Reliability and validity of the modified Toronto Clinical Neuropathy Score in diabetic sensorimotor polyneuropathy

Abstract

Introduction A reliable and valid clinical tool to capture symptoms and signs of diabetic sensorimotor polyneuropathy (DSP) for use in clinical research trials is urgently needed. The validated Toronto Clinical Neuropathy Score (TCNS) was modified to improve sensitivity to early DSP changes. We aimed to assess the reproducibility of this modified tool, the mTCNS and to determine its validity relative to the precursor TCNS. Methods Sixty‐five patients (six Type 1, 59 Type 2 diabetes) with diabetes duration 13 ± 8 years were accrued from four study sites and examined on 2 days for internal consistency and inter‐ and intra‐rater reliability of the mTCNS. In the absence of a single quantitative gold‐standard measure for DSP, results of the mTCNS were compared with the precursor TCNS for the purpose of estimating validity. Results Internal consistency of the two domains within the mTCNS was good (Cronbach's alpha 0.78). Very good inter‐rater reliability for the mTCNS was demonstrated by an intra‐class correlation coefficient for the mTCNS of 0.87 (95% confidence interval, 0.79–0.91), which was similar in magnitude to that of the TCNS (0.83; 95% confidence interval, 0.75–0.89). Intra‐rater reliability testing of the mTCNS showed moderate to good correlation for individual symptoms and sensory tests (Cohen's kappa values of 0.54–0.73). The mTCNS shared moderate correlation with the precursor TCNS (Pearson correlation coefficient, 0.58). Discussion The mTCNS, a clinical score with higher face validity for tracking mild to moderate DSP, has sufficient reliability and validity relative to its precursor TCNS for use in clinical research.