The Modified Rivermead Mobility Index: reliability and convergent validity in a mixed neurological population.

Abstract

To assess the inter-rater reliability, internal consistency and convergent validity of the Modified Rivermead Mobility Index (MRMI) in a mixed neurological population. The MRMI was scored for 30 consecutive patients (mean age 54.5+/-15.6 years) by two individual testers. Reliability was examined using intraclass correlation coefficients (ICC3,1) and Bland and Altman plots; internal consistency reliability using Cronbach's alpha (alpha) and convergent validity using Spearman's correlation coefficient (rho) test to compare the MRMI to the 10-m walk test as a gold standard of mobility. As the majority of patients had bilateral deficits, the MRMI was measured and added independently for both sides. The inter-rater reliability was excellent: ICC (95% CI)=0.93(0.86, 0.96). The Bland and Altman plots contained most data points and there was perfect agreement between raters bilaterally in 27% of cases, with a difference of one point in 60% of cases on the left and 63% of cases on the right. Internal consistency was good at alpha=0.72 (Rater 1) and 0.80 (Rater 2). The Spearman rho between MRMI and the 10-m walk test was high at 0.86. The MRMI was shown to have high levels of reliability in a mixed neurological population but we recommend that its psychometric properties are further investigated to establish the true clinical utility of this measure.