Objective: To describe the underlying dimensions for a range of functional measures across 3 neurologic diseases at different time points.
Design: Multiple cohort study.
Setting: Combination of public hospital wards and community.
Participants: Patients (N=308) from 3 cohorts: paralytic poliomyelitis (n=38), mean of 25 years previously, assessed once; acute stroke admitted to hospital and followed up for 12 months postdischarge (n=181); and traumatic brain injury (TBI), admitted to hospital and followed up for 12 months postdischarge (n=89).
Interventions: Not applicable.
Main outcome measures: The Barthel Index, FIM instrument, Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) physical component score (PCS) and mental component score (MCS), Community Integration Questionnaire (CIQ), and the London Handicap Score (LHS). Measures were compared at 2 time points: pre-event status for stroke, TBI and the polio cohort, and 12-month postdischarge status for stroke and TBI.
Results: The different measures generally correlated highly within disease states at each time point. Principal components analysis revealed 2 underlying dimensions, a physical dimension onto which loaded the Barthel Index, FIM, PCS, and LHS and a cognitive/emotional dimension onto which loaded the MCS and, for subjects with polio, the CIQ. These 2 dimensions accounted for 69% of the variance in measures at the pre-event time point and 85% of the variance at the 12-month time point.
Conclusions: These data suggest 2 basic underlying dimensions across a wide range of measures in 3 different neurologic conditions even at different time points. Most of the variation in the measures can be captured using the 2 component summary scores (PCS, MCS) of the SF-36.