Abstract P120: Neighborhood Factors Influence Cardiovascular Outcomes for Depressed and Non-Depressed Individuals With Heart Failure With Reduced Ejection Fraction

Abstract

Introduction: Neighborhood-level measures including Area Deprivation Index (ADI) and walkability are associated with increased cardiovascular disease (CVD) morbidity, higher utilization of health services, and mortality. We sought to understand the impact of these neighborhood-level factors on hospital readmissions and mortality among recently hospitalized depressed and non-depressed individuals with heart failure and reduced ejection fraction (HFrEF) who enrolled in the NHLBI-funded Hopeful Heart Trial of “blended” collaborative care for HFrEF and co-morbid depression. . Hypothesis: We hypothesized worse ADI and Walk Scores will be associated with increased 12-month incidence of all-cause or cardiac-specific hospital readmission and mortality. Methods: From 3/14-10/17, Hopeful Heart enrolled 756 hospitalized HFrEF patients (EF<45%) and NYHA class II-IV symptoms including 629 depressed (screen-positive on 2-item Patient Health Questionnaire (PHQ-2) for depression prior to discharge and scored ≥10 on the PHQ-9 administered via telephone two weeks post-discharge) and 127 non-depressed HFrEF patients (PHQ-2 negative and PHQ-9 <5). We determined participants’ ADI score (0-100 scale, higher scores indicating greater deprivation) and Walk Score (0-100 scale, higher meaning more walkable) using home addresses at the time of hospital discharge. We categorized participants into top quartile (Q4) ADI and Walk score groups versus all other quartiles combined (Q1-Q3), and then compared these two groups using unadjusted, nonparametric Kaplan-Meier analyses with log-rank tests for differences in 12-month mortality and unadjusted chi-square tests for differences in 12-month incidence of readmission. Results: Of the 756 HFrEF participants (44% female; 73% white; EF: 28.3±9.1% mean; ADI: mean 66.7±23.5; Walk Score: mean 33.4±25.9), the mean PHQ-9 score for both ADI and Walk Score Q4 groups was similar to the Q1-3 groups. Participants in the Q4 group for ADI had a higher incidence of readmission that those in Q1-Q3 (81% -144 out of 178 - versus 70% - 400 out of 572, p=0.004), while those in the Q4 group for Walk Score were readmitted at a similar level to those in Q1-Q3. Those individuals in the top quartile for ADI and Walk Score showed no differences in cardiovascular (CV) mortality (14% incidence, 24 out of 171, for ADI; 14% total incidence, 25 out of 175 for Walk Score) or all-cause mortality (23% total incidence, 171 out of 750, for ADI; 23% total incidence, 175 out of 756 for Walk Score) compared with all other quartiles. Conclusions: Worse ADI but not Walk Score is predictive of higher risk of 12-month hospital readmission, and neither measure is predictive of 12-month CV or all-cause mortality. Further studies are needed to elaborate how area deprivation and other neighborhood factors affect outcomes for individuals with heart failure and design interventions to address these disparities.