Google-driven search for big data in autoimmune geoepidemiology: analysis of 394,827 patients with systemic autoimmune diseases.

Systemic autoimmune diseases (SADs) are a significant cause of morbidity and mortality worldwide, although their epidemiological profile varies significantly country by country. We explored the potential of the Google search engine to collect and merge large series (>1000 patients) of SADs reported in the Pubmed library, with the aim of obtaining a high-definition geoepidemiological picture of each disease. We collected data from 394,827 patients with SADs. Analysis showed a predominance of medical vs. administrative databases (74% vs. 26%), public health system vs. health insurance resources (88% vs. 12%) and patient-based vs. population-based designs (82% vs. 18%). The most unbalanced gender ratio was found in primary Sjögren syndrome (pSS), with nearly 10 females affected per 1 male, followed by systemic lupus erythematosus (SLE), systemic sclerosis (SSc) and antiphospholipid syndrome (APS) (ratio of nearly 5:1). Each disease predominantly affects a specific age group: children (Kawasaki disease, primary immunodeficiencies and Schonlein-Henoch disease), young people (SLE Behçet disease and sarcoidosis), middle-aged people (SSc, vasculitis and pSS) and the elderly (amyloidosis, polymyalgia rheumatica, and giant cell arteritis). We found significant differences in the geographical distribution of studies for each disease, and a higher frequency of the three SADs with available data (SLE, inflammatory myopathies and Kawasaki disease) in African-American patients. Using a "big data" approach enabled hitherto unseen connections in SADs to emerge.