Purpose: The population-based National Down Syndrome Project combined epidemiological and molecular methods to study congenital heart defects in Down syndrome. Methods: Between 2000 and 2004, six sites collected DNA, clinical, and epidemiological information on parents and infants. We used logistic regression to examine factors associated with the most common Down syndrome-associated heart defects. Results: Of 1469 eligible infants, major cardiac defects were present in 44%; atrioventricular septal defect (39%), secundum atrial septal defect (42%), ventricular septal defect (43%), and tetralogy of Fallot (6%). Atrioventricular septal defects showed the most significant sex and ethnic differences with twice as many affected females (odds ratio, 1.93; 95% confidence interval, 1.40 -2.67) and, compared with whites, twice as many blacks (odds ratio, 2.06; 95% confidence interval, 1.32-3.21) and half as many Hispanics (odds ratio, 0.48; 95% confidence interval, 0.30 -0.77). No associations were found with origin of the nondisjunction error or with the presence of gastrointestinal defects. Conclusions: Sex and ethnic differences exist for atrioventricular septal defects in Down syndrome.Identification of genetic and environmental risk factors associated with these differences is essential to our understanding of the etiology of congenital heart defects. The National Down Syndrome Project (NDSP) seeks to investigate the etiology and phenotypic consequences of trisomy 21 Down syndrome (DS). 1 Aside from the universal findings of mental retardation and hypotonia, congenital heart defects (CHDs) are arguably the most important clinical sequelae of an extra chromosome 21. In 1998 the Atlanta Down syndrome Project (ADSP), a forerunner of the NDSP, reported that 41% of newborns with DS were born with one or more major heart defects, including atrioventricular septal defect (AVSD), secundum atrial septal defect (ASDII), ventricular septal defect (VSD), and tetralogy of Fallot (TOF). 2 Findings from the ADSP and other recent population-based studies of DS and CHDs 2-5 are summarized in Table 1.With the birth prevalence of major DS-associated CHDs well established by multiple studies using modern diagnostic methods, attention can now be directed toward understanding the etiology of these defects. Not only do infants with DS have a higher rate of CHDs than do infants without DS, but one defect, the AVSD, is particularly characteristic. To understand the etiology of CHDs in DS and of AVSD specifically, both genetic and environmental determinants must be explored. For example, several recent reports have suggested that the distribution of CHDs in DS varies by ethnicity (race/ethnicity), 6 -13 but most population-based studies have not had broad ethnic representation (Table 1). Drawing on our experience with the ADSP, we designed the multicenter NDSP to explore possible CHD risk factors singly and in combination. The NDSP is one of the largest population-based studies of CHDs in DS and the first to assemble clinical, demographic, a...