BACKGROUND: This longitudinal study examined the association between use of estrogen replacement therapy and incidence of self-reported, physician-diagnosed arthritis. METHODS: Data of 2,416 postmenopausal women who participated in the National Health and Nutrition Examination Survey Epidemiological Follow-Up Study were used in this study. Women, free of self-reported arthritis at entry into study and for 3 years thereafter, were questioned about use of estrogen and physician-diagnosed arthritis at each of the follow-up waves of study. Proportional hazard regression models were used for the analysis. RESULTS: Use of ERT was found to be associated with higher risk of incident arthritis, after adjusting for potential confounders (RR = 1.61, CI 1.37-1.89). Whenever use of ERT was replaced by duration of use in the regression model, ERT users for a year or less significantly increased their risk of incident arthritis (RR = 1.37, CI 1.07-1.74). The risk increased by 30 and 96% with hormone use for 1 to 4 and 4 to 10 years, respectively, and by 104% with hormone use for 10 or more years. CONCLUSION: Results suggest that users of ERT were at higher risk of developing arthritis and the longer the use of the hormone, the higher the risk. Copyright 1999 American Health Foundation and Academic Press.
BACKGROUND: This longitudinal study examined the association between use of estrogen replacement therapy and incidence of self-reported, physician-diagnosed arthritis. METHODS: Data of 2,416 postmenopausal women who participated in the National Health and Nutrition Examination Survey Epidemiological Follow-Up Study were used in this study. Women, free of self-reported arthritis at entry into study and for 3 years thereafter, were questioned about use of estrogen and physician-diagnosed arthritis at each of the follow-up waves of study. Proportional hazard regression models were used for the analysis. RESULTS: Use of ERT was found to be associated with higher risk of incident arthritis, after adjusting for potential confounders (RR = 1.61, CI 1.37-1.89). Whenever use of ERT was replaced by duration of use in the regression model, ERT users for a year or less significantly increased their risk of incident arthritis (RR = 1.37, CI 1.07-1.74). The risk increased by 30 and 96% with hormone use for 1 to 4 and 4 to 10 years, respectively, and by 104% with hormone use for 10 or more years. CONCLUSION: Results suggest that users of ERT were at higher risk of developing arthritis and the longer the use of the hormone, the higher the risk. Copyright 1999 American Health Foundation and Academic Press.