The utility of the behavioral risk factor surveillance system in testing quality of life theory: An evaluation using structural equation modeling

This study is a secondary data analysis that uses structural equation modeling (SEM) to assess the utility of the Behavioral Risk Factor Surveillance System (BRFSS)---the Centers for Disease Control's premier surveillance tool for monitoring behavioral risk factors---in predicting health related quality of life (HRQoL). Employing the statistical package SPSS/AMOS (version 7), the study utilizes New York State data extracted from the 2007 BRFSS national dataset to assess how well the observed pattern of variances and covariances of the state level data fit with a well-known HRQoL model developed by Wilson and Cleary (1995). Interaction effects are examined, utilizing a multigroup analytical approach to determine sub-population differences based on race, age, and current health status. The analysis represents both an exploratory study that seeks to identify new applications for this important epidemiological database, as well as a theoretical evaluation that examines the robustness of our current understanding of quality of life as an important health related concept. Overall, findings from the confirmatory factor analysis indicate support of the Wilson and Cleary (1995) model, with the final modified model producing fit indices well within the thresholds traditionally used as benchmarks of good fit. However, findings from the multiple group analyses found non-invariance in the measurement of basic health concepts across the following subpopulations: "old" versus "young;" "White" versus "African American" and "Hispanic;" and "Poor Health" versus "Good" and "Fair" health. Identified differences between groups tend to suggest the Wilson and Cleary Model is "clinically/medically" based and fits less well for those populations who are young and in good health. Furthermore, the examination of racial groups revealed major differences in the conceptualization of health and its impact on quality of life between White versus African-American and Hispanic subpopulations.