Validation of the EQ-5D in a general population sample in urban China.
ABSTRACT To evaluate the reliability and validity of the EQ-5D in a general population sample in urban China.
Thousand and eight hundred respondents in 18 communities of Hangzhou, China were recruited by multi-stage stratified random sampling. Respondents self-administered a questionnaire including the EQ-5D, the SF-36, and demographic questions. Test-retest reliability at 2-week intervals was evaluated using Kappa coefficient, the intraclass correlation coefficient. The standard error of measurement (SEM) was used to indicate the absolute measurement error. Construct validity was established using convergent, discriminant, and known groups analyses.
Complete data for all EQ-5D dimensions were available for 1,747 respondents (97%). Kappa values were from 0.35 to 1.0. The ICCs of test-retest reliability were 0.53 for the EQ-5D index score and 0.87 for the EQ VAS score. The SEM values were 0.13 (9.22% range) and 4.20 (4.20% range) for the EQ-5D index and EQ VAS scores, respectively. The Pearson's correlation coefficients between the EQ-5D and the SF-36 were stronger between comparable dimensions than those between less comparable dimensions, demonstrating convergent and discriminant evidence of construct validity. The Chinese EQ-5D distinguished well between known groups: respondents who reported poor general health and chronic diseases had worse HRQoL than those without. Older people, females, people widowed or divorced, and those with a lower socioeconomic status reported poorer HRQoL. Respondents reporting no problems on any EQ-5D dimension had better scores on the SF-36 summary scores than those reporting problems.
The Chinese version of the EQ-5D demonstrated acceptable construct validity and fair to moderate levels of test-retest reliability in an urban general population in China.
- SourceAvailable from: PubMed Central[Show abstract] [Hide abstract]
ABSTRACT: Chinese menopausal women comprise a large population and the women in it experience menopausal symptoms in many different ways. Their health related quality of life (HRQOL) is not particularly well studied. Our study intends to evaluate the influence of menopause on HRQOL and explore other risk factors for HRQOL in rural China. An interview study was conducted from June to August 2010 in Beijing based on cross-sectional design. 1,351 women aged 40-59 were included in the study. HRQOL was measured using the EuroQol Group's 5-domain (EQ5D) questionnaire. Comparison of HRQOL measures (EQ5D index and EQ5D-VAS scores) was done between different menopausal groups. Logistic regression and multiple regression analysis were performed to adjust potential confounders and explore other risk factors for health problems and HRQOL measures. Postmenopausal women who had menopause for 2-5 years (+/-1b stage) were more likely to suffer mobility problems (OR = 1.835, p = 0.008) after multiple adjustment. Menopause was also related to impaired EQ5D index and EQ5D-VAS scores after adjustment for age. Among menopausal groups categorized by menopausal duration, a consistent decrement in EQ5D index and EQ5D-VAS scores, that is, worsening HRQOL, was observed (p < 0.05). Multiple regression analysis revealed low education level and physical activity were associated with EQ5D index (beta = -0.080, p = 0.003, and beta = 0.056, p = 0.040, respectively). Cigarette smoking and chronic disease were associated with EQ5D index (beta = -0.135, p < 0.001 and beta = -0.104, p < 0.001, respectively) and EQ5D-VAS (beta = -0.057, P = 0.034 and beta = -0.214, p < 0.001, respectively). Reduction in physical function was found within the first five years after menopause. Worsening EQ5D index and EQ5D-VAS scores were related to menopause. Education level, physical activity, cigarette smoking, and chronic disease history were associated with HRQOL in middle aged Chinese rural women.BMC Women's Health 01/2014; 14(1):7. · 1.66 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Patients with major depressive disorder (MDD) may suffer from concomitant pain symptoms. The aim of this study is to determine whether the presence of painful physical symptoms (PPS) influences quality of life when taking into account baseline depression severity. Patients with a new or first episode of MDD (n = 909) were enrolled in a 3-month prospective observational study in East Asia. The Hamilton Depression Rating Scale, Clinical Global Impression-Severity score, Somatic Symptom Inventory, and EuroQoL questionnaire-5 Dimensions (EQ-5D) and EQ-Visual Analogue Scale (EQ-VAS) were assessed at baseline and 3 months' follow-up. The presence of PPS was defined as a mean score of ≥2 on the Somatic Symptom Inventory pain-related items. Regression analyses determined predictors of quality of life at 3 months, adjusting for age, sex, depressive symptoms, overall severity, and quality of life at baseline. PPS were present (PPS+) at baseline in 52% of patients. During the 3-month follow-up, EQ-VAS scores improved from 47.7 (standard deviation [SD] 20.6) to 72.5 (SD 20.4), and EQ-5D improved from 0.48 (SD 0.34) to 0.80 (SD 0.26). At 3 months, mean EQ-VAS was 66.4 (SD 21.2) for baseline PPS+ patients versus 78.5 (SD 17.6) for baseline PPS- patients, and mean EQ-5D was 0.71 (SD 0.29) versus 0.89 (SD 0.18). PPS+ at baseline was a significant predictor of quality of life at 3 months after adjusting for sociodemographic and baseline clinical variables. The presence of painful physical symptoms is associated with less improvement in quality of life in patients receiving treatment for major depression, even when adjusting for depression severity.Patient Preference and Adherence 01/2013; 7:463-70. · 1.33 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: The purpose of the study was to compare psychometric properties of the EQ-5D-5L (5L) and the EQ-5D-3L (3L) health outcomes assessment instruments in patients with hepatitis B in China. Patients, including hepatitis B virus carriers and those with active or inactive chronic hepatitis B, compensated cirrhosis, decompensated cirrhosis or hepatocellular carcinoma, answered a questionnaire composed of 5L, socio-demographic information, 3L, and the visual analog scale (VAS), respectively. After 1 week, a retest was conducted for inpatients. We compared acceptability, face validity, redistribution properties, convergent validity, known-group validity, discriminatory power, ceiling effect, test-retest reliability, and responsiveness of 5L and 3L. A total of 369 outpatients and 276 inpatients were recruited for the first interview. Of the inpatients, 183 were used in the retest. Most patients preferred 5L-3L. The 3L-5L response pairs had an inconsistency rate of 2.4 %. Correlation with the VAS was greater with 5L than with 3L. Age, education, and comorbidity were associated with health-related quality of life (HRQoL). 5L discriminated more infectious conditions than 3L. In all dimensions, the Shannon's index from 5L was larger while in three dimensions the Shannon's evenness index from 5L was slightly larger. The ceiling effect was reduced in 5L. In patients with stable health states, no significant difference was detected in the weighted kappa between 5L and 3L, but intraclass correlation coefficient of 5L was higher than that of 3L. In patients with improved health states, HRQoL was seen as increased in both 5L and 3L, without significant difference. The EQ-5D-5L was more suitable than the EQ-5D-3L in the patients with hepatitis B in China.Quality of Life Research 03/2014; · 2.86 Impact Factor