Test–retest reliability of the Cost for Patients Questionnaire

Abstract

To investigate the test-retest reliability of the Costs for Patients Questionnaire (CoPaQ). Through an online survey, individuals were invited to participate in a two-step study to assess the test-retest reliability of the CoPaQ. Participants to the first step were invited to complete the questionnaire a second time 2weeks after. Reliability was assessed by calculating Cohen's Kappa coefficients and intraclass correlation coefficients (ICC) for discrete and continuous data, respectively. A sensitivity analysis was carried out. From a total of 1,200 participants who completed the first test, 403 completed the second test. The ICC varied from -0.00 to 0.98 with poor, moderate, good, and excellent results. The Kappa coefficients varied from -0.004 to 0.65 and were poor, slight, fair, moderate, and substantial. The sensitivity analysis showed the median value of ICC and Kappa coefficients for each category before and after the outliers' exclusion. The median value of ICC changed from 0.30 (before) to 0.70 (after), and from 0.12 (before) to 0.04 (after), respectively, for each category. The median value of the Cohen's Kappa coefficient increased from 0.44 (before) to 0.46 (after) and decreased from 0.32 (before) to 0.30 (after), respectively. Test-retest reliability results indicated that the CoPaQ has a moderate reliability in terms of ICC and Kappa coefficients. The moderate reliability observed gives additional support for the applicability of this tool in economic evaluations of health interventions. Additional studies including on other properties and a cultural adaptation could further enhance the use of the tool.