The Statistical Background
INEVITABLY THE statistical information about the numbers of abortions, whether legal or illegal, is very sketchy. For legal abortions there is no central federal, state, or city registry to turn to for the information, which is buried in hospital charts of individual patients throughout the country. As for illegal abortions, only the patient and the abortionist know of the event. Surveys asking a truly representative cross-section of American women whether or not they have had an abortion have not been performed, and are not likely to yield accurate data. As a consequence not such information exists.
How then is it possible that we read in the press of the number of legal and illegal abortions performed yearly in the United States? Such figures as do exist are derived solely from questioning a sampling of women by asking them how many live births they have experienced, and how many abortions they have had.
For example, in one study, covering mainly the years 1957-1962, and concerned strictly with legal, or therapeutic, abortions it was shown that the women questioned had 1,039 therapeutic abortions as against 522,600 babies. This is a ratio of about 2 therapeutic abortions for every 1,000 deliveries. Based on this ratio, and extending it to about four million yearly deliveries in the United States, the ratio would tell us that there are about eight thousand therapeutic abortions performed in American hospitals every year.
One of the participants at the Conference pointed to the lack of statistics in this field and called for legislation establishing compulsory reporting any induced abortion to the health department.
As far as illegal abortions are concerned, obtaining accurate data is even more difficult. Is the information obtained accurate? If the information is accurate, is the sampling of the women who gave it truly representative of the American population as a whole? Who is willing to speak of the abortions she has had?
In addition it is important to know where the starting point of the study is. If one questions a thousand women who have just delivered a baby in a hospital, the number of illegal abortions they will have experienced is likely to be too low. After all, they have just demonstrated that they were willing to have a pregnancy go to term by not aborting the infant they have just delivered. If on the other hand one starts with a thousand who have just had a "miscarriage" the figure obtained is likely to be too high, since among these alleged "miscarriages" there will have been a number of pregnancies which were in fact interfered with. In brief, no study of a scientific validity even approaching a Gallup poll exists on illegal abortions.
However, the difficulties involved have not deterred some from at least attempting to study the subject. One study, coming from the five New York boroughs, showed 40 abortions per 1,000 deliveries. This would yield 160,000 illegal abortions a year for the United States as a whole. A second and much quoted study covered the experience of ten thousand women who attended the Margaret Sanger birth-control clinic in New York in the late 1920's. On the basis of their experience, the United States would have 1,200,000 abortions a year today. The problem is no different than with other studies: How representative of the United States as a whole were women who attended the Margaret Sanger birth-control clinic in New York in the late 1920's? Then, in the early 1950's, Alfred Kinsey and his associates reported on the sexual activities of American women. In the sample of women whom they questioned, 22 percent had had at least one, and on the average two, abortions by the age of forty-five. Such an incidence of illegal abortion would yield a figure of at least 600,000 illegal abortions per year. Again, however, the group of women questioned was largely white, urban, college-educated, and had a higher incidence of divorce and separation, and fewer children, than the population of American women as a whole.
Another study done in Indianapolis showed that 2.6 percent American women in the study had had an abortion. This study was not more representative of the nation than that of Dr. Kinsey in which 22 percent of women had had illegal abortions. In general none of the above studies provide data describing the practices of four major groups in our society: rural people, Negroes, the Spanish-speaking, and Roman Catholics. What little data are available on Negroes have tended to show that they have fewer abortions than white women.
It is not surprising then that a committee, appointed in 1955, reached the conclusion that there might be in the United States anywhere from 200,000 to 1,200,000 abortions per year -an extremely wide variation indeed. The report of this com- mittee was tempered by the statement: "The figures on abor tion frequently used throughout the conference by various individuals were based on personal estimates by the individuals themselves. No way has yet been found of obtaining reliable statistics that would give an exact figure for the total population."
Since no additional meaningful knowledge has been added since that report, the medical panel at the International Con- ference on Abortion could add no conclusions beyond those given in 1955. Whether the development, since 1950, of contraceptive methods such as the pill and intrauterine devices has decreased the number of abortions is not known. The general assumption has been made that the greater the contra- ceptive knowledge, and the greater the effectiveness of the contraceptive, the smaller the need for abortion will be; but here again no evidence has been adduced to substantiate the assumption.
Based on the same studies, the Conference participants did make some comments on who seeks an abortion. They pointed out that in the studies cited four out of five abortions are performed on married women who already have children, but who Iwant no more. This is not to say that the married pregnant woman is more likely to undergo an abortion than the non- triarried pregnant woman. Indeed, the opposite is true. But since there are many more married than single pregnant women in the United States, four out of five abortions are performed on married women.
If the number of abortions performed is difficult to come by, the number of deaths resulting from abortion is easier to obtain. As one conference participant, Dr. Christopher Tietac, pointed out, those who die from abortion do so mainly as a result of hemorrhage or infection. In either circumstance they are likely to be seen in a hospital, where the condition will be diagnosed. Those who die on arrival in hospitals, or shortly thereafter, usually undergo autopsies.
In summarizing the discussions of the medical panel, Dr. Andre' Hellegers, Professor of Obstetrics and Gynecology at Georgetown University, reported that in 1964 there were a total of 247 known deaths from abortion in the United States. In 1965 there were; 235. These figures include death from' spontaneous miscarriages, legal therapeutic abortions, and illegal abortions. The vast majority, however, occur as a result of illegal abortion. The question may be raised how many women annually die from abortion with the fact going unrecognized or unreported. Obviously, an accurate assessment of this is impossible.
