**4. Discussion**

This study surveyed a large and diverse sample of workplaces to examine the extent that OHS and wellness activities co-occur and identified the workplace characteristics most likely to be associated with the co-occurrence of the activities. Most workplaces surveyed reported having few OHS and wellness activities. Large workplaces, those in the electrical and utilities sector, and with a high rating for people-oriented culture were factors most associated with a high co-occurrence of OHS and wellness activities. Large and medium-size workplaces, those in the electrical and utilities, and municipal sectors were associated with a moderate co-occurrence of activities.

This study found a fewer number of wellness activities in Ontario workplaces compared to US studies where at least three-quarters of workplaces reported one or more wellness activity [17,18]. This difference may be due to the fewer incentives for Canadian employers to invest in wellness activities as most medically-necessary services are covered by Canada's public healthcare system. Furthermore, employer contributions to healthcare costs are comparatively modest. The higher number of EAPs compared to other wellness activities is unsurprising as small workplaces represented the

highest proportion of respondents in the study. The outsourcing of resources to EAPs can represent a better investment for small workplaces compared to more costly investments in onsite wellness activities [20]. Small organizations also experience several obstacles to implementing wellness activities such as constraints in assigning resources and dedicated staff for wellness initiatives, perceiving a lack of employee interest in participating in wellness activities, or have poor access to health promotion resources and wellness providers [20]. While stress management, physical activity promotion, and flexible hours were reported frequently, others such as healthy food choices and shower facilities were less frequently reported. This suggests that workplaces might be primarily focusing on encouraging their workers to change their own behaviors. Yet only focusing on changing individual behaviors is unlikely to lead to meaningful worker health improvements since a small percentage of workers participate in wellness activities without workplace policies and environmental supports also in place [3].

A small proportion of workplaces indicated a moderate or high co-occurrence of OHS and wellness activities, and similar findings have also been reported elsewhere [17,18]. Medium and large organizations were more likely to provide moderate to high co-occurrence of OHS and wellness activities compared to small workplaces, while having a JHSC did not show meaningful differences in wellness activities among small workplaces. This suggests that medium- to large-size workplaces are more likely than small workplaces to have the resources and supports in place to promote these efforts concurrently. Workplaces in the service, agriculture, and pulp and paper sectors were least likely to report co-occurring activities, while the electrical and utilities and municipal sectors were most likely. These differ from findings previously reported among employers in Massachusetts [18]. Whether the differences in our study reflect true differences in the employer population in Ontario compared to Massachusetts or reflects selection bias between the two studies, requires closer inspection. Further research is also needed to examine the workplace practices in sectors pertaining to low and high OHS and wellness activities to better understand the reasons for the implementation differences in Canada. Our findings also suggest that a people-oriented culture can at the very least support a higher implementation of wellness activities. Employer efforts to create a workplace culture of trust and respect might enhance workers' receptivity and openness to messages and programs designed to change behaviors and improve health [34,35].

Some limitations need to also be acknowledged when interpreting these findings. First, this was a cross-sectional study and causal relationships cannot be directly inferred. Second, the response rate was low, although the study's large initial sample size can facilitate the detection of more robust and reproducible statistical relationships than previous research with smaller sample sizes. We also reduced nonresponse bias by statistically weighting all modelling estimates to infer responses from a population of comparable organizations in the Ontario labor market. Third, our findings are only generalizable to the industry sectors we sampled and future studies need to examine how our findings relate to other industries such as the finance, information, professional, and entertainment sectors. Fourth, our use of a self-reported survey is prone to recall and social desirability biases. Differential measurement error is also possible across OHS and wellness activities. Respondents to the survey were selected based on their knowledge of OHS activities in the workplace, not on wellness activities. As such, it is possible that respondents could estimate OHS activities more accurately than wellness activities. Fifth, it is possible that some wellness activities were counted more than once if they were also provided as part of an EAP service (e.g., education, risk management tools, and self-care materials). Nonetheless, we conducted a sensitivity analysis and found that co-occurrence profiles did not meaningfully change when wellness activities that might be part of an EAP were removed. Lastly, the OLIP survey was not designed to collect detailed wellness information or the extent that these are integrated and coordinated with OHS activities.

