PREGNANCY AND PREGNANCY TERMINATION

Pregnancy is one of the most important events in a woman's life. For many women, pregnancy defines, enhances, or determines their economic, social, or personal value. Pregnancy and childbearing can be highly desired for personal reasons, or to meet social expectations. In some circumstances, childlessness may be grounds for suspicion, divorce, or worse. Pregnancy in the wrong situation may be undesirable to an equivalent degree. Traditionally, the woman with an unacceptable pregnancy may be abandoned, lose her prospects for marriage, be forced into prostitution, or suffer an even worse fate. Other pregnancies, even though they occur within a socially sanctioned circumstance, may result in loss of opportunities for social or economic advancement. Control over the occurrence and outcome of pregnancy is integral to women's control over their lives. This concept was strongly stated at the International Conference on Population and Development in Cairo in 1994.

Menarche, the first menstrual period, marks the beginning of a woman's reproductive period, and menopause, the last menstrual period, marks the end. Menarche occurs between ages 8 and 14 for most well-nourished women, but may come later if there is malnutrition or chronic disease. If the first menstrual period follows ovulation (release of an egg or ovum from the ovary), the woman may be fertile before menarche has occurred. For most women, fertility increases rapidly in the first year after menarche, as ovulation becomes more regular, so that as many as 95 percent of women in their late teenage years may be fertile. Infertility in the first years after menarche may be due to malnutrition, malformations of the uterus and reproductive organs, hormone abnormalities leading to anovulation, or genetic diseases. After the early twenties, fertility declines, as sexually transmitted diseases; endometriosis (occurrence in abnormal sites of tissue normally found in the lining of the uterus); and other inflammatory, infectious, and vascular diseases affect reproductive organs. Menopause occurs between ages 40 and 55 in most women, but most women over 40 are infertile, and very few live births occur to women over 45.

PREGNANCY

Pregnancy is typically divided into trimesters, which are uneven in length. Obstetricians invariably refer to "weeks" of pregnancy, meaning the number of weeks from last menstrual period, or when the last menstrual period should have occurred. The first trimester lasts from 3 to 12–14 weeks, the second from 12–14 to 24–26, and the third from 24–26 to delivery. The terminology is confusing, as menstrual age (the time since the last period) is not the same as gestational age (the time since fertilization, the union of the sperm and ovum). Therefore gestational age will be 2 weeks less than menstrual age. The system evolved in an era when many or most women could recall a last menstrual cycle, but few knew when conception or fertilization occurred. Now, more women may be aware of the date when conception occurred, and some pregnancies occur after assisted reproductive technologies (ART), when the time of fertilization may be known exactly. Either way, pregnancy itself begins immediately after implantation. This occurs about 3 weeks after the menstrual period and about 1 week after fertilization (therefore the first trimester starts at 3 weeks). A blood pregnancy test becomes positive within a day or two of implantation, and a urine test becomes positive a few days later. By the time the menses are missed, 4 weeks after the last menses, a urine pregnancy test will be positive if fertilization occurred at the expected time.

Between 10 percent and 40 percent of pregnancies result in spontaneous abortion in the first trimester. The wide variation in incidence reflects difficulty in recognizing very early abortion as well as differences in abortion rates among different age groups. Most spontaneous abortions are the result of chromosomal or developmental abnormalities of the pregnancy and are unavoidable; 1–2 percent of pregnancies abort spontaneously in the second trimester, often due to infection, and about 1 percent of pregnancies have serious anomalies. About 1 percent of pregnancies are ectopic, occurring outside of the uterus. Most ectopic pregnancies require medical or surgical treatment to avoid hemorrhage. Ectopic pregnancy is one of the major causes of maternal mortality; in the United States, it is still associated with about 10 percent of maternal deaths (Cunningham et al. 1997).

RESPONSE TO PREGNANCY

Women typically pass through several stages of reaction to pregnancy. The first reaction may be denial, particularly when the pregnancy is unwanted, or delight, when the pregnancy is wanted. Later, desired or planned pregnancy may still be accompanied by periods of ambivalence or anxiety (Affonso 1997). The concern may be about personal or work changes, financial stresses, health concerns, or concern about fetal health. At some point in the pregnancy, typically first or second trimester, the pregnancy is usually accepted. The fetus is visualized as a baby, and planning for birth and beyond ensues (Taylor 1980).

