Rasch-Transformed Total Neuropathy Score clinical version (RT-TNSc(C)) in patients with chemotherapy-induced peripheral neuropathy

Composite scales such as the Total Neuropathy Score clinical version (TNSc(C)) have been widely used to measure neurological impairment in a standardized manner but they have been criticized due to their ordinal setting having no fixed unit. This study aims to improve impairment assessment in patients with chemotherapy-induced peripheral neuropathy (CIPN) by subjecting TNSc(C) records to Rasch analyses. In particular, we wanted to investigate the influence of factors affecting the use of the TNSc(C) in clinical practice. TNSc(C) has 7 domains (sensory, motor, autonomic, pin-prick, vibration, strength, and deep tendon reflexes [DTR]) each being scored 0-4. Data obtained in 281 patients with stable CIPN were subjected to Rasch analyses to determine the fit to the model. The TNSc(C) did not meet Rasch model's expectations primarily because of misfit statistics in autonomic and DTR domains. Removing these two, acceptable model fit and uni-dimensionality were obtained. However, disordered thresholds (vibration and strength) and item bias (mainly cultural) were still seen, but these findings were kept to balance the assessment range of the Rasch-Transformed TNSc(C) (RT-TNSc(C)). Acceptable reliability findings were also obtained. A 5-domains RT-TNSc(C) may be a more proper assessment tool in patients with CIPN. Future studies are needed to examine its responsive properties.