The objective of this study was to describe the relationships between functional status at discharge and intensity of therapies received during inpatient medical rehabilitation. The sample was comprised of 140 patients with traumatic brain injury and 106 patients with spinal cord injury at eight hospitals that subscribe to the Uniform Data System for Medical Rehabilitation. Data included linear measures of motor and cognitive ability derived from the Functional Independence Measure at admission to and discharge from rehabilitation. Multiple regression was used to predict intensity of therapies, discharge motor and cognitive function, the extent to which potential functional gains were achieved, and the efficiency of gains. Intensities of occupational, physical, and speech therapies were not significant predictors of outcomes for either group, controlling linearly for admission function, psychology intensity, length of stay, onset to admission interval, age, and interrupted stays. Only intensity of psychology services seemed to have any relation to functional gain (in cognition for patients with traumatic brain injury). The apparent lack of benefit related to intensity of therapies may be due to factors such as spontaneous recovery, goals not measured by the Functional Independence Measure, limited modulation of therapy intensity according to likely patient responsiveness, or therapies focused on impairment or other goals rather than disability. We suggest that efficiently staged rehabilitation should vary the intensity and nature of services according to patients' functional status, impairments, comorbid conditions, and other clinical factors.