Abortion in the United States: A Statistical Study

WHO COLLECTS ABORTION DATA?

There are two major sources for abortion statistics. The Centers for Disease Control and Prevention (CDC) of the U.S. Department of Health and Human Services collects abortion statistics for the U.S. government. The Alan Guttmacher Institute (AGI), a private organization that studies reproductive health issues and strongly supports the position that abortion is an acceptable option, conducts periodic surveys of abortions performed in the United States and throughout the world.

The CDC compiles abortion information collected by state health departments, hospitals, and other medical facilities. This data comes from fifty-two reporting areas—the fifty states, the District of Columbia, and New York City. The AGI, in contrast, directly contacts all known abortion providers for its periodic surveys and follows up its inquiries by letter and telephone. Thus, the AGI data is considered to be the most accurate data available.

The total number of abortions reported to the CDC by the individual states generally is lower than that collected by the AGI. The CDC believes that the number of abortions performed in physicians' offices probably is underreported more often than are those done in hospitals and other medical facilities. Because most abortions in physicians' offices usually are performed in the early stages of pregnancy, the CDC's early-abortion counts are very likely less than the actual numbers.

HOW MANY ABORTIONS?

CDC Data

The latest CDC survey data of legal induced abortions are compiled in "Abortion Surveillance—United States, 2001" (Morbidity and Mortality Weekly Report, Surveillance Summaries, vol. 53, no. SS-09, November 26, 2004). There has been a consistent decrease in the number of abortions per year since 1996. For 2001, 853,485 abortions were reported. It is difficult to compare this number with years prior to 1998, because the more recent data lacks information from certain states, as noted. (See Figure 4.1 and Table 4.1.)

The CDC began its abortion surveillance in 1969, two years after Colorado became the first state to liberalize its abortion statute. From 1970 through 1982 the reported number of abortions increased each year, with the largest percentage of increase occurring between 1970 and 1971. From 1976 through 1982 the annual increase slowed and then dropped slightly in 1983. From 1983 through 1990 the number of abortions increased again, with year-to-year fluctuations of 5% or less. The annual number of abortions has decreased since 1990 (the year in which the number of abortions was the highest reported by the CDC). (See Figure 4.1 and Table 4.1.)

The abortion ratio is the number of legal abortions for every one thousand live births in a given year. In 2001 the abortion ratio was 246 legal abortions per one thousand live births, or about one abortion for every four babies born alive. (See Table 4.1.)

The abortion ratio increased steadily from 1970 through 1980 and then remained somewhat stable for most of the next decade. The CDC reported the highest ratio (364 per one thousand live births) in 1984. Since 1987 the abortion ratio has declined steadily, appearing to increase slightly in 1996. However, the source of birth data changed that year. The ratio for 2000 was the lowest recorded since 1974. (See Figure 4.1 and Table 4.1.)

The abortion rate refers to the number of abortions performed per one thousand women ages fifteen to forty-four years (the primary childbearing years). The abortion rate rose from five abortions per one thousand women in 1970 to twenty-five per one thousand in 1980. From 1981 through 1993 the rate remained stable at twenty-three to twenty-four abortions per one thousand women. The abortion rate declined to twenty-one in 1994 and to twenty in 1995. From 1995 to 1997 the rate remained stable at around twenty abortions performed per one thousand women. Since then the rate further declined to seventeen in 1998 and 1999, and then to sixteen in 2000 and 2001. (See Figure 4.1 and Table 4.1.)

The CDC suggests that the overall decreasing abortion rates may be the result of several factors:

• A shift in the age distribution of women of childbearing age toward the older and, subsequently, less fertile ages
• A decreased number of unplanned pregnancies
• A reduced access to abortion services
• Changes in birth control practices, including an increased use of contraception, particularly in the use of long-acting hormonal methods and condoms by young women

Alan Guttmacher Institute Data

The AGI started collecting data on abortion in 1973, the year of the Roe v. Wade ruling. Because the AGI directly surveys abortion providers, its abortion counts generally have been higher than those reported by the CDC. For instance, the total number of abortions reported to the CDC in 1997 was 12% lower than the number AGI reported. According to the AGI, the abortion rate for women ages fifteen to forty-four was 20.9 in 2002 and 21.1 in 2001. (See Figure 4.2.) The CDC's most recent data, from 2001, calculated the rate at sixteen abortions per one thousand women in the same age group. (See Table 4.1.)

