Community-level cardiovascular risk factors impact geographic variation in cardiovascular disease hospitalizations for women

Abstract

Prior work has shown significant geographic variation in cardiovascular (CV) risk factors including metabolic syndrome, obesity, and hypercholesterolemia. However, little is known about how variations in CV risk impact cardiovascular disease (CVD)-related hospitalizations. Community-level CV risk factors (hypertension, dyslipidemia, hyperglycemia, and elevated waist circumference) were assessed from community-wide health screenings sponsored by Sister to Sister (STS) from 2008 to 2009 in 17 major US cities. Using data from the Healthcare Cost and Utilization Project's Nationwide Inpatient Sample (HCUP-NIS), CVD hospitalizations were identified based on ICD-9 codes for acute myocardial infarction (AMI), congestive heart failure (CHF), and stroke. We linked STS data with HCUP-NIS hospitalizations based on common cities and restricted the analysis to women discharged from hospitals inside the STS cities. Using hierarchical models with city as the random intercept, we assessed the impact of city-specific CV risk factors on between-city variance of AMI, CHF, and stroke. Analyses were also adjusted for patient age and clinical comorbidities. Our analysis yielded a total of 742,445 all-cause discharges across 70 hospitals inside of 13 linked cities. The overall city-specific range proportion of AMI, CHF, and stroke hospitalizations were 1.13 % (0.75-1.59 %), 2.57 % (1.44-3.92 %), and 1.24 % (0.66-1.84 %), respectively. After adjusting for city-specific CV risk factors, between-city variation was no longer statistically significant for all CVD conditions explored. In conclusion, we found that geographic variations in AMI, CHF, and stroke hospitalizations for women may be partially explained by community-level CV risk factors. This finding suggests that interventions to reduce CVD should be tailored to the unique risk profile and needs of high-risk communities.