RA Linked to Incident Anxiety and Mental Disorders in ELSA-Brasil

11/17/2026
Key Takeaways
- Among Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) participants, the rheumatoid arthritis (RA) case group had more incident generalized anxiety disorder (GAD) than a randomly selected reference sample of Brazilian civil servants (25.2% versus 13.3%), with a significant adjusted association.
- Incident common mental disorders (CMD) were more frequent in the RA group than the reference sample (36.4% versus 20.9%), with a significant adjusted association.
- Incident depressive episode was more frequent in the RA group, but its adjusted association was not statistically significant.
- At baseline, CMD remained associated with RA after adjustment, unlike depressive episode and GAD.
Investigators conducted an analysis of rheumatoid arthritis and mental health within the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil), a multicenter cohort of civil servants recruited in six Brazilian cities. The 1,795 eligible participants included 252 RA case-group members and 1,543 members of an independently randomly selected reference sample. Among reference participants, 27 also met the RA definition and remained in that sample. RA classification required at least two of self-reported physician diagnosis, disease-modifying antirheumatic drug use during the preceding 14 days, or positivity for rheumatoid factor or anti-cyclic citrullinated peptide antibodies. The Clinical Interview Schedule-Revised (CIS-R) identified depressive episode and GAD using International Classification of Diseases, Tenth Revision (ICD-10) algorithms; CMD instead meant a CIS-R score ≥12, not a specific ICD-10 diagnosis. Depressive episode was assessed at baseline and years 4, 8, and 12; GAD and CMD were assessed at baseline and years 8 and 12. Baseline and outcome-specific cumulative incident analyses were separate. The latter excluded participants with the respective baseline condition or missing follow-up, and incident meant first detected at a scheduled visit, not a known onset date. Complete-case logistic regression adjusted for age, sex, race/ethnicity, education, and baseline antidepressant use without sampling weights.
Incident GAD was associated with RA case-group membership after full adjustment (odds ratio [OR] 2.04, 95% confidence interval [CI] 1.30–3.18). This comparison involved outcome-specific at-risk groups without GAD at baseline, not all eligible participants. A parallel analysis considered newly detected, broader psychiatric morbidity as CMD.
Incident CMD was also associated with RA case-group membership after full adjustment (OR 2.06, 95% CI 1.33–3.18). Incident depressive episode was more frequent in the RA group, but its adjusted association was not statistically significant. At baseline, CMD remained significantly associated with RA after full adjustment (OR 1.41), whereas depressive episode and GAD did not. Baseline and incident associations describe different observations and do not establish causal effects.
RA status came from an epidemiologic algorithm rather than systematic rheumatologist assessment, leaving room for misclassification. RA duration, activity, pain, disability, and longitudinal treatment were unmeasured; residual confounding and missing follow-up could also affect interpretation. Healthy-worker selection and the Brazilian civil-servant setting limit generalizability beyond this occupational cohort. Because psychiatric outcomes were detected at scheduled visits, their onset dates were unknown, and assessment schedules differed by outcome. The investigators interpreted the pattern as extending beyond diagnosis-specific depression to anxiety and broader psychiatric morbidity, without establishing why those associations arose.
The authors concluded that longitudinal associations with RA were more consistent for GAD and broad CMD than for depressive episode. At cohort entry, the association with symptom-defined CMD likewise differed from those with the diagnosis-specific conditions.
Clinician Questions
Which age group was represented in the Brazilian rheumatoid arthritis and mental health cohort?
ELSA-Brasil recruited Brazilian civil servants aged 35–74 years at baseline in six cities. These rheumatoid arthritis–mental health findings come from an occupational cohort, and generalizability to other populations was not established.
Can baseline rheumatoid arthritis and common mental disorders measurements establish which condition came first?
No. The ELSA-Brasil analysis measured rheumatoid arthritis case-group membership and common mental disorders at cohort entry, but the timing of rheumatoid arthritis onset relative to psychiatric symptoms was unavailable. Their baseline association cannot establish temporal direction.
This content is intended for healthcare professionals for educational purposes. The article and its image were developed with AI assistance and reviewed by members of the ReachMD team prior to publication. Please refer to the original source linked for the complete and most up-to-date information.
