Tsai, Shan P. PhD; Bhojani, Faiyaz A. MD, DrPH; Wendt, Judy K. MPH
Over the years, occupational medicine has gradually shifted from an interest in how occupational exposures pose acute health threats, to an emphasis on chronic diseases and, more specifically, to looking at the occupational setting as a potential site for delivering health interventions. The work setting is a particularly well-suited point for intervention because a large number of adults who may not normally take an active role in reducing their health risk factors can be reached relatively easily and inexpensively.1 Worksite health promotion and disease prevention programs are also particularly appealing to the bottom line as they pay for themselves in reduced illness-absence and medical expenditures.2–5
Although research has long established the association between health risks and various health endpoints, only more recent research has explored the relationship between risk factors and business outcomes, primarily in the loss of productivity resulting from illness-related absences.6–11 Although the potential for improving long-term health, ie, mortality, in the workplace through worksite health promotion programs seems obvious, scientific evidence remains very limited.
Long-term mortality in a working population is a complex phenomenon, influenced by many factors such as age, gender, education, personal health risk factors, and work-related factors. Numerous studies have examined the relationship between exposure to specific chemicals and cancer.12 Although the relationship between health risk factors and mortality, especially cardiovascular disease (CVD), have been examined extensively, most were based on nonoccupational populations.13,14 In the current study, a prospective study design was used to examine the influence of health risk factors on long-term mortality. The purpose of this study is to use prospectively collected employee physical examination and mortality data to assess and quantify the impact of selected health risk factors on overall mortality and CVD mortality in a large industrial population. In addition, the study will also examine the impact of accumulating risk factors on mortality.
Materials and Methods
The study population consisted of all company employees aged 30 years or older who were hired and had one or more physical examinations before January 1, 1985, and worked between January 1, 1985, and December 31, 2005. Most were employed at petrochemical manufacturing facilities. A total of 12,896 employees were identified through the Shell Health Surveillance System (HSS), the system used by the company to store health-related data and monitor employee health.15 Eligible employees who did not participate in any physical examination programs were excluded (n = 15,798). Demographic information were derived from company payroll and personnel systems and included, but was not limited to, date of birth, race, gender, date of hire, date of retirement or last separation, date of death, job title, and pay status (hourly and salaried).
Employee biometric and health risk factor data extracted from the HSS included all employee pre-placement and periodic physical examinations since January 1978. These examinations included those mandated by Occupational Safety and Health Administration and those sponsored by the company. Proportionally, more employees from refinery and petrochemical plants participated in these examinations than from crude oil exploration and production operations. For employees who had more than one examination, the most current examination data prior to the start of the study period, January 1, 1985, were used. Smoking history was used to identify current smokers, ex-smokers, and nonsmokers. Smokers were defined as those who were current smokers or those who had quit smoking for <1 year. Ex-smokers were those who had quit smoking for ≥1 year. Overweight was defined as a body mass index (weight [kg]/height [m2]) between 25 and 29.9 kg/m2, and obesity was defined as a body mass index of 30 kg/m2 or above. High total cholesterol (hypercholesterolemia) was defined as a serum cholesterol level of 240 mg/dL or greater. High triglycerides (hypertriglyceridemia) were those 200 mg/dL or higher. Hypertension was defined as a systolic blood pressure reading greater than or equal to 140 mm Hg or a diastolic reading greater than or equal to 90 mm Hg. High glucose (hyperglycemia) was defined as fasting blood glucose values equal to 126 mg/dL or above. Current smoking, obesity, hypertension, hypercholesterolemia, hyperglycemia, and hypertriglyceridemia were considered health risk factors in the analyses.
Vital status of the study subjects as of December 31, 2005, was established through computer matching to the Social Security Administration’s Death Master File and the National Death Index. Employees or surrogates were not contacted and employees not found to be deceased were assumed to be alive. Details of this process have been reported elsewhere.16 Causes of death were obtained for 98% of all decedents, and coded according to rules of the International Classification of Diseases in effect at the time of death. The International Classification of Diseases-9 equivalent codes, 390–459, were used to identify CVD.
