No drug tragedy is more appalling than babies handicapped by their mothers' cocaine abuse in pregnancy. The figure of 375,000 "crack babies" has been widely reported in the media and recited by drug warriors such as William Bennett to warn of the perils of legalized drugs. Columnists Jack Anderson and A.M. Rosenthal speak of 300,000- 400,000 "addicted" or "crack-addicted" babies.

These are mind-boggling numbers: one in 10 births. They are also incredible, given that the 1988 National Household Survey on Drug Abuse found that the number of women who used crack in the year was just 474,000, or less than 1 percent of the childbearing population, while just one million used cocaine of any kind on a monthly basis! No doubt the survey, conducted annually by the National Institute on Drug Abuse, underreports drug use. But there are also good grounds to investigate the source of the supposed "375,000 crack babies."

The source is a 1988 survey of 36 hospitals by Dr. Ira Chasnoff, director and founder of the National Association for Perinatal Addiction Research and Education. Dr. Chasnoff makes it clear that 375,000 is not a count of crack-addicted babies, but rather a rough estimate of babies exposed to illicit drugs during pregnancy. The hospitals surveyed included both private and public institutions across the country, but were not a scientific, demographic sample. In fact, a glance at the list shows that it is weighted toward leading metropolitan areas where drug abuse is more prevalent. They reported rates of prenatal exposure to illicit substances ranging from 0.4-27 percent with an average of 11 percent — in round numbers, one in 10 — which works out to about 375,000 nationwide.

The meaning of Chasnoff s estimate is unfortunately obscured by the fact that different hospitals reported different statistics in their surveys. Some surveyed mothers' own self-reports of drug usage; others looked at urine samples of mothers (which are sensitive to cocaine for 2-3 days) or of babies (which are sensitive up to a week). A few hospitals counted marijuana and other illicit drugs along with cocaine, though Dr. Chasnoff maintains that cocaine accounted for most drug exposure. Like most studies, the survey did not distinguish between cocaine and crack, which, according to NIDA, accounts for about one-fifth of cocaine use. Estimates from Associate Professor Richard Barth of the School of Social Work at Berkeley have put the number of crack-exposed babies at around 1-4 percent.

Yet, however many babies are exposed to cocaine, the crucial question remains: how many are actually harmed by it? After all, one might well speak of a million-plus alcohol and tobacco babies. In fact, while crack-exposed babies run a high risk of premature birth and other problems, the U.S. Department of Health and Human Services estimates that over 2 in 3 suffer no obvious problems at birth.

However, there is concern that many may be at risk of more subtle, behavioral problems that are difficult to detect at birth and become evident only in later years. According to findings released last September by Dr. Chasnoff and Dr. Judith Howard of the University of California at Los Angeles, cocaine-exposed infants may suffer deficits in their ability to concentrate, interact socially and cope with unstructured environments at the age of 2-3 years. Such infants have been widely publicized in the media as "crack babies," though the nature and severity of their affliction is not entirely clear.

Professor Barth released a study in February that disputes Chasnoffs findings on long-term effects. Barth found that these babies, who were observed from 1-4 years, compete quite well and will live among their peers in a healthy way. Unlike fetal alcohol babies, they are not born mentally retarded and are as bright as normal children. The study involved children in Los Angeles, San Francisco and Oakland.

Chasnoff's observations on the effect of maternal cocaine use on infants have been well documented. He says that physiologically, the most noticeable reproductive hazard of cocaine is a markedly higher risk of preterm delivery, low birthweight and growth retardation, occurring in up to 25-30 percent of exposed infants. He also says cocaine has also been linked to an increased incidence of miscarriage, sudden infant death syndrome, fetal strokes and certain deformities. The severity of these problems is highly variable. Preterm and low-birthweight babies are at greater risk of numerous life-threatening and debilitating complications, but in most cases their growth deficits are reversible with reasonable care. Fetal strokes and other birth defects may cause lasting impairment, but are more rare. Women who stop using cocaine early in pregnancy reduce the risk of fetal harm, though permanent damage may conceivably occur from a single intensive session. Fetal "addiction" to cocaine is not a major issue; rather, the danger is that cocaine may harm the fetus by constricting blood vessels, raising blood pressure, cutting fetal blood flow and interfering with the brain's neurotransmitters.

In addition, of course, some say there is good reason to worry about cocaine's effects on maternal behavior. Dr. Robert ten Benzel, a pediatrician at the University of Minnesota, says there is enough anecdotal evidence that proves some women use crack to induce abortions.

A major difficulty in judging the risk of prenatal cocaine use is that its symptoms can also be caused by other factors that are seldom fully controlled in studies. For example, low-birthweight and preterm babies can be caused by inadequate nutrition, poor prenatal care and polydrug abuse, all of which are more Common in cocaine-abusing mothers. An important, new factor is syphilis, which has become a problem among crack users because some trade sex for drugs, according to Dr. Richard Fulroth, chief of newborn services at Highland Hospital in Oakland, California.

