Socioeconomic disparities in long-term mortality after infective endocarditis in Denmark: a nationwide cohort study.

Abstract

Socioeconomic position (SEP) influences several determinants of infective endocarditis (IE) progression. Whether IE mortality differs by SEP remains unclear. We examined 5-year mortality after IE by individual-level SEP. Using nationwide Danish registries, we identified patients with first-time IE (2010-2022). SEP was assessed by educational and affluence level, categorised as low, medium, and high. The Kaplan-Meier estimator provided 30-day, 1-, 3-, and 5-year mortality risks, risk differences, and risk ratios. Time-varying hazard ratios were derived from flexible parametric models. We estimated attendance at outpatient and dental follow-up visits (indicating adherence to follow-up) using the Aalen-Johansen estimator and modelled hospitalisation rate during follow-up (indicating comorbid disease burden) using a joint frailty model, treating death as a competing event in all models. The 5-year mortality by educational level was 63.1% (95% CI: 60.8%-65.2%) for low, 53.1% (95% CI: 51.3%-54.9%) for medium, and 45.8% (95% CI: 42.5%-48.8%) for high education. The excess mortality was particularly pronounced within the first 2 years. Within one year after discharge, 65.2% (95% CI: 62.9%-67.4%) of low, 73.6% (95% CI: 72.0%-75.2%) of medium, and 74.5% (95% CI: 71.8%-77.2%) of high-education patients had an outpatient contact. The corresponding five year dental visit proportions were 44.0% (95% CI: 41.6%-46.5%), 64.0% (95% CI: 62.2%-65.9%), and 74.0% (95% CI: 71.2%-76.9%), respectively. The hazard ratio for hospitalisations was 1.35 (95% CI: 1.25-1.49) when comparing low <i>vs.</i> high education and 1.19 (95% CI: 1.10-1.27) for low <i>vs.</i> medium. Estimates and time-dependent patterns were similar for affluence level. Patients with low SEP had higher IE mortality, particularly within the first 2 years. Reduced adherence to follow-up care and comorbid diseases may contribute to this. Determining how the excess mortality is directly linked to the IE episode warrants further investigation. Independent Research Fund Denmark (grant no. 3101-00102B) and Center for Population Medicine, Department of Clinical Epidemiology, Aarhus University.