Construct Validity of the EQ-5D-5L, Time Trade-off, and Standard Gamble in Response to Body Mass Index and Obesity Status: a Cross-Sectional Study - ความเทียงตรงเชิงโครงสร้างของเครืองมือ EQ-5D-5L, Time Trade-off, และ Standard Gamble ในการตอบสนองต่อดัชนีมวลกายและภาวะอ้วน: การศึกษาแบบภาคตัดขวาง

Abstract

                EuroQol 5 Dimensions (EQ-5D) is a widely accepted measure for health utility. However, con-cerns about outcome validity remain when using EQ-5D in some diseases. This study aims to assess the construct validity of 3 instruments, EQ-5D-5L, Standard gamble (SG), and Time trade-off (TTO), in measuring health utility according to body mass index (BMI) and obesity status in the Thai population. A cross-sectional study was conducted in patients aged 15 and over who visited the out-patient department at Suratthani Regional Hospital and 4 other community hospitals in Suratthani during May - July 2015. Samples’ demographic characteristics and underlying diseases were collected from the medical records. The direct utility instruments – SG and TTO – were employed to measure the utility values; questions were asked respecting to their weight-related concerns. Samples were also asked to complete to the EQ-5D-5L questionnaire, which comprises EQ-5D-5L and EQ-VAS. Obesity status was categorized into 4 groups: underweight – BMI<18.5; normal-weight – 18.5<BMI<25; overweight - 25<BMI<30; and obesity – BMI>30. Construct validity was determined by examining the correlation between utility values [EQ-5D-5L, EQ-VAS, TTO, and SG] and BMI and obesity status. A total of 479 patients were included. Results showed that BMI was not correlated with utility values obtained from any instruments. In contrast, modest correlations were observed between obesity status and utility values from SG (r=-0.264, p<0.001) and TTO (r=-0.233, p<0.001); the EQ-5D-5L and EQ-VAS scores were not correlated with obesity status. Despite their difficulty to utilize, SG and TTO remain decent instruments in measuring health utility in obesity research. On the other hand, the unadjusted utility values from the EQ-5D-5L and EQ-VAS appeared to be insensitive to discriminate the severity of obesity in the samples. These results are contra-dictory to that reported in previous studies, indicating the need to further explore on this issue.

Key words: EQ-5D-5L, standard gamble, time trade-off, utility, obesity