The methods used by the Latin American Collaborative Study of Congenital Malformations (ECLAMC) for monitoring the birth prevalence of multiply malformed infants are based on a clinical-epidemiological approach oriented to the early detection of teratogenic agents. They consist of three steps: 1) the analysis of observed vs expected rates of all congenital anomalies (CA), including their isolated and associated forms; 2) the same type of analysis applied to each multiple congenital anomaly (MCA) pattern; and 3) a clinical case presentation reserved only for those considered as true MCA because of presence of three or more independent CA. During the period 1982-1983 299,231 infants were examined. Multiply malformed infants, excluding Down syndrome cases, were born at a rate of 4/10,000, 40% having syndromes (two or more interrelated CA), 30% anomaly pairs, ie, two independent CA, and 30% true MCA cases. In a program with 150,000 births per year, as in ECLAMC, this means about five true MCA cases per week, a number easily handled individually on a clinical basis.