Sociodemographics and HRQL: Mapping the SF-12 and EQ-5D

点击翻译按钮获取中文摘要

DESCRIPTION (Provided by Applicant): Background: Subjective measures of health status and health-related quality of life (HRQL) hold great promise because they realign how the health care community evaluates its programs and services and assess outcomes that are likely to be of most relevance to patients and populations. Few studies, however, have examined the performance of these measures in the U.S. general population and even less is known about their performance in racially heterogeneous, socioeconomically diverse populations. In 2000 MRS administered both the SF-12 (a health profile) and the EQ-5D (a preference-weighted health index) to a large nationally representative sample of the U.S. general population. The inclusion of both questionnaires in the same survey enables an exploration of (1) how each measure captures the effect of sociodemographic factors on HRQL and, (2) how well the SF12 can be mapped onto the EQ-5D in this representative sample. Such a mapping may allow the SF-12, an instrument known to be useful in assessing clinical outcomes in groups of patients, also to be used in cost-effectiveness studies. Specific Aims: Aim 1: (a) To examine the scores on the SF-12 and EQ-5D index and visual analogue scale (VAS) with regard to specific sociodemographic variables, including income and education, age, gender, race/ethnicity, and health insurance coverage; (b) to summarize the population decrement in HRQL (health status burden) associated with income and education, gender, race/ethnicity, and health insurance status. Aim 2: To examine the degree to which the SF-12 can predict scores on the EQ-5D index and EQ VAS. Methods: The first aim will be accomplished through a series of ordinary linear least squares regression analyses. The four dependent variables for each regression analysis will be the PCS-12, MCS-12, EQ-5D index, and EQ VAS. The independent variables will be the sociodemographic characteristics of interest. The second aim will be accomplished through regressing the EQ5D index and EQ VAS onto the MCS-12 and PCS-12 in separate regressions. We initially will use the main effects of the MCS-12 and PCS-12 together with polynomials. The validity of the predicted scores will be examined through the bootstrap approach and use of data from both low-income primary care patients and the York Population Health Laboratory project. Conclusions: The EQ-5D primarily was designed for use in economic analyses while the SF-12 was designed for measurement of clinical health status. Understanding how sociodemographic factors affect HRQL for each measure provides valuable information with which to inform both quality of care and cost-effectiveness analyses (CEA). In addition, understanding the relationship between the two measures could allow data from studies initially undertaken for one purpose (i.e. clinical trials, quality assessment) to be used for the other (i.e. CEA).