Associations Between Substance Abuse, Socioeconomic Status and Mortality Among Burn Patients

Purpose   

     Prior research has found substance abuse and socio-economic status (SES) to be important predictors of health outcomes (Pickett & Pearl, 2001). Although there are several factors that may explain how socio-economic status increases the risk of injury, research findings have been conflicting (Cubbin & Smith, 2002). The aim of this study was to understand the implications of chronic substance abuse and SES in predicting mortality in burn patients. The three primary hypotheses tested were: (1) there will be a significant interaction between SES and chronic substance abuse; (2) there will be a significant interaction between gender and chronic substance abuse; (3) there will be a significant interaction effect between race and SES in predicting mortality.

Methods

     The total cross-sectional sample was comprised of 1256 burn patients admitted to the burn unit at a large urban hospital between 2005 and 2010. The sample includes males (57.5%) and females (42.5%) and has the following ethnic demographic: White (55.0%) and Non-White (45.0%). The age distribution across categories include: 0-29 years (21.5%), 30-59 years (42.5%), 60+ years (36.0%). Gender, race, mortality and diagnoses for chronic alcohol abuse and chronic drug abuse were included in the patient’s medical history. In addition, a composite variable was created to identify any chronic substance abuse. SES was measured by the principle payer variable. Principle payer has been used in prior research as a proxy for SES (e.g., Carlisle & Leake, 1998). In order to test the primary hypotheses, logistic regression analyses were used to measure the significance of interaction effect between chronic substance abuse (independent variable) and SES and/or gender (moderators) in predicting mortality (dependent variable). In addition, logistic regression was used to test the interaction effect of SES and race in the prediction of mortality.

Results

     There was no significant interaction between SES and chronic substance abuse and mortality. Second, the interaction effect between chronic substance abuse and gender was not significantly associated with mortality outcomes, indicating that gender did not moderate the association between chronic substance abuse and mortality. Third, there was a significant interaction between race and SES in predicting mortality. The results suggested that non-white patients of low SES were at a higher risk of mortality than white patients of low SES. An additional analysis that adjusted for age was also conducted which resulted in a non-significant association between SES and mortality.

Implications

     The results of this study indicated that the interaction effect between SES and race was significantly associated with increased odds of mortality. Specifically, non-white burn patients of low SES were at 2x the risk of death than white burn patients of low SES; however, when SES was adjusted for age, there was no longer a relationship between SES and mortality. Given these findings, it would be useful for future research to consider age when examining the strength of associations between SES and mortality. Also, future research examining type of insurance and health outcomes over time would be helpful.