FRI129 Adrenal Insufficiency In Olmsted County, MN, USA: A Population-Based Study

Abstract

Abstract Disclosure: M. Moazzami: Grant Recipient; Self; Rochester Epidemiology Project. S. Achenbach: None. E. Atkinson: None. I. Bancos: None. Background: Epidemiology of adrenal insufficiency (AI) is incompletely characterized with most data originating from referral centers and registries. We aimed to determine the incidence, morbidity index, risk of hospitalizations, and mortality in patients with AI living in Olmsted County, Minnesota. Methods: We assessed the standardized incidence rate, morbidity index, risk of hospitalizations and mortality of AI in residents (aged 18+) in Olmsted County, MN, USA, from Jan 1, 2005, to December 31, 2021. The data was collected using the Rochester Epidemiology Project (REP), a database of medical records for all Olmsted County residents since 1966. Incidence rates were standardized for age and sex according to the 2020 US white population. Sex- and age- matched referent subjects were selected from the REP. AI was defined based on presence of oral chronic glucocorticoid therapy (hydrocortisone, prednisone, prednisolone, or dexamethasone) defined as (1) any hydrocortisone use or (2) 6 months of consecutive use of prednisone, prednisolone, or dexamethasone or (3) two AI diagnosis codes 6+ months apart plus at least one glucocorticoid prescription within 1 year of the second diagnosis. Elixhauser morbidity index was used. Findings: Between 2005 and 2021, 5419 patients were diagnosed with AI (median age of 62 years, range 18-103), 56% women). Overall standardized incidence rates for AI increased from 206 (95% CI 174-238) in 2005 to 252 (223-280) per 100 000 person-years in 2021. At baseline, patients with AI had a higher morbidity index (median 3 vs median 1, p <0.001) and a higher history of hospitalizations within 5 years prior to baseline (57% vs 30%; p<0.001) when compared to referent subjects. When comparing patients with AI to referent subjects, and after adjusting for age, sex, morbidity index and prior history of hospitalizations, patients with AI were more likely to be hospitalized during follow-up (aHR 1.4, 95% CI 1.3-1.5). When comparing patients with AI to referent subjects, and after adjusting age, sex, and morbidity index, patients with AI had higher mortality (aHR 1.7, 95% CI 1.6-1.9). Patients with AI demonstrated a higher risk of dying from musculoskeletal disorders (aHR 4.9, 95% CI 2.2-10.8), respiratory disorders (aHR 4.7, 95% CI 3.3-6.6), neoplasms (aHR 2.4, 95% CI 2.0-2.9), and infectious disorders (aHR 2.3, 95% CI 1.2-4.3) than referent subjects. Interpretation: We demonstrate an increase in AI incidence between 2005 and 2021. Surprisingly, even after controlling for morbidity and prior hospitalizations, patients with AI were more likely to be hospitalized. In comparison to referent subjects, mortality was higher in patients with AI, even after adjustment for morbidity index. Presentation: Friday, June 16, 2023