Determinants of diagnosis and disease course in primary Sjögren's syndrome: Results from datamining of electronic health records.

Background: Determinants of diagnosis in primary Sjögren's syndrome (pSS) in tertiary care settings is not well understood.

Methods: Patients were screened by tracing reports of anti-SSA (anti-Ro) antibody assays between January 2008 and October 2015. Electronic health records (EHR) were reviewed. Patients fulfilling the 2016 American College of Rheumatology/European League Against Rheumatism (EULAR) classification criteria were included. Variables including the specialties of first consultation, initial clinical presentations, investigations ordered at first consultation, number of hospital visits prior to reaching the final diagnosis of pSS and the baseline EULAR SS Disease Activity Index (ESSDAI) were noted.

Results: A total of 275 patients with pSS consulted 24 different specialties at first visit. Rheumatology accounted for 128 (46.55%) patients. At first consultation, initial suspicion for pSS was 48.4% for all specialties together and 64.84% for the rheumatologist. Median number of visits prior to arriving at the final diagnosis was 1 (1-6), when the initial impression was pSS and 3 (1-14), if the initial clinical impression was a non-SS differential (P < 0.001). A first impression of pSS, enquiry about sicca symptoms and ordering anti-SSA (anti-Ro) antibody test at first consultation were strong predictors of early diagnosis with odds (95% CI) of 5.01 (1.72-14.55) P < 0.001, 4.79 (1.16-19.84) P = 0.03 and 9.60 (3.0-30.67) P < 0.0001, respectively. None of the clinical variables proved to be useful predictors of early diagnosis.

Conclusions: Diagnosis of pSS is challenging even in tertiary care centers as patients present with myriad features to several specialties. Initial suspicion was limited to 48.4% for all specialties together and 64.84% for the rheumatologist. High suspicion of pSS along with ordering anti-SSA (anti-Ro) antibody could hasten diagnosis.