Integrated OHS and wellness activities are widely promoted as an effective approach to chronic disease prevention [36]. This is partly explained by the emergence of evidence supporting the idea that workplace factors contribute to adverse health outcomes traditionally considered to be unrelated to work (such as stress, heart disease, and mental health) [10,11,37]. While distinguishing workplaces by their implementation of OHS and wellness activities may not reflect a truly integrated worker health approach [23], our findings provide a better understanding of the workplace factors associated with having suitable resources to support an integrated approach in the Canadian labor market. In 2016, a panel report from the National Institute of Health Pathways to Prevention Workshop identified small workplaces as a priority area for supporting integrated approaches through Total Worker Health® [38]. However, as our study and others have shown [15–18], there is a lack of demonstrated effectiveness in smaller workplaces in the concurrent adoption of health protection and wellness programming. Smaller workplaces might not integrate their OHS and wellness resources not because of a lack of support or motivation per se, but because of a lack of resources, including personnel, which might make it challenging to just perform traditional OHS hazard control alone [39]. Findings show that larger workplaces, with a people-oriented culture, and in the electrical and utilities, and municipal sectors are associated with more health protection and wellness resources that can be streamlined into integrated approaches. The next logical step is to examine intermediate and long-term health and productivity changes for these workplaces expected to benefit the most from co-occurring and integrated OHS and wellness activities. Subsequent findings can inform studies that are extended or scaled to other industries and smaller workplaces. Actionable recommendations whereby Canadian workplaces can integrate their existing OHS and wellness activities and ingrain these within a workplace's culture is also an area of research that needs to be explored further. Integration can also be enabled by an integrated safety and wellness committee, shared budgets and resources, and incentivizing employees in health protection and health promotion efforts [23].

#### **5. Conclusions**

This study provides valuable information on the co-occurrence of OHS and wellness activities and identifies workplace demographic factors most associated with their implementation in Canadian workplaces. Large workplaces, those in the electrical and utilities sector, and with a high rating for people-oriented culture are factors strongly related to the implementation of both OHS and wellness and might benefit most from integrated worker health activities. Future research needs to understand how to facilitate the uptake of OHS and wellness activities in workplaces with fewer concomitant organizational resources to increase OHS and wellness implementation. Furthermore, our findings need to be verified in other workplace contexts that were not explored in this study, and the factors that influence organizational change and worker participation need also to be better understood. Finally, it will be important to understand how to streamline OHS and wellness activities in workplaces for an integrated worker health approach.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1660-4601/15/12/2739/ s1, Table S1: Ontario Leading Indicators Project (OLIP): survey variable names.

**Author Contributions:** Conceptualization, A.B. and B.C.A.III; Methodology, A.B.; Software, A.B.; Validation, A.B. and B.C.A.III; Formal Analysis, A.B.; Investigation, A.B.; Resources, A.B., P.M.S., and B.C.A.III; Data Curation, C.N.S. and B.C.A.III; Writing—Original draft preparation, A.B.; Writing—Review and editing, C.N.S., P.M.S., L.S.R., I.A.S., and B.C.A.III; Visualization, A.B.; Supervision, B.C.A.III; Project Administration, C.N.S.; Funding Acquisition, C.N.S., P.M.S., L.S.R., I.A.S., and B.C.A.III.

**Funding:** This research was supported by a grant from the Workplace Safety and Insurance Board's Research Advisory Council (#09032) and the Province of Ontario, Canada.

**Acknowledgments:** The authors thank the other members of the 'Ontario Leading Indicators Project' research team: Sheilah Hogg-Johnson, Cameron Mustard, Michael Swift, Selahadin Ibrahim, and Sara Macdonald. The authors also provide a special thank you to our health and safety association partners in Ontario who helped us recruit workplaces, and to the workplaces themselves for participating.

**Conflicts of Interest:** The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
