Temporal Dynamics, Demographic Determinants and Clinical Complexity of Heart Failure Hospitalizations in Italy Across the COVID-19 Pandemic Era

Abstract

Introduction. Heart failure (HF) is one of the leading causes of hospitalization, disability and healthcare resource use in high-income countries, particularly in ageing populations such as Italy. The COVID-19 pandemic profoundly disrupted the organization of healthcare services, raising concerns about the continuity of chronic disease management and the true long-term epidemiological impact on HF admissions.

Objectives. To analyse the long-term temporal trends of HF hospitalizations in Italy over a fifteen-year period (2008–2022), assessing the contribution of demographic factors, seasonality and clinical complexity. As a secondary objective, we tested whether the pandemic produced a sustained, structural shift in the epidemiological curve of admissions.

Methods. Retrospective observational study based on national Hospital Discharge Records (SDO) from both public and private facilities. Admissions with HF as the primary diagnosis were identified through ICD-9-CM codes. Cases were stratified by age, sex, season, DRG relative weight and clinical severity according to the Elixhauser Comorbidity Index. Temporal trends were estimated using a Poisson Inverse Gaussian Generalized Linear Mixed Model, with a random intercept for region and the time component modelled through a segmented regression with a structural break in 2020, distinguishing the pre-pandemic trend, the level change and the slope change in the late-pandemic phase. The Annual Percent Change (APC) and the Average Annual Percent Change (AAPC) were also computed. The analysis is a refinement of a recently published manuscript (Amodio et al., 2026). All analyses were conducted with R (version 4.5.0).

Results. Over the study period, 3,162,075 HF admissions were recorded, corresponding to a crude rate of 35.11 per 10,000 person-years. A slight male preponderance emerged, and age was confirmed as the strongest determinant of risk, with rates increasing exponentially up to 228.31 per 10,000 person-years in subjects aged ≥75 years. A marked seasonal pattern was observed, peaking in winter. The largest share of admissions (59.3%) fell within the intermediate comorbidity tier (Elixhauser 13–20). Temporal analysis revealed a long-term structural decline, with a sharp drop in 2020 (−24%; RR 0.80) followed by a partial rebound in 2021–2022 (RR 1.06). Highly complex patients (Elixhauser >20) showed no signs of recovery, with an accelerated decline in the late-pandemic phase, whereas resource-intensive admissions (DRG weight >1.05) displayed a steadily increasing trend. The model incorporated region-specific random slopes, demonstrating that regional temporal trajectories were highly consistent with national aggregated estimates across all comorbidity profiles.

Conclusions. The burden of HF hospitalizations in Italy remains substantial and evolving, driven by population ageing, sex differences, seasonality and rising clinical complexity. Population-level data show no sustained, structural increase in admissions during the late-pandemic phase: the observed fluctuations appear attributable to organizational and behavioural factors, such as service disruption and care avoidance, rather than to a true epidemiological shift. These findings highlight the need for integrated and resilient care models capable of ensuring continuity of HF management even during health emergencies.