**3. Results**

Demographic characteristics: Forty-two percent of the sample were women. Descriptive statistics for workers stratified by gender and trust are presented in Table 1. For both women and men, the largest percent of respondents (72.0% and 72.2%, respectively) were White, Non-Hispanic with a mean age of about 42. More than half of the respondents were married and had either technical training or some college.

Mistrustful environment: Approximately 22% of women and 20.3% of men indicated that their supervisor did not always create an open and trusting environment (Table 1). As shown in Table 1, for both women and men, the highest prevalence of mistrust reported was among workers ages 45–64, (women = 24.4%, men = 23.0%) followed by workers ages 30–44 (women = 22.3%, men = 20.5%). Black women (23.2%), followed by White women (22.4%) had the highest prevalence of reporting that their supervisor does not create an open and trusting environment, compared with women of other races/ethnicities (Asian, Hispanic, Other). White men (21.1%) followed by Black men (20.6%) reported higher prevalence of a mistrustful environment. Prevalence of a mistrustful environment was higher for women with increasing levels of education (highest for those with college or post-graduate education, 25.2%). Men with technical training or some college/associate degree had a slightly higher prevalence (20.9%) of a mistrustful environment. Divorced women (26.8%) and men (24.7%) had the highest prevalence of reporting a mistrustful work environment. Prevalence by income was similar for both women and men, with slightly higher prevalence of a mistrustful environment for those earning \$3000–\$4999 per month among women (23.0%) and for those earning \$5000–\$7499 among men (21.7%).


**Table 1.** Weighted prevalence (%) and standard errors (SE) of open and trusting and open work environment factors by sociodemographic characteristics and gender (Gallup-Sharecare Well-Being Index, 2010–2012).

\* *Z*-test for difference between women and men *p* < 0.05.

Table 2 shows the weighted prevalence of LS7 risk factors by trust. The prevalence of all LS7 risk factors were higher for both women and men who reported working in a mistrustful environment, compared to those whose working environment was not mistrustful.


**Table 2.** Weighted prevalence (%) and standard errors (SE) for LS7 CVD risk factors by work environment characteristics and gender (Gallup-Sharecare Well-Being

*Z*-test for difference between women and men significant *p*

Multivariate results for trust with LS7 CVD risk factors, after adjustment for demographic factors and having any health insurance, are shown in Table 3. Confidence intervals excluded 1.0, indicating statistical significance at *p* < 0.05 for each outcome. Trust was associated with the LS7 CVD risk factors in both men and women after adjustment for covariates. Due to the large sample size, we report effect sizes with a focus on those ≥10%. We found that workers who did not work in an open and trusting environment had greater odds of having high blood pressure (women = 15%, men = 20%), high cholesterol (women = 18%, men = 22%), and diabetes (women = 15%, men = 18%) compared to those who reported having an open and trusting environment with their supervisor. Both women and men workers had greater odds of being a current smoker (both 15%), having a poor diet (women = 10%, men = 11%), and being obese (both women and men = 18%). Women reporting a mistrustful environment also had greater odds of having a low physical activity level (10%). Odds ratios for having four or more LS7 CVD risk factors were elevated for those working in a mistrustful environment (women = 22%, men = 29%).

**Table 3.** Multivariate associations between LS7 CVD risk factors and open and trusting work environment stratified by gender (Gallup-Sharecare Well-Being Index, 2010–2012) a.


<sup>a</sup> Models are adjusted for age, race/ethnicity, education, marital status, family income, and any health insurance.

#### **4. Discussion**

The findings of this study suggest that lower workplace social capital, as measured by the WBI, is associated with higher odds of having one or more of the LS7 CVD risk factors.

Our findings are consistent with others who have found associations between social capital and health [39,46,48,65,81–83]. Previous research reported that working in a negative environment and having low social support could lead to stress and psychosocial distress [84]. Workplace stress can directly increase CVD risk through biological pathways (e.g., inflammation) or CVD risk factors [12,13]. Studies have found that workers experiencing job stress were more likely to have diabetes [85]. Associations between work environment and high blood pressure and high blood cholesterol are mixed, with some studies reporting results similar to our findings [86,87] and others reporting no association [85].

Furthermore, workplace stress can indirectly affect CVD risk through at-risk health behaviors [85,87]. These behaviors include poor diet, insufficient physical activity [85,86], smoking [85], high alcohol consumption, and lack of sleep [88]. Workers who reported a lack of support from supervisors were more likely to be heavy smokers [89]. In addition, men with low workplace social support were more likely to be obese [90].

Although we adjusted for potential demographic confounders in our models, we examined each of the LS7 CVD risk factors separately. Health behaviors are often interrelated and can affect the presence or absence of other health behaviors. For example, smoking was found to increase caloric intake [91], while a healthy lifestyle (diet and exercise) was negatively associated with smoking [92]. Because of the potential co-occurrence, we conducted sensitivity analyses to see whether the odds ratios increased if we selected having four or more of the LS7 risk factors instead of only one. We used

this as the dependent variable in our regression models. Both women (22%) and men (29%) showed an increase in odds for having four or more LS7 risk factors if they indicated that their supervisor did not create an open and trusting environment.

Analyses were presented separately by gender, not only due to the differences in CVD risk between women and men, but also due to the importance of gender in the social capital and managerial psychology literature [93]. Odds ratios were similar for both genders when the LS7 factors were looked at individually, and slightly higher among men when the dependent variable was having four or more LS7 factors.

Improvements to the work environment are needed to reduce CVD risk among workers. Social modification to the work environment, such as adjusting managerial style to create an open and trusting environment, can decrease work stress. Considering managerial trust from a Total Worker Health® framework meets the goals of illness prevention to advance worker well-being. Efforts can also be made to target the health behaviors themselves. There are a range of possible strategies for addressing the LS7 risk factors in the workplace. For example, physical modification to the work environment, such as installing sit/stand desk stations and walking workstations, can reduce sedentary behavior and may increase physical activity. Additionally, increased access to nutritious food in the workplace may improve diet. Supervisors who support workplace wellness may help in reducing CVD risk factor in workers [94].
