Abstract

Objectives

This study aimed to explore the impact of recall period on the measurement properties of the EuroQol 5-Dimension 5-Level (EQ-5D-5L) and EuroQol Health and Well-Being 9 items (EQ-HWB-9) instruments.

Methods

An online cross-sectional survey was conducted with a general population sample in Hong Kong (n = 1262). Respondents completed both a “today (TD)” and a “last week (LW)” version of both the EQ-5D-5L and EQ-HWB-9. Psychometric properties (eg, distributional characteristics, ceiling, informativity, convergent validity with conceptually related items, or domains of ReQoL-10 and Patient-Reported Outcomes Measurement Information System − 29 + 2, known-group and structural validity) were assessed for both recall period versions of both instruments.

Results

Considering all measurement properties, differences between LW and TD versions were small but slightly larger for the EQ-HWB-9 than for the EQ-5D-5L. LW versions showed lower ceilings, captured more problems, and demonstrated greater informativity. They had comparable or stronger correlations with ReQoL-10 and Patient-Reported Outcomes Measurement Information System − 29 + 2 items or domains than TD, especially for EQ-HWB-9. Known-group validity was slightly better for LW versions, particularly in mental health subgroups, whereas structural validity of both measures was largely consistent across recall periods.

Conclusions

In this general population sample, although the LW recall versions exhibited slightly superior measurement properties compared with the TD versions, our findings overall support the original recall periods for the EQ-5D-5L and EQ-HWB-9. Future research should evaluate the impacts of varying recall periods across populations and cultural contexts to further refine guidelines for instrument selection in the intended areas of use for each instrument.