Related Experiment Videos
Use of principal components analysis to develop a composite score as a primary outcome variable in a clinical trial.
W G Henderson1, S G Fisher, N Cohen
1Cooperative Studies Program Coordinating Center, Hines VA Hospital, Hines, Illinois.
Abstract:
This article describes the use of principal components analysis to derive a composite score from a battery of 24 audiologic tests. The composite score is being used as the primary outcome variable in a clinical trial comparing the efficacy of three cochlear implant devices for people with bilateral, profound hearing loss. The first principal component from the within-class pooled variance-covariance matrix over four time periods was chosen to establish the coefficients for the composite score. This component accounted for 61% of the total variance of the 24 audiologic tests. The first principal component had its largest coefficients associated with the most difficult audiologic tests. The mean composite score of all patients improved over time; some patients showed dramatic improvement. The changes in the composite score over time were also closely related to subjective impressions of implant performance by the patient, audiologist, and otolaryngologist.