Contemporary Rates of Hospitalization for Heart Failure in Young and Middle-Aged Adults in a Diverse US State

Risk of heart failure (HF) increases with age and is well-described as the leading cause of hospitalization in older adults (≥65 years [y]). However, over the past decade, younger adults (<65y) in the US have experienced the greatest increases in HF-related deaths with worsening Black-white disparities.¹ Contemporary data on the burden of, and racial/ethnic disparities in, HF hospitalizations among younger adults are limited, and historical estimates largely precede the worsening of HF-related death rates.^{2, 3} Therefore, we sought to examine contemporary rates of HF hospitalizations stratified by age, race/ethnicity, and sex.

METHODS

The data that support the findings of this study are available from the corresponding author upon reasonable request. We analyzed claims data from all hospitals included in the Illinois Health and Hospital Association’s Comparative Health Care and Hospital Data Reporting Services (N=204). We included non-Hispanic (NH) white, NH Black, and Hispanic Illinois residents aged ≥35y hospitalized between 2016-2018 with HF (International Classification of Diseases, Tenth Revision: I09.81, I11.0, I13, I50) listed as the principal diagnosis. We excluded adults who identified as Asian or other/unknown race/ethnicity due to small sample sizes and those <35y to minimize the effect of congenital heart disease. We estimated the Charlson Comorbidity Index (range: 1-30) for each patient. We applied age, race/ethnicity, and sex-stratified 5-year population estimates from the American Community Survey (Illinois: 2013-2017) to negative binomial regression models to calculate annualized rates of HF hospitalization; in adjusted models, age group was included as a categorical variable. We calculated incidence rate ratios (IRR [95% CI]) to compare NH Black and Hispanic adults with NH white adults, respectively, within each age and sex category. This analysis of deidentified administrative data was exempt from institutional review board approval. Analyses were completed using SAS v.9.4. (Cary, NC).

RESULTS

Of 137,582 hospitalizations with a principal diagnosis of HF, 50% were men, 28% were NH Black, 6% were Hispanic. Median Charlson Comorbidity Index score (IQR) was 4 (3, 6) for the overall sample, and overall, 27% of HF hospitalizations occurred in adults <65y. Among NH Black men and women, 59% and 43% of those hospitalized for HF were <65y; among Hispanic men and women hospitalized for HF, 41% and 25% were <65y. In contrast, 21% and 12% of NH white men and women hospitalized for HF were <65y.

Rates of hospitalization for HF were 5.9 (95% CI: 4.3, 8.2) times higher in NH Black compared with NH white and not different between NH white and Hispanic adults (IRR = 1.0 [95% CI: 0.8, 1.4]) after adjustment for age and sex. When stratified by age-category and race/ethnicity-sex, rates of HF hospitalizations (per 10,000) were higher for NH Black compared with NH white men and women in every age category, but the greatest differences were in those 35-44y (IRR: 10.8 [9.2, 12.7] and 10.0 [8.2, 12.3], respectively). Rates of HF hospitalization were similar between Hispanic and NH white adults (Figure 1).

Figure 1. Incidence rate ratio (95% CI) and rates (per 10,000) of heart failure hospitalizations among a) men and b) women by age category and race/ethnicity, 2016-2018.

Rates of annualized heart failure hospitalization (per 10, 000 population) are shown below figures for a) men and b) women; annualized rates of HF hospitalization for white men and women used as reference. Points represent incidence rate ratio estimates and bars represent 95% confidence intervals. CI: confidence intervals; y: years; NH non-Hispanic.

DISCUSSION

Our findings expand upon prior analyses and demonstrate a significant contemporary burden of early onset HF, most prominent among NH Black adults. In our sample, approximately half of NH Black men and women hospitalized for HF were <65y leading to significantly higher disability-adjusted life years and years of potential life lost. Indeed, the median Charlson Comorbidity Index score in our study was 4, which in a previous HF cohort was associated with a 3-year mortality rate of >50%.⁴ While our study identifies that a significant proportion of patients hospitalized for HF are <65y, younger adults in this age range have not been been well represented in HF clinical trials or registries, highlighting a critical evidence gap. The most pronounced Black-white disparities in our data are in the youngest age categories and are similar in magnitude to the 20-fold increased incidence of HF in Black compared with white adults <50y reported in Coronary Artery Risk Development in Young Adults (CARDIA) in 2009.⁵

A key strength of our study is the use of a large group of hospitals with a diverse population that is fairly representative of the US population, however results may not apply to other geographic areas. Limitations include the use of billing codes which may lead to misclassification. As the dataset did not include federal hospitals, this analysis may have under- or over-estimated true HF hospitalization rates. We were unable to exclude the contribution of re-admissions, but it is unlikely to disproportionately lead to higher HF hospitalization rates in younger age groups. Finally, this analysis was not able to adjust for key confounders (e.g. socioeconomic status) and does not identify the root causes that underlie Black-white disparities in HF (e.g. structural and systemic racism).

In summary, these data highlight the critical emerging problem of HF in the young, particularly among NH Black men and women, and its significant contribution to overall contemporary rates of HF hospitalizations. Prevention efforts targeting key HF risk factors as well as clinical trial design and HF disease management will need to reorient to individuals earlier in the life course in order to mitigate disparities, and promote health.