The impact of occupational injury on the health of family members of injured workers
Keywords: Occupational injury and illness, family health status, propensity score matching
Many studies have examined the indirect economic and social consequences of occupational injuries and illnesses for workers. However, indirect impacts on the health of family members have not been studied. We hypothesize that family members of injured workers are more likely to experience ill health and need medical treatment than family members of uninjured workers. This could be due, for example, to increased physical burdens resulting from care giving and other work to make up for loss of worker income and contribution to household chores. These burdens, and the psychological stress on the worker, could also lead to stress in the family.
One problem in testing the hypothesis is that we cannot make a valid comparison of the family health status of injured and uninjured workers, since the occurrence of injury is likely to be associated with a variety of other factors that predict family health. Therefore, we restricted our study sample to include injured workers only. We divided these workers into two categories: severely injured, as indicated by the presence of indemnity payments for loss of work, and non-severely injured. We then hypothesize that families of severely injured workers will need more medical treatment than the families of non-severely injured workers. However, directly comparing the family health status of severely injured and non-severely injured workers can also cause bias, because severely injured workers are not a random sample of the population of injured employees. For instance, severely injured employees may have lower education and income levels and be exposed to more job hazards. To address this problem, we use a propensity score matching (PSM) method. Using this method, we compare family hospitalization rates and expenses of severely injured workers with those of non-severely injured workers who have similar characteristics. MarketScan data for 2002-2005 is used for the analysis. The health and productivity management module of the MarketScan data is linked with medical claims data for injured workers and their family members. The robustness of the PSM results is also examined by comparing family hospitalization rates and costs before and after injury. Our findings indicate that family members of severely injured workers are more likely to be hospitalized and incur higher inpatient expenditures than family members of non-severely injured workers.