CHARACTERISTICS OF WOMEN SEEKING ABORTION

According to the CDC's "Abortion Surveillance—United States, 2001," slightly more than half (51.5%) of the women who had abortions in 2001 were younger than twenty-five years of age. Most were white (55.4%), unmarried (81.6%), and in the first two months of pregnancy (59.1%). (See Table 4.2.)

Age

Women in the twenty- to twenty-four-year-old age group obtained the greatest proportion of abortions in 2001—one-third (33.6%) of the total. Nearly one-quarter (22.8%) were obtained by those ages twenty-five to twenty-nine, and slightly more than one-sixth (17.0%) were obtained by women ages fifteen to nineteen years. Less than 1% were obtained by women younger than fifteen years. Women ages thirty to thirty-four had 14.7% of abortions in 2001, and women thirty-five years and older had 11.3%. (See Table 4.3.)

In 2001 the majority of abortions (nearly 60%) were early abortions—those obtained during the first eight weeks of pregnancy. The percentage of early abortions had grown rather steadily over the previous twenty-eight years, from slightly over 36% in 1973. (See Table 4.2.) As Figure 4.3 shows, the percentage of women who obtained an early abortion in 2001 increased with age. Approximately 66% to 67% of women ages thirty and over obtained early abortions.

In 2001 although less than 1% of all abortions were obtained by teenagers younger than fifteen years of age (see Table 4.3), their abortion ratio (number of abortions per one thousand live births) was the highest. (See Figure 4.4.) Put simply, within this age group the percentage of abortions was small, but of those adolescents who got pregnant, most chose abortion.

In 2001 the abortion ratio was lowest for women ages thirty to thirty-four years (147 per one thousand live births). (See Table 4.4.) Women ages thirty-five to thirty-nine and women twenty-five to twenty-nine had the second- and third-lowest ratio of abortions to live births (180 and 200 per one thousand live births, respectively).

With the exception of the thirty-five- to thirty-nine-year-old age group, the abortion ratio increased from 1974 through the early 1980s. It declined thereafter, especially for the youngest and oldest groups, with abortion ratios for the twenty- to thirty-four-year-old age group remaining relatively stable since the mid-1980s. (See Figure 4.5.)

Race and Ethnicity

In 2001 more than half (54.1%) of women in the United States who had an abortion were white, about one-third (35.7%) were African-American, and 7.8% were of other races. (See Table 4.5.) The abortion ratio for African-American women (491 per one thousand live births) was approximately three times the ratio for white women (165 per one thousand live births). Similarly, the abortion rate for African-American women (twenty-nine per one thousand women) was also approximately three times the rate for white women (ten per one thousand women).

The abortion ratio for Hispanic women in 2001 was only slightly lower than that for non-Hispanic women (230 versus 232 respectively per one thousand live births). However, the abortion rate for Hispanic women was considerably higher than that for non-Hispanic women (twenty-two versus fourteen abortions, respectively, per one thousand women). (See Table 4.6.)

For women in all age groups, fertility was higher for Hispanic women at eighty-two births per one thousand women than for non-Hispanic women: 61.0 for white women, 56.5 for white, non-Hispanic women, 64.6 for African-American women, and 55.4 for Asian and Pacific Islander women. (See Table 4.7.)

Marital Status

In 2001, as in past years, unmarried women (divorced, widowed, or never married) comprised the greatest proportion of those who had an abortion (81.6%). Married women, including those who are separated from their husbands, comprised only 18.4% of those who had an abortion in 2001. (See Table 4.2.)

