Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Chapter 4 ABORTION AND THE RISK OF DEATH This chapter describes the risk of death faced by a woman obtaining an abortion. It documents that the method of abortion and the trimester in which it is obtained are important determinants of the mortality. There is a comparison of the mortality risk from legal abortion with that from full-term pregnancy and other surgical procedures. Data available on mortality associated with illegal abortion also are presented, along with an explanation for the declining trend in deaths from illegal abortion. United States Annual data on deaths associated with abortion (legal and other) are collected by two agencies within the Department of Health, Education, and Welfareâthe National Center for Health Statistics (NCHS) and the Center for Disease Control (CDC). NCHS has compiled annual data on the reported cause of death from death certificates as part of its vital statistics function; data are now available through 1973. Since 1972, the Center for Disease Control has also collected mortality data, in cooperation with the state public health departments, in an effort to obtain more comprehensive and specific information. Both of these data sources have been used in preparing this chapter. Deaths related to legal abortion are reported in two ways. Deaths "attributed" to abortion exclude deaths assigned to causes other than abortion; deaths "associated" with abortion include deaths resulting from pre-existing complications. Data from NCHS are based on attributed deaths and CDC data are based on associated deaths. Thus, except for data before 1972, the analyses in this chapter are based on deaths associated with legal abortionâthe category that includes the greatest possible number of deaths. Very few deaths are associated with legal abortion in the United States: 20 in 1972 and 22 in 1973.â¢!/ The total number of legal abortions reported to the Center for Disease Control during these two years was approximately 1,202,600, which yields a mortality ratio (the number of deaths per 100,000 abortions) of 1.7 in the first trimester and 12.2 in the second trimester.2/ However, the number of reported deaths is so small that 72
73 a shift of a death from one week to the next, or the addition of another death, may have a marked effect on the size of those ratios. For this reason, comparisons among mortality ratios must be treated with care. As shown in Table 12 the risk of death associated with abortion increases dramatically with the duration of pregnancy, from 0.5 per 100,000 abortions obtained at eight weeks or less, to 16 deaths per 100,000 abortions obtained at 16 weeks or later. When examined by method of termination, the combined mortality ratio for suction and D&C during 1972-1973 was 1.6 deaths per 100,000 legal abortions, and for saline, 15.4 deaths per 100,000 abortions.3/ The former ratio is substantially lower than, and the latter about equal to, the maternal mortality ratio from complications of pregnancy and childbirth (excluding abortion) which in 1973 was 14.1 deaths per 100,000 live births.4/ The mortality ratio for hysterotomies and hysterectomies combined was 61.3 per 100,000 such procedures, four times as great as the saline ratio. However, hysterotomy is infrequently used, is usually performed on high-risk women (women with pre-existing complications, who are older and of high parity), and is used only when the other methods cannot be applied. It also is performed when a saline abortion fails to expel the fetus. Thus, higher mortality ratios would be expected with hysterotomy.
74 Table 12 Reported Deaths Associated with Legal Abortion in the United States, by weeks of Gestation and Method of Abortion, 1972 and 1973 Number of Abortions Total 1,202,563 Number of Deaths 42 Mortality Ratio (Deaths per 100,000 abor- tions)^ 3.5 Weeks of Gestation a/ 8 or less 421,896 9-10 361,885 11-12 212,981 13-15 87,573 16 or more 118,228 2 6 9 6 19 0.5 1.7 4.2 6.9 16.1 Method of Abortion Suction/D&C Saline Hysterotomy/hys- terectomy Other 1,065,338 123,684 8,161 5,380 17 19 5 1.6 15.4 61.3 18.6 a/ Distribution of abortions based on gestation of pregnancy known for 449,709 abortions reported during 1972 (77 percent of the total for that year) and for 453,535 abortions reported for 1973 (74 percent of the total for that year). Source: U.S. Department of Health, Education, and Welfare, Public Health Service, Center for Disease Control. Morbidity and Mortality Weekly Report 24, January 24, 1975.
75 FIGURE 8 Abortion Mortality Ratios by Weeks of Gestation, United States, Combined 1972-1973 Data WEEKS OF GESTATION Source: Table 12. New York State The non-restrictive abortion statute that went into effect in New York on July 1, 1970, combined with a comprehensive reporting system, has made New York state a valuable source for data on the health effects of legal abortion. The Supreme Court's decision in January, 1973, was guided, in part, by the low number of abortion-related deaths from legal abortion reported from New York after the enactment of its non- restrictive laws. In New York state (including New York City) from July 1, 1970 to June 30, 1972, 446,052 abortions were reported. Forty percent were obtained by state residents and 60 percent by non-residents, with the majority of non-resident abortions being performed in New York City.
