Allergic Diseases and Long Term Risk of Autoimmune Disorders: longitudinal cohort study and cluster analysis

Background The association between allergic diseases (ADs) and autoimmune disorders (AIDs) is not well established. Objective To determine incidence rates of AIDs in allergic rhinitis/conjunctivitis (ARC), atopic eczema (AE) and asthma, and investigate for co-occurring patterns. Methods Design: Retrospective cohort study (1990–2018) employing “The Health Improvement Network” (UK primary care database). Exposure group: ARC, AE and asthma - all ages. Controls: For each exposed patient, up to 2 randomly selected age- and gender-matched controls with no documented AD. Adjusted incidence rate ratios (aIRRs) were calculated using Poisson regression. A cross-sectional study was also conducted employing Association Rule Mining (ARM) to investigate disease clusters. Results 782 320, 1 393 570 and 1 049 868 patients with ARC, AE and asthma, respectively, were included. aIRRs of systemic lupus erythematosus (SLE), Sjogren9s syndrome (SS), vitiligo, rheumatoid arthritis (RA), psoriasis, pernicious anaemia, inflammatory bowel disease (IBD), coeliac disease (CD) and autoimmune thyroid disease were uniformly higher in the 3 ADs compared to controls. Specifically, aIRRs of SLE (1.45) and SS (1.88) were higher in ARC; SLE (1.44), SS (1.61) and myasthenia (1.56) higher in asthma; SLE (1.86), SS (1.48), vitiligo (1.54) and psoriasis (2.41) higher in AE. There was no significant effect of the 3 ADs on multiple sclerosis and ARC and AE on myasthenia. ARM: ADs clustered with multiple AIDs. Three age- and gender-related clusters were identified, with relatively complex pattern in females ≥55 years. Conclusion The long-term risk of AIDs are significantly higher in patients with ADs. ADs and AIDs show age- and gender-related clustering patterns.