk levels defined as zero to two risks, three to four risks, and five or more risks, respectively.
Approximately 55% of employees completed an HRA in 2009 and 2010. A total of 7252 employees who were enrolled in the company's medical plan also completed an HRA in both 2009 and 2010 and were the population of interest in this study for assessing risk change over time. The medical and pharmacy claims for 2009 and 2010, which were paid by the company, were summed for each employee. These amounts do not include out-of-pocket expenses such as copays and deductibles paid by the employee.
The t test and chi-square tests were performed for continuous variables and categorical variables, respectively. Differences in the prevalence of health risks from 2009 to 2010 were tested with McNemar's test. General linear multiple regression models were applied to adjust for confounding demographic characteristics when comparing differences among different ethnicities. All analyses were conducted using SAS 9.0 Software (SAS Institute Inc, Cary, NC).
The UMHMRC analyzed de-identified data including HRA and personnel data including ethnicity for employees who completed an HRA in both 2009 and 2010. Demographics of the HRA participants compared with the employee population as a whole can be found in Table 1. A slightly larger percentage of the two-time HRA participant group was female compared with the employee population as a whole (67.8% vs 64.6%; P < 0.0001). Similarly, the HRA participants were more likely to be whites (66.5%) than the entire employee group (63.9%) (P < 0.0001). The demographics of respondents included the following: whites (N = 4823; 66.5%), African American (N = 1009; 13.9%), Hispanic (N = 771; 10.6%), Asian (N = 565; 7.8%), and self-reported multiracial or other (N = 84; 1.2%). Because the number of employees in the multiracial/other group is so small, we have not presented their data regarding health risks.
Table 2 shows the demographics and health risks of each ethnic group in 2009 and 2010. As shown in the table, a large majority of the African American employees are female (83.0%), which is significantly different from all other groups. Asian employees have the smallest percentage of female employees compared with the other groups (58.7%). There are also significant differences in the average age of each ethnic group, with Asians being the youngest (38.0 years) and whites being the oldest (43.4 years).
The health risk prevalence, when displayed by ethnic group, shows many interesting differences. For almost every measured health risk, Asians had the smallest percentage of employees at high risk in 2009 (exceptions are safety belt use and alcohol use). This is reflected in their average number of risks, which was 1.41 in 2009. African Americans had the highest prevalence of 4 health risks at baseline compared with other ethnic groups (eg, obesity, blood pressure, safety belt use, and physical activity). Whites had the highest prevalence of cholesterol, smoking, alcohol, disease, and stress risks in 2009 compared with the other ethnicities. Hispanic employees had the greatest prevalence of the job dissatisfaction and life dissatisfaction risks. As reflected in the differences in health risk profiles, whites, African Americans, and Hispanic ethnic groups also had a different number of average health risks at baseline (1.67, 2.04, and 1.66, respectively). The average number of risks among African American employees was significantly higher than that among other ethnic groups (P < 0.0001) after adjusting for age and sex.
When examining the changes in health risks from 2009 to 2010, all ethnic groups experienced an increase in the following health risks: blood pressure, cholesterol, disease, and job dissatisfaction. Health risks that showed improvement for all groups during the study period were smoking, safety belt use, and stress. Stress had the highest magnitude of improvement of any of the health risks.
In 2010, the percentage of employees at high risk for blood pressure varied by ethnicity, with African Americans, whites, Hispanics and Asians having 30.4%, 18.0%, 16.9%, and 10.3% at high risk, respectively. The percentage of employees at risk for obesity also varied by ethnicity, with African Americans, whites, Hispanics, and Asians having 51.5%, 30.4%, 31.0%, and 36.6% at high risk, respectively. Physical inactivity was a high risk for 18.3% of African Americans, 14.6% of whites, 15.0% of Hispanics, and 15.2% of Asians.
