The Buffalo Cardio-Metabolic Occupational Police Stress Study was a longitudinal study initiated to investigate associations between stressors unique to the law enforcement profession and psychological and physiological health outcomes. The study participants were police officers who worked in the Buffalo, New York Police Department. Three examinations over a period of 12 years were completed (Fig. 1): a baseline examination with 464 officers (2004–2009), the first follow-up examination with 300 officers (2011–2015), and the second follow-up examination with 240 officers (2015–2019). Data for all exams were collected at the Center for Health Research, School of Public Health and Health Professions, University at Buffalo, State University of New York. A written informed consent was collected from each participant at each of the three examinations. This study was reviewed and approved by the internal review board of the State University of New York at Buffalo (internal review board approval number/ID: See 45 C.F.R. part 46.101(c); 21 C.F.R. part 56). For the baseline examination (Exam 1), a total of 710 police officers who worked with the Buffalo, New York Police Department were invited to participate; 464 (65.4%) active-duty and retired police officers agreed to participate and were examined during 2004–2009. No specific inclusion criteria were indicated for the study, only that participants be sworn police officers and willing to participate. The first follow-up examination (Exam 2) was conducted nearly 7 years later when a total of 300 officers were examined between 2011 and 2015; of the 300, 276 were part of the original cohort evaluated at the baseline examination while the remaining 24 were new recruits added at Exam 2 (Fig. 1). The second follow-up examination (Exam 3) was conducted approximately 12 years from baseline when a total of 240 officers were examined; of the 240, 191 officers were part of the original cohort evaluated at the baseline examination. At each examination a comprehensive evaluation of the participants was performed to collect data on demographic and lifestyle factors, occupational exposures, psychosocial factors, and physical health outcomes. Female officers who were pregnant at the time of examination were excluded. Further details on recruitment, data collection, and variables assessed are described elsewhere.²⁰ This article was prepared following the STROBE reporting guidelines (see STROBE checklist as Supplementary Data 1, http://links.lww.com/JOM/B818).

Demographic characteristics (age, gender, race/ethnicity, education, marital status, years employed as a police officer, and rank) and lifestyle behaviors (smoking status, alcohol intake, and hours of physical activity) were obtained using self-administered questionnaires. Alcohol consumption was ascertained using data from the Food Frequency Questionnaire related to drinking patterns. Physical activity level was assessed using the Seven-Day Physical Activity Recall questionnaire, an interviewer administered questionnaire developed in the Stanford Five-City Project.²¹ Medication use was ascertained through self-report and by inventory of current medications that the participants brought to the clinic (exam center) on their date of interview.

A fasting blood sample (20 mL) was obtained from all participants by a trained phlebotomist. Blood parameters (HDL cholesterol [mg/dL], triglycerides [mg/dL], and glucose [mg/dL]) were measured by standard laboratory techniques. After the participants had rested in the supine position for 5 minutes, three BP measurements were taken 2 minutes apart. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were derived as the mean of the three respective measurements. Waist circumference was measured as abdominal girth at the highest point of the iliac crest and the lowest point of the costal margin in the midaxillary line. Three measurements of waist circumferences were taken, and the average value was used as the participant’s waist circumference.

Assessment of MetSyn and its components were based on criteria by the National Cholesterol Education Program Adult Treatment Panel III guidelines with recent modifications from the American Heart Association and the National Heart, Lung, and Blood Institute.¹¹ Further description of the assessment of MetSyn for the Buffalo Cardio-Metabolic Occupational Police Stress study was provided by Hartley et al (2011).²² The following five individual MetSyn components included: 1) abdominal obesity (gender-specific waist circumference ≥102 cm in males, ≥88 cm in females); 2) elevated triglycerides (fasting triglycerides ≥150 mg/dL, or reported treatment with nicotinic acid or fibrates); 3) reduced HDL-C (gender-specific fasting HDL-C < 40 mg/dL in men, <50 mg/dL in women, or reported treatment with nicotinic acid or fibrates); 4) glucose intolerance (fasting serum glucose ≥100 mg/dL, or reported treatment for diabetes); and 5) hypertension (SBP ≥130 mmHg, diastolic blood pressure ≥85 mmHg, or reported physician-diagnosed hypertension and anti-hypertensive treatment). The five components each coded 0/1 (0 = for absence of the component, 1 = for presence of the component) were summed to get the total count of MetSyn components (range 0–5). MetSyn was considered present if an individual had at least three components (sum of components ≥3).

