Background/Purpose

RA has been associated with increased mortality compared to general population estimates. Previous studies were limited due to the inability to directly compare RA patients to controls, short follow-up, and lack of detailed data on clinical, lifestyle, and serologic factors. We evaluated mortality among women followed prospectively prior to RA diagnosis, directly comparing to women without RA.

Methods

We conducted a study of RA and mortality among 121,700 women followed from 1976 to 2010 in the Nurses’ Health Study (NHS). Incident RA was validated by medical record review according to the 1987 ACR RA criteria and classified by serostatus. Women who reported RA or other connective tissue diseases before the start of NHS were excluded. Women were followed from cohort entry to death or end of follow-up and were censored for loss to follow-up. Deaths were validated by the National Death Index; death certificate and medical record review determined cause of death. Cox regression models estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for all-cause, cardiovascular disease (CVD), cancer, and respiratory mortality for women with RA compared to women without RA. We obtained HRs for mortality by RA duration and serologic RA phenotype. Models were adjusted for age, demographics and other mortality factors, including physical activity, smoking, obesity, comorbidities, and family history of cancer, CVD, and diabetes.

Results

We validated 960 incident RA cases and identified 25,699 deaths in 34 years of NHS follow-up. Of the 261 deaths among women with RA, 75 (29%) were from cancer, 58 (22%) were from CVD, and 43 (16%) were from respiratory causes. Compared to women without RA, women with RA had increased all-cause mortality that remained significant after adjusting for age and other mortality factors (HR 2.07, 95% CI 1.83-2.35, Table). Mortality was significantly increased for seropositive (HR 2.33, 95% CI 2.00-2.71) and seronegative RA (HR 1.60, 95% CI 1.30-1.98) compared to non-RA women. Each five years of RA duration conferred a 32% (95% CI 27-36%) increased mortality compared to non-RA. Women with RA had significantly increased risk for mortality from CVD (HR 1.87, 95% CI 1.44-2.43), cancer (HR 1.35, 95% CI 1.07-1.69) and respiratory (HR 4.50, 95% CI 3.28-6.17) causes compared to women without RA. Respiratory mortality for women with seropositive RA was six-fold higher than non-RA women (HR 6.23, 95% CI 4.38-8.85).

Conclusion

In 34 years of prospective follow-up, women diagnosed with RA had a two-fold increased risk of death from any cause compared to women without RA. Respiratory mortality was six-fold higher in seropositive RA and women with RA were significantly more likely to die from CVD and cancer than women without RA. Respiratory mortality appears to be an important but understudied cause of death in RA. These findings provide evidence of high RA mortality burden that is unexplained by traditional mortality predictors.

Table. Hazard ratios for all-cause and cause-specific mortality in RA serologic phenotypes among women in the Nurses’ Health Study, 1976-2010 (n = 119,264).
	All-cause mortality		CVD-specific mortality	Cancer-specific mortality	Respiratory-specific mortality†
	Age-adjusted HR (95% CI)	Multivariable HR (95% CI)*	Multivariable HR (95% CI)*	Multivariable HR (95% CI)*	Multivariable HR (95% CI)*
All RA
No RA	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)
RA	1.42 (1.26-1.61)	2.07 (1.83-2.35)	1.87 (1.44-2.43)	1.35 (1.07-1.69)	4.50 (3.28-6.17)
No RA	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)
RA duration (per 5 years)	1.16 (1.13-1.20)	1.32 (1.27-1.36)	1.24 (1.15-1.33)	1.22 (1.14-1.30)	1.55 (1.42-1.69)
Seropositive RA
No RA	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)
RA	1.54 (1.33-1.79)	2.33 (2.00-2.71)	1.80 (1.27-2.55)	1.25 (0.92-1.70)	6.23 (4.38-8.85)
No RA	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)
RA duration (per 5 years)	1.20 (1.15-1.25)	1.38 (1.33-1.44)	1.26 (1.15-1.39)	1.23 (1.13-1.34)	1.69 (1.53-1.85)
Seronegative RA
No RA	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)
RA	1.15 (0.93-1.41)	1.60 (1.30-1.98)	1.84 (1.23-2.73)	1.40 (0.99-1.98)	1.97 (0.97-3.98)
No RA	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)	1.00 (ref)
RA duration (per 5 years)	1.08 (1.02-1.15)	1.20 (1.14-1.28)	1.19 (1.06-2.73)	1.18 (1.06-1.31)	1.23 (1.00-1.51)
*Adjusted for age, questionnaire period, US region, race/ethnicity, education, husband’s education, body mass index (<18.5, 18.5-24.9, 25-29.9, ≥30), cigarette smoking pack-years (never, 0-10, 10.1-20, >20), post-menopausal hormone use, physical activity, healthy eating index, cancer, cardiovascular disease, diabetes mellitus, family history of diabetes, family history of cancer, family history of myocardial infarction <60 years of age, and aspirin use through follow-up. Modifiable factors were adjusted up to RA diagnosis (cigarette smoking pack-years, physical activity, and body mass index). †Among women with RA, there were a total of 43 respiratory deaths that were due to pneumonia (11), emphysema (8), chronic interstitial lung disease (5), asthma (1), and other respiratory diseases (18).

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/incident-rheumatoid-arthritis-and-risk-of-mortality-among-women-followed-prospectively-from-1976-to-2010-in-the-nurses-health-study/