The Association Between Atopic Dermatitis and Select Disease Events in Adults in the United States: A Retrospective...

Previous studies have reported positive associations between atopic dermatitis (AD) and select disease events; however, definitive conclusions are lacking. The aim of this study was to characterize demographic and clinical characteristics of patients with and without AD and estimate crude incidence rates (IRs) and adjusted hazard rate ratios (HRRs) of select disease events in these cohorts.

This retrospective observational analysis used Optum® Market Clarity Electronic Health Records, a US administrative healthcare database. Adults with and without AD between 2016 and 2021 with ≥ 12 months of database activity prior to index date (based on diagnosis or first treatment) were included. Each patient with AD was matched on age and index date with five patients without AD. The AD cohort was stratified by treatment prescribed within 30 days of AD diagnosis: no AD treatment; topical corticosteroids (TCS)/topical calcineurin inhibitors (TCI) only ("TCS/TCI only"); phototherapy with/without TCS/TCI ("phototherapy"); and systemic therapy, including oral corticosteroids with/without phototherapy and with/without TCS/TCI ("systemics"). Crude IRs and adjusted HRRs of infections, malignancies, cardiovascular events, mental health outcomes, asthma, fracture, and death across patient subgroups were estimated.

The total cohort comprised 161,646 adults. Among the 25,915 patients with AD, 8384 (32.35%), 13,459 (51.94%), 366 (1.41%), and 3706 (14.30%) were included in the "no AD treatment," "TCS/TCI only," "phototherapy," and "systemics" subgroups, respectively. Crude IRs/1000 patient-years were higher among patients with AD versus those without AD for serious infections (38.35 vs. 19.21), major adverse cardiovascular events (16.51 vs. 11.82), venous thromboembolism (11.12 vs. 6.89), malignancies (excluding nonmelanoma skin cancer; 17.51 vs. 15.25), and depression (63.50 vs. 44.18); similar trends were observed after controlling for potential confounding variables. When stratified by treatment, IRs and HRRs of these events were generally higher in patients in the "systemics" subgroup versus other treatment subgroups.

Overall, certain disease event rates were higher in the population with AD versus the population without AD, with a trend towards higher rates among patients prescribed systemic therapies compared with those prescribed TCS/TCI only. This study further informs the relationship between AD and the risk of select disease events.