We conducted two sets of analysis to estimate incidence over the 7- and 12-year follow-up periods. To estimate incidence over the 7-year follow-up period, a participant must have been examined at both the baseline and the first follow-up examination (n = 276) and participants were excluded if they met the following two conditions: (a) prevalent cases (those with MetSyn or any component at baseline) and (b) those with missing status for MetSyn or its components at either examination (see Table 1 for inclusion/exclusion criteria and analytic sample size). Similar inclusion and exclusion criteria were used to estimate incidence over the 12-year follow-up period (Table 1). Descriptive statistics were used to tabulate baseline characteristics of the participants at risk of MetSyn and sex differences in baseline covariates were tested using chi-square tests of independence or Fisher’s exact test for categorical covariates and analysis of variance for continuous covariates.

The risk and risk ratio (RR) of MetSyn and its components by sex were calculated using the Poisson regression with robust variance estimation, often called the modified Poisson regression.²³ A RR was considered statistically significant if the 95% CI did not include 1. All statistical analyses were performed using the SAS (SAS Institute) software version 9.4.²⁴ Significance level was set at 5%. Comparison of incidence across sex was adjusted for covariates (age, race/ethnicity, education, marital status, police rank, smoking status, alcohol consumption, and physical activity) that may be related to the dependent variable (MetSyn and its components) or the independent variable (sex).

The 7-year unadjusted risk of MetSyn and its components was higher in males than females except for abdominal obesity where the incidence in males was 26% lower compared to females (RR = 0.74, 95% CI: 0.47, 1.17) but it was not statistically significant (Table 4). The overall incidence of MetSyn was 33% higher in males than females (RR = 1.33, 95% CI: 0.73, 2.43), but it was not statistically significant. The sex difference in risk of elevated triglyceride was marginally significant where the incidence in males was 2.5 times as high as in females (RR = 2.49, 95% CI: 0.98, 6.30; P value = 0.0539). Incidence of glucose intolerance was 69% higher in males than females but did not reach statistical significance (RR = 1.69, 95% CI: 0.90, 3.18; P value = 0.1050). After adjustment for demographic and lifestyle factors, sex difference in glucose intolerance attained statistical significance where the incidence in males was two-fold higher (RR = 2.16, 95% CI: 1.11, 4.20, P = 0.0234). The risk of hypertension was three-fold higher in males relative to females (RR = 3.21, 95% CI: 1.18, 8.76; P value = 0.0226). On the other hand, covariate adjustment attenuated the sex difference in incidence of hypertension (RR = 2.08, 95% CI: 0.70, 6.18) rendering it statistically nonsignificant.

The 12-year unadjusted risk of MetSyn and its components was higher in males than females except for reduced HDL cholesterol where the incidence in males was 34% lower compared to females (RR = 0.66, 95% CI: 0.27, 1.161), but it was not statistically significant (Table 4). Although it did not attain statistical significance, male officers had a notably higher 12-year incidence of glucose intolerance by 86% (RR = 1.86, 95% CI: 0.83, 4.17), hypertension by 45% (RR = 1.45, 95% CI: 0.73, 2.90), and elevated triglyceride by 54% (RR = 1.54, 95% CI: 0.71, 3.34) compared to their female counterparts. Covariate adjustment did not impact statistical significance of sex differences in 12-year incidence, but there were notable changes in effect size before and after multivariate adjustment; the incidence of MetSyn in males was only 10% higher before adjustment compared to 53% higher following adjustment and the incidence of glucose intolerance was 86% higher in males before adjustment compared to 159% higher following adjustment.

