Reliability and Minimum Detectable Change of the Gait Deviation Index (GDI) in post-stroke patients

The Gait Deviation Index (GDI) is a summary measure that provides a global picture of gait kinematic data. Since the ability to walk is critical for post-stroke patients, the aim of this study was to determine the reliability and Minimum Detectable Change (MDC) of the GDI in this patient population. Twenty post-stroke patients (11 males, 9 females; mean age, 55.2 ± 9.9 years) participated in this study. Patients presented with either right- (n = 14) or left-sided (n = 6) hemiparesis. Kinematic gait data were collected in two sessions (test and retest) that were 2 to 7 days apart. GDI values in the first and second sessions were, respectively, 59.0 ± 8.1 and 60.2 ± 9.4 for the paretic limb and 53.3 ± 8.3 and 53.4 ± 8.3 for the non-paretic limb. The reliability in each session was determined by the intra-class correlation coefficient (ICC) of three strides and, in the test session, their values were 0.91 and 0.97 for the paretic and non-paretic limbs, respectively. Between-session reliability and MDC were determined using the average GDI of three strides from each session. For the paretic limb, between-session ICC, standard error of measurement (SEM), and MDC were 0.84, 3.4 and 9.4, respectively. Non paretic lower limb exhibited between-session ICC, standard error of measurement (SEM), and MDC of 0.89, 2.7 and 7.5, respectively. These MDC values indicate that very large changes in GDI are required to identify gait improvement. Therefore, the clinical usefulness of GDI with stroke patients is questionable.