Figure 1 shows the overall risk status (low, medium, and high) distribution of the study population by ethnicity. Employees are classified as being at low-, medium-, and high-risk categories on the basis of the presence of 0 to 2, 3 to 4, or 5 or more health risk factors, respectively. Time 1 (2009) and time 2 (2010) results are compared for each ethnic group so that we can see the change in overall health status during the study period.
African American employees had the smallest percentage of participants at low risk at baseline (66.5%), with Asians having the highest percentage of low risk (82.3%). Nevertheless, African American employees had the greatest improvement in the percentage of employees at low risk, with 69.8% in 2010, an increase of 3.3 percentage points. African American employees also had a significant improvement in the percentage of employees with 0 health risks (from 15.7% in 2009 to 19.0% in 2010; P < 0.05; data shown in Table 3). White and Asian employees also showed an increase in the percentage of employees in the overall low-risk category (whites: 73.8% in 2009 to 75.0% in 2010; Asian: 82.3% in 2009 to 82.7% in 2010). Hispanic employees had a slight decrease in the percentage of employees in the low-risk group (down 0.3 percentage points).
Table 3 shows the average annual health care and pharmacy costs for each ethnic group in 2009 and 2010. The total health care cost is the sum of the medical and pharmacy costs paid by the corporation and does not include employee HSA or out-of-pocket expenditures. For the entire employee population, the average total health care costs increased about $351 (10.5%), from $3355 to $3706. All ethnic groups saw an increase in overall medical and pharmacy expenditures. Asians, with the lowest baseline medical and pharmacy costs in 2009, had the highest percentage increase in total costs from 2009 to 2010 from $2161 to $2482, a difference of $321 or 14.9%. White employees had the highest baseline health care costs ($3613), which increased to $4005 in 2010. This was an increase of $392 or 10.8%. African American employees experienced an 8.5% increase in total costs, and Hispanic employees' costs increased 8.1% from 2009 to 2010. There is no significant difference in the percentage increase among ethnicity groups after adjusting for age, sex, and baseline costs.
The corporation is committed to helping employees maximize their health and improve their health behaviors and has many programs and strategies spearheaded by the Benefits and Medical departments. Similarly, this corporation has long been a leader in corporate diversity efforts and has regularly been noted in the DiversityInc Top 50 list of companies. Their Corporate Diversity Council meets quarterly and has developed employee networks, which are used for recruitment, to give new employees a sense of community, mentoring programs, sponsorship initiatives, talent development opportunities, and business outreach. Along with that, the organization provides top-rated benefits and work/life programs and is consistently ranked as one of the top companies to work for in America. The company is committed to helping employees be as healthy as possible and is concerned about any possible health disparities among employees.
This study of health risks and health risk changes found significant differences in the health risks of employees by ethnic group. The average number of health risks was similar for whites and Hispanics (1.67 and 1.66 in 2009, respectively), but the prevalence of different health risks varied greatly among those groups. Asian employees were found to have the lowest average number of health risks (1.41 in 2009) using the modified obesity risk cut of 25 kg/m2 or more, which is used for Asian populations. After age and sex adjustment, Asians' average number of health risks was not significantly different from whites or Hispanics in this employee population. African American employees had the highest average number of health risks (2.04 in 2009).
The efforts of the organization's Healthy Living programs may have yielded some positive results. In each ethnic group, the percentage of employees with 0 health risks increased from 2009 to 2010, and the average number of health risks improved for all groups. Furthermore, the percentage of employees at high risk for smoking, safety belt use, and stress all improved during the study time for each ethnicity. Studies have shown that as employees move to lower risk categories, their medical, pharmacy, and disability claims decline and there is improvement in on-the-job productivity (“presenteeism”).21–23 The percentage of employees at high risk for obesity decreased or remained the same for all four ethnicities from 2009 to 2010. As the employee population ages and as the US population as a whole continues to gain weight, maintaining and not gaining weight is a positive finding.
