Swedish Version of the Distress Thermometer: Validity Evidence in Patients With Colorectal Cancer.

Abstract

Background: The objective of this study was to validate the NCCN Distress Thermometer (DT), including the accompanying Problem List (PL), in a Swedish population of patients diagnosed with colorectal cancer (CRC). Methods: A total of 488 patients diagnosed with CRC completed the DT/PL and EORTC core quality-of-life questionnaire (QLQ-C30) before surgery. Construct validity of the PL was analyzed using a confirmatory factor analysis. Internal consistency reliability (ICR) was tested using Cronbach's alpha coefficient. Correlations between the reported PL areas and QLQ-C30 function scales were used to explore convergent validity. Discriminant validity was examined by evaluating associations between the DT and QLQ-C30 measures of overall health-related quality of life (HRQoL). Results: Findings showed that the Swedish translation of the DT/PL is consistent with the original English version. The DT has good ICR, with the total number of reported problems significantly correlating with DT scores (r=0.67; P<.001). Analysis of convergent validity indicated that the PL areas significantly correlated with QLQ-C30 function scales, with emotional problems showing the highest correlation (r=0.76; P<.001), and item-level correlation analyses showed significant correlations between symptoms. There was also good discriminant validity between the DT and the QLQ-C30 in terms of HRQoL, including overall health status (r=-0.49; P<.001) and overall quality of life (r=-0.57; P<.001). Furthermore, there was good discriminant validity between the DT and QLQ-C30 regarding poor, moderate, and excellent HRQoL. Conclusions: These findings provide validity evidence regarding the DT, including the PL. Findings also show that the DT has good potential for screening distress-related practical, family, emotional, and physical problems during the cancer trajectory in Swedish-speaking patients. Additionally, the DT seems to be an effective screening tool to detect patients with poor, moderate, and excellent HRQoL.