Effect Modification of Cigarette Smoking on the Relationship between Post‐traumatic Stress Disorder, Other Mental Health Conditions and Rheumatoid Arthritis Risk: A Large Case‐Control Study in U.S. Veterans

Objective We examined associations of post‐traumatic stress disorder (PTSD) and other mental health disorders (OMH) with rheumatoid arthritis (RA), accounting for effects of smoking. Methods We conducted a matched case‐control study, identifying incident RA cases and controls using national Veteran Health Administration data (2006‐2019). PTSD/OMH exposures prior to RA diagnosis were identified using administrative/diagnostic codes. Conditional logistic regression examined associations of RA (vs. no‐RA) with PTSD (±OMH) and OMH. Relative Excess Risk of Interaction was calculated between OMH/PTSD and smoking using standard multivariable logistic regression with standard errors clustered within match groups stratified by ever/never smoking. Analyses were repeated excluding individuals without a visit in the previous year. Results A total of 357,361 Veterans (mean age 53 years, 87% male, 70% ever smoked) were studied. Both PTSD (19% vs. 13%) and OMH (31% vs. 26%) were more frequent in RA than controls. Multivariable analyses demonstrated associations of PTSD and OMH with RA with evidence of PTSD/OMH‐smoking interactions; associations were stronger in never (aORs 1.54/1.40, respectively) vs. ever smokers (aORs 1.20/1.08). Excluding individuals without a healthcare visit, PTSD and OMH remained associated with seronegative RA in ever (aOR 1.17; 95% CI 1.09‐1.26 and aOR 1.14; 95% CI 1.06‐1.21) and never smokers (aOR 1.37; 95% CI 1.16‐1.62 and aOR 1.27; 95% CI 1.10‐1.47); similar associations of PTSD/OMH with seropositive RA were not observed. Conclusion This study demonstrates that the relationships of PTSD (and OMH) with RA differ by smoking status. The strongest associations between PTSD/OMH and incident RA were among never smokers and in seronegative disease.