Among all risk factors, the decline in stress showed the greatest magnitude of improvement. The HRA did not inquire about the reason for stress, whether it was job or nonjob related. A national survey has found that work is a source of stress for more than 70% of US adults.24 Other major contributors of stress include finances, the economy, relationships, family responsibilities, health, and others. The baseline HRA was conducted in 2009 when there was a significant global financial crisis. Financial services companies, including the one used for this study, laid off employees, and no salary increases were awarded among other expense control measures during this period. In 2010, employees were again eligible for salary increases and hiring resumed.
African American employees had the greatest percentage improvement in the overall low-risk category (0 to 2 total risk factors) from 2009 to 2010 but remained the ethnicity with the largest percentage of medium- and high-risk employees. This group reported an increase from 66.5% of their population at low risk in 2009 and 69.8% or 3.3 percentage points net improvement compared with 1.2 percentage points net change in low risk for the white employee population. Asian employees had the greatest percentage at low risk, 82.3% in 2009 and 82.7% in 2010.
This corporation, like most companies in the United States, experienced increases in health care costs among employees from 2009 to 2010. The average annual increase in costs ranged from 8.1% for Hispanic employees to 14.9% for Asian employees. These costs are not adjusted for inflation, which was 3.4% in 2010. Therefore, the inflated-adjusted increases in costs ranged from 4.7% to 11.5%. The company has ongoing efforts to improve benefit plan structures and Healthy Living initiatives to help moderate those cost increases.
The current data analysis provided by the UMHMRC has identified a number of opportunities to address health disparities for this employer through the company's worksite health clinics, their wellness program, and in partnership with the health plans offered to employees and workplace employee diversity groups for employees of various ethnicities. Future research can evaluate the impact of those programs on mitigating the different health risks and health risk improvements we observed among employees of different ethnicities.
There are several limitations to this study that should be noted. The health risk factors, with the exception of biometrics, were self-reported by employees. For example, the percentage of respondents who report smoking is significantly lower than that reported in government studies for the overall US adult population. This may be the result of several factors including the employer policy of a smoke-free workplace, free smoking cessation programs offered for employees, and the socioeconomic status of this workforce. Underreporting of smoking may also be related to the nonsmoker lower health insurance premium offered to employees.25 Certain workforce demographic factors were not available for this study including education and income levels. The income level of employees in a consumer-directed health plan may affect health care utilization.26,27 Finally, the health care costs have not been adjusted for the region of the country. Certain cities may have different health care utilization and expenditures for the same procedures, testing, and medical care.28 Furthermore, although the reliability of this instrument has been tested in general employee populations, no ethnicity-specific testing was performed to determine any differences by ethnic or sociodemographic groups.
In the US-based employee population of this large financial services corporation, we observed several differences in the health risks and outcomes of health care costs among different ethnicities. This employer has an active diversity council that seeks to eliminate disparities among different ethnic groups, and the difference in health risks is something they will seek to remediate. It is likely that other large employers may also find health risk differences among employees belonging to various ethnicities, just as we have long been aware of age and sex differences in health risks among employees.29–31 The future of population health management research may seek to understand the reasons behind those differences and how best to address health disparities so that all employees can achieve a high level of health and wellness.
1. Dan D, Finch R, Harrison D, Kendall D. An Employer's Guide to Reducing Racial and Ethnic Disparities in the Workplace. Washington, DC: National Business Group on Health; 2011.
2. US Department of Health and Human Services, Agency for Healthcare Research and Quality. 2010 National Healthcare Disparities Report. AHRQ Publication No. 11-0005. Rockville, MD. Published 2011. Available at http://www.ahrq.gov/qual/qrdr10.htm
. Accessed September 2012.
5. DeLaet DE, Shea S, Carasquillo O. Receipt of preventive services among privately insured minorities in managed care versus fee-for-service insurance plans. J Gen Int Med. 2002;17:451–457.
6. Simpson SM, Krishnan LL, Kunik ME. Racial disparities in diagnosis and treatment of depression: a literature review, 2006. Psychiatr Q. 2007;78:3–14.
