Some factors affecting the secondary sex ratio in a Latin American sample.

ABSTRACT

PIP: A sample based on hospital birth records from the Latin American Collaborative Study on Congenital Malformations (ECLAMC) was used. ECLAMC, which covers Uruguay, Chile, Argentina, Brazil, Bolivia, Peru, Paraguay, Ecuador, Venezuela, Colombia, and Costa Rica, registered 1,037,272 live births in the period 1982-1986. Weighted stepwise multiple regression analyses were used, and in all models the sex ration was the dependent variable. Maternal ages were grouped in categories of 5-year intervals, from mothers up to 19 years of age to mothers over 45 years of age. The 1st analysis of the sex ratio used biological (maternal age, birth order), temporal (year), and spatial (country) factors as independent variables. The sex ratio was significantly affected by secular, spatial, biological (maternal age, birth order, and ethnic group), and socioeconomic (evaluated by hospital payment) variables. The black ethnic component removed the spatial effect (Brazil and Venezuela) in certain cases. The Amerindian admixture effect on the sex ratio was negative and significant. Sex ratios were highest in mothers under 17 years old, gradually declined with age up to about 27 years, and thereafter increased slightly. Sex ratios decreased as birth order increased. In addition, a positive secular trend was observed (regression coefficient = 0.00082/yr). Other analyses included black and Amerindian components to verify whether the observed spatial effects were caused by different ethnic compositions. A final model with all the variables together with ethnic group indicated that ethnic stratification and form of payment had a negative correlation to sex ratio without any dramatic change compared to the previous analysis. Both the black and the Amerindian components are capable of removing the spatial effects of Brazil and Venezuela on the sex ratio.