Person-years at risk were calculated for each employee beginning January 1, 1985, and ending at the closing date of the study (December 31, 2005) or the date of death, whichever came first. Hazard ratios (HRs) for cumulative number of risk factors were calculated using a Cox proportional hazards model (SAS PHRGEG procedures) adjusted for age and gender. HRs for each individual risk factor (ie, cigarette smoking, obesity, hypertension, hypercholesterolemia, hyperglycemia, and hypertriglyceridemia), adjusted for age (continuous value) and gender, were also calculated by including each risk factor in the Cox regression model. All statistical analyses were done using SAS System Software PC Version 8.2.17
The distribution of age, gender, and the health risk factors are displayed in Table 1. The vast majority of study subjects were men (88.6%) whereas women accounted for only 11.4%. The study population was relatively young with a mean age at the beginning of the study (January 1, 1985) of 42 years, and an average follow-up of 20 years. More than half (57%) were hired between 1970 and 1985 and 64% were white-collar salaried employees. The average duration of employment was 25 years. A total of 1551 deaths were identified, with approximately two thirds (62%) occurring at age 60 and older.
Among the study population, 29% were current smokers. The average number of years since cessation among ex-smokers was 11.8 (data not shown). Nearly 18% of employees were obese whereas 22% of employees were hypertensive. The prevalence of hypercholesterolemia and hypertriglyceridemia was 20% and 15%, respectively. The prevalence of hyperglycemia was relatively low, with only 3% of employees having fasting glucose levels greater than or equal to 126 mg/dL.
The age distribution of subjects who did not participate in any physical examination programs did not differ significantly from that of participants (data not shown) and the duration of employment was similar (25.8 years). However, nonparticipants were more likely to be women (18%) and with salaried pay status (72%).
When compared to employees who never smoked, mortality risk for ex-smokers was similar although slightly elevated, with HR = 1.12 (95% CI = 0.96 to 1.31) and HR = 1.13 (95% CI = 0.86 to 1.49) for all causes and CVD, respectively (Table 2). However, mortality risks among smokers were significantly increased, with HR = 2.77 (95% CI = 2.40 to 3.21) for all causes and HR = 2.81 (95% CI = 2.17 to 3.63) for CVD.
Each risk factor was evaluated independently for effects on all-cause and CVD mortality (Table 3). All six risk factors were significantly associated with increased CVD mortality, and in addition to smoking, the strongest effects were associated with hyperglycemia (HR = 2.36, 95% CI = 1.80 to 3.10) and obesity (HR = 1.71, 95% CI = 1.40 to 2.09). Similar patterns were seen with univariate analysis of each factor with all-cause mortality, although risk estimates were lower, and failed to reach statistical significance for hypertension and hypercholesterolemia.
The multivariate adjusted relative mortality risks for all causes and CVD are displayed in Table 4. The relative risks for CVD were significantly higher among cigarette smokers (HR = 2.65, 95% CI = 2.12 to 3.30), obese persons (HR = 1.72, 95% CI = 1.34 to 2.22), and those with hypercholesterolemia (HR = 1.43, 95% CI = 1.13 to 1.80) or hyperglycemia (HR = 1.88, 95% CI = 1.36 to 2.62). Similar elevations of risk were seen for all causes, although the impact of hypercholesterolemia was substantially reduced.
As shown in Table 5, CVD and all-cause mortality increased with an increasing number of risk factors when compared to those with none of the risk factors. The HR for CVD increased from 2.04 (95% CI = 1.45 to 2.88) for employees with one risk factor to 5.13 (95% CI = 3.39 to 7.76) for employees with four or more risk factors, and for all causes from 1.40 (95% CI = 1.18 to 1.65) to 2.61 (95% CI = 2.07 to 3.29). We also examined the impact of risk factors on a subset of CVD, ie, coronary heart disease (CHD). The cumulative effect of these risk factors on CHD is greater than the effect on CVD with corresponding HRs increasing from 2.80 (95% CI = 1.20 to 6.54) to 7.52 (95% CI = 2.85 to 19.81; data not shown).