Generally, the sex differences in incidence during the two follow-up periods (7- vs 12-year) are consistent except for abdominal obesity and reduced HDL cholesterol. Over 7-year follow-up incidence of abdominal obesity was 26% lower in males than females, but the difference was not statistically significant (Table 4, unadjusted model) while during the 12-year follow-up the incidence of abdominal obesity was identical in males and females. On the other hand, during the 7-year follow-up, the incidence of reduced HDL cholesterol was nearly identical in males and females while during the 12-year follow-up the incidence of reduced HDL in males was 34% lower compared to females, yet the difference did not attain statistical significance.

Despite sufficient scientific evidence of the prevalence of CVD risk factors in the occupation of policing, to date, there are no longitudinal studies among US police officers that examined the risk of MetSyn and its individual components and potential sex differences in incidence. Understanding the magnitude of the risk of MetSyn and its components in police officers is an important step for mitigating future CVD morbidity and mortality. This longitudinal study addressed the research gap by estimating the 7- and 12-year risk of MetSyn and its components. The 7-year risk for developing MetSyn among officers without any prevalent MetSyn component was 20.4% and the 12-year risk was 22.2%. This means 20% of the sample, who at baseline did not have any MetSyn components, developed at least three of five the components in the span of 7 years. Similarly, 22% of the sample who were free of any MetSyn component at baseline developed three or more components over a 12-year period. During both follow-up periods, individual MetSyn components with lowest and highest incidence were reduced HDL cholesterol and abdominal obesity, respectively.

Previous studies of incident MetSyn are limited. In their study of male and Italian police officers, Garbarino and Magnavita (2015) reported the 5-year MetSyn incidence was 17.5% compared to 22.2% for male officers at 7 years in the current study.²⁵ There were also differences in the incidence of each component compared to the current study; some of which may be explained by the different MetSyn definitions applied. Incident MetSyn cases using ICD-9 and ICD-10 diagnostic codes were 7.5 per 100,000 person-years for active duty US Armed Forces personnel between 2002–2017 with the incidence being twice as high for female vs male service members (13.9 vs 6.4 per 100,000 person-years).²⁶

The development of MetSyn among police may be due to many factors, including stress. Chronic stress leads to hyperactivity of the hypothalamic pituitary adrenal axis, thereby elevating cortisol levels and leading to the development of visceral adiposity, hypertension, and dyslipidemia, components of the MetSyn.^25,27,28 Police work is a noted high stress occupation.^29–34 Sources of chronic stress in police work include shift work, long work hours, organizational stressors, and repeated exposure to traumatic events and threats to physical and psychological health.^35–38 In addition, acute, sudden, and intense exposure to traumatic events and threats to self lead to unpredictable bursts of physical response, which may place a high demand on the cardiovascular system including elevations in heart rate and BP.³⁹ Trauma may also lead to development of posttraumatic stress disorder, and posttraumatic stress disorder has been associated with increased risk of hypertension, obesity, dyslipidemia, and CVD.⁴⁰

Studies have examined the association between stress and CVD in police officers. Garbarino and Magnavita reported increased adjusted risks of developing MetSyn and hypertriglyceridemia among police officers reporting high levels of stress.²⁵ A study by Hartley et al²² found that the adjusted number of MetSyn components increased significantly in women police officers across tertiles of organizational pressure and lack of support, while no associations were found in male officers. Abdominal obesity and reduced HDL-C were the components consistently associated with police stress in women.²²

