HISPANIC ORIGIN AND NEURAL-TUBE DEFECTS IN HOUSTON HARRIS-COUNTY, TEXAS .2. RISK-FACTORS

Several investigators have reported Hispanics to be at elevated risk f or neural tube defects (anencephaly acid spina bifida). Factors contri buting to this risk have not been established. The authors conducted a case-control study of neural tube defects (NTDs) among births occurri ng in Harris County, Texas, from April 1, 1989, through December 31, 1 991. Through the use of multiple ascertainment methods, 59 cases of an encephaly and 32 cases of spina bifida were detected. Controls (n = 45 1) were sampled for the same time period from Harris County vital reco rds. Regardless of how Hispanic ethnicity was classified, having a His panic parent was a risk factor for both anencephaly and spina bifida. The primary etiologic question was whether increased NTD risk in Hispa nics is explained by maternal diabetes or by other factors (e.g., mate rnal birthplace, prenatal care, reproductive history, age, socioeconom ic status). Mexico-born Hispanics were no more likely than Texas-born Hispanics to deliver a fetus or infant with an NTD. Having a Hispanic mother was a risk factor for anencephaly among infants born to women w ith early prenatal care (odds ratio (OR) = 4.54, 95% confidence interv al (Cl) 2.21-9.40) but not for those born to latecomers. Earlier prena tal care seemed ''protective'' for non-Hispanics (OR = 0.18, 95% Cl 0. 06-0.65) but not for Hispanics. After simultaneous adjustment for eigh t variables in multivariate analysis, having a Hispanic (versus non-Hi spanic) mother remained a strong risk factor for both anencephaly (OR = 2.58, 95% Cl 1.19-5.61) and spina bifida (OR = 3.71, 95% Cl 1.48-9.3 1). Any previous pregnancy termination/fetal loss was also associated with anencephaly in a final logistic regression model (OR = 2.48, 95% CI 1.20-5.10), and having a teenage mother (aged <20 years) approached significance (OR = 2.21, 95% Cl 0.92-5.31). ''Hispanic mother'' was t he only study variable significantly associated with spina bifida in m ultivariate analysis. Results for diabetes suggested no association wi th anencephaly (OR = 1.24, 95% CI 0.25-6.17). An increased risk of NTD s among Hispanics remained after controlling for other factors. For an encephaly, this risk might be partially explained by economic and cult ural differences between Hispanics and non-Hispanics, and the effect o f these factors on rates of prenatal diagnosis acid elective pregnancy termination.