Development of a simplified Cold Intolerance Symptom Severity questionnaire in patients with peripheral nerve injury.

The aim of this study was to analyse the Cold Intolerance Symptom Severity (CISS) questionnaire in its Italian validated version, using Rasch analysis, to gain insights for a possible refinement of the questionnaire. The CISS was administered to a convenience sample of 96 consecutively recruited outpatients with upper limb peripheral nerve injury. Data were analysed using Rasch analysis. According to rating scale diagnostics, response options of items 3 and 5 did not comply with the pre-set criteria for an optimal category functioning. After collapsing the malfunctioning categories, all items fitted the measured construct. Principal component analysis of standardized residuals showed local dependence between two items (one of them was considered redundant and deleted); after this deletion, unidimensionality of the 12-item questionnaire (CISS-12) was achieved. The reliability indices of CISS-12 were high (>0.85). Some clearer item wording was introduced in response to comments from an expert panel and patient feedback. Overall, Rasch analysis provided the rationale for improving the measurement qualities of the questionnaire, refining its rating scales, identifying those items most useful for measuring the intended construct and confirming the high reliability of its person-ability and item-difficulty estimates. In conclusion, the new simplified CISS-12 presents robust psychometric properties for measuring cold intolerance in patients with upper limb peripheral nerve injury and represents a solid basis for clinical studies aimed at a precise (interval level) measurement of cold-induced symptoms in these patients.