Contraceptive Use Prior to Pregnancy

Rachel K. Jones, Jacqueline E. Darroch, and Stanley K. Henshaw, in "Contraceptive Use among U.S. Women Having Abortions in 2000–2001" (Perspectives on Sexual and Reproductive Health, vol. 34, no. 6, November/December 2002), reported that 46.3% of women receiving abortions in 2000 had not used a contraceptive method in the month they conceived, primarily due to perceived low risk of pregnancy (33% of nonusers) and concerns about contraception (32% of nonusers). More than half of women receiving abortions (53.7%) did use a contraceptive method in the month of conception, however. Of these women, most used the male condom (27.6%) followed by the pill (13.6%). (See Table 4.8.) Some women using the condom and the pill (14% and 13%, respectively) believed that they had used the method perfectly, yet had become pregnant. Nevertheless, most women cited inconsistent use of the method as the reason for becoming pregnant—49% of condom users and 76% of pill users gave this reason.

Previous Live Births and Previous Abortions

In 2001 more than half of all women who had an abortion had never obtained one before (54.6%), whereas about one-fourth (25.5%) had obtained one previous abortion and 18.3% had obtained two or more abortions. (See Table 4.9.)

WHY DO WOMEN HAVE ABORTIONS?

The AGI, in "Induced Abortions in the United States: Facts in Brief," (http://www.agi-usa.org/pubs/fb_induced_abortion.html, May, 18, 2005), notes that each year almost half (49%) of the 6.3 million pregnancies in the U.S. are unplanned. Half of the women with unplanned pregnancies obtain an abortion. On average, women give at least three reasons for choosing abortion:

• About three-quarters say that having a baby would interfere with their work, school, or other responsibilities.
• About two-thirds say they cannot afford a child.
• About one-half say they do not want to be a single parent or are having problems with their husband or partner.

WHERE DO WOMEN GO FOR AN ABORTION?

In 2001, as in previous years, the largest reported numbers of abortions were performed in New York City (91,792), Florida (85,589), and Texas (77,409). New York and Texas are two of the most populated states; California, which is the most populous state and in previous years had the largest numbers of abortions performed there, did not report the number of abortions for 1999 or 2001. The fewest abortions were performed in Idaho (738), South Dakota (895), and North Dakota (1,216). (See Table 4.10.)

Abortion rates by state of occurrence do not necessarily reflect the number of abortions obtained by residents. Many women travel out of state for abortions because of a lack of providers or because of restrictive laws, such as required parental notification and consent (see chapters 10 and 11), waiting periods, and required counseling that involves more than one visit to the provider. See Table 4.11 for a list of states that require a waiting period before an abortion can be obtained and for those that require biased counseling. Biased counseling means that the woman is counseled in a way that promotes choices other than abortion.

For women obtaining an abortion in 2001 whose state of residence was reported, 90.3% had their abortion in the state where they lived. That is, 8.7% of reported abortions (on average) were obtained by out-of-state residents, with a wide range from 0.3% in Hawaii to 56% in the District of Columbia. (See Table 4.10.)

AVAILABILITY OF ABORTION PROVIDERS

The AGI notes that in 2000, 87% of U.S. counties had no abortion providers, up from 86% in 1996 and 77% in 1978. About one-third (34%) of women of childbearing age (fifteen to forty-four years old) lived in counties with no known providers. In North Dakota, South Dakota, West Virginia, Mississippi, and Wyoming, approximately four of five women of reproductive age lived in counties with no abortion providers. (See Table 4.12.) Nearly one-third of the country's metropolitan areas lacked abortion services, and for many women in rural areas, obtaining an abortion entailed traveling hundreds of miles from their residence.

According to the AGI, the number of abortion providers in the United States has decreased from a high of 2,908 in 1982 to about 2,000 in 1996 to 1,819 in 2000—11% fewer than in 1996. (See Table 4.12.) To compare, this is less of a decline than from the years 1992 to 1996, which saw a 14% decline in the number of abortion providers. The number of providers in 2000 was 37% lower than the high in 1982.

Between 1996 and 2000 nine states had an increase in the number of providers and thirty-eight states and the District of Columbia saw a decrease. The number of providers did not change in the three remaining states. (See Table 4.12.) A survey conducted in 1997 of obstetricians and gynecologists who perform abortions revealed that more than half of the providers were ages fifty or older, which some believe indicates that the number of providers will decline as current providers retire. However, according to the AGI in "Abortion Incidence and Services in the United States in 2000" (Stanley K. Henshaw and Lawrence B. Finer, Perspectives on Sexual and Reproductive Health, vol. 35, no. 1, January/February 2003), the approval and use of medical abortion (abortion pills) began to play a "small but significant role in abortion provision."