We further evaluated the impact of multiple risk factors on mortality by two smoking status categories, ie, nonsmokers and smokers (Table 6). Ex-smokers were combined with nonsmokers in view of the similar mortality risks between the two, and to minimize loss of statistical power. Among nonsmokers, the HR for CVD increased with an increasing number of risk factors, from 1.55 (95% CI = 1.04 to 2.33) for employees with one risk factor, to 1.92 (95% CI = 1.25 to 2.96) for employees with two risk factors, and to 3.21 (95% CI = 1.68 to 6.14) for employees with four or more risk factors. A similar increasing pattern was observed for all-cause mortality but significantly so only when subjects had three (HR = 1.49, 95% CI = 1.13 to 1.97) and four or more (HR = 1.72, 95% CI = 1.17 to 2.54) risk factors.
Among smokers, CVD risk increased substantially from 3.27 (95% CI = 2.07 to 5.17) for those who had one risk factor to 14.35 (95% CI = 7.23 to 28.49) for those who had four or more risk factors (Table 6). All-cause mortality also increased in a stepwise pattern with increasing number of risk factors, with risk of dying increasing from 2.67 (95% CI = 2.14 to 3.32) for those who had only one risk factor to 6.13 (95% CI = 3.99 to 9.41) for those who had four or more risk factors. Results for the above analyses were virtually the same when limited to male employees only (data not shown).
This 21-year follow-up study showed that employees with a variety of health risk factors had an increased risk of death compared to those who did not have these risk factors. Smoking, obesity, hypercholesterolemia, and hyperglycemia were independent and significant risk factors for CVD mortality. These findings were not unexpected because these risk factors have consistently been associated with risk of CVD in many cohort studies of nonoccupational populations. In addition, CVD and all-cause mortality increased with the accumulation of risk factors. The impact was particularly marked for CVD regardless of smoking status. Among nonsmokers, the risk of dying from CVD was significantly increased with only one risk factor. Among smokers, the risk of death was more than double that of nonsmokers with the same number of risk factors.
High-blood pressure and triglycerides were predictive of CVD mortality in the univariate analysis. Reasons for the apparent lack of association with these two variables in the multivariate analysis are not immediately clear. One explanation may be the interaction between these two variables and other risk factors. Because they are highly statistically correlated, their degree of independent predictive power may be difficult to assess, as their independent contribution to CVD mortality may be obscured by that of the other risk factors.
Cigarette smoking was associated with increased all-cause and CVD mortality in our study and has been associated with an increased risk of various diseases in other studies.18 However, for those who quit smoking, the risk of death decreased substantially and differed little from that of nonsmokers, a finding reported by others as well.19
Obesity is a well-established risk factor for death caused by CVD, especially CHD.20–22 Obesity typically raises blood pressure and cholesterol levels23–25 and also adversely affects triglycerides25 and glucose.26 We have shown that obesity is an independent risk factor of death regardless of whether other risk factors were adjusted for in the regression analysis. Given the rising prevalence in this population5 and in the American workforce in general,27,28 and the future impact on mortality that is expected to result, obesity should be a direct target for intervention.
Hyperglycemia is another independent risk factor for cardiovascular and all-cause mortality, although its attributable mortality was relatively small due to its low prevalence in this population (3.2%). Increased glucose level is acquired largely through obesity, although genetic and epigenetic components are also thought to play a role.29–31 Thus, in addition to therapeutic treatment, weight reduction and regular physical activity could be a productive means of reducing glucose level and minimizing its impact on long-term mortality. Given the serious adverse health outcomes of hyperglycemia, it is important to reduce the prevalence of employees with impaired fasting glucose (ie, 110 to 125 mg/dL). Approximately 9% of employees fell into this category, three times the number with hyperglycemia in this population. If glucose levels of these employees were to remain unchecked and uncontrolled, this larger group would be expected to become hyperglycemic over time. Thus, early treatment for employees with impaired fasting glucose could have a positive impact on employee health.
Identification and quantification of risk factors is an important first step of worksite health promotion programs designed to reduce employee morbidity and mortality. Every risk factor examined in this study had a negative impact on cardiovascular mortality, and the risk was compounded when multiple risk factors existed in a single individual. This finding suggests that worksite health promotion programs should address all of these risk factors, particularly among high-risk individuals.
