False-positive reporting of Down syndrome on Ohio and New York birth certificates.

Although correction for underreporting of congenital malformations on birth certificates is included in most studies, inaccuracy of reporting has not been widely examined. Two separate investigations were conducted on the inaccuracy of Down syndrome (DS) reporting on birth certificates; ie, false-positive cases in which an individual coded as DS did not in fact have DS. In Ohio, 824 individuals were coded as DS on their birth certificate during 1970-1981. Of these, a definitive determination as to whether or not they had DS was made on 778 by using cytogenetic data, medical records, the state's birth defects registry, school records, and by questioning physicians. Fifty-seven false-positives were found, indicating a 7.8% level of coding inaccuracy for all races and 6.9% for whites only. Nine of these arose from miscodings during data processing; 48 were misdiagnosed as DS. This can be contrasted with false-negatives also studied in Ohio, where 66.1% of DS cases were not reported on the birth certificate. No statistical differences were observed between false-positives and true DS in the distribution of sexes, in population size of county of birth, or in year of birth (although there was a declining false-positive rate over the 12 year period). The percentage of DS false-positives, however, was significantly higher for younger maternal ages (greater than or equal to 30 years) than older ones (greater than or equal to 30 years) and for nonwhites compared to whites. Further, there was a strong negative correlation between the percentage of false-positives and the degree of certainty expressed in reporting DS on the birth certificate.