Finer and Henshaw also reported that clinics, including physicians' offices, perform most (93%) of the abortions in the United States, an increase of 3% from 1996. Of the 1,819 providers in 2000, 46% were clinics. Hospitals made up one-third of abortion providers in 2000, nearly the same as in 1996. However, the proportion of abortions performed in hospitals decreased from 7% to 5% from 1996 to 2000. One-fifth (21%) of providers were physicians' offices, down from 23% in 1996. In contrast, in 1973, the year of the Roe v. Wade decision, hospitals made up 81% of all providers. In 2000 physicians' offices (23% of providers) accounted for only 2% of the abortions performed, a steady decline from 4% in 1992 and 3% in 1996.

Abortion Medical Training

Abortion continues to be the most frequently performed surgical procedure in the United States. The AGI states, "Traditionally, residency programs in obstetrics and gynecology have been the only programs to formally train residents in abortion procedures and in the management of complications from spontaneous and elective abortions." The legalization of abortion in 1973 did not bring about any marked increase in obstetrics and gynecology resident training in abortion. In fact, from the late 1970s the number of residents trained in abortion was on a steady decline, which caused a major shortage of abortion providers. However, this decades-long trend seems to have changed, and opportunities for training have increased.

In 1998 the National Abortion Federation (NAF), the professional association of abortion providers in the United States and Canada, conducted a survey, asking the 261 accredited U.S. residency programs in obstetrics and gynecology about the opportunities available to medical students for first- and second-trimester abortion training. The results were published in a report, "Abortion Training in U.S. Obstetrics and Gynecology Residency Programs, 1998," by Rene Almeling, Laureen Tews, and Susan Dudley (Family Planning Perspectives, vol. 32, no. 6, 2000). Of the 179 programs that responded to the survey, 81% reported that they offer first-trimester abortion training—46% offer it routinely and 34% offer it as an elective. (See Table 4.13.) Second-trimester training is offered by 74% of programs—44% routinely and 29% as an elective. Some programs that do not offer training

allow residents to obtain abortion medical training elsewhere. Although 26% of programs indicated that all residents in their programs receive abortion training, 40% said that fewer than half are trained (including 14% that train no residents).

Almeling, Tews, and Dudley note that there is reason to be "cautious" when interpreting the results of the NAF study. Programs that do not offer abortion training may be less likely to respond, or they may feel pressure to report the availability of abortion training because of guidelines from the Accreditation Council for Graduate Medical Education (ACGME). This organization oversees medical education in the United States. It had noticed the trend in the shortage of abortion providers. As a result of the impact of provider shortage on the already dwindling number of abortion services, the ACGME decided that, effective January 1996, induced abortion "must be part of residency training, except for residents with moral or religious objection." Teaching hospitals that did not train or arrange for abortion training risk losing accreditation and, consequently, federal reimbursements of services rendered by medical residents. Under pressure from Catholic bishops and antiabortion groups, the ACGME requirement was amended by Congress. The policy has become a recommendation that elective training in induced abortion be offered in residency training. So far, no medical school has lost accreditation for failing to do so.

In July 2002 New York City became the first U.S. city to include abortion training as a standard element of obstetrics-gynecology (ob-gyn) residency programs. The plan requires that all ob-gyn residents at the city's eleven public hospitals be trained to provide abortion services, including the administration of mifepristone for medical (nonsurgical) abortion. These residents can refuse the abortion training if they have moral objections to abortion, in accordance with ACGME guidelines.

The American Academy of Family Physicians (AAFP) holds that physicians should work toward decreasing the number of unwanted pregnancies by providing patient education and counseling, as stated in "Reproductive Decisions" (American Academy of Family Physicians, http://www.aafp.org/x7053.xml, March 2005). If a woman becomes pregnant, the AAFP believes that it is her legal right to make reproductive decisions. However, the AAFP also states that a physician shall not be compelled to perform any act that goes against his or her "good judgment or personally held moral principles." The AAFP holds that abortion should be performed only by a "duly-licensed physician in conformance with standards of good medical practice as determined by the laws and regulations governing the practice of medicine in that locale."

