Birth Defects: Are We Doing Enough?

by Clifford J. Sherry

Dr. Sherry is a teratologist and neurophysiologist living in Skrivaner, Texas.

This article appeared in the Christian Century  December 5, 1984, p. 1150. Copyright by the Christian Century Foundation and used by permission. Current articles and subscription information can be found at This material was prepared for Religion Online by Ted & Winnie Brock.


Birth defects do not respect ethnic, religious or socioeconomic backgrounds. The first and most basic step in research leading to their prevention is accurate knowledge of how often, when and where they occur. Since the existing system for collecting such data, even if it were working perfectly, would still miss two out of every three babies born with defects, it seems wise to modify it.

Major birth defects or congenital anomalies occur in approximately 8.3 per cent of all live births. One out of every 12 people one meets is likely to have a congenital anomaly. It might be something visible like a cleft palate, or something invisible like diabetes.

Records of infant mortality have been kept since about 1850 for Massachusetts, and since about 1915 for the nation as a whole. The peak occurred in 1870-1874, when deaths reached a level of 170 per 1000 live births. During the next 100 years. infant mortality in the United States decreased to its present level of 20 per 1000 live births. This decrease is due primarily to the development of vaccines and antibiotics, which help control infectious diseases. Currently, the major cause of infant mortality is birth defects and congenital anomalies. Since the United States is a rich, powerful, humanitarian country, it might seem that the rate of 20 deaths per 1,000 live births were a threshold that could not be lowered. But Sweden. the Netherlands, Iceland, Norway. Finland, Switzerland, France, Taiwan, Israel, West Germany, Belgium and Canada have all done a consistently better job than we have of reducing that rate.

The federal agency charged with primary responsibility for collecting incidence data about birth defects is the Birth Defects Branch of the Center for Disease Control in Atlanta. It uses two different but complementary data bases. The first, the Birth Defects Monitoring Program, is a voluntary data-collection system using approximately 882,660 live births in approximately 1,049 hospitals in the United States. This system is not based on population or on a random sample of births. Since its sample represents only 29 per cent of all births in only 16 per cent of all United States hospitals, its usefulness as illustrative of the whole country is limited.

The other data base used by the Birth Defects Branch is obtained from birth certificates by the National Center of Health Statistics. Theoretically, this, information should be more nearly complete (and therefore more representative and useful), but there are some significant problems with it.

Although its accuracy has never been examined nationwide, three recent studies conducted in Iowa, New York and Texas reveal some of the data base’s weaknesses. In the most comprehensive analysis, conducted in Iowa, the Public Health Services Congenital Anomalies Section (a predecessor of the Birth Defects Branch) compared 1963 birth certificates with hospital records. It found that only 39 per cent of the major defects reported in the records were also entered on the birth certificates. Of this number, only 72 per cent were recorded accurately. Thus, only 28 per cent of defects detected at birth and recorded in hospital records were accurately reported on birth certificates, the data source for the National Center for Health Statistics and the Center for Disease Control. Two out of every three children born with major defects, then, are missed by this system. The New York study, conducted in 1972, found that only 12 per cent of the defects reported in hospital records also appeared on birth certificates. The Texas study, conducted in 1981, traded the level of reporting of only one particular birth defect, spina bifida, Although spina bifida (“open spine”) is easy to detect, less than half of the birth certificates reported the condition.

Second, when newborns have multiple anomalies, the most severe is often the only one recorded. The level of reporting varies widely from locale to locale and from time to time. For example, a recent study by the National Center for Health Statistics showed Delaware recording 364 birth defects per 100,000 live births in 1973, while New Mexico reported 1,943 per 100,000 in 1974. During this period, New Mexico claimed a 68 per cent increase in the incidence of birth defects while 11 other states reported increases of 20 per cent or more, and 12 reported ten to 20 per cent.

Third, even if the current system for collecting these data were working efficiently, we would still miss the majority of children with congenital anomalies since many are not detectable at birth. A recent study by the National Foundation -- March of Dimes found that only 33.6 per cent of major defects were recognizable at birth. Approximately two out of every three children born with major anomalies are missed. When added to the inefficiency of our reporting system, this means that we actually have records of only from four (based on the 12.1 per cent rate of reporting found in the New York study) to nine (based on the 27.9 per cent accuracy of reporting found in the Iowa study) out of every 100 children suffering from birth defects.

