Psychometric evaluation of EQ-5D-5L in OHCA survivors from the TTM2 trial: a post hoc analysis

Abstract

Aims: Our aim was to investigate the psychometric properties of the health assessment instrument EQ-5D-5L in OHCA survivors.

Methods: We included survivors from the Targeted Hypothermia versus Targeted Normothermia after OHCA (TTM2) trial, who completed EQ-5D-5L at 6 months. Confirmatory factor analysis was used to evaluate the hypothesised unidimensional latent structure of EQ level sum score (EQ LSS), summarizing scores across Mobility, Self-care, Usual activities, Pain/discomfort, and Anxiety/depression. Differential item functioning was evaluated for age. We examined internal consistency and precision for the EQ LSS. We evaluated construct validity of EQ LSS, EQ value and EQ VAS, using the modified Rankin Scale and Montreal Cognitive Assessment, representing functional outcome and cognitive function-two common health challenges experienced by OHCA survivors.

Results: 783 of 939 (84%) eligible survivors were included. Confirmatory factor analysis showed good model fit and strong factor loadings for all dimensions (0.61-0.90). We observed a significant but negligible effect of age on Mobility (β = 0.29, p < 0.001, ΔR² = 0.019). Internal consistency was 0.88. The floor effect was 35%. Survivors with more functional dependency and/or cognitive problems reported significantly worse health by EQ LSS, EQ value, and EQ VAS (all, p < 0.001).

Conclusion: The psychometric properties of EQ LSS support its use to measure health status in OHCA research. The strong association between health and functional dependency indicate robust and comparable construct validity for EQ LSS, EQ value, and EQ VAS in this sample.

Keywords: Cross-sectional studies; Health status; Heart arrest; Patient reported outcome measures; Psychometrics; Validation study.

Fig. 1

The figure illustrates the confirmatory factor analysis (CFA) of EQ-5D-5L amongst OHCA survivors from the TTM2-trial (n = 783). The CFA was based on polychoric correlations with Weighted Least Square Mean and Variance (WLSMV) adjusted estimation. The model represents an all-effect model with standardised factor loadings and error variances (ε_1–5). Abbreviations: CFI – Comparative Fit Index; CI – confidence interval; df – degrees of freedom; OHCA – out-of-hospital cardiac arrest; RMSEA – Root Mean Square Error of Approximation; SRMR – Standardized Root Mean Square Residual; TLI – Tucker-Lewis Index; TTM2-trial – the Targeted Hypothermia versus Targeted Normothermia after Out-of-Hospital Cardiac Arrest trial.

Fig. 2

Illustration of the association between health by EQ-5D-5L and modified Rankin Scale (mRS) categories (n = 783). A. EQ level sum score (EQ LSS) ranges between 5 and 25, with higher values indicate more health problems. B. EQ value was based on the Swedish value set, which ranges between 0.243 and 0.976 (indicated by the horizontal dashed lines), with higher values indicating less health problems. C. EQ VAS ranges between 0 and 100, with higher values indicating better overall health.