Agreement between patient and proxy responses during recovery after hip fracture: Evidence for the FIM instrument

Objective: To evaluate the agreement between patient and proxy responses of the FIM instrument at 4 different periods of time during the first 6 months after hip fracture. Design: Prospective cohort study. Setting: A large urban health region with 2 tertiary hospitals that treat hip fractures. Participants: Patients (n= 137) who were 65 years or older, admitted to the health region with a primary diagnosis of hip fracture, who had Mini-Mental State Examination scores greater than 17. Family caregivers (n= 137) participated as proxy respondents. Interventions: Not applicable. Main Outcome Measure: The FIM instrument. Agreement was evaluated at each of the 4 assessments during the 6-month follow-up after hip fracture using intraclass correlation coefficient. Results: FIM scores improved over the 6 months with the greatest improvement occurring within the first month of recovery. Agreement was higher for more observable activities than less observable ones. The magnitude of agreement improved over the 6 months with the proportion of clinically important systematic differences decreasing over time. Agreement for change scores was lower than the agreement at each of the 4 assessments. Conclusions: Patient-proxy agreement levels are acceptable; the agreement varies with the subscale and the recovery phase. Substitution of proxy for patient responses across time may be used guardedly when patient responses are missing.