The “no problems”-problem: an empirical analysis of ceiling effects on the EQ-5D 5L

Abstract

To analyze the association between ceiling effects on the EQ-5D 5L and morbidity in a general population sample. We used a cross-sectional sample of the German general population (n = 5007) to describe the frequency of health state "11111" and "no problems"-answers on the five single dimensions stratified by the number of diseases for which participants utilized health care during the last 6months. For the five single dimensions we also used specific criteria to analyze their discriminative ability. A logit-model was applied for a multivariate analysis of ceiling effects. 31% of participants reported the health state "11111." This percentage strongly decreased with increasing morbidity, down to 4.9% if four or more diseases were present. The dimensions "mobility," "usual activities," and "pain/discomfort" showed good discriminative abilities. The dimensions "anxiety/depression" and "self -care" were able to discriminate between different levels of morbidity, but nevertheless showed strong ceiling effects, in particular "self-care." When analyzing ceiling effects of the EQ-5D 5L, one has to draw attention to morbidity since high proportions of participants indicating the best health state might result from being healthy regarding the dimensions assessed by the EQ-5D, in particular in general population datasets.