Epidemiology of twinning in the National Birth Defects Prevention Study, 1997 to 2007.

BACKGROUND: Our objective was to evaluate associations between twinning and maternal demographic factors and periconceptional exposures among infants with and without orofacial clefts. METHODS: We used data from the National Birth Defects Prevention Study; 228 twins and 8242 singletons without birth defects (controls), and 117 twins and 2859 singletons with orofacial clefts, born 1997 to 2007, were included in the analyses. Because of the occurrence of twinning due to the use of assisted reproductive technologies, logistic regression models were computed to estimate odds ratios and 95% confidence intervals for each exposure, stratified by fertility treatment use. To evaluate factors by zygosity, we used sex-pairing data and a simulation approach to estimate the zygosity of like-sex twin pairs for unassisted conceptions. RESULTS: Among control mothers who did not use fertility treatments, predictors of twinning included non-Hispanic black maternal race (adjusted odds ratio, 1.6; 95% confidence interval, 1.0-2.4), and tobacco smoking (adjusted odds ratio, 1.6; 95% confidence interval, 1.1-2.4). Among control mothers who used fertility treatments, older maternal age, higher income, and state of residence were associated with twinning. Associations were generally stronger among mothers of dizygotic (estimated) twins than monozygotic (estimated) twins. Results for mothers of infants with isolated orofacial clefts were similar to those of controls. CONCLUSION: We observed an increased twinning frequency with increasing maternal age, but factors such as maternal race/ethnicity and socioeconomic status may also contribute. Among women receiving fertility treatments, factors associated with twinning suggested a relation with treatment specifics (e.g., treatment type and number of embryos implanted) and availability of insurance coverage.