The economic impact of employee health on the organization.

This dissertation study includes four manuscripts. The health-related measures were collected from health risk appraisals completed by the volunteers employed by a manufacturing company. The economic measures included absenteeism records provided by the employer and medical claim data collected by the insurance company. The first two manuscripts focus on bivariate relationships. As reported in the first manuscript, demographic measures, as well as 20 of the 25 health related measures (adjusted for demographic differences) were significantly related to self-reported physical activity levels (p $<$.05). In the second manuscript, as compared to employees with low health risks (in 1985), the employees with high health risks in 18 health-related measures were higher in average annual medical insurance costs (1985 to 1987; in 1985 dollars: excluding pregnancy cost). The high risk group also had a significantly higher percentage of employee whose costs were above the group mean (according to age and sex) in 11 of the 18 measures (p $<$.05). In addition, both the average annual medical costs and the percent of high cost employees increased as the total number of high risk factors per employee increased, regardless of sex and age. Multivariate analyses were performed in the last two manuscripts. As demonstrated in the third manuscript, the selected health-related measures significantly predicted the prospective (1985 to 1987) average annual medical costs (p $<$.0001). The multiple R$\sp2$ for the total, male and female employee groups were.18,.17, and.22, respectively. Using the same predictors and coefficients of the regression formula, the average annual costs (log10 form) could also be successfully predicted for a different sample who had similar employee characteristics. In the last manuscript, the selected health-related measures could significantly predict the three major economic measures (medical claim costs, absenteeism, and both) for the hourly employees by three age/sex sub-groups, respectively. The multiple R for the nine regression models ranged from.478 to.357 (p $<$.001). In conclusion, these four manuscripts lead to the major conclusion "employees with better health result in economic savings for the organization.".PhDPhysical Education: KinesiologyUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/104416/1/9034548.pdfDescription of 9034548.pdf : Restricted to UM users only