Pain is a frequent symptom in patients who require inpatient rehabilitation. For example, 40-50% of patients hospitalized for stroke rehabilitation report pain. Further, 82% of those hospitalized for postsurgical orthopedic rehabilitation complain of at least moderate pain, and among those hospitalized for rehabilitation after acute traumatic spinal cord injury, 97% report pain at least once during the rehabilitation stay. Pain has been described as an immediate and long-term predictor of less functional recovery at discharge from rehabilitation units. The objective of this study was to investigate the association between pain and functional independence in older adults after post-acute rehabilitation.


A prospective, observational cohort study of patients aged ≥ 65 years was performed in one community and one Veterans Administration inpatient rehabilitation center. Pain was assessed using the Geriatric Pain Measure (GPM, score 0-100). Functional independence was measured with the motor component of the Functional Independence Measure (mFIM, score 13-91). Both scores were obtained on admission, discharge, and 3-, 6- and 9-month follow-up. In bivariate analyses, discharge GPM and persistent pain (i.e., pain lasting more than 3 months) were evaluated as predictors of mFIM at 9-month. Applying a multilevel modeling (MLM) approach, individual deviations in GPM scores were used to predict variations in mFIM.


245 participants were enrolled (mean age 80.6 years, 72% male). At admission, 210 participants (87.9%) reported pain (16.3% mild [GPM<30]; 49.3% moderate [GPM: 30-69]; and 22.1% severe [GPM>70]). 21.3% of participants reported persistent pain after discharge. The bivariate analyses did not find statistically significant associations between discharge GPM or persistent pain and mFIM at 9-month. However, in the MLM analysis, deviations in GPM were significant predictors of deviations in mFIM, suggesting that when an individual experienced above-average levels of pain (i.e., GPM > her/his personal mean GPM) they also experienced worse functional independence (i.e., current mFIM < her/his personal mean mFIM, figure 1).


Twenty-one percent of older adults undergoing post-acute rehabilitation reported persistent pain after discharge from rehabilitation. The bivariate analysis did not find association between pain and functional independence; however, the MLM analysis showed that when a participant was experiencing more pain than her/his average, s/he also had less functional independence.