This retrospective cohort study evaluated the effects of different frequencies of physical therapy intervention on the total knee arthroplasty (TKA) and total hip arthroplasty (THA) risk of osteoarthritis (OA) patients.We sampled 438,833 insurants from Taiwan National Health Insurance Research Database for patients diagnosed as having OA during 2000 to 2013. OA who received physical therapy within in the first year of OA diagnosis were divided based on the number of sessions they received in that first year: >24, 13-23, and <12 sessions.The results revealed that the TKA and THA incidence rates among patients aged 60 to 80 years were respectively 3.5% and 0.9% in the >24 cohort and 4.9% and 1.4% (all P < .001) in the comparison cohort. Moreover, the HRs of TKA and THA in the >24 cohort were 0.77 (0.67-0.87, P < .001) and 0.71 (0.53-0.96, P = .024), respectively. By contrast, no significant differences were noted between the 13-23 and <12 cohorts and their respective comparison cohorts.In conclusion, our study results indicated that elderly patients aged 60 to 80 years who underwent >24 physical therapy sessions within 1 year of receiving an OA diagnosis exhibited reduced of TKA and THA risks.