Abstract
OBJECTIVE: To calculate mortality rate associated with rheumatoid arthritis (RA), to estimate the effect of initial functional status and of change in functional status on mortality among persons with RA, and to compare the mortality experience of such persons to that of the US population. METHODS: The study used a prospective panel of 1269 persons followed for a mean of 8.4 years (median 7 yrs, interquartile range 3-12, maximum 18). Mortality status was ascertained from contacts with next of kin, study physicians, and search of the National Death Index. The Kaplan-Meier method was used to calculate the proportion dying in each time interval, with and without stratification for initial functional status [Health Assessment Questionnaire (HAQ) score] or average change in functional status. Cox proportional hazards regression was used to establish the effect of functional status, demographic characteristics, and health status on mortality risk. RESULTS: There were 270 deaths among the 1269 persons with RA. After 18 years of followup the overall death rate was 39%. The death rates in the best through worst initial quartiles of HAQ score were 29, 33, 44, and 54%. The death rate was 51% among persons with declining HAQ score versus 31 and 32% among those with no change or improvement in this measure, respectively. Demographic and health status did not reduce the effect of HAQ or average change in HAQ on mortality risk. Compared to the US population, the persons with RA had a standardized mortality rate of 1.32. CONCLUSION: The persons with RA in this study had elevated mortality rates. Poor initial functional status and declining functional status significantly increased mortality risk among these persons with RA.