Background/Purpose: Smoking is a major preventable cause of death and cessation is recommended for the general population. However, it is unclear whether being diagnosis with RA affects smoking cessation and whether continuing to smoke contributes to the excess mortality of RA. We aimed to describe smoking behavior changes after RA diagnosis and to evaluate the effect of smoking on mortality after RA diagnosis.

Methods: We investigated smoking and mortality among women diagnosed with RA during follow-up in the Nurses’ Health Study (NHS). The NHS is composed of 121,700 women aged 30-55 years at baseline in 1976 and followed with biennial questionnaires. RA diagnosis was validated by medical record review according to the 1987 ACR criteria by two rheumatologists who identified the date of RA diagnosis. Smoking status and intensity (never, current, or past; cigarettes/day) were reported biennially during follow-up and cumulative smoking pack-years were calculated for each follow-up cycle. We categorized smoking at each follow-up cycle as occurring before or after the date of RA diagnosis. Deaths were identified by the National Death Index up to 2012. Cox regression models estimated HRs for all-cause mortality according to smoking pack-years after RA diagnosis, adjusting for sociodemographic, behavioral, and clinical factors.

Results: We analyzed 923 women diagnosed with RA during follow-up in the NHS with detailed prospective data available on smoking before and after RA diagnosis. There were 288 deaths in 16,111 person-years of follow-up. At RA diagnosis, mean age was 59.5 (SD 9.9) years, 52% were obese, and 60% consumed alcohol. In the cycle prior to RA diagnosis, 36% were never smokers, 43% were past smokers, and 21% were current smokers. Among current smokers just prior to RA diagnosis, 16% immediately quit smoking after RA diagnosis and maintained cessation, 21% continued to smoke throughout follow-up, 49% continued smoking for at least two cycles after RA but quit later, and 1% quit smoking after RA diagnosis then started smoking again later. Compared to never smokers after RA diagnosis, women who smoked >5 pack-years after RA diagnosis had significantly increased mortality (HR 2.69, 95% CI 1.33-5.46, Table) after adjustment for age, smoking pack-years prior to RA diagnosis, and other confounders. When analyzing only ever smokers at RA diagnosis, smoking >5 pack-years remained significantly associated with mortality compared to those who never smoked after RA diagnosis (HR 4.35, 95% CI 1.81-10.44).

Conclusion: Despite the known harmful effects of smoking in chronic diseases, only 16% of smokers quit after diagnosis with RA and maintained smoking cessation during follow-up. Smoking >5 pack-years after RA diagnosis was associated with increased mortality, independent of smoking before RA diagnosis. Interventions promoting cessation of smoking for patients newly diagnosed with RA may diminish the excess mortality of RA.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/smoking-behavior-changes-after-rheumatoid-arthritis-diagnosis-and-risk-of-mortality-during-36-years-of-prospective-follow-up/