Clearly, the usefulness of these data is limited in estimating the actual incidence of birth defects; establishing the temporal and/or spatial increases or decreases in these incidences; determining if there are isolated epidemics” of specific anomalies that might be due to certain environmental agents; and providing useful data for finding the causes of and/or preventatives for birth defects.

Although we do not know what brings about most birth defects, we do know that our environment is becoming increasingly contaminated, and that some defects are caused by environmental agents called teratogens. Most teratogen-induced imperfections are caused by exposure to the agent during the first trimester of pregnancy, when most of the major organs and systems are being formed. Probably the most infamous teratogen is thalidomide, a tranquilizing, anti-nausea and sleep-inducing drug. In 1956, when it was introduced in Europe, it was heralded as safe, effective and almost without side effects. While thalidomide apparently is relatively safe for adults, it can have disastrous effects on developing embryos. Doses as low as 54 milligrams in an average-sized adult female (far less than a quarter of a teaspoon) can cause an increased risk of phocomelia --  a decrease in the size of the upper limbs. For some time, no one suspected that thalidomide was the cause of the significant increase in the incidence of phocomelia.

It required the brilliant retrospective studies of Drs. W. Lenz and N. G. McBride to determine that the drug was at fault. That thalidomide was not approved for sale in the United States was more of a bureaucratic fluke than anything else. If it were to appear today it might be approved for use, since the animals that we use to screen for teratogens are relatively insensitive to its effects. Of the thousands of chemicals that we release into our environment every year. most have not been screened as possible carcinogens, or as possible teratogens or mutagens.

We do know that several very common drugs and environmental agents are potential teratogens. For example, alcohol, even in relatively modest amounts (two or three mixed drinks per day), can cause the fetal alcohol syndrome, characterized by abnormalities of the face and head, growth disturbances and mental deficiency. Smoking can also bring on problems. Children born to mothers who smoke tend to be smaller than their peers and may have abnormal reflexes. Something as ordinary as aspirin may be a teratogen. When more than the recommended dosage is taken, it is associated with an increased risk of cleft palate.

Ionizing radiation (as in X rays) can harm the fetus during the first and last trimesters. In the first trimester, it acts as a teratogen (the exact defect is determined by the time that exposure occurs); in the last, it acts as a mutagen.

Unfortunately, we do not know how most teratogens work. Although thalidomide has been the subject of hundreds of studies, we still do not know exactly how it causes its effects. Furthermore, we do not know the causes of several hundred of the birth defects that have been described in scientific literature. And congenital anomalies vary widely in incidence. Some have occurred only once or twice, while others, such as spina bifida, are relatively common. Based on the best statistics available, in the United States spina bifida appears approximately once in every thousand live births. In some countries. such as Great Britain and South Africa, the incidence is as high as one in 250 to 500 live births.

Birth defects do not respect ethnic, religious or socioeconomic backgrounds. The first and most basic step in research leading to their prevention is accurate knowledge of how often, when and where they occur. Since the existing system for collecting such data, even if it were working perfectly, would still miss two out of every three babies born with defects, it seems wise to modify it. The current method is handled state by state and is essentially voluntary. Although providing inaccurate or incomplete information on a birth certificate is a class C misdemeanor in Texas, it is unlikely that a county or district attorney would actually bring a physician to court over such an issue.

The first step in improving the system is to fund a carefully designed multistate study to determine the level of mis- and/or underreporting. At the same time, federal legislation that would mandate the collection of complete and accurate information should be enacted. This legislation should stipulate realistic penalties for noncompliance by the responsible physician, hospital or county clerk. It should also require enforcement officials to make random. unannounced comparisons of hospital records and birth certificates.

The second step should call for a one-year certificate to be filed for each child on the anniversary of his or her birth. A responsible professional, such as the, primary physician, would be required to complete a form for each child with a congenital anomaly, giving specifics about the defect, the place of birth and other pertinent information. Although we might miss some people, such as those who die or do not visit a physician, we would greatly increase the accuracy and comprehensiveness of our information.

Collecting accurate data about the spatial and temporal distribution of birth defects is the first step in isolating causes and beginning prevention. For example, one could detect localized areas of high incidence and search for teratogens. Possessing accurate data could help the United States to take its place as a leading nation on the World Health Organization’s list of countries with a low incidence of perinatal morbidity and mortality. That is where we should be.