76 During this two-year period 29 deaths were reported,* including four which occurred in other states. Ten of these deaths followed abortions performed in the first trimester** (2.9 deaths per 100,000 abortions) and 19 deaths followed abortions performed in the second trimester (21.1 deaths per 100,000 abortions).5/ International Data A comparison of recent abortion mortality ratios from selected foreign countries with the 1972-1973 abortion mortality ratios of the United States provides further evidence of the differential between first- and second- trimester mortality ratios. Although only 25 abortion deaths were reported during 1970-1971, England and Wales experienced higher abortion mortality ratios than other developed nations, with a first-trimester ratio of 10 deaths per 100,000 abortions and a second-trimester ratio of 31 deaths. During 1972-1973 the first-trimester mortality ratio declined to 2.7 deaths per 100,000 abortionsâa ratio that is comparable with data from the United States. The major reasons for the much higher mortality ratios from England and Wales in the 1970-1971 period are as follows: 1) A higher proportion of second-trimester abortions were performed in England because of bed shortages and procedural delays arising from the need to have authorization from two physicians; 2) First-trimester methods (suction and D&C) were often used to terminate second-trimester pregnancies, which sometimes resulted in retained fetal and placental tissue, leading to infection and hemorrhage; 3) More pregnancies in England were terminated by hysterotomy (with its attendant high mortality ratio) than in the United States; *The quality of the mortality data associated with these abortions is considered to be good because non-resident as well as resident deaths were accurately identified in this study. A questionnaire was sent to all gynecologists in the United States to ascertain the number of abortion- related deaths of women who had received abortions in New York but who lived in other states. Physicians reporting deaths from abortions per- formed in New York were contacted personally to verify the death and eliminate duplication, and clinical and/or autopsy records were reviewed in all cases. Because most deaths in this intensive survey were reported by several physicians, the investigators concluded that all deaths from abortions performed in New York were probably identified. There is no way to verify whether, in fact, this assumption is correct and all abortion deaths were reported by all physicians. **-Includes one suicide of a young woman who was falsely diagnosed as pregnant and killed herself before that information could be relayed to her after the surgical procedure.
77 4) More sterilization was performed concurrently with the abortion procedure in England than in the United States; and 5) Most first-trimester methods were performed with general anesthesia, which increases the risks.bj Sweden and Denmark have also experienced relatively high abortion mortality ratios until recent years, (Table 13)* reflecting to a large extent the procedural requirements which have tended to delay many abortions until the second trimester where the risk of medical complications is much greater.** In Denmark, the 1966-1970 abortion mortality ratio was six for all trimesters, and in Sweden, seven deaths were reported during 1964-1971, for an abortion mortality ratio of eight deaths per 100,000 abortions.^/ As the procedural requirements have been relaxed in Sweden and the number of abortions has increased, the percentage of abortions obtained in the first trimester has also increased, rising from nearly 10 percent in 1964 to 80 percent in 1973.8/ Reflecting this change, the abortion mortality ratio has declined significantly; for the period 1954-1963 it was 60, decreasing to eight in the period 1964-1971. Factors other than trimester that also have affected abortion mortality ratios include differences in the type of patient, improvements in the techniques of abortion, and greater skill and experience of physicians.*** The lowest mortality ratios associated with legal abortion are found in Czechoslovakia and Hungary. Czechoslovakia reported only 18 deaths out of almost 800,000 legal abortions from 1962 to 1970, and Hungary reported only 17 deaths from approximately 1.5 million abortions *Because the mortality ratios in this table are not itemized by trimester, the significant difference in mortality risk between the first and second trimesters is not shown for individual countries. However, since the major- ity of abortions performed in Denmark and Sweden before the mid 1960s were in the second trimester, and those in Czechoslovakia and Hungary during all the periods shown in the table were first-trimester abortions, a compari- son among these countries permits a crude approximation of this differential. **Denmark legislated abortion on request during the first trimester in 1973; similar legislation became operative in Sweden in January, 1975. ***Data reported by Tietze and Dawson, and by Moore-Cavar help explain the secular decline in abortion mortality ratios. From 1963 to 1972 the abortion mortality ratio in the U.S. declined from 72 to three. Since only therapeutic abortions (usually associated with high risk patients) could be performed until the late 1960s, the abortion mortality ratios were correspondingly high. When non-restrictive laws were passed, allowing more first-trimester abortions to be performed, the proportion of low-risk to high-risk abortions increased and the maternal mortality ratios declined. The JPSA study on early medical complications of legal abortion suggest that increased experience on the physician's part in performing abortions also contributed to the decline.