These statistical inadequacies emphasize the extreme care with which all available figures should be used. But despite these problems, all the doctors at the International Conference on Abortion reached a consensus that a total of 500 abortion deaths per year would be a reasonable figure based on the current data.
The table on page 47 gives the latest available figures on abortion deaths for every state in the union (1965). The data come from Vital Statistics of the United States, Volume 2, Part B, and divide the deaths from abortions by race. If the table tells the statistical story for the population as a whole, it still does not fully describe who gets the abortions and who dies from them. The medical panel agreed that there were wide variations in practice from hospital to hospital and from state to state. With the presence of panelists from abroad, it became clear that practices there also differ from those in the United States. It was generally agreed that in the United States it is the wealthiest who are most likely to get the safest abortion, and that this abortion, if performed in a hospital, is most. likely 1to be done for alleged psychiatric reasons. Data from New York City, where all fetal deaths have to be reported, show that in proprietary hospitals 3.9 abortions were performed for every 1,000 deliveries. On the private-patient service of non- profit hospitals the comparable figure was 2.4. It was 0.7 on the ward services of nonprofit hospitals, while in municipal hospitals it was only 0.1. This difference in practice is further seen in the fact that white women had 2.6 abortions per 1,000 deliveries, Negro women, 0.5, and Puerto Rican women 0.1. One of the panelists, Dr. Andre' Hellegers, has written that this difference is due in part to the fact that the private white patient is more likely to register for prenatal care early in pregnancy, so that abortion is still possible. On the other hand nonwhite or ward-service patients are more likely to seek pre- natal care after the twelfth week when abortion can no longer be performed by simple dilatation and curettage. One study has revealed that private patients register before the twelfth week of pregnancy with seven times greater frequency than ward patients.
There is general agreement, however, that this by no means explains the whole difference. Not only does the white private but there is little doubt that she is more likely to have the patient know where to go, when to go, and whom to go to, funds to obtain a legal abortion when all other circumstances are equal. Municipal hospitals, subject to the scrutiny of the public, are much more likely to be strict in their interpretation of the abortion laws. Some will go to the length of establishing abortion quotas, simply in order to create the impression that it is not easy to get an abortion in the institution. It was repeatedly pointed out at the Conference that many of the eight thousand so-called "therapeutic" or "legal" abortions performed yearly considerably stretch the letter of the law.
The discrepancy between poor and wealthy in legal abortions also exists with respect to illegal abortions. The wealthy are more likely to find a physician who, for a fee, will perform an illegal abortion, and who will do so competently. We have already noted that the poor get fewer abortions than the rich. Yet when it comes to dying from abortions, the discrepancy in the quality of the abortions performed can be seen in death statistics. Thus, between 1960 and 1962, for every 10,000 live births in New York City, 1.0 white women, 4.7 Puerto Rican women, and 8.0 Negro women died from abortion.
Returning for a moment to legal abortions, some clear trends are visible. The annual number of therapeutic abortions has declined considerably in the past few decades. Thus in New York City there were about 700 therapeutic abortions per year in the period 1943-1947, and this declined to about 300 per year in the years 1960-1962. Per 1,000 live births, the ratio thus fell from 5.1 to 1.8. This overall trend conceals a sharp and continuing decline in the traditional medical indications for abortions, i.e., abortions to protect the mother from physically harmful effects of pregnancy. This is attributed to rapid advances in medicine. At the same time there has been a slight increase in the incidence of abortion on psychiatric grounds. The slight increase in psychiatric indications, together with the marked decrease in physical reasons, has led to the proportion of therapeutic abortions performed for psychiatric indications increasing approximately fivefold.
Taking a closer look at the data in the table of page 47, several things are obvious. First, the data clearly show the discrepancy in the death rate of white and nonwhite women. Although there are many times more white than nonwhite women in the United States there were 129 deaths in the non- white group compared to 106 among the white women, in 1965. Secondly, it is obvious that there are many more deaths from abortion in the highly urbanized than in the rural areas. This. underscores the question, already raised, whether statistics on the incidence of abortion in New York and other urban areas are indeed applicable to rural states. Thus it will be seen that New York, California, Michigan, and Texas together account for half the abortion deaths in the United States. Twenty- three less urban states, with a combined population as large as the four previously mentioned states, accounted for only ten deaths. Indeed, eighteen states reported no abortion deaths at all!
It should not be thought that this urban-rural difference only exists in the United States. One Conference participant from abroad, Professor Jan Otto-Ottosson of Sweden, wrote: "It is known that women in big cities and unmarried and deserted women resort to illegal abortions more frequently than others." In another passage he wrote: "Among other social data it may be mentioned that the frequency of abortion is much higher in large towns than in rural areas. Thus, in 1960, Stockholm had an abortion frequency of 12.6 per 10,000 women, while Daiarna, with the lowest in the country, had 1.7. This is a difference in ratios of more than 7 to 1." Similar observations have been reported from other countries.
In summary, some abortion facts are obvious, while on others we are almost completely in the dark. On one thing there was uniform agreement among the panelists: There is an urgent need for extensive studies to get further facts on abortion. Without such facts it is difficult and risky, if not impossible, to reach sound decisions on legal or other public-policy questions.
Chapter 5- The Law in the United States and Elsewhere