SHIFTING THE LENS FROM EVENT RATES TO DAYS ALIVE AND OUT OF HOSPITAL (DAOH) AS A PATIENT-CENTERED PERFORMANCE METRIC: 30-DAY AND 1-YEAR TRENDS AND EQUITY AFTER HEART FAILURE HOSPITALIZATION IN SINGAPORE
Author(s)
Sibo Liu, MS1, Nicholas Graves, PhD1, Audry Lee, MRCP, MBBChir, MA2, Chun Fan Lee, PhD3, Sameera Senanayake, PhD1, Khung Keong Yeo, MBBS2, Derek Hausenloy, PhD4, Sanjeewa Kularatna, PhD1;
1Duke-NUS Medical School, Health Services Research & Population Health (HSRPH), Singapore, Singapore, 2National Heart Centre Singapore, Department of Cardiology, Singapore, Singapore, 3Duke-NUS Medical School, Centre for Quantitative Medicine, Singapore, Singapore, 4Duke-NUS Medical School, Cardiovascular & Metabolic Disorders Program, Singapore, Singapore
1Duke-NUS Medical School, Health Services Research & Population Health (HSRPH), Singapore, Singapore, 2National Heart Centre Singapore, Department of Cardiology, Singapore, Singapore, 3Duke-NUS Medical School, Centre for Quantitative Medicine, Singapore, Singapore, 4Duke-NUS Medical School, Cardiovascular & Metabolic Disorders Program, Singapore, Singapore
OBJECTIVES: Although temporal trends in readmission and mortality are routinely monitored after heart failure (HF) hospitalizations, it remains unclear whether changes in these conventional metrics translate into more days alive and out of hospital (DAOH) over time. We quantified trends in post-discharge DAOH, readmission, and mortality, and assessed whether improvements were shared across population subgroups.
METHODS: We used the Singapore Cardiovascular Longitudinal Outcomes Database (SingCLOUD) to identify patients with a first HF hospitalization discharged alive between 2011-2019. DAOH was calculated by subtracting total hospitalized days and days lost to death from 30-day or 1-year time windows. Risk-adjusted trends were estimated using multivariable zero-one-inflated-beta and logistic regression models. Trend heterogeneity by age, sex, and ethnicity was assessed via year×subgroup interaction terms.
RESULTS: 10,147 patients were included. Mean 30-day DAOH was 27.8 ± 5.3 days. Risk-adjusted 30-day DAOH showed no temporal change [annual change 0.04 percentage-points/yr (pp/yr), 95% confidence interval (CI), -0.10 to 0.18; P trend=0.554], although HF-specific readmissions declined while other outcomes remained stable. Mean 1-year DAOH was 308.5 ± 102.2 days. Risk-adjusted 1-year DAOH improved from 79.9% to 82.5% (0.32 [0.16-0.48] pp/yr; P trend<0.001), equivalent to an increase from 291.7 to 301.0 days (1.16 [0.57-1.75] days/yr). Contemporaneously, 1-year HF-specific readmissions, all-cause mortality, and cardiovascular mortality declined. Subgroup analyses showed steeper reductions in HF-specific readmission among men versus women at both horizons, and greater 1-year DAOH improvements among Malay than Chinese patients.
CONCLUSIONS: There was improvement in 1-year but not 30-day DAOH after HF hospitalization between 2011-2019, paralleling reductions in readmission and mortality. DAOH provides an easily communicable, patient-centered summary of progress, while subgroup differences in its trends and in HF-specific readmissions highlight potentially inequitable benefits. DAOH appears sensitive to subtle but important heterogeneity in system-level changes dispersed across components of care, supporting its use for health-system performance assessment, policy evaluation, and population-level outcome monitoring.
METHODS: We used the Singapore Cardiovascular Longitudinal Outcomes Database (SingCLOUD) to identify patients with a first HF hospitalization discharged alive between 2011-2019. DAOH was calculated by subtracting total hospitalized days and days lost to death from 30-day or 1-year time windows. Risk-adjusted trends were estimated using multivariable zero-one-inflated-beta and logistic regression models. Trend heterogeneity by age, sex, and ethnicity was assessed via year×subgroup interaction terms.
RESULTS: 10,147 patients were included. Mean 30-day DAOH was 27.8 ± 5.3 days. Risk-adjusted 30-day DAOH showed no temporal change [annual change 0.04 percentage-points/yr (pp/yr), 95% confidence interval (CI), -0.10 to 0.18; P trend=0.554], although HF-specific readmissions declined while other outcomes remained stable. Mean 1-year DAOH was 308.5 ± 102.2 days. Risk-adjusted 1-year DAOH improved from 79.9% to 82.5% (0.32 [0.16-0.48] pp/yr; P trend<0.001), equivalent to an increase from 291.7 to 301.0 days (1.16 [0.57-1.75] days/yr). Contemporaneously, 1-year HF-specific readmissions, all-cause mortality, and cardiovascular mortality declined. Subgroup analyses showed steeper reductions in HF-specific readmission among men versus women at both horizons, and greater 1-year DAOH improvements among Malay than Chinese patients.
CONCLUSIONS: There was improvement in 1-year but not 30-day DAOH after HF hospitalization between 2011-2019, paralleling reductions in readmission and mortality. DAOH provides an easily communicable, patient-centered summary of progress, while subgroup differences in its trends and in HF-specific readmissions highlight potentially inequitable benefits. DAOH appears sensitive to subtle but important heterogeneity in system-level changes dispersed across components of care, supporting its use for health-system performance assessment, policy evaluation, and population-level outcome monitoring.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
HSD29
Topic
Health Service Delivery & Process of Care
Disease
SDC: Cardiovascular Disorders (including MI, Stroke, Circulatory)