The pregnancy may not be accepted; the woman may choose to abort the pregnancy, or she may deny existence of the pregnancy, or she may never accept it even though she eventually gives birth. The lack of acceptance may follow a situation where abortion was unavailable; this may be an individual situation or a matter of political policy. During wartime, impregnation of "enemy" women (which may be accompanied by denial of abortion) is a form of terrorism (Swiss 1993).

Pregnancy itself does not impart any ability to deal with stress, nor does it decrease ability to manage stress. Women who have difficulty caring for themselves adequately when not pregnant, such as adolescents and women with severe psychiatric or developmental handicaps, may have difficulty caring for themselves while pregnant, to the detriment of the fetus. Pregnancy may be an incentive to stop self-abusive behavior such as drug use, but while pregnancy may be the motivating factor, it does not supply the emotional organization to change behavior. Domestic abuse of women does not usually stop with pregnancy; it often escalates (American College of Obstetricians and Gynecologists 1995b).

PREGNANCY AND BIRTHRATES

The pregnancy rate is the number of pregnancies occurring per 1,000 women of reproductive age (considered to be ages 15–45 inclusive), per year. The fertility rate is the number of all births (liveborn and stillborn) per 1,000 women age 15–45 per year; this is the birth rate for reproductive-age women. The term birthrate, unmodified, usually means the number of live births per 1,000 population (male and female of all ages).

Pregnancy rates increase during the teenage years and generally peak during the late twenties in most societies, before declining in the thirties and forties. Birthrates follow a similar trend, although the peak birthrate may occur later than the peak pregnancy rate, and the difference between pregnancy rates and birthrates is more pronounced at the extremes of reproductive life. At the beginning and the end of reproductive life more pregnancies are likely to be terminated by abortion.

More pregnancies than are wanted occur. Some are mistimed, occurring earlier than wanted, and some are unwanted at any time. In the United States, between 40 percent and 50 percent of pregnancies are unintended (Horton 1995; Henshaw 1998a). Overall, mistimed pregnancies are more frequent than unwanted pregnancies, but the relationship may be reversed in some groups, such as women over 40. Rates of unplanned pregnancy are generally lowest in countries with wide availability of effective contraception combined with public education, such as western Europe (Paul 1999). Unplanned pregnancy may occur because of contraceptive failure or non-use of contraception. In the United States noncontracepting women, about 8 percent of women at risk for unintended pregnancy, account for about 50 percent of unplanned pregnancies (Gold 1990).

Unplanned pregnancy may result in birth, spontaneous abortion, or induced abortion. The abortion ratio (ratio of pregnancies aborted to pregnancies occurring) in the United States for unplanned pregnancy is about 0.5 (Gold 1990). In other countries the rates may be lower, where unplanned pregnancies are accepted or abortion is not readily available, or higher, where unplanned pregnancy is not tolerated personally, socially, or economically (Henshaw et al. 1999). Teenage pregnancy rates have been of concern because of adverse medical and economic outcomes, which are intertwined (Fraser et al. 1995). For most of the century, pregnancy rates for U.S. teenagers have been declining. Rates began to rise in the 1980s, reaching a recent high of 117 per 1,000 (women ages 15–19) in 1990 (Alan Guttmacher Institute 1999). By 1997 rates had declined about 10 percent to 97 per 1,000. Birthrates have shown a parallel decline from 60 per 1,000 to 54 per 1,000, and abortion rate declined as well.

In contrast to teenage women, women in their twenties have higher pregnancy and birthrates. Birthrates are currently about 110 per 1,000 women in this age group (National Center for Health Statistics 1999). Crude abortion rates are similar to those of adolescents, 20–30 per 1,000 women. Since adolescents have fewer pregnancies and births, the abortion ratio is higher than for women in their twenties or thirties. Overall, more than 40 percent of teenage pregnancies resulted in induced abortion. There are large variations in pregnancy, birth, and abortion rates by state, ethnic group, and age (Horton 1998). Nineteen-year-olds have a much higher pregnancy rate than 15-year-olds. Nonwhite and black teenage pregnancy rates are approximately twice those of white teenage rates. Nonwhite and younger teenagers are more likely to have induced abortions (National Center for Health Statistics 1999).