7. Hertz RP, McDonald M, Unger AN, Lustik M. Racial and ethnic disparities in the prevalence and management of cardiovascular risk factors in the United States workforce. J Occup Environ Med. 2007;49:1165–1175.
8. Hertz RP, Unger AN, McDonald M, Lustik MB, Biddulph-Krentar J. The impact of obesity on work limitations and cardiovascular risk factors in the U.S. workforce. J Occup Environ Med. 2004;46:1196–1203.
9. Hertz RP, Unger AN, Ferrario CM. Diabetes, hypertension, and dyslipidemia in Mexican Americans and non-Hispanic whites. Am J Prev Med. 2006;30:103–110.
10. Hertz RP, Unger AN, Cornell JA, Saunders E. Racial disparities in hypertension prevalence, awareness, and management. Arch Intern Med. 2005;165:2098–2104.
11. Charles H, Good CB, Hanusa BH, Chang CC, Whittle J. Racial differences in adherence to cardiac medications. J Nat Med Assoc. 2003;95:17–27.
13. Centers for Disease Control and Prevention. CDC health disparities and inequalities report—United States. Morb Mortal Wkly Rep. 2011;60:35–37.
14. Murray CJL, Klukarni SC, Michaud C. Eight Americas: investigating mortality disparities across races, counties, and race-counties in the United States. PLoS Med. 2006;3:1513–1524.
15. Wong MD, Shapiro MF, Boscardin WJ, Ettner SL. Contribution of major diseases to disparities in mortality. N Engl J Med. 2002;347:1585–1592.
16. Cooper R, Cutler J, Desvigne-Nickens P, et al. Trends and disparities in coronary heart disease, stroke, and other cardiovascular diseases in the United States: findings of the national conference on cardiovascular disease prevention. Circulation. 2000;102:3137–3147.
17. Burton WN, Chen CY, Li X, Schultz AB, Kasiarz D, Edington DW. Evaluation of a comprehensive employee wellness program at an organization with a consumer-directed health plan from 2009 to 2010. J Occup Environ Med. In press.
18. WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004;363:157–163.
19. Deurenberg P, Deurenberg-Yap M, Guricci S. Asians are different from Caucasians and from each other in their body mass index/body fat per cent relationship. Obes Rev. 2002;3:141–146.
20. Razak F, Anand SS, Shannon H, et al. Defining obesity cut points in a multiethnic population. Circulation. 2007;115:2111–2118.
21. Pelletier KR. A review and analysis of the clinical and cost-effectiveness studies of comprehensive health promotion and disease management programs at the worksite: update VII 2004–2008. J Occup Environ Med. 2009;51:822–837.
22. Pelletier KR. A review and analysis of the clinical and cost-effectiveness studies of comprehensive health promotion and disease management programs at the worksite: update VIII 2008 to 2010. J Occup Environ Med. 2011;53:1310–1331.
23. 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.
25. Connor GS, Schofield-Hurwitz S, Hardt J, Levasseur G, Tremblay M. The accuracy of self-reported smoking: a systematic review of the relationship between self-reported and cotinine-assessed smoking status. Nicotine Tob Res. 2009;11:12–24.
26. Ross JS, Bradley EH, Busch SH. Use of health care services by lower-income and higher-income uninsured adults. JAMA. 2006;295:2027–2036.
29. Liang W, Shediac-Rizkallah MC, Celentano DD, Rohde C. A population-based study of age and gender differences in patterns of health-related behaviors. Am J Prev Med. 1999;17:8–17.
30. Verbrugge LM. Gender and health: an update on hypotheses and evidence. J Health Soc Behav. 1985;26:156–182.
31. Macintyre S, Hunt K, Sweeting H. Gender differences in health: are things really as simple as they seem? Soc Sci Med. 1996;42:617–624.
Copyright © 2013 by the American College of Occupational and Environmental Medicine