Studying the Associations Between Socioeconomic Disparities and Myocardial Infarction Onset in a French Urban Area - Archive ouverte HAL Access content directly
Journal Articles Epidemiology Year : 2007

Studying the Associations Between Socioeconomic Disparities and Myocardial Infarction Onset in a French Urban Area

(1) , (2) , (3) , (4) , , (5)
1
2
3
4
5

Abstract

Objective: To assess the associations between the distribution of myocardial infarction onset rates and the contextual socioeconomic status (SES), using a specifically designed deprivation index, in an eastern French urban area and at a small geographic scale (IRIS, 2000 inhabitants on average) (PAISIM Project, for Atmospheric Pollution, Socioeconomic Disparities, and Myocardial Infarction). Material and Methods: One thousand one hundred and ninety-three myocardial infarction events in people aged 35 to 74, recorded between January 1, 2000, and December 31, 2003, were obtained from the local population-based coronary heart disease registry. Each case was allocated to its IRIS of residence. We calculated incidence rates by sex and by age group and age-standardized incidence ratios (SIR) by IRIS. To reduce the instability of extreme rates due to a sometimes very low number of events by IRIS, we smoothed our data using a local empirical Bayesian approach. We constructed, at the IRIS resolution, a census-based deprivation index that underlines strong socioeconomic disparities between geographic areas. The associations between the distribution of health outcome rates and deprivation index were estimated using an ordinary least squares (OLS) model. The spatial autocorrelation in data was assessed by Moran's index. Results: We found a significant relation between Age-SIR of myocardial infarction cases in men and SES index. However, the contribution of SES to explain the Age-SIR is relatively weak (R2 = 0.025; P < 0.05). The same conclusions can be drawn for the incidence rates (all ages) in women (R2 = 0.1094; P < 0.01). After smoothing, a positive and significant spatial autocorrelation was detected (Moran's Index about round 0.60; P < 0.01). Conclusions: The OLS model does not appear to satisfy to test associations between our datasets. More adequate models as hierarchical Bayesian models should be considered to study these associations. These models use a best smoothing health data and can take into account the spatial autocorrelation.
Not file

Dates and versions

hal-00689816 , version 1 (20-04-2012)

Identifiers

Cite

Sabrina Havard, Séverine Deguen, D. Arveiler, O. Laurent, D. Eilstein, et al.. Studying the Associations Between Socioeconomic Disparities and Myocardial Infarction Onset in a French Urban Area. Epidemiology, 2007, 18 (5), pp.S85. ⟨10.1097/01.ede.0000288982.72134.1c⟩. ⟨hal-00689816⟩
45 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More