Relative efficiency of the EQ-5D, HUI2, and HUI3 index scores in measuring health burden of chronic medical conditions in a population health survey in the United States.

Department of Community, Occupational and Family Medicine, Yong Loo Lin School of Medicine, National University of Singapore, Singapore.
Medical care (Impact Factor: 2.94). 02/2009; 47(1):53-60. DOI: 10.1097/MLR.0b013e31817d92f8
Source: PubMed

ABSTRACT We sought to compare the ability of the EQ-5D, Health Utilities Index Mark 2 (HUI2), and HUI Mark 3 (HUI3) index scores to discriminate between respondents based on the presence or absence of chronic medical conditions in a population health survey.
Secondary analyses were conducted with data from a probability sample (n = 3480, mean age: 42.5 years, male: 42.4%, Hispanic: 28.6%) of the 2001 noninstitutionalized US general adult population. F-statistic ratios were used to evaluate the relative efficiency of the EQ-5D, HUI2, and HUI3 in differentiating respondents with or without each of 18 chronic medical conditions, and differentiating respondents with low- or high-burden conditions.
In comparing respondents with and without chronic medical conditions, the F-statistic values of these 3 indices were not significantly different, except for EQ-5D versus HUI2 [mean F-statistic ratio: 0.79, 95% confidence interval (CI): 0.59-0.98]. In comparing respondents with a low-burden condition with those with a high-burden condition, the F-statistic values of EQ-5D and HUI2 index scores were similar, while those for EQ-5D versus HUI3 (mean: 0.79; 95% CI: 0.66-0.92) and for HUI2 versus HUI3 (mean: 0.83; 95% CI: 0.71-0.95) were significantly less than 1.0. The overall ceiling effects of the EQ-5D, HUI2, and HUI3 index scores were 48.9%, 15.4%, and 15.3%, respectively.
Although the EQ-5D seems to be marginally less informative, the EQ-5D, HUI2, and HUI3 index scores were generally comparable in determining health burden of chronic medical conditions in this population health survey data.

  • Source
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose To investigate the relation between age and HRQoL indicators in a community-dwelling population aged 65 years and older. Methods Data were collected within a sample stratified by age (65–69; 70–74; 75–79; 80–84; 85–89; 90 years and above) and sex and randomly selected in the population records in Switzerland. The EQ-5D was used to assess HRQoL. Analyses were conducted on the entire available sample (N = 3,073) and on the subsample with no missing data in the EQ-5D (N = 2,888), considering age, gender, education and region. Results Results of multiple regression analyses showed different age-related patterns across the EQ-5D. The proportion of respondents reporting no problems ranged from 51 % in the 65- to 69-year age group to 20 % in the 90 years and above age group. Odds ratio (OR) for Mobility problems increased from 2.04 in the 75- to 79-year age group to 13.34 in the 90 years and above age group; OR for Usual Activities increased from 1.76 to 11.68 and from 1.55 to 2.32 for Pain/Discomfort; OR for Self-Care increased from 5.26 in the 80- to 84-year age group to 30.36 in the 90 years and above age group. Problems with Self-Care remained low, increasing from 6.22 % in the 80- to 84-year age group to 26.21 % in the oldest age group. The magnitude of the gender, region and education effects was much lower than that of age. Conclusion HRQoL is globally preserved in older adults in Switzerland, even if substantial impairment is reported in very old age affecting mainly functional health dimensions. Anxiety/Depression and Pain/Discomfort did not appear to be affected by age; high rates of difficulties were reported for Pain/Discomfort but not for Anxiety/Depression.
    Quality of Life Research 12/2014; DOI:10.1007/s11136-014-0894-2 · 2.86 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The EQ-5D and SF-6D are 2 health-related quality-of-life indexes that provide preference-weighted measures for use in cost-effectiveness analyses. The National Cancer Institute's Cancer Care Outcomes Research and Surveillance (CanCORS) Consortium included the EQ-5D and SF-12v2 in their survey of newly diagnosed lung cancer patients. Utilities were calculated from patient-provided scores for each domain of the EQ-5D or the SF-6D. Utilities were calculated for categories of cancer type, stage, and treatment. There were 5015 enrolled lung cancer patients with a baseline survey in CanCORS; 2396 (47.8%) completed the EQ-5D, and 2344 (46.7%) also completed the SF-12v2. The mean (standard deviation) utility from the EQ-5D was 0.78 (0.18), and from the SF-6D (derived from SF-12v2) was 0.68 (0.14). The EQ-5D demonstrated a ceiling effect, with 20% of patients reporting perfect scores, translating to a utility of 1.0. No substantial SF-6D floor effects were noted. Utilities increased with age and decreased with stage and comorbidities. Patient-reported (EQ-5D) visual analog scale scores for health status had a moderate (r = 0.48, p < 0.0001) positive correlation with utilities. A subset (n = 1474) completed follow-up EQ-5D questionnaires 11-13 months after diagnosis. Among these patients, there was a nonsignificant decrease in mean utility for stage IV and an increase in mean utility for stages I, II, and III. This study generated a catalog of community-weighted utilities applicable to societal-perspective cost-effectiveness analyses of lung cancer interventions and compared utilities based on the EQ-5D and SF-6D. Potential users of these scores should be aware of the limitations and think carefully about their use in specific studies. © The Author(s) 2015.
    Medical Decision Making 02/2015; 35(3). DOI:10.1177/0272989X15570364 · 2.27 Impact Factor