The goal of this paper is to illustrate problems with the comparability of health indicators used in cross-national research and how cultural and validity biases may confound the interpretation of results. In particular, the authors address the comparability issues by using self-reported health in two different contexts—Mexico and the United States. The study design of MHAS allows for the first time to compare differences in self-reported health between a developing country (Mexico) and an industrialized country (the U.S.). Our results show that many of the observed patterns of self-reported diseases are sensitive to demographic phenomena such as, for example, differences in population age structure. In addition, our results show that Mexicans tend to down-grade their health status as being worse compared to non-Hispanic white Americans, and these differences cannot be explained by differences in diseases load or other objective health indicators. Thus, the analyses suggest that self-reported health of Mexicans may be culturally biased and determined by a general tendency to report worse health status.