Healthcare Disparities In Heart Failure Patients And Use Of Sodium Glucose Co-Transport Inhibitors (SGLT2-i)

Abstract

Introduction Co-existence of heart failure (HF) and diabetes presents myriad of challenges in patientcare which are further complicated by socioeconomic disparities that disproportionately affect patients of color. To understand and address the unique challenges faced by socioeconomically disadvantaged HF-patient population in the Bronx county of New York City, we sought to phenotype the burden of diabetes, and understand any disparities including the use of SGLT2-i. Methods Consecutive adult(18years and older) HF patients [reduced and preserved left-ventricular ejection fraction (LVEF)] seen in the advanced-HF clinics were included in the analysis. Patients were categorized based on race/ethnicity and compared for baseline characteristics and associated outcomes including worsening HF event, all-cause hospitalization in the previous 1-year, and medication adverse-events. Results Of the 1218 HF-patients [n=556 non-Hispanic-Black (nH-Black), n=485 Hispanic, n=177 non-Hispanic-White (nH-White)]; 541(44.4%) had type-2 diabetes-mellitus (44.1% nH-Black, 47.0% Hispanic, 38.4% nH-White, p=0.14). In the HF-DM cohort, substantial proportion (20.9%) had poorly controlled diabetes (HgA1c 9% or higher) that did not vary with race/ethnicity. The median-household-income based on residential address was significantly lower for nH-Black $41,500 (33,490-58,392) and Hispanic $37,804 (29,579-54,278) cohorts compared to nH-White $65,399 (52,091-92,232) cohort. Overall use of SGLT2-i was low (5.7% overall, 0.9% in HF-NoDM and 11.7% HF-DM). On further analysis of HF-DM patients based on SGLT2-i use, patients on SGLT2-i were younger [62(54-69)vs.66(58-75)years], more nH-White (22.2%vs.11.3%); but were also more likely to have HF with reduced ejection fraction (HFrEF)(69.8%vs.54.2%), more poorly controlled DM (31.8%vs.19.5%) and higher median HgA1c levels[7.9(6.9-9.2)vs.7.0(6.3-8.4)%], all p<0.05. Based on race/ethnicity, while the three groups had statistically similar rates of worsening HF events (30.9% nH-Black, 26.0% Hispanic, 24.3% nH-White, p=0.104), the rate of all-cause hospitalization was higher in nH-Black(38.9% nH-Black, 32.2% Hispanic, 31.2% nH-White, p=0.040) . Using logistic-regression model, nH-Black had higher Odds-ratio (OR)=1.46(1.01-2.11), p=0.043 when compared to nH-White for all-cause hospitalization adjusted for age, gender, DM, hypertension, ischemic-cardiomyopathy and LVEF. However, this association was substantially attenuated (p=0.14) when the model was adjusted for median-household-income. Conclusion These findings highlight the complex relationships between income inequality, social determinants of health (including language barriers) and observed racial/ethnic differences in HF and diabetes. Co-existence of heart failure (HF) and diabetes presents myriad of challenges in patientcare which are further complicated by socioeconomic disparities that disproportionately affect patients of color. To understand and address the unique challenges faced by socioeconomically disadvantaged HF-patient population in the Bronx county of New York City, we sought to phenotype the burden of diabetes, and understand any disparities including the use of SGLT2-i. Consecutive adult(18years and older) HF patients [reduced and preserved left-ventricular ejection fraction (LVEF)] seen in the advanced-HF clinics were included in the analysis. Patients were categorized based on race/ethnicity and compared for baseline characteristics and associated outcomes including worsening HF event, all-cause hospitalization in the previous 1-year, and medication adverse-events. Of the 1218 HF-patients [n=556 non-Hispanic-Black (nH-Black), n=485 Hispanic, n=177 non-Hispanic-White (nH-White)]; 541(44.4%) had type-2 diabetes-mellitus (44.1% nH-Black, 47.0% Hispanic, 38.4% nH-White, p=0.14). In the HF-DM cohort, substantial proportion (20.9%) had poorly controlled diabetes (HgA1c 9% or higher) that did not vary with race/ethnicity. The median-household-income based on residential address was significantly lower for nH-Black $41,500 (33,490-58,392) and Hispanic $37,804 (29,579-54,278) cohorts compared to nH-White $65,399 (52,091-92,232) cohort. Overall use of SGLT2-i was low (5.7% overall, 0.9% in HF-NoDM and 11.7% HF-DM). On further analysis of HF-DM patients based on SGLT2-i use, patients on SGLT2-i were younger [62(54-69)vs.66(58-75)years], more nH-White (22.2%vs.11.3%); but were also more likely to have HF with reduced ejection fraction (HFrEF)(69.8%vs.54.2%), more poorly controlled DM (31.8%vs.19.5%) and higher median HgA1c levels[7.9(6.9-9.2)vs.7.0(6.3-8.4)%], all p<0.05. Based on race/ethnicity, while the three groups had statistically similar rates of worsening HF events (30.9% nH-Black, 26.0% Hispanic, 24.3% nH-White, p=0.104), the rate of all-cause hospitalization was higher in nH-Black(38.9% nH-Black, 32.2% Hispanic, 31.2% nH-White, p=0.040) . Using logistic-regression model, nH-Black had higher Odds-ratio (OR)=1.46(1.01-2.11), p=0.043 when compared to nH-White for all-cause hospitalization adjusted for age, gender, DM, hypertension, ischemic-cardiomyopathy and LVEF. However, this association was substantially attenuated (p=0.14) when the model was adjusted for median-household-income. These findings highlight the complex relationships between income inequality, social determinants of health (including language barriers) and observed racial/ethnic differences in HF and diabetes.