Comparison of birth and abortion rates can be cumbersome. Distribution of women's age, ethnic group, geographic location, time of collection of statistics, event definition, and accuracy of reporting will affect rates.

CONDUCT OF PREGNANCY

A comprehensive discussion of pregnancy is outside the scope of this encyclopedia; for more complete information, an obstetrics textbook (e.g., Cunningham et al. 1997; Gabbe et al. 1996) should be consulted.

The first trimester of pregnancy is often accompanied by nausea and vomiting, which is typically short-lived but may be severe enough to interfere with daily activities. Some women have hyperemesis, vomiting which is severe enough to result in dehydration or even death; it is treated with intravenous fluids, antinausea drugs, and sometimes intravenous feeding. Although a link between hyperemesis and ambivalence or anxiety about pregnancy has been postulated, the strength and the relevance of any such link is uncertain.

Women often note changes in food likes and dislikes, but there is no medical reason to restrict types of food. There are often cultural restrictions and prescriptions, and there is seldom any reason to interfere with custom. In the second half of pregnancy many women have heartburn, because of the pressure of the pregnancy on the bowel and stomach, and because of relaxation of the esophagus. Changes in diet or eating habits, or using antacids, may help.

Women need additional calories during the second 2 trimesters; the amount is about 300 kcal per day for most women, the equivalent of a modest sandwich. The additional calories should consist mostly of complex carbohydrates; most Americans have adequate protein intake. Women who receive protein supplements may have higher rates of preterm birth than women receiving carbohydrate supplements or no supplements at all. The U.S. Department of Agriculture administers a food supplement program called WIC (Women, Infants, and Children), which provides food to low-income pregnant and lactating women, infants, and children up to 5 years of age. Supplementation is associated with small decreases in the rate of low-birthweight delivery and a decrease in cost of caring for newborns (Merkatz 1990).

Folic acid, a vitamin, has been shown to decrease the incidence of neural tube defects, a complex of birth defects of the brain or spine. Since the neural tube fuses in the fifth menstrual week, often before pregnancy is noticed, women should take adequate folic acid before pregnancy (Czeizel and Dudas 1992; Rosenberg 1992). The recommended daily allowance of folic acid prior to pregnancy, 400 mg, is easily achieved with a well-balanced diet. Since bread and cereals are now fortified with folic acid, it is becoming harder to avoid adequate folic acid. However, residents of closed communities, producing their own food may not have access to fortified food. Overconsumption of most vitamins is harmless, and the excess is excreted in urine. However, fat-soluble vitamins may accumulate and reach levels that are toxic to the fetus. Iron is useful for women who are anemic because of iron deficiency.

Prenatal medical care is available to most women in the United States through a patchwork of public and private funding (Alan Guttmacher Institute 1994). Public funding has increased over the last decades as it became evident that the cost of caring for low-birthweight infants could be decreased by providing prenatal care (Merkatz 1990). The relative importance of individual facets of prenatal care is not certain (e.g., Crane et al. 1994; Higby et al. 1993), and some usual practices are probably useless but are persistent (AAP and ACOG 1997; Merkatz 1990). Efforts to assess prenatal care are complicated by the observation that women who actively seek prenatal care have better outcomes than those who do not, even when the latter group receives the same care.

OUTCOMES OF PREGNANCY

A normal pregnancy lasts 38–42 menstrual weeks (term). Infants over 37 weeks are considered full-term. Infants under 37 weeks are considered pre-term. Low birthweight (or under 2,500 grams at birth), includes both preterm infants and full-term infants who weigh less than expected. The lowbirthweight rate is the percentage of births under 2,500 grams.