Greater perceived stress has been associated with increased prevalence of CVD and its risk factors.^4,10 It was estimated that 41% of police officers have a stress-related physical health problem, such as hypertension.⁴¹ Police officers under stress have a more adverse CVD risk factor profile and higher CVD mortality rates compared to other occupations and the general population.^7,10 Police officers in the highest quartile of stress had significantly higher mean levels of triglycerides than their colleagues in the lowest quartile.²⁵ A study of Polish police officers reported positive correlation between higher perceived stress on the job and the prevalence of MetSyn.⁴²

Police-related lifestyle factors may also influence MetSyn. Although officers come into policing as physically fit, the sedentary nature of police coupled with stress contributes to the development of MetSyn.⁴¹ The majority of police work time involves sitting in patrol cars, writing reports, or interviewing persons.⁴³ Increased levels of obesity and the higher waist size, component of MetSyn, are likely related to this lack of physical activity and increased sitting time.⁴⁴ Anderson and colleagues (2016) found a two-fold increase in adjusted odds for MetSyn among officers reporting low versus high levels of physical activity.⁴⁵ Lack of regular physical activity was found to be one of the occupational risk factors contributing to the higher prevalence of elevated BP, MetSyn, and CVD among emergency responders including police officers.⁴⁶

Shift work also may increase the risk for components of MetSyn. Irregular work hours, specifically those at night and early morning, were reported to be associated with adverse levels of waist circumference and triglycerides in men and triglycerides in women who participated in a British birth cohort study.⁴⁷ In a small pilot study of police officers, those working the night shift combined with short sleep duration had significantly more MetSyn components than those working day shift with short sleep duration.¹⁶

Dietary patterns have been associated with MetSyn. Lutsey et al (2008) found in a large cardiovascular disease study of middle-aged men and women that a Western dietary pattern, and consumption of meat, fried foods, and diet soda were negatively associated with incident MetSyn.⁴⁸ While no studies of dietary patterns and metabolic syndrome in police officers were identified, dietary patterns in officers have been associated with other MetSyn risk factors, including shift work, inactivity, and long work hours. Bonnell and colleagues found that while overall energy intake was similar between day and night shift officers, night shift workers consumed higher energy dense foods.⁴⁹ Gibson et al (2018) found that male officers working 49 hours or more per week compared to less than 40 hours per week and males reporting high versus low job strain had significantly poorer diet quality than men officers. In the same study, low levels of weekly physical activity compared to high levels were significantly associated with poorer diet quality in male and female officers.⁵⁰

In the present study, the 7-year risk of glucose intolerance was two-fold higher in males compared to females after covariate adjustment while the risk of hypertension was 3-fold higher in males than females, but the effect was attenuated after covariate adjustment. Sex differences in overall Metsyn, abdominal obesity, elevated triglyceride, and reduced HDL cholesterol during both follow-up periods were not statistically significant. Although these findings did not reach statistical significance, they point toward potential differences in MetSyn risk by sex. Future studies with more female police officers are desirable.

The use of a specific police department as our study population may limit generalizability of our findings to other departments that have somewhat different characteristics. Additional limitations include the fact that the participants at baseline were volunteers which lends itself to a potential self-selection bias and that there was a significant attrition during both the 7-year follow-up (40.5% = 188/464) and the 12-year follow-up (58.8% = 273/464) period. Even with appropriate recognition of this limitation, similar CVD health problems are seen among police in other cities, states, and countries. Police work is a specific occupation and officers are exposed to unique stressors likely not seen in other work. This may affect comparability to the general population. However, findings in the present study may be relevant to other occupations characterized as first responders. A potential confounder in this type of analysis is the healthy worker effect.⁵¹ Those who graduate from the police academy are necessarily young, healthy individuals who must meet the rigorous physical, medical, and emotional standards required of police work. Less healthy individuals who fail these standards are excluded. Therefore, the healthy worker effect may lower the relative occupational mortality risk of law enforcement personnel compared with the general population, particularly among younger officers. The present study has several strengths. Because of the longitudinal design, we were able to calculate risk estimates, change, or progression in subclinical markers, regarding indicators of MetSyn that are not available in cross-sectional or retrospective data.