Hospital Variation of Spironolactone Use in Patients Hospitalized for Heart Failure in China—The China PEACE Retrospective Heart Failure Study

BACKGROUND: Although aldosterone antagonists improve outcomes in select individuals with heart failure and reduced ejection fraction, studies in the United States have raised concerns about underuse and overuse. Variations in the prescription of aldosterone antagonist in China are unknown. METHODS AND RESULTS: In the multicenter, hospital‐based, retrospective China PEACE (China Patient‐Centered Evaluative Assessment of Cardiac Events) study, we identified a nationally representative cohort of admissions for heart failure in a nationally representative sample of Chinese hospitals in 2015. Patients were classified into 1 of 3 groups according to their eligibility for spironolactone—“ideal” (left ventricular ejection fraction <40% and without contraindications), “contraindicated” (a documented contraindication, irrespective of left ventricular ejection fraction), and “uncertain‐benefit” (all others). We measured hospital variation of spironolactone prescriptions at discharge in the “ideal” and “contraindicated” group and calculated the median odds ratio (MOR), a measure of institution‐level variation for 2 individuals with similar characteristics discharged at 2 randomly selected hospitals. Hospital characteristics associated with spironolactone use were identified using multivariable linear regression model. Among 1222 ideal patients from 97 hospitals, the median rate of spironolactone prescription was 78.6% (interquartile range [IQR], 42.8%–89.6% [range, 0%–100%], MOR, 3.4 [95% CI, 2.7–4.0]) at discharge. Among 900 contraindicated patients from 83 hospitals, the median rate of spironolactone prescription was 30.0% (IQR, 9.1%–50.0% [range, 0%–100%], MOR, 3.1 [95% CI, 2.4–3.9]) at discharge. Hospitals with independent departments of cardiology and located in Eastern China were associated with a 38.0% (95% CI, 18.7–57.3; P<0.001) and a 14.6% (95% CI, 2.3%–26.9%; P=0.020) higher rate of spironolactone use for ideal patients. CONCLUSIONS: In this national study of hospitals in China, the use of spironolactone among ideal patients and the inappropriate use of spironolactone among patients with contraindications was substantial, with rates that varied markedly by institution. REGISTRATION: URL: https://www.clinicaltrials.gov. Unique identifier: NCT02877914.