The Shell HSS has been an important tool in evaluating employee health, formulating preventive strategies, and assessing the effectiveness of health programs.4 The magnitude of increased risks is consistent with those reported in the literature from studies based on general populations.32–35 Similar data have been reported on an occupational cohort from the Chicago Heart Association Detection Project in Industry study.36 Although the cohort definition and study design differed from the present study, the conclusions in terms of increasing risk with accumulation of risk factors was similar. One of the strengths of this study is the use of physical examination data to identify employee health risks. These objective and reliable measures minimize the potential for reporting and recall bias that may be introduced when using self-reported health risk data. Furthermore, mortality data were prospectively collected prior to the end of study, thus limitations inherent in the retrospective collection of health endpoint data are avoided. In addition, the follow-up was 21 years, and at the end of the study more than half of the study subjects were 60 years and older when CVD prevalence was high. A limitation of this study is that the risk factors were based on single baseline measurements, which may have changed during the course of follow-up.
In summary, this study shows a positive association between a number of employee health risk factors and mortality from all causes and from CVD. A greater number of health risk factors corresponded to a higher rate of death. Reductions of employee health risk factors may be an effective means of improving employees’ health and increasing a company’s productivity. The first step in addressing this issue is to improve awareness, among both management and employees, of the impact health risk factors have on long-term mortality. Companies should establish a healthy culture in the work environment and promote the transition from just “thinking about health” to making sustainable lifestyle changes, such as increased promotion of physical activity. Successes in reducing health risk factors should translate to the employee level as well, for example by redesigning health insurance programs to promote and reward wellness. Results of this study could have significant implications for assessing the economic ramifications of an unhealthy workforce.
1.Pelletier KR. Clinical and cost outcomes of multifactorial cardiovascular risk management interventions in worksites: a comprehensive review and analysis. J Occup Environ Med. 1997;39:1154–1169.
2.Aldana SG, Pronk NP. Health promotion programs, modifiable health risks, and employee absenteeism. J Occup Environ Med. 2001;43:36–46.
3.Marshall AL. Challenges and opportunities for promoting physical activity in the workplace. J Sci Med Sport. 2004;7 (1 Suppl):60–66.
4.Tsai SP, Wendt JK, Ahmed FS, Donnelly RP, Strawmyer TR. Illness absence patterns among employees in a petrochemical facility: impact of selected health risk factors. J Occup Environ Med. 2005;47:838–846.
5.Tsai SP, Donnelly RP, Wendt JK. Obesity and mortality in a prospective study of a middle-aged industrial population. J Occup Environ Med. 2006;48:22–27.
6.Riedel JE, Lynch W, Baase C, Hymel P, Peterson KW. The effect of disease prevention and health promotion on workplace productivity: a literature review. Am J Health Promot. 2001;15:167–191.
7.Burton WN, Conti DJ, Chen CY, Schultz AB, Edington DW. The role of health related risk factors and disease on worker productivity. J Occup Environ Med. 1999;41:863–877.
8.Boles M, Pelletier B, Lynch W. The relationship between health risks and work productivity. J Occup Environ Med. 2004;46:737–745.
9.Cawley J, Rizzo JA, Haas K. Occupation- specific absenteeism costs associated with obesity and morbid obesity. J Occup Environ Med. 2007;49:1317–1324.
10.Schmier JK, Jones ML, Halpern MT. Cost of obesity in the workplace. Scand J Work Environ Health. 2006;32:5–11.
11.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.
12.Greenberg RS, Mandel JS, Pastides H, Britton NL, Rudenko L, Starr TB. A meta-analysis of cohort studies describing mortality and cancer incidence among chemical workers in the United States and Western Europe. Epidemiology. 2001;12:727–740.
13.Flegal KM, Graubard BI, Williamson DF, Gail MH. Cause-specific excess deaths associated with underweight, overweight, and obesity. JAMA. 2007;298:2028–2037.
14.Stamler J, Stamler R, Neaton JD. Blood pressure, systolic and diastolic, and cardiovascular risks. US population data. Arch Intern Med. 1993;153:598–615.
