Background/Purpose: Depression is associated with elevated systemic inflammation and risk of several chronic diseases including lupus, psoriasis, and inflammatory bowel disease. However, the association between depression and subsequent risk for developing RA is less clear. Therefore, we hypothesized that depression may increase RA risk. Since depression may develop in the period prior to clinical RA diagnosis, we decreased the potential for reverse causation by including a time separation between the assessment of depression and RA risk window.

Methods: We performed a prospective cohort study investigating depression and incident RA. We pooled data from the Nurses’ Health Study (NHS, 1992-2014) and NHSII (1993-2015). As in previous studies, depression was defined using a composite definition: report of clinician diagnosis, regular antidepressant use, or Mental Health Inventory-5 score < 60. Women reported RA diagnosis and 2 rheumatologists performed medical record review to confirm diagnosis using objective criteria and identify the date of diagnosis and serologic phenotype (seropositive: RF+ or CCP+). All incident RA cases met 1987 ACR or 2010 ACR/EULAR criteria. Measures of depression and covariates including smoking, body mass index, dietary quality, physical activity, and menopausal factors were obtained using biennial questionnaires. Cox regression estimated HRs and 95%CIs for RA (overall and by serologic phenotype) according to depression status, adjusted for potential confounders. All analyses included a lag that separated depression status and the window for RA risk by at least 4 years to lower the possibility that depressive symptoms preceding RA diagnosis explained the results.

Results: We analyzed n=194,700 women; mean age at baseline of 46.8 (SD 11.5) years and 24% of women had depression. A total of 834 incident RA cases (64% seropositive) occurred during 2,994,244 person-years of follow-up. In the age-adjusted analysis, women with depression had HR for all RA of 1.41 (95%CI 1.22-1.64, Table). After adjusting for covariates, the HR for all RA was attenuated, but still showed a modest but significant association (multivariable HR 1.28, 95%CI 1.28, 95%CI 1.10-1.49). After adjustment for covariates, depression was not statistically associated with seropositive RA (multivariable HR 1.10, 95%CI 0.91-1.32). However, depression was strongly associated with risk for incident seronegative RA (multivariable HR 1.70, 95%CI 1.32-2.19), independent of potential confounders including smoking, dietary intake, body mass index, physical activity, and menopause.

Conclusion: In this large prospective study, depression was associated with increased risk for incident RA, using stringent methods for RA case identification. The association of depressive symptoms with seronegative RA was not explained by measured lifestyle factors. We accounted for potential reverse causation of early RA symptoms leading to depression. Future studies are needed to replicate these findings and to elucidate potential mechanisms linking depression and seronegative RA, such as systemic inflammation or aberrant pain sensitivity.

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/depression-and-subsequent-risk-for-incident-seronegative-rheumatoid-arthritis-among-women/