Treatment-resistant depression and risk of autoimmune diseases: evidence from a nested case-control study

Abstract

<p>Background: Recent literature increasingly suggests the predictive role of depression in autoimmune diseases due to immune activation, but the role of treatment-resistant depression (TRD), an indicator of non-responsive depression that is more prone to dysregulated inflammation, is lesser-known in the development of autoimmunity.</p><p>Objectives: In this population-based nested case-control study, we examined the association between TRD and subsequent risk of autoimmune diseases and explored the potential sex-specific difference.</p><p>Methods: Using the territory-wide electronic medical records in the public healthcare setting of Hong Kong, we identified 24,566 incident depression patients between 2014 and 2016 without autoimmune history and followed up from diagnosis of depression to death or December 2020 to identify TRD development (exposure) and autoimmune incidence (outcome). TRD was defined as having received at least two antidepressant regimens for adequate duration and the third regimen to confirm previous treatment failures. We matched cases and controls 1:10 using incidence density sampling with replacement based on age, sex, and year of first depression diagnosis. We then estimated the odds ratios (ORs) using conditional logistic regression, adjusting for baseline history of physical and psychiatric conditions. The first diagnosis dates of autoimmune diseases for cases (index date) were assigned to be the same for the matched controls, who were alive and free of autoimmune diseases on the index date.</p><p>Results: Grouping 22 types of autoimmune diseases with 71,232 person-years of follow-up, the cumulative incidence in the TRD group was generally higher compared with the non-TRD group (21.5 vs 14.4 per 10,000 person-years). Conditional logistical model suggested a significant association between TRD status and the development of autoimmune diseases (OR:1.67, 95%CI: 1.10–2.53, p=0.017). Subgroup analysis showed that the effect sizes were generally higher among men compared with women, and the association was significant in organ-specific diseases instead of systemic diseases.</p>