The low-birthweight rate is an indicator of maternal health and of the effectiveness of prenatal and intrapartum care. However, the low birthweight numbers may be affected by the definition of live births. In some locations, live births are considered to be births over 28 weeks and/or 750 or 1,000 grams. In the United States the usual definition is any birth over 500 grams or showing movement or cardiac activity after birth; birth of a fetus under 500 grams without movement is classified as a spontaneous abortion. However, the definition varies by state and some states have adopted 350 grams as the threshold (Horton 1998). A 350-gram threshold will appear to raise the rate of low birthweight and the mortality rate of low-birthweight infants, as virtually no fetus born between 350 and 500 grams will survive (see ACOG 1995a).

Low birthweight is the leading cause of perinatal death. Perinatal deaths consist of antepartum deaths (in which the fetus dies before delivery), intrapartum death (in which the fetus dies during labor or delivery), and neonatal deaths (in which a liveborn infant dies in the first 28 days). In contrast to the low-birthweight rate, which is usually expressed as a percentage, the death rate is usually expressed as the rate per thousand births (both live and stillbirths). Death rates vary according to policies on inclusion of the smallest fetuses and infants.

A viable fetus is a fetus that can survive outside the uterus. Viability may affect a decision to use an intrauterine treatment versus delivery and treatment of the infant. Viability has also played a role in some debates about abortion, as the gestational age of viability has decreased with advances in perinatal care.

For most healthy women, the physical stress of pregnancy is easily managed. Pregnancy is much more dangerous to women with underlying anemia, heart, or kidney disease, who cannot manage the necessary increase in blood supply and circulation, and to women who do not have adequate nutrition. Family planning, by allowing adequate nutrition before pregnancy and recovery between pregnancies, is critical to women's health status during pregnancy (Alan Guttmacher Institute 1997). Women who are poorly nourished or ill have a much higher chance of delivering a lowbirthweight baby, who in turn is more likely to have chronic illnesses and disabilities.

Delivery has additional risks to all women. In the nineteenth century, with the best available care, maternal mortality was about 1 percent per pregnancy, and it remains at that level, or higher, in some parts of the world. It is estimated that each year 600,000 women die from pregnancy and childbirth (Berg et al. 1996; WHO 1998a). The three most common causes of morbidity and mortality are hemorrhage, infection, and pre-eclampsia (high blood pressure and blood vessel disease unique to pregnancy). Morbidity and mortality from all 3 of these situations are less likely where there are trained birth attendants (midwives or physicians) with access to a short list of medications (Rooks 1997). Relatively simple interventions may make a major difference in outcome (e.g., Wallace et al. 1994). In contrast, the incremental increase of complex technology is relatively small, although it may be dramatic in some situations. The widespread application of some technologies has been harmful by creating additional problems, such as interference with normal

labor by restriction of movement (Butler et al. 1993) or by diversion of resources from other beneficial programs.

Births are classified as spontaneous (vaginal), operative vaginal, or operative abdominal. Spontaneous birth includes births that are assisted by a birth attendant's hands. Operative vaginal births include forceps and vacuum extractor (suction cup applied to the baby's head); rates range from as low as 1 percent to 30 percent or higher in some settings. Operative abdominal births are "cesarean sections," and rates vary widely. In western Europe rates are generally 8–15 percent of total births. In the United States, rates had risen steadily for several decades before decreasing very slightly in the last several years to about 22 percent of all births (National Center for Health Statistics 1999). In some cities throughout the world there is a local demand for elective abdominal delivery as an alternative to labor, generally restricted to women who can afford to pay for the request. With such widely varying rates of operative intervention there is no consensus on optimal rates, although there are many opinions. For instance, there has been no decline in the rate of mental retardation in the United States, although both fetal monitoring and operative delivery were purported to prevent at least some retardation (ACOG 1992; Rosen and Dickinson 1992).

After birth, women and babies do best when they are kept together and allowed to establish lactation (WHO 1998b). Separation of mother and baby as practiced in many hospitals is not only unpleasant, but inappropriate in terms of infection control and infant nutrition. The effect of labor and delivery routines on "bonding" or attachment has been debated; whether any simple intervention can influence parenting is questionable at this point. The social and educational support of trained lay women has measurable success in some studies on labor outcome and breastfeeding success (Rooks 1997). More intensive programs of pregnancy, delivery, and postpartum peer care have not been fully evaluated (O'Connor 1998).

