The risks confronting twins: a national perspective.

Our objectives were twofold: (1) to report the relative risks and population-attributable risks of twins compared with singletons for several adverse pregnancy outcomes and (2) to describe the association between having been of low or very low birth weight and death in the neonatal, postneonatal, and infant periods for twins compared with singletons. We performed population-based analysis of all live births and infant deaths from 1985 to 1986 birth cohorts, as reported in the U.S. Linked Birth/Infant Death Data Sets. With singletons as the referent group, twins of all races had relative risk for very low birth weight, low birth weight, and neonatal, postneonatal, and infant death of 9.97, 8.61, 7.06, 2.75, and 5.43, respectively. Although twins make up only 2.09% of live births, the population-attributable risks of twins (the proportion of the population's adverse outcome associated with being a twin) for very low birth weight, low birth weight, and neonatal, postneonatal, and infant death was 15.8%, 13.7%, 11.2%, 3.4%, and 8.4%, respectively. These population-based data show that although twins are relatively infrequent they account for a disproportionately large share of adverse pregnancy outcomes. Given the relative ease with which twins can be identified early in the course of pregnancy, development and testing of interventions to postpone preterm delivery in twin pregnancy should become a national public health priority.