Marí-Dell'Olmo, Marc, Gotsens, Mercè, Palència, Laia, Rodríguez-Sanz, Maica, Martínez-Beneito, Miguel A., Ballesta, Mónica, Calvo, Montse, Cirera, Lluís, Daponte, Antonio, Domínguez Berjón, María Felicitas, Gandarillas, Ana, Izco, Natividad, Martos, Carmen, Moreno-Iribas, Conchi, Nolasco, Andreu, Salmerón, Diego, Taracido, Margarita, Borrell, Carme
Trends in socioeconomic inequalities in mortality in small areas of 33 Spanish cities
BMC Public Health. 2016, 16:663. doi:10.1186/s12889-016-3190-y
URI: http://hdl.handle.net/10045/57606
DOI: 10.1186/s12889-016-3190-y
ISSN: 1471-2458
Abstract: 
Background: In Spain, several ecological studies have analyzed trends in socioeconomic inequalities in mortality from all causes in urban areas over time. However, the results of these studies are quite heterogeneous finding, in general, that inequalities decreased, or remained stable. Therefore, the objectives of this study are: (1) to identify trends in geographical inequalities in all-cause mortality in the census tracts of 33 Spanish cities between the two periods 1996–1998 and 2005–2007; (2) to analyse trends in the relationship between these geographical inequalities and socioeconomic deprivation; and (3) to obtain an overall measure which summarises the relationship found in each one of the cities and to analyse its variation over time. Methods: Ecological study of trends with 2 cross-sectional cuts, corresponding to two periods of analysis: 1996–1998 and 2005–2007. Units of analysis were census tracts of the 33 Spanish cities. A deprivation index calculated for each census tracts in all cities was included as a covariate. A Bayesian hierarchical model was used to estimate smoothed Standardized Mortality Ratios (sSMR) by each census tract and period. The geographical distribution of these sSMR was represented using maps of septiles. In addition, two different Bayesian hierarchical models were used to measure the association between all-cause mortality and the deprivation index in each city and period, and by sex: (1) including the association as a fixed effect for each city; (2) including the association as random effects. In both models the data spatial structure can be controlled within each city. The association in each city was measured using relative risks (RR) and their 95 % credible intervals (95 % CI). Results: For most cities and in both sexes, mortality rates decline over time. For women, the mortality and deprivation patterns are similar in the first period, while in the second they are different for most cities. For men, RRs remain stable over time in 29 cities, in 3 diminish and in 1 increase. For women, in 30 cities, a non-significant change over time in RR is observed. However, in 4 cities RR diminishes. In overall terms, inequalities decrease (with a probability of 0.9) in both men (RR = 1.13, 95 % CI = 1.12–1.15 in the 1st period; RR = 1.11, 95 % CI = 1.09–1.13 in the 2nd period) and women (RR = 1.07, 95 % CI = 1.05–1.08 in the 1st period; RR = 1.04, 95 % CI = 1.02–1.06 in the 2nd period). Conclusions: In the future, it is important to conduct further trend studies, allowing to monitoring trends in socioeconomic inequalities in mortality and to identify (among other things) temporal factors that may influence these inequalities.
Keywords:Disease mapping, Multilevel analysis, Geographical inequalities, Bayesian methods, Trends, Urban areas, Small areas, Mortality, Inequalities in mortality, Socioeconomic inequalities
BioMed Central
info:eu-repo/semantics/article