OBJECTIVE To evaluate the predictive value of the Activity subscale of the Braden Scale for Predicting Pressure Sore Risk in assessing mobility impairment and recovery among hospitalized older adults. DESIGN Retrospective cohort study. SETTING UF Health Shands Hospital, University of Florida, Gainesville, Florida. PATIENTS 19,769 older adults (≥65 years) hospitalized between January 2009 and April 2014. MEASUREMENTS Incident mobility impairment and recovery were assessed with the Braden Activity subscale (BAS) score that nurses use to grade patients at every shift change (~3 times/d). Posthospital mortality rate and discharge disposition were used to assess the prognostic value of the BAS. RESULTS Of the 10,717 study patients observed "walking frequently" at admission, 2218 (20.7%) developed incident mobility impairment. Of the other 9052 study patients, who were impaired at admission, 4734 (52.3%) recovered to a state of walking occasionally or frequently. Older adults who developed mobility impairment during hospitalization had an odds of death higher than that of those who remained mobile (odds ratio [OR], 1.23; 95% confidence interval [CI], 1.08- 1.39). This effect predominately occurred within the first 6 follow-up months. Older adults who recovered from mobility impairment had an odds of death lower than that of those who did not recover mobility in the hospital (OR, 0.54; 95% CI, 0.49-0.59). This effect was slightly stronger within the first 6 months after hospitalization. CONCLUSIONS Nurses' BAS assessment of mobility status during hospitalization provides substantial prognostic value in hospitalized older adults. The BAS could be an efficient and valuable source of information about mobility status for targeting posthospital care of older adults. Journal of Hospital Medicine 2017;12:396-401.