Abstracts

Rationale: Self-management is an imperative for patients with epilepsy, since the most important strategy to control seizures is taking prescribed medications and the individual nature of epilepsy makes it impossible to predict every possible contingency. This study provides further understanding of how epilepsy self-management and the antecedents of self-management differ in different socioeconomic groups, and the relationship between self-management and health care use. Several hypotheses are tested about relative importance of antecedent factors in predicting self-management behavior. Methods: The project builds on an existing study that is following a cohort of 450 epilepsy patients under treatment at two clinical sites in Houston, Texas. The participants in the existing study will be recruited for a supplemental survey covering various self-management domains and the antecedent factors associated with self-management. The survey will be self-administered by the patients. We will test if knowledge, control, outcome expectations, self-efficacy, social support, and patient satisfaction are positively associated with self management, and anxiety/depression and stigma are negatively associated with self management. We will use multivariate modeling to test if the antecedent factors have a differential relationship with self-management comparing high and low socio-demographic groups. We will also examine the relationship between self-reported self-management and health care use. Results: We have identified significant socio-demographic disparities in health care for people with epilepsy. Minority patients with low incomes, Medicaid or no coverage, were more likely to receive generalist physician visits, go to the hospital ER, and be admitted to the hospital for care compared to middle and upper income white patients with private coverage. Their use of specialist visits was lower and their use of new generation AEDs was similar. We were able to show that the socio-demographic disparities in health care use were largely eliminated or reduced when controlling for site of care. We expect these same variations to exist with respect to self-management and its antecedents. We also expect there to be an inverse relationship between self-management and health care use. Conclusions: Preliminary results from the existing study suggest that health care patterns could be modified with strategies designed to change the practices of providers at university-affiliated safety net sites with a goal to equalize the service patterns among patient cohorts. The results also suggest that patient education could be provided to improve the knowledge, attitudes, expectations, self-management, and communication skills of patients. The supplemental study described here is designed to determine if self-management education should be a major focus of intervention strategies aimed at equalizing service patterns across socio-demographic groups.