International variation in the hospital burden of ANCA-associated vasculitis: a multinational ecological analysis, 2016-2022.

OBJECTIVES: To compare rates and characteristics of hospital episodes in which ANCA-associated vasculitis (AAV) was recorded as the primary discharge diagnosis across eight jurisdictions, including subtype distribution, demographic patterns, temporal trends, and in-hospital mortality. METHODS: We conducted a multinational ecological cross-sectional study using national hospital discharge databases from eight jurisdictions (Spain, Australia, Germany, England, Wales, Chile, Mexico, and the USA) between 2016 and 2022. AAV episodes were identified using ICD-10 primary diagnosis codes for granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA), and eosinophilic granulomatosis with polyangiitis (EGPA). Crude and age-adjusted hospitalisation rates, hospitalisation rate ratios and in-hospital mortality were analysed. Temporal trends were primarily assessed using Pearson-scaled Poisson models, with negative binomial models as sensitivity analyses. RESULTS: Among 758,257,987 hospital episodes, 116,039 were AAV (0.015%). GPA was the most frequent subtype (63.6%), followed by MPA (25.6%) and EGPA (10.9%). Crude AAV hospitalisation rates ranging from 0.04/100,000 in Mexico to 9.61/100,000 in Germany; age-adjusted rates were similar. Mortality data from Spain, Mexico, and the USA showed the highest AAV in-hospital mortality proportion in the USA (6.75%). In Pearson-scaled Poisson analyses, AAV hospitalisation rates increased in Spain (HRR 1.12, 95% CI 1.09-1.15; p < 0.001) and Australia (1.08, 1.06-1.10; p < 0.001), and decreased in Wales (0.86, 0.80-0.93; p < 0.001) and Chile (0.92, 0.86-0.98; p = 0.012). These trends persisted after exclusion of 2020 but were not statistically significant in negative binomial models. CONCLUSIONS: AAV hospital burden varied substantially across jurisdictions in rates, subtype distribution, and outcomes, warranting cautious interpretation in light of differences in population structure and health system capture.