TERMINATION OF PREGNANCY

Abortion is an event, other than a birth, that terminates a pregnancy. Abortion may be spontaneous, if it begins without intervention from the woman and without medical intervention, or induced, if some agent or procedure is used to cause the abortion. Induced abortion may also be classified as legal or illegal. Other terminology is neither uniform nor clear. Therapeutic abortion may refer to all legal induced abortions performed under medical supervision, or to those performed for medical indication such as severe illness in the woman. Spontaneous abortion may be classified as inevitable if it has not yet occurred but will occur, as incomplete if the pregnancy has been partially passed, or as complete if all tissue has been expelled or removed.

A viable pregnancy may refer to a pregnancy in which the baby is apparently healthy and growing; the pregnancy would be expected to continue if there were no intervention. A nonviable pregnancy will result in spontaneous abortion. Nonviable pregnancies are also called blighted ovum (which is technically incorrect since the ovum or egg has already divided), empty sac, or fetal demise. Abnormal pregnancies are sometimes detected by ultrasound or by blood testing, before there are any symptoms of spontaneous abortion.

Most abortions are requested because the pregnancy was unwanted, but there are other reasons, some of which are listed in Table 2.

In the United States, 90 percent of abortions occur in the first trimester; half of all women request abortion before 8 menstrual weeks. Ten percent of women request abortion after 12 weeks. Between 1 and 2 percent of all abortions performed are for fetal malformations; these are almost all second-trimester procedures. Fewer than 1 percent of abortions are performed in the third trimester. Generally, as the length of gestation increases, the cost of abortion increases, and the number of providers decreases (Gold 1990).

There are several types of procedures in use (for a more complete discussion, see Paul 1999). In the first trimester, there are both medical and surgical techniques. Surgical techniques consist of variations of suction curettage. In this procedure the cervix is stretched open if necessary; in very early pregnancy (fewer than 6 menstrual weeks) no opening may be necessary. The inside of the uterus is suctioned using a plastic tube and a vacuum created by an electric or manual pump. "Sharp" curettage, the traditional dilatation and curettage (D&C) is more traumatic and has been largely replaced by suction curettage.

Several medications are used for early medical abortion, depending on availability. Methotrexate is a medication that blocks folic acid, a vitamin. Several days later the woman takes a second drug, misoprostol (a prostaglandin drug), which makes the uterus contract and expel the pregnancy. The process is similar in both timing and feeling to a spontaneous abortion. The exact time sequence is difficult to predict; about 75 percent of women abort within 1 week. Vaginal bleeding occurs for a mean of about 10 days, but may last longer. The medication may fail to produce an abortion about 1 percent of the time, and about 1 percent of women need a suction curettage because of heavy bleeding. Methotrexate can be used up to 7 or 8 menstrual weeks, and is the most common agent in use in the United States, because it is FDA approved for other uses, and therefore easily available.

Mifepristone (formerly called RU 486) has been used in millions of women in China, France, Sweden, and the United Kingdom for medical

abortion since 1986, but has had to overcome numerous political roadblocks in the United States. Mifepristone, a different class of medication from methotrexate, blocks progesterone hormone binding sites. It requires use of a second drug, a prostaglandin drug such as misoprostol, several days later to expel the pregnancy. In contrast to methotrexate, 75 percent of women abort within 2 days, and 90–95 percent abort within 1 week. Mifepristone can be used up to 7 to 9 menstrual weeks depending on the selection of the second drug.

After 7 to 9 menstrual weeks, suction curettage is the main technique. At the end of the first trimester and in second trimester, opening the cervix sufficiently becomes more challenging. Osmotic dilators are used over several hours or several days; osmotic dilators are placed in the cervix, where they absorb water and swell. Preparation for abortion in the mid- and late second trimester may take several days. Surgical procedures, often called dilatation and evacuation (D&E), are variations of early pregnancy techniques using suction and extraction instruments. After 16 weeks, medical (induction) techniques can be used to induce labor. All the agents—prostaglandin, oxytocin, and saline—are unpredictable, may take several hours to several days, and may involve a curettage to remove placenta if it is not completely expelled. Hospitalization is the rule, and therefore induction techniques are more costly than surgical techniques in an out-patient setting.

