Predictors of self-reported knee osteoarthritis in community-dwelling older women in Japan: A cross-sectional and longitudinal cohort study

Purpose: This study aimed to determine the predictors of knee osteoarthritis in community-dwelling elderly Japanese women. Methods: In this prospective cohort study, The Tokyo Metropolitan Institute of Gerontology collected baseline data in 2008 and follow-up data in 2012 for participants from the Itabashi Ward of Tokyo, Japan. Participants were asked at each time point if they had been diagnosed with knee osteoarthritis. The baseline evaluation was conducted with 1289 community-dwelling women aged 75-85 years, of which 992 reported no history of knee osteoarthritis. The follow-up survey targeted these 992 participants; we obtained history of knee osteoarthritis from 867 of these participants. The baseline evaluation also included collection of anthropometric, fitness, hematologic, and lifestyle data. Results: We performed logistic regression analysis of the cross-sectional data at baseline. Participants who reported fewer light exercise sessions (<= 2-4 days/week) had lower odds ratios for history of self-reported knee osteoarthritis than those who reported more frequent exercise (>= 5-6 days/week). Logistic regression analysis of the longitudinal data revealed that slow walking speed (< 65.22 m/min), low serum albumin levels (< 4.10 mu g/dL), and low frequency of soy product consumption (<= 1 time per 2 days) at baseline resulted in higher odds ratios for incidence of self-reported knee osteoarthritis during the 4-year follow-up period. Conclusions: The results suggest that slow walking speed, low serum albumin, and insufficient consumption of soy products are predictors for knee osteoarthritis in elderly Japanese women. These results could help in the design of knee osteoarthritis prevention programs for elderly women.