Reliability, Validity and Responsiveness of the EQ-5D-5L in Assessing and Valuing Health Status in Adolescents and Young Adults with Posttraumatic Stress Disorder: a Randomized Controlled Trail

Abstract

To validate the EQ-5D-5L health-related quality of life (HrQoL) questionnaire in adolescents and young adults with posttraumatic stress disorder (PTSD) after sexual and/or physical abuse. We used data from a randomized controlled trial of 87 participants with PTSD aged 14-21 years. The discriminative ability was evaluated by comparing participants scores on the descriptive system of the EQ-5D-5L with general population scores. Furthermore, the discriminative ability between subgroups of participants with different disease severity levels was estimated. Correlations between the EQ-5D index and clinical parameters were used to measure construct validity. Test-retest reliability was measured by intraclass correlation coefficients (ICC) between baseline, posttreatment and 3-month follow-up scores of participants with stable symptoms. Finally, the responsiveness of the EQ-5D was calculated by mean differences, effect sizes and receiver operating characteristic (ROC) analyses for participants with improved symptoms. Participants reported significantly more problems on the dimensions "mobility", "usual activities", "pain/discomfort" and "anxiety/depression" than the general population. The EQ-5D-5L was able to discriminate between different disease severity levels. The EQ-5D index and clinical scales were significantly correlated with absolute values of correlation coefficients varying between 0.21 and 0.59. Furthermore, ICCs between 0.65 and 0.91 indicated good test-retest reliability for the EQ-5D index. The ICCs for the EQ-VAS between baseline and 3-month follow-up, and posttreatment and 3-month follow-up were statistically significant with 0.71 and 0.87, respectively, whereas the ICC between baseline and posttreatment was 0.08 and not statistically significant. Effect sizes to measure the responsiveness ranged between -0.008 and 0.40 for the EQ-5D index and - 0.32 and 0.40 for the EQ-VAS. Furthermore, the area under the curve in ROC analyses was between 0.40 and 0.64 for the EQ-5D index and 0.60 and 0.70 for the EQ-VAS. Discriminative ability, test-retest reliability and construct validity of the EQ-5D-5L were good, whereas the responsiveness was rather weak. Nevertheless, the EQ-5D-5L can be used to measure HrQoL in adolescents and young adults with PTSD. German Clinical Trials Register identifier: DRKS00004787; date of registration: 18th March 2013; https://www.drks.de