Legal abortion has very few serious complications. Term pregnancy has a death rate at least ten times higher than first-trimester abortion (10–12 deaths per 100,000 versus 0. 5–1.0 deaths per 100,000 women in the United States) (Berg et al. 1996; Paul et al. 1999), and a morbidity rate (serious medical outcomes such as major operations) hundreds of times higher. There is virtually no situation in which it is safer for a woman to continue a pregnancy than to abort a pregnancy, although there are sometimes situations in which abortion should be delayed briefly. However, abortion is safest if it is performed early (Gold 1990; Paul et al. 1999). Early abortion by suction curettage does not have an effect on future fertility. Abortion itself is not associated with adverse psychological sequelae; unwanted pregnancy may have adverse associations regardless of whether the pregnancy is aborted or continued. In second trimester the medical risks are higher than in first trimester (Paul et al. 1999).

It is difficult to ascertain the numbers and types of abortions performed in the United States. In most states abortion is a reportable procedure; however, many procedures are not reported, probably for reasons of confidentiality of both provider and patient (Fu et al. 1998). Some abortions may be misclassified as treatment of a spontaneous abortion, since the suction procedure is identical; or the reverse may be true. In hospitals, some induced abortions may be classified as spontaneous abortion. The Alan Guttmacher Institute estimates abortion procedures using both governmental and nongovernmental sources. The CDC also publishes "Abortion Surveillance" as a supplement to Mortality and Morbidity Weekly Review. Small variations in abortion rates may be related to reporting and surveillance issues as well as varying rates of pregnancy and abortion.

Currently, early medical abortion accounts for fewer than 1 percent of all abortions in the United States, as methotrexate is the only agent available. In other countries, medical abortion may account for up to a third of all abortion procedures. In France, half of all women seeking abortion request abortion early enough to be eligible for medical abortion, and two-thirds of them choose medical abortion. This shift to medical abortion in France has been accompanied by a shift to earlier abortion in general, while overall abortion rates have remained the same (Paul et al. 1999). Medical abortion is highly acceptable to women in many countries. Acceptability studies consistently estimate that 60–80 percent of women would choose medical abortion over surgical abortion were it available (Winikoff 1994). However, despite the preference for medical abortion over surgical abortion, surgical abortion remains the only method available to most women, if they have access to any method at all. In the United States the number of abortion providers has dropped over the last two decades. Abortion is available in a minority of counties in the United States, most of them in urban areas (Lichter et al. 1998). The majority of abortion procedures take place in free-standing medical facilities, many of them primarily devoted to abortion provision (Henshaw 1998b). Planned Parenthood clinics provide about 12 percent of all abortions. About 10 percent of abortions occur in hospitals or hospital-affiliated clinics. At least 5 percent of abortions occur in doctors' offices that are not identified as providing primarily abortion-related care.

Most physicians providing abortions in the United States are obstetrician-gynecologists, but other physicians provide abortions. Only about a third of gynecologists provide abortions, so most women needing abortion are referred to an unfamiliar caregiver. First-trimester abortions are also provided in several states by physicians' assistants (PAs) and midwives. Most states have "physician-only" statutes that limit the provision of abortion to physicians. However, these statutes were written before the increase in utilization of advanced practice nurses, such as midwives, and "physician extenders," such as PAs, who currently perform other procedures of comparable complexity and skill (Freedman et al. 1986). In a few countries (e.g., Bangladesh) midwives provide most abortion services.

Costs for abortions range form several hundred dollars for a first-trimester procedure in an office or clinic setting, to thousands of dollars for procedures performed in hospitals. Many insurance policies cover abortion, although reimbursement rates vary. Some forms of insurance do not cover abortion at all, for example, insurance provided to U.S. federal employees. Publicly funded insurance for indigent women covers "medically indicated" abortions in some states but not others. Finally, many women requesting abortion do not have any insurance at all, most commonly because their employment does not provide insurance. Women who need specialized procedures because of underlying medical illnesses may have additional barriers; there are several states in which no hospital will allow an abortion to be performed.