78 TABLE 13 Historical Trends on the Number of Legal Abortions, Number of Deaths, and Mortality per 100,000 Abortions, Selected Areas and Periods Mortality Ratio (Deaths per Legal Abortions Number of deaths 100,000 abortions) Denmark 1956-1960 19,900 18 90.5 1961-1965 21,800 9 41.3 1966-1970 34,700 2 5.8 Sweden 1949-1953 28,000 27 96.4 1954-1963 35,200 21 59.7 1964-1971 87,900 7 8.0 Czechoslovakia 1957-1961 330,400 19 5.8 1962-1966 400,900 9 2.2 1967-1970 398,600 9 2.3 Hungary 1957-1969 421,400 23 5.5 1960-1963 669,700 21 3.1 1964-1967 739,000 9 1.2 1968-1970 787,600 8 1.0 United States 1963-1968 9,700 a/ 7 72.2 1972 586,800 20 3.4 1973 615,803 22 3.6 a/ Hospitals participating in PAS (Professional Activity Survey). Source: Christopher Tietze and Deborah Dawson. "Induced Abortion: A Fact- book," Reports on Population/Family Planning, New York: The Popu- lation Counci1, December 1973, pp. 45-46, and U.S. Department of Health, Education, and Welfare, Public Health Service, Center for Disease Control. Morbidity and Mortality Weekly Report 24, January 24, 1975.
79 during 1964-1970.9/ Although some skepticism could be expressed at the low level of the mortality ratios in these countriesâ2.3 and 1.0 respectivelyâ part of the explanation for these figures is the fact that more than 99 percent of the legal abortions in these countries are performed during the first 12 weeks of pregnancy; second-trimester abortions are permitted only when the mother's life is in danger.!£/ The recent experience of both the United States and England and Wales with first-trimester abortions serves as a partial confirmation of the low figures reported from Eastern Europe; the 1972-1973 first-trimester mortality ratio for the U.S. was 1.7 deaths per 100,000 abortions and for England and Wales, 2.7 deaths per 100,000 abortions. A Comparison of the Mortality from Abortion, Full-Term Pregnancies, and Other Surgical Procedures The above data indicate that the risk of death for a woman obtaining a legal abortion under nonrestrictive legislation is very low on an absolute scale. Also, legal abortion in the first trimester is far less dangerous to a woman's life than is carrying a pregnancy to term or undergoing most common surgical procedures. During 1973, the number of deaths in the United States from complications of pregnancy and childbirth (excluding abortion) was 441; the mortality ratio per 100,000 live births was 14.1.11/ Table 14 compares the mortality risks of legal abortion with those of eight other commonly performed operations in the United States. A first- trimester legal abortion carries about the same risks as a tonsillectomy with or without adenoidectomy, both of which are less risky than the other surgical procedures listed. Although these comparisons are somewhat crude, inasmuch as they do not take into account the physician's skill and experience, the equipment used, or the condition of the patient at the time of the operation (abortion is usually an elective procedure and the others are performed for therapeutic reasons), they do serve to illustrate the very minor risk of death from legal abortion in contrast with other operations.
80 TABLE 14 Mortality Ratios of Selected Surgical Procedures United States, 1969 Operation Mortality Ratio (Number of Deaths per 100,000 procedures) Legal Abortion First trimester 1.7a/ Second trimester 12.2a/ Tonsillectomy without Adenoidectomy 3 Tonsillectomy with Adenoidectomy 5 Ligation and Division of Fallopian Tubes 5 Partial Mastectomy 74 Cesarean Section (low cervical) 111 Abdominal Hysterectomy (not abortion) 204 Appendectomy 352 a/Based on 1972-1973 data from the Center for Disease Control. Source: Charles G. Child III. "Surgical Intervention", Life and Death and Medicine, San Francisco: W. H. Freeman and Company, 1973, p. 65. This book originally appeared as the September 1973 edition of Scientific American. Illegal Abortion It is difficult to find credible estimates of the number of deaths associated with illegal abortion. One estimate, which has been frequently quoted, is between 5,000 and 10,000 deaths per year.12/ That is hardly plausible, considering that the total number of deaths of women aged 15-44 from all causes in the United States is approximately 50,000 annually, and the total number of deaths due to abortion reported by the National Center for Health Statistics (NCHS) has been below 500 since 1958 and below 100 since 1971.