The organization also recommends that medical students and family practice residents be trained in counseling and referral skills regarding all options available to pregnant women. This would include contraceptive methods (including sterilization), adoption services, abortion services, financial assistance, and other assistance available to the mother or the child. They also support language in the "Program Requirements for Residency Training in Family Practice" of the Residency Review Committee for Family Practice that calls for providing opportunities for residents to learn skills they think will be part of their future practices.

HOW MUCH DOES AN ABORTION COST?

Abortion fees vary, depending on the stage of pregnancy, where the abortion is performed, the kind of procedure, and the anesthetic used. There may be other costs if the abortion cannot be done locally. Expenses might include travel costs, costs for overnight stays, or lost wages in states that require a waiting period between counseling and the abortion.

In "Get 'In the Know': Questions about Pregnancy, Contraception and Abortion—Cost of Abortion" (http://www.guttmacher.org/in-the-know/cost.html), the AGI noted that, in 2001, the average charge for an abortion performed at ten weeks of pregnancy in a clinic was $468. The AGI noted, however, that most abortions in the U.S. are performed at low-cost clinics, and the procedure cost, on average, $372. The fees generally cover the examination, laboratory tests, local anesthesia, the procedure, and the follow-up visit. In clinics offering medical abortions, the average cost was $487. In most cases medical abortions (those using abortion pills) cost slightly more than other first-trimester abortions because of the cost of mifepristone and the follow-up required.

The cost of an abortion increases after the first trimester. Hospital abortions cost more than those performed in abortion clinics and doctors' offices. In addition to the fees of the surgeon and anesthesiologist, costs include the hospital stay, anesthesia, tests, and medications.

The "Cost of Abortion" fact sheet notes that in 2001 about 74% of women funded their own abortions (but may be reimbursed by their insurance provider), 13% were covered by Medicaid (in cases of life endangerment, rape, and incest), and 13% had private insurance that paid abortion costs directly. The fact sheet noted, however, that as of March 2004, seventeen states used their own funds to subsidize abortion for poor women.

WHEN ARE ABORTIONS PERFORMED?

Table 4.2 shows that in 2001 well over half (59.1%) of abortions were performed at eight weeks or less of development. Figure 4.6 shows that in that same year 59.1% of abortions were performed at nine weeks or less. (Such data can differ slightly depending on the source of the data and how the data was calculated.) Both sources show that 19% of abortions were performed at nine to ten weeks of pregnancy. Both sources also show that few abortions were performed after fifteen weeks of gestation—4.3% at sixteen to twenty weeks and 1.4% at or after twenty-one weeks of pregnancy.

As in past years, in 2001 a larger percentage of women ages nineteen years and younger obtained abortions later in pregnancy than did older women. In addition, a larger percentage of African-American women obtained abortions later in pregnancy than did white women and women of other races. (See Table 4.14.)

ABORTION METHODS

Abortion can be performed using surgical or medical (drug) methods. The type of abortion a woman obtains depends on her choice of method, her health, and the gestational age of the fetus. Methods of abortion are described in Table 1.1 in Chapter 1.

Percentages of Methods Used

The CDC has collected data since 1973 on the types of procedures used each year to perform abortions. The CDC has determined that there has been a shift in the proportion of which procedures are used. From 1973 to 1991 the percentage of abortions performed by curettage (which here includes dilation and evacuation) increased from 88.6% to 99.1%. The percentage then dropped somewhat during the 1990s, and more sharply in 2000 and 2001 to 97.9% and then to 95.5%. Nevertheless, the percentage of curettage performed has increased dramatically since 1973. Conversely, the percentage of abortions performed by instillation (induction) declined sharply from 10.3% in 1973 to 0.2% in 1999. This figure increased slightly to 0.4% in 2000 and to 0.6% in 2001. The CDC suggests that the increased use of curettage over induction is likely due to the decreased risk of complications using curettage. In addition, a greater proportion of women obtain abortions earlier in their pregnancy now than in past decades, thus allowing curettage over induction. (See Table 4.2.) Hysterectomy (removal of the woman's uterus), hysterotomy (an incision, or cut, to the uterus, usually to deliver a fetus), and medical (nonsurgical) procedures, which were included in the "Other" procedure category, were used in 3.9% of abortions. For 2001 the CDC reports a total of 20,093 medical abortions performed.

