![[ Visit Client Website ]](images/banner.jpg)
BACKGROUND: Neighborhood (i.e. small area) context is an important predictor of various health outcomes. While the effects of area-level demographic and socioeconomic measures on certain individual health outcomes have been well studied, little research has been done that examines the association between small-area characteristics and occupational health outcomes. This study investigated the relationship between county-level demographic and socioeconomic characteristics and high rates of occupational injury and illness, using establishments (i.e. workplaces) as the unit of analysis.
METHODS: Data on 40,249 establishments were obtained from the 2009 OSHA Data Initiative (ODI) database, including establishments’ injury and illness rate, industry, number of employees and address. Establishments were designated high-rate establishments if their 2009 rate of injuries and illnesses that result in lost work days, restricted duties or job transfer (DART) was twice the sample mean or greater.
County-level demographic and socioeconomic measures for the period 2005-2009 were obtained from the American Community Survey. Establishment locations were mapped using an online geocoding service provided by the University of Southern California. County-level characteristics were linked to establishments using the point-in-polygon spatial join capabilities available in ArcGIS 9.3.
Because this analysis involved measurements made at both the county and establishment levels, a multi-level approach was employed to take into account the hierarchical structure of the data, where data are nested within counties. A logistic regression model was fit to estimate the multivariable-adjusted odds that an establishment would have a high DART rate for all county and establishment-level independent variables (fixed effects), while specifying county as a random effect.
RESULTS: The mean DART rate for the sample was 3.93 (95% CI 3.88-3.97) per 100 full-time equivalent workers. Of the establishments in the sample, 15.85% (n=6,378) were identified as high-DART rate establishments. Controlling for the effect of establishments’ industry and size (number of employees), as well as the random effect of county, county poverty rate was positively associated with high DART rates (OR 1.24, 95% CI 1.08-1.43). Higher county levels of income inequality (OR 0.77, 95% CI 0.68-0.87) and the percentage of the county population that was foreign born (OR 0.88, 95% CI 0.81-0.96), of non-white race (OR 0.93, 95% CI 0.91-0.95) or living in owner-occupied housing (OR 0.92, 95% CI 0.87-0.97) were negatively associated with high DART rates.
CONCLUSIONS: The demographic and socioeconomic characteristics of counties may influence occupational injury and illness rates at workplaces located in them. Research is needed to elucidate possible underlying mechanisms.