81 The NCHS data reflect under-reporting, and represent a minimum estimate of mortality due to all types of abortionâspontaneous, lega1, and illegal. Figure 9 compares the number of deaths due to legal abortions with those due to "other" abortions over the past 18 years in the United States. While the "other" category includes miscarriages and abortions of undetermined origin, various observers conclude that most of the deaths included in the "other" category are from illegally induced abortions.* For example, of the 47 deaths reported in 1973, 22 were from legal abortions, 16 were from illegal abortions, seven were from spontaneous abortions, and the remaining two deaths were from unknown causes.13/ The total number of reported deaths due to abortion has steadily declined from 324 in 1961 to 47 in 1973. Contributing to this decline were the advances in medical treatment of gynecological complications of abortion. These advances also affect non-abortion statistics: maternal mortality in general has dropped from 30 deaths per 100,000 live births in 1958-1962 to 19 deaths per 100,000 live births in 1973, (standardized for age and delivery) a reduction of 37 percent over a 15-year period.14y Another factor could be a reduction in the number of unwanted pregnancies in the United States because of more effective use of contraception.!^/ A national decline in deaths due to "other" abortions accelerated as women began to shift to legal abortions after the changes in abortion laws and practices which occurred around 1970. New York City Figures from New York City also indicate a decline in "other" abortion deaths, most sharply between 1970 and 1972 following the implementation of non-restrictive abortion practices in New York in July 1970. These data are shown in Table 15. As the table indicates, the performance of legal abortions has caused some maternal deaths, but the legalization of abortion itself has led to a very sharp decline in the total number of deaths related to pregnancy termination. *See, for example, Donald P. Swartz, "The Harlem Hospital Center Experience," The Abortion Experience, Howard J. Osofsky and Joy D. Osofsky, eds., Hagerstown, Md: Harper and Row, 1973; and Emily Moore-Cavar. International Inventory of Information on Induced Abortion, New York: International Institute for the Study of Human Reproduction, Columbia University, 1974.
82 FIGURE 9 Abortion Deaths By Type of Abortion 1958-1973 350 300 250 200 w S 150 100 Deaths Attributed to All Types of Abortions Deaths Associated with Legal Abortion 1960 1965 1970 1973 Source: U.S. Department of Health, Education, and Welfare. Vital Statistics of the United States, Mortality, Part A.; 1972 and 1973 data are from U.S. Department of Health, Education, and Welfare, Public Health Service, Center for Disease Control. Morbidity and Mortality Weekly Report 24: January 24, 1975.
83 Period 1966-1968 1968-1970 1970-1972 1972-1974 TABLE 15 Abortion-Related Deaths in New York City, Resident Women, July 1966 - June 1974 Legal Other 11 6 46 45 14 5 Total 46 45 25 11 Source: Christopher Tietze, Jean Pakter, and Gary S. Berger. "Mortality with Legal Abortion in New York City, 1970-1972," Journal of the American Medical Association 225: 507-509, July 30, 1973. Jean Pakter, Donna O'Hare, M. Halpern, and Frieda Nelson. "Impact of the Liberalized Abortion Law in New York City on Deaths Associated with Pregnancy: A Two-Year Experience," Bulletin of the New York Academy of Medicine 49: 804-818, September 1973. Supplemented by unpublished data obtained from the New York City Department of Health. Romania Data from Romania show that when non-restrictive abortion laws are tightened, deaths associated with illegal abortion increase. Abortion on request was legal in Romania from 1957 to October 1966, when restrictive legislation was adopted.^6_/ After this change there was a substantial increase in the number of deaths associated with abortion (Table 16), most of which after 1966 are attributed to illegal abortions; nearly all abortions were illegal after late 1966. The number of abortion-related deaths in Romania during 1971 was greater than the 324 abortion deaths reported in the United States during 1961, notwithstanding that the U.S. population in 1961 was nine times greater than the Romanian population of 20,000,000 in 1971.
84 TABLE 16 Deaths Attributed to Illegal Abortion in Romania 1965-1971 Year Number of Deaths 1965 64 1966 83 1967 170 1968 221 1969 258 1970 314 1971 364 Source: World Health Organization. Statistics Annual, Selected years, 1966-1971. Mortality Risks by Race Differences in the mortality risks by race for other-than-legal abortions can be seen in Table 17. Although abortion mortality ratios cannot be defined (there is no reliable estimate of the number of spontaneous and illegal abortions), abortion mortality ratesâ number of deaths per 100,000 women aged 15-44 yearsâcan be computed. It should be noted that these figures are based only on the illegal abortion deaths reported to the National Center for Health Statistics, which are an under-estimation of the actual number of deaths caused by illegal abortions. Rates were not computed for deaths associated with legal abor- tion because the absolute number of deaths was so small. TABLE 17 Abortion Mortality Rates Associated with Other-than-Legal Abortions, by Race, United States, 1968-1973 -(Deaths per 100,000 women aged 15-44) Year All Women White Women Other than White Women 1968 0.30 0.16 1.29 1969 0.31 0.18 1.23 1970 0.28 0.14 1.21 1971 0.18 0.07 0.84 1972 0.14 0.09 0.40 1973 0.06 0.03 0.29 Source: U.S. Department of Health, Education, and Welfare. National Center for Health Statistics. Vital Statistics of the United States, Morbidity Part A.; and U.S. Department of Commerce, Bureau of the Census. Estimates of the Population of the United States, by Age, Sex and Race: April 1, 1960 to July 1, 1973. Series P-25, No. 519, April 1974.
