The study provides evidence that, even with statistically significant associations, finding variability, but no clear pattern, between trends and associations and/or numbers and rates might in fact suggest relatively spurious relationships; this is a result of not controlling for confounders.

BackgroundIn this study we propose improvements to the method of elaborating deprivation indexes. First, in the selection of the variables, we incorporated a wider range of both objective and subjective measures. Second, in the statistical methodology, we used a distance indicator instead of the standard aggregating method principal component analysis. Third, we propose another methodological improvement, which consists in the use of a more robust statistical method to assess the relationship between deprivation and health responses in ecological regressions.MethodsWe conducted an ecological small-area analysis based on the residents of the Metropolitan region of Barcelona in the period 1994–2007. Standardized mortality rates, stratified by sex, were studied for four mortality causes: tumor of the bronquial, lung and trachea, diabetes mellitus type II, breast cancer, and prostate cancer. Socioeconomic conditions were summarized using a deprivation index. Sixteen socio-demographic variables available in the Spanish Census of Population and Housing were included. The deprivation index was constructed by aggregating the above-mentioned variables using the distance indicator, DP2. For the estimation of the ecological regression we used hierarchical Bayesian models with some improvements.ResultsAt greater deprivation, there is an increased risk of dying from diabetes for both sexes and of dying from lung cancer for men. On the other hand, at greater deprivation, there is a decreased risk of dying from breast cancer and lung cancer for women. We did not find a clear relationship in the case of prostate cancer (presenting an increased risk but only in the second quintile of deprivation).ConclusionsWe believe our results were obtained using a more robust methodology. First off, we have built a better index that allows us to directly collect the variability of contextual variables without having to use arbitrary weights. Secondly, we have solved two major problems that are present in spatial ecological regressions, i.e. those that use spatial data and, consequently, perform a spatial adjustment in order to obtain consistent estimators.