Measuring the health state of individuals is important for the evaluation of health interventions, monitoring individual health progress, and as a critical step in measuring the health of populations. Self-report responses in household survey data are widely used for assessing the non-fatal health status of populations. The object of this document is to elaborate on several statistical models used in the analysis of survey data. First, the paper focuses on off-the-shelf models that are widely available as part of any standard statistical software. In particular, the authors demonstrate the problems of inference that arise from these standard methods when the underlying data are not cross-population comparable. In later sections, the authors introduce methods that modify these standard routines to enhance the cross-population comparability of survey analyses.

   One of the World Health Organization’s longest standing mandates is the collection and routine reporting of information on population health. This paper reports on the average level of population health, focusing on the self-reported health status in six domains assessed through 66 population-based surveys conducted in 57 countries included within the WHO Multi-Country Survey. Based on the best analytical methods currently available, the results presented within this paper and incorporated within subsequent analyses to estimate Healthy Life Expectancy, provide more comparable information on the self-reported average level of population health across countries than was previously possible from survey data.

   Murray et al. have outlined a series of different strategies for enhancing cross-population comparability of survey results through the formal analysis of systematic cutpoint shifts. One way to address this problem, whether it arises in self-reported or physician-assessed data, is by fixing the levels of the unobserved latent variable of interest in order to isolate cutpoint differences as the source of variation in assessments of these levels. In combination with new statistical models, the incorporation of this exogenous information allows estimation of variation in cutpoints attributable to socio-demographic or other factors. This paper we describes the application of this new approach to the publically-available National Health and Nutrition Examination Survey dataset. The objective of this paper is to examine whether sex, race/ethnicity and income affect self-reports and physician-assessments of mobility through predictable differences in the use of categorical responses.