Ethnopharmacological relevance: To determine whether adjuvant Chinese herbal medicine (CHM) treatment is associated with the risk of joint replacement in osteoarthritis (OA) patients. Materials and methods: This retrospective study used a population-based national health insurance (NHI) database from 2000 to 2012 in Taiwan. A total of 125,023 newly diagnosed OA patients were selected from one million beneficiaries of longitudinal health insurance database. Based on applying ten selected frequently used CHM formulas for OA, patients were divided into CHM user and non-CHM user. One-CHM to four-non-CHM user were propensity score matched with age, gender, monthly income, urbanization, comorbidities, Nonsteroidal antiinflammatory drugs (NSAIDs) and index year were adjusted to reduce selection bias and confounding. Cox regression model was used for comparing the hazard ratios (HR) for the risk of joint replacement and KaplanMeier curve for the proportion of joint replacement. Results: OA patients who were female, younger (20-60 years), higher income and lived in urbanization location were found to preferred using CHM. Younger CHM users had a lower adjusted HR (0.63) of the risk of joint replacement (95% confidence interval (CI) = 0.42-0.94). Compared to non-CHM user, HR among CHM users (>= 225 days annually) is 0.48 (95% CI = 0.31-0.76). The proportion of joint replacement in younger non-CHM user began to rise notably with time (log-rank test, p = 0.026). However, this benefit by CHM did not apply to older (over 60 years) OA patients. Conclusion: This study suggested that adjuvant CHM might be associated with a lower rate of joint replacement in OA patients. CHM therapy might be considered in OA patients to reduce the need of joint replacement.
