Behçet's syndrome incidence and prevalence in Sardinia: implications of a latent class analysis combining...

To estimate the epidemiology of Behçet's syndrome (BS) in Sardinia using a combined administrative and clinical data latent class analysis (LCA).

Cases with a diagnosis of BS were retrieved from 2006-2016 in the Hospital Discharge Records (HDRs), 2006-2016 Rare Diseases Regional Register (RDRR), and 2014-2016 in the Specialist Outpatient Database (SOD). Medical records from regional rheumatology clinics were reviewed and classified by the International Criteria for Behçet's Disease (ICBD) and International Study Group (ISG) criteria. Statistical analysis involved cross-referencing databases and applying LCA to estimate the probability of a BS diagnosis. Prevalence and incidence of BS in Sardinia were calculated, as well as the sensitivity and specificity of each database source.

Administrative databases analysis retrieved 271 unique cases. A medical record review of 133 patients identified 116 BS cases, of which 107 matched administrative records and 9 were new. After excluding two deaths, 280 cases were considered for analysis (164 administrative-only, 107 mixed, 9 clinical-only). Using ICBD, the LCA confirmed 193 (68.9%) as BS cases (68.3% female, mean age 47). This yielded a 2016 Sardinian BS prevalence of 11.7 per 100,000 inhabitants (15.7) for women and 7.6 for men), with annual incidence rates ranging from 0.24 to 0.48 per 100,000 inhabitants (2014-2016). Prevalence and incidence were lower using the ISG criteria. Database sensitivity varies from 40.6% to 70.5%, while specificity ranges from 16.3% to 99.0%.

BS in Sardinia is a rare disease. Relying on a single data source to estimate.