To explore the association between therapy minutes per length of stay (LOS) day (TMLD), functional outcomes, and rate of functional recovery among older adults after elective hip or knee replacement surgery across postacute (PAC) settings. Secondary analysis of data collected for an observational cohort study from 2005 to 2010. Four inpatient rehabilitation facilities (IRF) and 7 skilled nursing facilities (SNF). Adults aged 65 years or older (N=162) with Medicare fee-for-service insurance and a primary diagnosis of elective hip or knee replacement. Not applicable. FIM mobility and self-care measures at discharge. The TMLD was divided into high, medium, and low categories. Participants were grouped into low, medium, and high gain rate groups based on their average change in mobility and self-care FIM measures per LOS day. Gain rate and TMLD groups were crossmapped to create 9 gain-TMLD groups separately for mobility and self-care. There were no significant differences in admission mobility or self-care measures by gain rate and TMLD trajectory or by facility type (IRF or SNF). TMLD was not significantly associated with discharge mobility measures. Participants in high gain trajectories attained independence with mobility and self-care tasks at discharge regardless of TMLD. Those in low gain trajectories needed supervision or assistance on all mobility tasks. Older age and greater pain at discharge were significantly associated with lower odds of being in the medium or high gain rate groups. For clinicians and facility managers who must care for patients with constrained resources, the shift to value-based reimbursement for rehabilitation services in PAC settings has reinvigorated the question of whether the duration of therapy provided influences patient outcomes. Three hours of daily therapy after joint replacement surgery may exceed what is necessary for recovery. Postsurgical pain management remains a significant challenge in older adults. Published by Elsevier Inc.