National Health Measurement Study (NHMS)
Unique Identification Number: QG-074-001-1-2-United States-DISC-2005
Summary: The National Health Measurement Study (NHMS) was to surveyed older US adults with a suite of health-related quality-of-life (HRQoL) indexes to allow comparison and cross-calibration of these instruments. The design oversampled African Americans and older individuals to allow subgroup analyses.
Background: Several preference-weighted indexes measuring self-reported generic HRQoL are used widely in population surveys and clinical studies in the U.S. and around the world. These indexes are used to evaluate individual and population health. Because they have been developed using econometric methods to elicit utility weights for their scoring systems they are generally accepted for use in cost-effectiveness analyses of health interventions. Each index uses a multidimensional representation of health, but each index covers the dimensions of health (e.g., physical function, mental function, social function, pain, other symptoms, etc.) differently and they use questionnaires with different psychometric properties. Each index is scored so that perfect health is represented as 1.0 and dead is represented as 0.0, but they are known to have different scaling properties.
Rarely have two or more of these instruments been included in a population survey. So there are few opportunities to directly compare how they describe and measure health using multi-instrument data.
Methodology: In 2005-06, US adults aged 35-89 were contacted by the National Health Measurement Study (NHMS) staff using random digit dialed telephone survey methods; computer-assisted interviewers administered the EuroQol EQ-5D, Health Utilities Index, SF-36v2 TM , and QWB-SA questionnaires in random order to every respondent. Demographic, socioeconomic, and additional health data were elicited. People aged 65+ and telephone exchanges with high proportions of African American households were over-sampled. Approximately 40% of telephone numbers were associated with street addresses and an introductory letter with $2 cash was sent as pre-incentive before the household was contacted by surveyors. Respondents were offered $25 for completing the survey. The data were de-identified, and extensive documentation was developed.
The NHMS collected data on 3844 adults in the continental US (1641 males and 2203 females; 1086 African Americans). This represented a simple response rate of 56%. Post-stratified survey weights and strata to adjust for clustering are supplied to allow national reference estimates for community-living adults ages 35-89.
HRQoL indexes in the data are EQ-5D, HALex, HUI2, HUI3, SF-6D (from both SF-36 and SF-12 questions), QWB-SA, and the 4-item Centers for Disease Control and Prevention HRQOL-4 (“Healthy Days”) instrument, as well as all the SF-36v2 TM scales and summary scores. Individual items subsuming each HRQoL index are available except HUI2 and HUI3, for which only domain levels are archived. Respondents are characterized by census region of residence, age, gender, marital status, race, ethnicity, education, household income and assets, health insurance, weight, height, smoking status, psychological well-being scales, and everyday and lifetime discrimination items. Respondents indicated whether they had been diagnosed with coronary heart disease, stroke, diabetes, arthritis, eye disease, sleep disorder, chronic respiratory disease, clinical depression or anxiety disorder, gastrointestinal ulcer, thyroid disorder, and/or severe chronic back pain. Census tract is not identified; however race composition, education levels, economic factors, and urbanicity of each respondent's census tract of residence are included as contextual variables.
Subject Area: Medical and Health
Data Format: SAS system file
Date of Study: June 2005 - August 2006
Geographic Coverage: Continental United States
Descriptors: Health-related quality-of-life, HRQOL, Health Survey, Self-reported Health, Aging, Health of African Americans, EQ-5D, Health Utilities Index, HUI2, HUI3, Quality of Well-being Index, QWB-SA, SF-36, SF-6D, HALex
Bibliographic Citation: The National Health Measurement Study, 2005-2006. [machine-readable data file] / Dennis G. Fryback [principal investigator] / Data and Information Services Center, University of Wisconsin-Madison [distributor].
Access Status: Unlimited access.
Documentation: Five PDF document files: NHMS codebook, a dataset overview, NHMS Weights, NHMS Summary of Computed Variables and NHMS CATI Script (public version).
Publications: Fryback DG, Dunham NC, Palta M, Hanmer J, Buechner J, Cherepanov D, Herrington S, Hays RD, Kaplan RM, Ganiats TG, Feeny D, Kind P. U.S. norms for six generic health-related quality-of-life indexes from the National Health Measurement Study. Medical Care, 2007; 45(12):1162-1170.
Acknowledgement of Funding: This research was made possible by a grant from the National Institute on Aging, NIH: P01 AG020679.
Version Note: Due to some errors in the values of 3 variables (QWBSCORE, BMI and BMICAT), an updated version of the NHMS datafile was made available in June 2009, with corresponding updates to the codebook file and the summary of computed variables report.
Additional Web Site: Health Measurement Research Group