In the United States the majority of people polled supported the availability of abortion, but many were in favor of some restrictions. Parental consent laws for minors have been passed in nearly half the states. Any such law must contain a "judicial bypass" to be considered constitutional, so that minors can petition a judge if they cannot tell a parent (Paul et al. 1999). In practice, most minors do involve a parent with or without parental consent laws. The main effect of such laws is to delay abortion for those minors without good family support.

In the United States there is a vocal and well-funded minority with the goal of criminalizing abortion. Mandatory waiting periods and mandatory consent processes have also been passed. These increase the amount of time and expense necessary for a woman to obtain an abortion, particularly where there are few providers. Other quasi-legal attempts to decrease abortion include passage of laws that are either unenforceable or unconstitutional. These include restrictions on the type of facility that can perform abortion and "partial-birth abortion" laws. The resulting legal challenges are expensive, and this tactic can be considered a sustained economic assault (Reproductive Freedom News 1999). Antiabortion groups in the United States have been increasingly involved in terrorist activities, including harassment, arson, violence, and even murder. Several thousand incidents or harassment and arson are reported annually to the National Abortion Federation (National Abortion Federation 1999).

Abortion is generally legal in most European countries and in much of Asia, but is illegal in most of South America and Africa (WHO 1998b). The effect of government-sanctioned denial of abortion rights is not to decrease the number of abortions significantly. Abortion rates are correlated with multiple factors such as patterns of sexual activity, type of contraceptive use, desired family size, and tolerance of unwanted pregnancy. In the United States it has been estimated that the number of abortions performed annually before widespread legalization was 600,000 to 1.2 million per year (Gold 1990). The number reported the first year after legalization was 615,831 (Koonin et al. 1996). The effect of criminalizing abortion is to delay abortion for some women and to make it riskier for almost all women, resulting in increased death rates (Population Reports 1997). As an example, after abortion was criminalized in Romania, the death rate from abortions quintupled; while maternal mortality doubled (WHO 1998b). In South American and East Africa, the rates of illegal abortion are similar, about 35 per 1,000 women of childbearing age, but death rates are dissimilar (0.3 percent and 1.5 percent, respectively, of women undergoing abortion), related to underlying health status and access to postabortion care (WHO 1998b).

Increases in the provision of contraceptive services are associated with decreases in abortion rates (Henshaw et al. 1999; Estrin 1999). However, the United States has consistently refused to fund contraceptive services adequately, particularly services in developing countries, because of right-wing political pressure (Alan Guttmacher Institute 1996).

Pregnancy carries an intrinsic risk to women's health, which can be minimized by appropriate medical care to women who are in good health to start. Family planning is essential to women's health during pregnancy, and essential to the health of their infants. Women who do not want to be pregnant will risk health and life to end the pregnancy. Denying or criminalizing abortion care results in additional health risks for women while diverting health resources that could be used for maternity services for other women.

references

Affonso, Dyanne D., Chong-Yeu Lui-Chang, and Linda J. Mayberry 1997 "Worry: Conceptual Dimensions and Relevance to Childbearing Women." Health Carefor Women International 20:227–236.

Alan Guttmacher Institute 1994 Uneven and Unequal:Insurance Coverage and Reproductive Health Services. New York: Alan Guttmacher Institute.

——1996 Endangered: U.S. Aid for Family PlanningOverseas. New York: Alan Guttmacher Institute.

——1997 Family Planning Improves Child Survival andHealth. New York: Alan Guttmacher Institute.

——1999 Teenage Pregnancy: Overall Trends and State-by-State Information. New York: Alan Guttmacher Institute (April).

American Academy of Pediatrics and American College of Obstetricians and Gynecologists 1997 Guidelinesfor Perinatal Care, 4th ed. Elk Grove Village, Ill.: AAP.

American College of Obstetricians and Gynecologists 1992 Fetal and Neonatal Neurologic Injury. Technical Bulletin. Washington, D.C.: ACOG.

——1995a Committee Opinion: Perinatal and InfantMortality Statistics. Washington, D.C.: ACOG.

