Knowledge of the epidemiology of congenital heart disease depended on:

• The development of pediatric cardiology as a discipline and specifically on the increasing attention to newborns and to small sick infants, who in earlier days may have died without recognition of their heart disease.
  
  
• The development of good population study methods to identify this young case population and to select appropriate controls.
  
  
• The availability of computing resources to deal with extensive data collections.
  

The BWIS included infants with cardiac defects confirmed in special studies who were residents in a defined Mid-Atlantic region of the United States. This was a case-control study which evaluated all live-born infants with congenital heart disease in comparison to an annually ascertained control population which was representative of all regional live births.

Over a nine year period all regional pediatric cardiology centers enrolled all infants with confirmed congenital heart disease and provided detailed coded diagnoses of all abnormalities as determined by echocardiography, cardiac catheterization, surgery and at autopsy in deceased infants. Cases were also found through searches of pathology records of the 53 regional hospitals, and of death certificates. Controls were selected annually by random selection of resident newborns in all participating hospitals and designated for interview at certain ages to conform to the age-distribution of cases. These controls were representative of the regional birth cohort. All families were visited by a trained interviewer, who obtained the parents' consent to complete a detailed questionnaire regarding the family history, parental health, maternal reproductive history, medical therapies, occupational and home exposures to potentially harmful substances. About 200 covariates were defined for the analysis of case-control differences.

The first analyses compared the data on all cases to all controls. It was then necessary to examine diagnostic subsets of congenital heart disease to establish whether or not the presumed risk factors affected cardiac development generally or whether they were specific for certain chambers or valves of the heart or for the great vessels. This required a categorization of cardiovascular malformations appropriate for the search for possible causes.

Classification of diagnostic subsets:

Clinical differentiation of cardiovascular anomalies according to the anatomic abnormality of the heart and its great vessels was considered in relation to the presumed embryonic time at which development may have been disturbed.

• Early anomalies i.e. defects of 'primary cardiogenesis' were those in which the four-chambered heart was not properly formed.
• In presumably 'late' anomalies the four-chambered structure of the heart is present, but defects might have occurred during the embryonic growth period.

Case-control analyses of diagnostic groups:

All cases

'Early' cardiac defects:	'Late' cardiac defects:
Defects of laterality and looping	Ventricular septal defect: muscular type
Defects of the cardiac outflow tract	Left-sided obstructive lesions
Atrioventricular septal defects	Right-sided obstructive lesions
Ventricular septal defects : membranous type	Atrial septal defects
Total anomalous pulmonary venous return	Patent arterial duct
	Ebstein's malformation of the tricuspid valve
	Cardiomyopathies

For each of these categories the review included details of the cardiac and non-cardiac anomalies, descriptive analysis of prevalence by time, season and area of residence, gender, race and twinning, and a search for genetic and environmental risk factors, including the family history, maternal age and reproductive history, maternal illnesses and medications, lifestyle exposures to potentially toxic substances of both parents at home and in their work.

The methods and results of the BWIS were published in:

• Ferencz,C, Rubin, JD, Loffredo,CA, Magee,CM. The Epidemiology of Congenital Heart Disease, The Baltimore-Washington Infant Study (1981-1989), Perspectives in Pediatric Cardiology, vol.4. MountKisco, N.Y: Futura Publishing Co.Inc, 1993.
  
  
• Ferencz,C Loffredo,CA, Correa-Villasenor, Wilson,PD. Genetic and Environmental Risk Factors of Major Cardiovascular Malformations, The Baltimore-Washington Infant Study, (1981-1989), Perspectives in Pediatric Cardiology, vol.5. Armonk, N.Y: Futura Publishing Co.Inc, 1997.

Findings of the Baltimore-Washington Infant Study are shown in the context of the various topics: Occurrence, Mortality and Risk Factors.

The New England Regional Infant Cardiac Program

This was the first large multicenter population-based cardiac registry which ascertained infants in whom the diagnosis was confirmed by cardiologic studies including findings at cardiac catheterization, surgery and/or autopsy. This was a descriptive clinical study of cases without a study of controls which provided information on the frequency of specific diagnoses, on associated non-cardiac defects, survival and certain family characteristics (Fyler, 1980).

Fyler, DC Report of the 1980 New England Regional Infant Cardiac Program. Pediatrics 1980: 65(suppl):375-461.

Regional Registries of Birth Defects

Nationally and internationally population-based data collections were set up in response to concerns with environmental causes of birth defects. These data collections focused initially on visible malformations, such as oral clefts and neural tube defects, but eventually included also other major outcomes such as cardiovascular and renal disorders.

Special attention to the epidemiology of cardiovascular diseases was given in the following programs:

• Metropolitan Atlanta Congenital Defects Program (MACDP)
• California Birth Defects Monitoring Program
• EUROCAT: European Surveillance of Congenital Anomalies
  (formerly European Concerted Action on Congenital Anomalies and Twins)
• National Birth Defects Prevention Network
• National Center for Birth Defects and Developmental Disabilities