2544 - A Prospective Study of Alcohol Consumption and Risk of Rheumatoid Arthritis

Tuesday, November 8, 2011: 5:15 PM
W195 (McCormick Place West)
Bing LU1, Daniel H. Solomon2, Karen H. Costenbader3 and Elizabeth W. Karlson3, 1Brigham & Women's Hospital and Harvard Medical School, Boston, MA, 2Brigham and Women's Hospital, Boston, MA, 3Brigham and Women's Hospital and Harvard Medical School, Boston, MA
Presentation Number: 2544

Background/Purpose: Several case-control studies suggested that moderate alcohol consumption may reduce risk for RA. However, early symptoms of RA may change alcohol intake and assessing alcohol after RA onset may thus over-estimate the effect of alcohol exposure. No published studies have assessed the association between repeated assessment of alcohol consumption over many years and future RA risk in the setting of a large prospective cohort study.

Method: The Nurses' Health Study (NHS) was started in 1976, enrolled 121,700 registered female nurses aged 30-55 years from 11 US states, Lifestyle and environmental exposures and outcomes are collected through biennial questionnaires. Alcohol consumption was assessed on the food frequency questionnaire (FFQ) that was completed every 4 years, from 1980 through 2006. Among the 95,516 women who completed the FFQ at baseline, women who reported prevalent RA or other connective tissue disease were excluded and the diagnosis of incident RA through 2006 was confirmed using the connective tissue disease screening questionnaire and medical record review for the American College of Rheumatology criteria. We assessed the exposure variable using cumulative average alcohol consumption. We excluded the most recent record of alcohol intake and calculated the cumulative average alcohol including all other alcohol measures prior to RA onset. We categorized alcohol consumption into 5 groups (none, <5g/d, 5-9 g/d, 10-19 g/d, ≥20g/d). The estimated alcohol content of each alcohol beverage was 13.2 g per bottle or can of beer, 10.8 g per glass of wine, and 15.1 g per standard drink of liquor. Multivariable Cox proportional hazards models were used to estimate hazard ratios (HR), after adjusting for age, smoking, body mass index, menopausal status, hormone use, parity and breastfeeding.

Result:  718 incident cases of RA developed during 2,340,730 person-years of follow-up. The incidence rates of RA were 31, 30, 26, 33 and 34 /100,000 person-years across increasing levels of cumulative alcohol consumption (Table). After adjusting for other potential confounding factors, women who drank <5 grams alcohol per day had a 20% reduced risk of developing RA compared to women who did not drink any (P=0.022) (Table). Among women who drank 5-10 grams alcohol per day, the risk was reduced by 37% (p=0.001). Higher levels of alcohol consumption (≥20 grams per day) were not protective against the development of RA.

Conclusion:   In this long-term prospective cohort study of women, with alcohol intake data collected many years prior to the onset of RA, evidence of a protective association between low levels (1-9 grams per day) of alcohol intake and risk of developing RA was found. Higher levels of alcohol intake were not protective. Future studies are needed to confirm and investigate the mechanism of this effect.

Table. Alcohol consumption and risk of RA

Alcohol (gram/day)

No. Cases/ Person-Yrs

Crude Incidence*

HR (95% CI)



152/ 484365























*Per 100,000 person-years.

Adjusted for age, smoking, BMI, menopausal status, hormone use, parity and breastfeeding.

Keywords: alcohol use and rheumatoid arthritis (RA)

Disclosure: B. LU, This work is supported by NIH 1R03AA020100-01, 2 ; D. H. Solomon, Amgen and Abbott, 2 ; K. H. Costenbader, None; E. W. Karlson, NIH 1R03AA020100-01, 2 .