15.Joyner RE, Pack PH. The Shell Oil Company’s computerized health surveillance system. J Occup Med. 1982;24:812–814.
16.Tsai SP, Gilstrap EL, Cowles SR, Snyder PJ, Ross CE. Long-term follow-up mortality study of petroleum refinery and chemical plant employees. Am J Ind Med. 1996;29:75–87.
17.SAS Institute, Inc. SAS Procedures Guide Release 8.12. Gary, NC: SAS Institute, Inc.; 1999.
18.Ezzati M, Lopez AD. Estimates of global mortality attributable to smoking in 2000. Lancet. 2003;362:847–852.
19.U.S. Department of Health and Human Services. The Health Benefits of Smoking Cessation. Rockville, MD: U.S. Department of Health and Human Services, Public Health Service. Center for Disease Control. Center for Chronic Disease Prevention and Health Promotion. Office on Smoking and Health. DHHS Publication No. (CDC) 90-8416; 1990.
20.Wilson PWF, D’Agostino RB, Sullivan L, Parise H, Kannel WB. Overweight and obesity as determinants of cardiovascular risk: the Framingham Heart Study. Circulation. 1983;67:968–977.
21.Calle EE, Thun MJ, Petrelli JM, Rodriguez C, Heath CW Jr. Body mass index and mortality in a prospective cohort of US adults. N Engl J Med. 1999;341:1097–1105.
22.McGee DL; The Diverse Population Collaboration Investigators. Body mass index and mortality: a meta-analysis based on person-level data from twenty-six observational studies. Ann Epidemiol. 2005;15:87–97.
23.Krieger DR, Landsberg L. Mechanisms in obesity-related hypertension: role of insulin and catecholamines. Am J Hypertens. 1988;1:84–90.
24.Denke MA, Sempos CT, Grundy SM. Excess body weight: an under-recognized contributor to high blood cholesterol levels in white American men. Arch Intern Med. 1993;153:1093–1103.
25.Denke MA, Sempos CT, Grundy SM. Excess body weight: an under-recognized contributor to dyslipidemia in white American women. Arch Intern Med. 1994;154:401–410.
26.Abate N, Garg A, Peshock RM, Stray-Gundersen J, Grundy SM. Relationships of generalized and regional adiposity to insulin sensitivity in men. J Clin Invest. 1995;96:88–98.
27.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.
28.Flegal KM, Carroll MD, Kuczmarski RJ, Hohnson CL. Overweight and obesity in the United States—prevalence and trends 1960–1994. Int J Obes. 1998;22:39–47.
29.Chien LY, Liou YM, Chen JJ. Association between indices of obesity and fasting hyperglycemia in Taiwan. Int J Obes. 2004;28:690–696.
30.Prokopenko I, Langenberg C, Florez JC, et al. Variants in MTNR1B influence fasting glucose levels. Nat Genet. 2009;41:77–81.
31.Woo M, Patti ME. Diabetes risk begins in utero. Cell Metab. 2008;8:5–7.
32.Daviglus ML, Stamler J, Pirzada A, et al. Favorable Cardiovascular risk profile in young women and long-term risk of cardiovascular and all-cause mortality. JAMA. 2004;292:1588–1592.
33.van Dam RM, Li T, Spiegelman D, Franco OH, Hu FB. Combined impact of lifestyle factors on mortality: prospective cohort study in US women. BMJ. 2008;337:a1440.
34.Khaw KT, Wareham N, Bingham S, Welch A, Luben R, Day N. Combined impact of health behaviours and mortality in men and women: the EPIC-Norfolk prospective population study. PLoS Med. 2008;5:e12.
35.Eberly LE, Prineas R, Cohen JD, et al. Multiple Risk Factor Intervention Trial Research Group. Metabolic syndrome: risk factor distribution and 18-year mortality in the multiple risk factor intervention trial. Diabetes Care. 2006;29:123–130.
36.Lowe LP, Greenland P, Ruth KJ, Dyer AR, Stamler R, Stamler J. Impact of major cardiovascular disease risk factors, particularly in combination, on 22-year mortality in women and men. Arch Intern Med. 1998;158:2007–2014.