Abstracts

Rationale: Previous national reports by McClelland et al (data: 1988-2003), Englot et al (1990-2008) and Schiltz et al (2005-2009) depicted remarkable racial disparities in access to surgical care for epileptic patients. With the implementation of the Affordable Care Act in 2010, rigorous attempts have been made to provide healthcare access to the uninsured US population across all races. In this report, we re-visit the question of racial disparity in: (1) access to surgical management of intractable temporal lobe epilepsy (TLE); and (2) short-term post-surgical outcomes post 2010 era as applicable to epilepsy surgery. Methods: Study design and data source: Population-based, observational, cohort study using the recent National Inpatient Sample (NIS) databases for the years 2012 and 2013. Cohort definition: Adult patients (>18 years) with a diagnosis of intractable TLE (ICD-9 diagnosis code: 344.41 or 344.51) were identified. To define access to surgical management, patients were dichotomized into those undergoing lobectomy (ICD-9 procedure code: 01.53) versus those without any procedure. Outcome endpoints: Access to surgical management for epilepsy across entire cohort, and post-operative outcomes [Unfavorable discharge (non-routine), prolonged length of stay (LOS) (>75th percentile) and high hospital charges (>75th percentile)] in surgical cohort only. Exposures: Primary exposure of interest is racial differences (African Americans and also other races with reference to Caucasians) for defined endpoints. Statistical methods: Various multivariable log-binomial models are were constructed to identify association of race for individual outcomes by adjusting for demographics (age, sex, insurance, socioeconomic status); Hospital characteristics (bedsize, region); comorbidities as stratified by Charlson Comorbidity Index as modified by Deyo et al to use on ICD-9 codes and total number of inpatient procedures. For robust estimates, we fitted models with generalizing estimating equations using sandwich-covariance matrix estimator to restrict clustering of similar outcomes within hospitals. For missing exposures ( < 10%), a model-based multiple-imputation approach as an alternative to traditional deletion methods was performed. Sensitivity analysis was performed by evaluating the estimates using 1000-bootstrapped replacement samples based upon nested clusters within hospitals. Results: Of the 4,062 patients with intractable TLE, 3.6% (n=148, AA=11) underwent surgery. Overall, the mean age of the cohort was 42.35 16.33 years (39.59 13.09 in surgical cohort), and 54% were female. In surgical cohort, a steep rise in surgeries that were funded by Medicare and Medicaid (41.9%) was noted compared to the pre 2010 data reported by Englot and colleagues ( 31.4%) et al. In an analysis unadjusted for covariates, African-Americans were less likely to undergo epileptic surgery for TLE (OR: 0.42; 95% CI: 0.23-0.79; p=0.007) in comparison to whites. In multivariable regression analysis adjusted for demographic variables, no disparities were noted in access to surgical care in these patients (Figure 1, forest plot). In the surgical cohort we noted no racial disparity in unfavorable discharge, high-hospital charges and prolonged LOS. (Figure 2, forest plot) Conclusions: Our study reflects no racial disparity in access to surgical care or short-term post-operative outcomes in patients with intractable TLE post 2010 amendment of the ACA, though the sample of African American patients was small. The seismic changes to the US healthcare system may plausibly have accounted for addressing the gap in racial disparity for epilepsy surgery. Funding: None