2007 HSR&D National Meeting Abstract
1069 — Hospital Community Effects on Racial Disparities in Use of Cardiac Procedures
Urech TH (Division of Health Policy and Quality of Care and Utilization Studies, Michael E. DeBakey VAMC) , Petersen LA
(Division of Health Policy and Quality of Care and Utilization Studies, Michael E. DeBakey VAMC)
We sought to determine whether African American patients receiving post-acute myocardial infarction (AMI) care at VA hospitals located in predominantly African American (AA) communities are more likely to undergo a post-AMI procedure compared to AA patients receiving care at VA hospitals located in predominantly white areas.
Using administrative data, we identified veterans diagnosed with an index episode of AMI at VA cardiac surgery hospitals between fiscal years 1999 and 2001. We categorized each hospital as being located in a predominantly AA or predominantly white area using county-level 2000 US Census demographic data. We determined whether the type of community where the hospital was located was associated with the odds of receiving a post-AMI cardiac procedure (coronary angiography, coronary artery bypass graft surgery [CABG], and percutaneous coronary intervention [PCI]) and one-year mortality among AA patients and white patients using logistic regression. We controlled for age, illness burden, and Medicare reliance.
We examined 13,512 post-AMI patients (11,396 white and 2,116 AA) that received care in 40 VA cardiac surgery hospitals. 22 hospitals were in white areas and 18 were in AA areas. In adjusted analyses, AA patients cared for in AA communities were more likely to undergo coronary angiography compared to AA patients cared for in hospitals in white communities (odds ratio [OR]1.74, 95% confidence interval [CI] 1.44, 2.10). White patients cared for in AA areas were more likely to undergo coronary angiography compared to white patients in white areas (OR1.35, 95% CI 1.24, 1.46). The interaction effect of hospital community area and race was not statistically significant for CABG and PCI. Differences in one-year mortality were also not statistically significant.
In many studies, AA patients are less likely to receive post-AMI procedures compared to white patients. Hospital community effects may contribute to disparities in receipt of post-AMI coronary angiography among AA patients.
Researchers studying racial disparities in receipt of post-AMI procedures should consider incorporating hospital community area characteristics in addition to both patient-level and hospital-level characteristics in their analyses.