Deaths from Abortion

As Table 4.15 shows, the fatality rate from abortion procedures has dropped from 4.1 deaths per one hundred thousand reported legal induced abortions in 1972 to its lowest point of 0.3 in 1995. (See Table 4.15.) In 1997 the rate was 0.6. Fatality rates are not shown for years 1998–2000 due to missing data. The CDC reported that in 2002 the death rate related to pregnancy and childbirth was 0.1.

Although abortion-related deaths are rare, legal abortions are still surgical procedures and entail some risk. An abortion performed at or before eight weeks of pregnancy is the safest. Similarly, the abortion method used also has a bearing on risk. Vacuum aspiration, used during the first trimester of pregnancy, carries the least risk of the surgical methods.

Do Physicians Offer Medical Abortion?

A survey of health care providers conducted before mifepristone was approved by the FDA asked obstetricians/gynecologists and family practice physicians about their views on medical abortion ("A National Survey: Views of Women's Health Care Providers on Abortion," The Henry J. Kaiser Family Foundation, June 13, 2000). Most of the physician respondents in the survey were familiar with mifepristone—91% of obstetricians/gynecologists and 67% of family practice physicians reported that they were familiar with the drug. However, in a survey conducted by the Kaiser Family Foundation almost one year later and after the drug had been approved for use ("National Survey of Women's Health Care Providers on Reproductive Health: Views and Practices on Medical Abortion," May 15-August 28, 2001), a smaller percentage of physicians in these two groups were familiar with the drug: 82% of obstetricians/gynecologists and 61% of general practice physicians. (See Figure 4.7.) The 2000 survey found that, if the FDA approved mifepristone, 54% of obstetricians/gynecologists and 31% of family practice physicians would offer mifepristone abortions. However, in the 2001 survey, only 6% of gynecologists and 1% of general practice physicians reported that they had provided mifepristone to a patient since its approval. (See Figure 4.8.)

Only 27% of obstetricians/gynecologists and 1% of family practice physicians surveyed in 2001 provided surgical abortions to their patients; among that 27%, 12% of providers have used mifepristone since its approval. (See Figure 4.9.)

OPPONENTS AND PROPONENTS OF MIFEPRISTONE

The nation's largest pro-life group, the National Right to Life Committee (NRLC), threatened to boycott any drug company that sells mifepristone. In the early 1980s the NRLC successfully pressured the Upjohn Company to discontinue developing an abortion-inducing drug through a two-year boycott of its products. On FDA approval of mifepristone, Laura Echevarria, spokeswoman for the NRLC, said, "The Clinton-Gore Administration, which claimed it wanted to make abortion rare, has embraced an abortion pill that will result in more abortions and new risks to women" ("F.D.A. Approves Sale of Abortion Pill," National Right to Life, http://www.nrlc.org/press_releases_new/Release092800.html). Pro-life groups argue that "chemical" abortion would leave serious intangible scars on women who would experience real labor pains and see their aborted fetuses.

Pro-choice activists, in contrast, approved of the medical-abortion drug. Gloria Feldt, president of Planned Parenthood, stated, "Mifepristone, or the early abortion pill, is as significant a technological advance for women's health as the birth control pill was 40 years ago" ("Food and Drug Administration Approves 'Abortion Pill'," CNN.com, http://archives.cnn.com/2000/HEALTH/women/09/28/abortion.pill/, September 28, 2000). The American Civil Liberties Union issued a statement by Catherine Weiss, director of the group's Reproductive Freedom Project. She stated, "Mifepristone represents a significant breakthrough in reproductive health care by offering women, especially those who live far from an abortion provider, better access to a safe, private, and early option for ending a pregnancy" ("ACLU Hails FDA Approval of Safe, Early-Option Abortion Pill," American Civil Liberties Union, http://www.aclu.org/ReproductiveRights/ReproductiveRights.cfm?ID=8134&c=143, September 28, 2000).