*Strengths and Limitations*

This study has several strengths. The WBI is a large, nationally representative survey. Skopec et al. [95] found that the survey provided reasonably similar data when compared to established national surveys, such as the National Health Interview Survey (NHIS) and the Behavioral Risk Factor Surveillance System (BFRSS), on several important health-related measures. However, the Gallup sample was slightly older, had fewer minorities, and a higher educated sample than in other national surveys [95]. Outcomes examined included select health behaviors and health outcomes. Findings in our study are similar to those reported by adults who worked in the past 12 months in the 2010 National Health Interview Survey (NHIS) conducted by the National Center for Health Statistics. Weighted prevalence items were obesity (BMI ≥30) (Gallup = 27.7%; NHIS = 28.1%), current smoker (Gallup = 21.9%, NHIS = 19.7%), hypertension (Gallup = 22.3%; NHIS = 19.4%), insufficient exercise (Gallup = 76.3%; NHIS = 88.3%), and diabetes (Gallup = 6.1%; NHIS = 5.8%) [72]. Data on high cholesterol are not available for the 2010 NHIS, but are available for 2015 (Gallup = 21.3%; NHIS = 21.5%) [72].

The work environment question included in this study allows us to examine an important workplace psychosocial factor that is often difficult or expensive to study. It is unclear where the work environment question originated. Documentation provided by Gallup indicated that it was based upon findings from leading scientists in the areas of survey research, behavioral economics, and health. No information on the validity or reliability of the Gallup question is available.

This study also has several limitations. The survey is cross-sectional and therefore no conclusions can be made regarding causality. Data were collected via a telephone survey that has a low response rate, potentially affecting the representativeness of our findings. For each regression model, observations with missing values for included covariates were dropped. All data were self-reported at one point in time and are subject to response biases, such as recall and social desirability. Social desirability bias [96] is the tendency of respondents to present themselves in a socially desirable light, which may deviate from their true behaviors. Social desirability bias has been shown to affect the reporting of health behaviors, including underreporting negative behaviors and over reporting positive ones [97]. However, a recent study by Prather et al. [98] did not find confounding due to social desirability bias. Additionally, although we adjusted for potential confounders in our models, other non-measured confounders may have influenced our results. The WBI survey only touched upon a small number of components of an individual's work environment. Components of social

capital and the work environment such as occupation, organizational structure (e.g., work schedule, work arrangements), culture, job autonomy, job resources, job security, work engagement, workplace hostility, additional characteristics of the supervisor, and others are needed for a better-informed study. Findings by Oksanen et al. [45,46] suggest that the effects of low social capital might not be similar in all work units or groups of different socioeconomic structure. However, because social capital and socioeconomic status were measured at the individual level, we are unable to examine the effects of social capital in different work contexts. Additionally, stand-alone, single-item questions may not offer the precision needed to make an accurate assessment of supervisory style, and as Choi et al. [34] suggest, there is a lack of consensus on measurement of social capital. The measure of social capital used in this study deals with leadership trust. Researchers have also included differing measures of social capital that include employee networks and workforce norms [99].

Health behavior questions in the WBI were different from AHA's LS7 definitions, particularly diet that included only one of the five diet variables. The health factor metrics are markedly different in the WBI compared to the AHA's LS7 definitions, noticeably absent are the clinical measurements of blood pressure, blood cholesterol, fasting glucose, and medication used to treat these health factors. Lastly, the study's large sample size increases the probability of finding statistically significant associations; therefore, we focused on effect size rather than *p*-values. Despite these limitations, results show that more than 20% of workers report that their supervisor does not always create an open and trusting environment. This is associated with a 20% increase in odds for having four or more CVD risk factors, suggesting that this is an important factor when designing interventions to address worker cardiovascular health. Therefore, these results show support for the usefulness of this aspect of social capital to understand the work environment, supervisory behavior, and their association with worker cardiovascular health.

#### **5. Conclusions**

This study found that a negative work-environment characteristic representing an aspect of workplace social capital contributed to greater odds of having important CVD risk factors among full-time workers. Results suggest that supervisor behavior can play an important role in improving worker health. Workplace intervention programs for CVD and other chronic health conditions should consider addressing this aspect of workplace social capital, and supervisor competencies and behavior in particular, with proper training as a potential means to improve worker health. Thus, our results reinforce the notion voiced elsewhere [69,100] that supervisor support is essential to a comprehensive approach to worker safety and health; issues of managerial trust are worthy of inclusion in a Total Worker Health® framework.

**Author Contributions:** T.A. and R.T. contributed equally to the manuscript. T.A., R.T., J.J. and K.M.K. contributed to Writing-Review and Editing.

**Funding:** This research received no external funding. However, Kelly's effort was supported by Cooperative Agreement No. U19OH008868 from the Centers for Disease Control and Prevention (CDC), National Institute for Occupational Safety and Health (NIOSH) to the Healthier Workforce Center at the University of Iowa.

**Acknowledgments:** The authors would like to thank Chia-Chia Chang and Casey Chosewood, NIOSH Total Worker Health®, and Melanie Standish of the Gallup organization for their assistance in accessing and understanding the data. We would also like to thank Leslie McDonald and Candice Johnson, Division of Surveillance, Hazard Evaluation and Field Studies, National Institute for Occupational Safety and Health, for their valuable comments regarding the American Heart Association Life's Simple 7. Appreciation is extended to James Grosch, Tara Hartley, Sara Luckhaupt, and Marie Haring Sweeney for their review comments on this manuscript.

**Conflicts of Interest:** The authors declare that they have no conflict of interest. The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the National Institute for Occupational Safety and Health.
