ABSTRACT
BACKGROUND: The Gillette Gait Index (GGI) is a summary measure incorporating 16 clinically important kinematic and temporal parameters. The purpose of this study was to compare GGI scores from computerized gait analysis versus qualitative visual assessments of overall gait to assess the validity of the GGI as a summary score for gait analysis. METHODS: The GGI was calculated for 25 children with cerebral palsy who underwent computerized gait analysis before and 1 year after lower extremity surgery to correct gait problems. Twelve observers reviewed video recordings from the gait analysis to assess the severity of each patient's gait impairment preoperatively and postoperatively and the amount of preoperative to postoperative change. Variability of the video ratings was assessed, and GGI scores were compared with the mean video ratings. RESULTS: The individual ratings showed some variability, with moderate intrarater agreement (weighted kappa = 0.49-0.56) and slight to fair interrater agreement (kappa = 0.11-0.25). However, the mean scores from all raters were much more consistent, as demonstrated by a highly significant relationship in preoperative to postoperative change viewing the videos separately versus together (r2 = 0.62; P = 0.0001). GGI scores were correlated with these mean scores preoperatively (r2 = 0.34; P = 0.003), postoperatively (r2 = 0.30; P= 0.005), and in preoperative to postoperative change (r2 = 0.30, P = 0.006 for absolute change; r2 = 0.22, P = 0.02 for percentage change). CONCLUSIONS: These results support the validity of GGI as a gait analysis summary score and suggest that GGI may be a useful outcome measure in patients undergoing gait analysis. CLINICAL RELEVANCE: Clinicians and researchers should consider using the GGI as a quantitative outcome measure for assessing overall gait.
