AF/AFL-related hospitalizations in a provincial Chinese discharge-record database, 2016–2024: a retrospective observational study

Atrial fibrillation and atrial flutter (AF/AFL) commonly coexist with older age and multimorbidity. Temporal changes in discharge-record counts are difficult to interpret when database coverage and recurrent admissions also change. We analyzed 2016–2024 hospital discharge front-page records with an I48 code in any of 16 diagnosis fields. Seven procedure code-name pairs were reviewed. We quantified annual institution coverage, repeated admissions among valid de-identified patient identifiers, and outcomes stratified by primary versus secondary AF/AFL diagnosis. Stable-institution analyses included 1,514 institutions reporting in every study year. Patient-cluster-robust logistic models were fitted in two layers: a baseline model using calendar year, demographics, diagnosis position, and mutually defined diagnosis groups, and an extended descriptive model adding in-hospital ventilation, dialysis, and cardiac ablation-related procedures. The cohort comprised 2,868,578 AF/AFL-related hospitalizations from 1,098,442 valid unique patients and 5,069 institutions. Annual records increased from 142,305 in 2016 to 501,878 in 2024 (252.7%); records from the stable 1,514-institution subset increased from 137,063 to 443,780 (223.8%). Among admissions with valid patient identifiers, subsequent admissions increased from 26.4% in 2016 to 70.9% in 2024. AF/AFL was primary in 132,018 admissions (4.6%). Primary-diagnosis admissions had lower in-hospital death (0.32% vs. 2.47%), shorter median length of stay (7 vs. 8 days), and more cardiac ablation-related procedures (25.7% vs. 0.33%) than secondary-diagnosis admissions. In baseline models, calendar year was associated with lower odds of death (aOR 0.91, 95% CI 0.91–0.91) and death/DAMA (aOR 0.93, 95% CI 0.93–0.93). Associations changed after adding post-admission interventions, including calendar-year aORs of 0.88 (0.87–0.88) and 0.91 (0.91–0.91), respectively. AF/AFL-related discharge records increased in both the full and stable-institution cohorts, but the study cannot separate database expansion, repeated admissions, coding change, and population-level disease trends. The findings describe hospital-record patterns rather than incidence or population burden. Primary and secondary AF/AFL diagnosis admissions represent clinically different administrative strata, and adjusted associations should not be interpreted causally.