Socio-economic contextual determinants and behavioral changes during pregnancy: evidence from the "MAMI-MED" cohort

Pregnancy is a period marked by significant physiological and psychological changes in women and increased fetal nutritional requirements, necessitating maternal adaptation and behavior modifications. Clinicians and health institutions recommend pregnant women engage in healthy practices, such as smoking and alcohol cessation, folic acid consumption, vaccinations, and the like. As behavioral changes in general, the individual's conduct during pregnancy is also influenced not only by personal socio-economic status but also by the socio-economic conditions of the individual's area of residence. This mechanism is recognized by the social epidemiological approach and relates to the concept of neighborhood effect on individual health-related choices. Leveraging such considerations, the work aims to explore the association between selected behaviors recommended by clinicians during pregnancy and specific contextual variables in the residential areas where pregnant women live.

Data from the "MAMI-MED" cohort, recruiting pregnant women during the first prenatal visit at a hospital in Catania (Italy), were analyzed. The cohort provides a valuable resource for investigating the relationship between various exposures during pregnancy and the health outcomes of both mothers and infants. Geocoding techniques were employed to link individual-level data to selected contextual variables related to education, income, unemployment, and housing costs in the participants' residential areas. Mann-Whitney test, Kruskal-Wallis tests, logistic regressions and mixtures of regressions models with concomitant variables are implemented 1) to investigate the associations between contextual covariates and individual responses, 2) to assess the presence of latent sub-groups of the population reacting differently to the same contextual factors.

The results of Mann-Whitney and Kruskal-Wallis tests, and logistic regressions indicated that neighborhood's socio-economic characteristics, such as educational level and unemployment rate, are associated with women's behaviors during pregnancy, smoking cessation in particular. Results from the logistic regression for BMI showed that obese and overweight individuals tend to live in neighborhoods where the percentage of individuals holding at least a bachelor's degree is comparatively lower. A mixture of regressions predicting individual BMI detected the presence of two latent groups in the population under analysis. The main finding seems to suggest that people living in worse socio-economic environments have a higher sensitivity to changes in education conditions, with respect to individuals living in better-off neighborhoods.

These findings highlight the importance of considering social and contextual dimensions in understanding and promoting healthy behaviors during pregnancy.