85 The mortality rate associated with illegal abortion declined about 80 percent for white women and about 78 percent for non-whites from 1968- 1973. For non-whites, however, there has been a larger absolute decline in the rate, which may represent a greater shift of non-white women away from illegal abortions to the medically safer abortions provided in clinics and hospitals, particularly after 1970. Nevertheless, non-white women still face a much higher risk of death from other-than- legal abortion. Summary Two alternatives are available for a woman who is pregnant and wishes to terminate her pregnancyâlegal abortion or illegal abortion. As shown by the preceding data, legal abortion in the first trimester carries a relatively small risk of death. In the United States and other countries that have nonrestrictive abortion statutes, the risk of death associated with legal abortion in the first trimester is markedly lower than that associated with carrying a pregnancy to term, and much lower than the risk associated with frequently performed surgical procedures. Although data limitations preclude a direct comparison of the mortality ratios associated with legal and illegal abortion, available statistics from the United States indicate that the number of reported deaths from other-than-legal abortions declined steadily as less restrictive abortion legislation was passed and implemented throughout the country. The major factors that affect legal abortion mortality ratios are the trimester in which the abortion is performed and the method of abortion. The risk of death associated with first-trimester abortions performed by suction and D&C is low compared with that associated with saline abortions or carrying a pregnancy to term; the risk associated with hysterectomy or hysterotomy is substantially higher than any of the other methods, although still lower than the risk associated with surgical procedures such as cesarean delivery, for which the 1969 mortality ratio was 111.
86 REFERENCES 1. U.S. Department of Health, Education, and Welfare, Public Health Service, Center for Disease Control. Morbidity and Mortality Weekly Report 24, January 24, 1975. 2. Ibid. 3. U.S. Department of Health, Education, and Welfare, Public Health Service, National Center for Health Statistics. "Summary Report, Final Mortality Statistics, 1973," Monthly Vital Statistics Report 23: No. 11 (Supplement 2) Rockville, Maryland, February 10, 1975; National Center for Health Statistics. "Summary Report, Final Natality Statistics, 1973," Monthly Vital Statistics Report 23: No. 11 (Supplement) January 30, 1975. 4. Gary S. Berger, Christopher Tietze, Jean Pakter, and Selig H. Katz. "Maternal Mortality Associated with Legal Abortion in New York State: July 1, 1970-June 30, 1972," Obstetrics and Gynecology 43: 318, March 1974. 5. Christopher Tietze and Deborah A. Dawson. "Induced Abortion: A Factbook," Reports on Population/Family Planning, New York: The Population Counci1, 45, December 1973. 6. Ibid., pp. 44-45 7. Ibid. 8. Ibid., pp. 30-31 9. Ibid., p. 45. 10. Ibid., pp. 30-31 11. See sources listed in footnote 3 above. 12. Jerome M. Kummer, ed. Abortion: Legal and Illegal. A Dialogue Between Attorneys and Psychiatrists. Santa Monica: Jerome M. Kummer, M.D., 1967, 13. Morbidity and Mortality Weekly Report, January 24, 1975. 14. U.S. Department of Health, Education, and Welfare, Public Health Service, National Center for Health Statistics. Vital Statistics of the United States. Annuals for the years 1968-1972. Rockville, Maryland.
⢠87 15. Norman B. Ryder and Charles F. Westoff. "Wanted and Unwanted Fertility in the United States: 1965 and 1970," Charles F. Westoff and Robert Parke, Jr., eds. Demographic and Social Aspects of Population Growth, Washington, D.C.: Commission on Population Growth and the American Future, 1972; and Norman B. Ryder. "Contraceptive Failure in the U.S. Family Planning Perspectives 5: 133-142, Summer 1973. 16. Nicholas H. Wright. "Restricting Legal Abortion: Some Maternal and Child Health Effects in Romania," American Journal of Obstetrics and Gynecology 121: 246, January 15, 1975.