Controlling for other drug use, it has been estimated that women have a four times greater risk of low-birthweight babies if they use cocaine throughout pregnancy, or 1.8 times if they use it less often. The risk factor for smoking throughout pregnancy is three, and for drinking three or more drinks per day 1.7, according to the January issue of American Journal of Public Health. Thus cocaine seems comparable to other, legal drugs as a risk factor in problem pregnancies. Barth's study shows children with low birth weights catch up in overall growth by age two.

As for the mental and behavioral symptoms reported by Chasnoff and Howard, their incidence has not yet been statistically determined and remains controversial. Among the most worrisome evidence is a study published in the November issue of The Journal of Pediatrics that reported abnormal brain encephalograms in 35 percent of infants of cocaine-positive mothers, seven times the normal incidence. However, other investigators have failed to detect this, including Dr. Fulroth.

The horrors of crack baby syndrome have been sensationalized in the press. One researcher was quoted in Newsweek recently saying, "It's as if the part of the brain that makes us human beings capable of discussion or reflection is wiped out."

However, Chasnoff maintains that crack babies develop "within the normal range for cognitive development and are not, as some people have stated, brain damaged," whatever their behavioral differences. Experts note that cocaine-exposed babies show a wide range of outcomes, and cannot be automatically presumed to be defective or irremediable. Most agree that the prognosis for crack babies is hopeful, provided they receive attentive post-natal care and nurturing — which are, of course, apt to be sorely lacking within crack-using families.

Many observers think that the physical dangers of cocaine have been overblown by drug war hysteria. Some privately suggest that pressure for government funding has created an anti-cocaine bias. A study of research papers on prenatal cocaine use by Dr. Gideon Koren found that positive findings were significantly more likely to be accepted for publication than negative ones, even though the negative studies were better controlled and had more subjects. Professor Barth and other experts emphasize the many adverse economic, behavioral and cultural factors that put cocaine babies at risk.

Crack's greatest danger to most babies may come not from pharmacological harm, but rather from the lack of prenatal and postnatal care.

In Oakland's Mandela House, which provides addict mothers with attentive prenatal and postnatal care, Dr. Fulroth reports no problem births among its residents. Fulroth calls cocaine a "red flag" of broader problems. He says, "cocaine in my opinion is a symptom of what's happening to our society when you take away opportunity for education, housing [and] health care...."

Just how widespread is the problem of crack babies? A recent Household Survey of welfare agencies reported 8,974 crack-exposed babies in eight cities: New York, Los Angeles, Chicago, Miami, Phoenix, San Francisco, Tacoma and Fort Wayne. In his first National Drug Control Strategy, President Bush spoke of 100,000 crack babies — instead of 375,000 — making a somewhat loose extrapolation from eight cities to the entire nation and equating crack-exposed babies with crack babies.

A gauge of the HHS survey may be found in epidemiological birth statistics since crack was introduced in 1985. These show a true epidemic at some inner-city hospitals. At Highland Hospital, the rate of low-birthweight deliveries soared from 7 percent in 1985 to 11.2 percent in 1989. In Los Angeles from 1987-88, infant mortality among blacks jumped from 16.3-21.1 per 1000, and in Washington, D.C., from 21.4-25.1 — a worse rate than in some Third World countries. Looking at all races, Los Angeles County reported a 17 percent increase in low-birthweight infants from 5.3 percent-6.2 percent between 1988 and 89, which translates to an excess of 36,000 if projected nationally. This trend has been blamed on a combination of factors, including cocaine, syphilis and poor prenatal care. Nationwide, the most recent data from the Centers for Disease Control show a mere 2.2 percent increase in the rate of low-birth weight babies from 1985-87, corresponding to an excess of 6,000 births per year. The CDC data show no noticeable increase in other birth defects.

In sum, babies harmed by crack exposure may well represent the single greatest public health cost of cocaine, but the casualties are much less than the 375,000 commonly claimed in the media. In terms of fetal damage, cocaine may be comparable to alcohol, which claims some 4,000-12,000 cases of fetal alcohol syndrome per year, according to the National Institute on Alcohol Abuse and Alcoholism in 1987. On the other hand, as a cause of low birthweight, cocaine seems less troublesome than tobacco, which is somewhat less toxic but more widely used. Unfortunately, the pharmacological dangers of cocaine are greatly compounded by prenatal and postnatal neglect. This is plenty enough to be concerned about, without exaggerating the problem with spurious claims to defend the bankrupt, punitive policies that have brought it about.

Dale Gieringer, "How Many Crack Babies?" The Drug Policy Letter, March/April 1990, p. 4.