——1995b Domestic Violence. Washington, D.C.: ACOG

——1998 Committee Opinion: Vitamin A Supplementation during Pregnancy. Washington, D.C.: ACOG.

Berg, Cynthia J., Hani K. Atrash, Lisa M. Koonin, and Myra Tucker 1996 "Pregnancy-Related Mortality in the United States, 1987–1990." Obstetrics and Gyneclogy 88:161–167.

Crane, James P., Michael L. LeFevre, Renee C. Winborn, Joni K. Evans, Bernard G. Ewigman, Raymond P. Bain, Frederic P. Frigoletto, Donald McNellis, and the RADIUS Study Group 1994 "A Randomized Trial of Prenatal Ultrasonographic Screening: Impact on the Detection, Management, and Outcome of Anomalous Fetuses." American Journal of Obstetricsand Gynecology 171:382–399.

Butler, Jane, Barbara Abrams, Jennifer Parker, James M. Roberts, and Russell K. Laros 1993 "Supportive Nurse-Midwife Care is Associated with a Reduced Incidence of Cesarean Section." American Journal ofObstetrics and Gynecology 168:1407–1413.

Cunningham, E. Gary, Paul C. MacDonald, Norman F. Gant, Kenneth J. Leveno, Larry C. Gilstrap, Gary D. V. Hankins, and Steven L. Clark 1997 Williams Obstetrics, 20th ed. Stamford, Conn.: Appleton and Lange.

Czeizel, Andrew E., and Istvan Dudas 1992 "Prevention of the First Occurrence of Neural-Tube Defects by Periconceptional Vitamin Supplementation." NewEngland Journal of Medicine 327:1832–1835.

Estrin, D. J. 1999 "In Three Former Soviet States, Rates of Abortion Are Declining as Contraceptive Prevalence Increases." International Family Planning Perspectives 25:49–50.

Fraser, Alison M., John E. Brockert, and R. H. Ward 1995 "Association of Young Maternal Age with Adverse Reproductive Outcomes." New England Journalof Medicine 332:1113–1117.

Freedman, Mary Anne, David A. Jillson, Roberta R. Coffin, and Lloyd F. Novick 1986 "Comparison of Complication Rates in First Trimester Abortions Performed by Physician Assistants and Physicians." American Journal of Public Health 76:550–554.

Fu, Haishan, Jacqueline E. Darroch, Stanley K. Henshaw, and Elizabeth Kolb 1998 "Measuring the Extent of Abortion Underreporting in the 1995 National Survey of Family Growth." Family Planning Perspectives 30:128–133, 138.

Gabbe, S., J. R. Niebyl, and J. L. Simpson 1996 Obstetrics:Normal and Problem Pregnancies, 3rd ed. New York: Churchill-Livingston.

Gold, Rachel Benson 1990 Abortion and Women's Health. New York: Alan Guttmacher Institute.

Henshaw, Stanley K. 1998a "Unintended Pregnancy in the United States." Family Planning Perspectives 30:24–29, 46.

——1998b "Abortion Incidence and Services in the United States, 1995–1996." Family Planning Perspectives 30:263–270, 287.

——, Susheela Singh, and Taylor Haas 1999 "Recent Trends in Abortion Rates Worldwide." InternationalFamily Planning Perspectives 25:44–48.

Higby, Kenneth, Elly M. J. Xenakis, and Carl J. Pauerstein 1993 "Do Tocolytic Agents Stop Preterm Labor? A Critical and Comprehensive Review of Efficacy and Safety." American Journal of Obstetrics and Gynecology 168:1247–1259.

Horton, Jacqueline A. (ed.) 1995 The Women's HealthData Book, A Profile of Women's Health in the UnitedStates. Washington, D.C.: Jacobs Institute of Women's Health.

——1998 State Profiles of Women's Health. Washington, D.C.: Jacobs Institute of Women's Health.

Koonin, Lisa M., Jack C. Smith, Merrell Ramick, and Clarice A. Green 1996 "Abortion Surveillance—United States, 1992." Morbidity and Mortality Weekly Review, vol. 45, supp. 3. Atlanta, Ga.: Center for Disease Control.

Lichter, Daniel T., Diane K. McLaughlin, and David C. Ribar 1998 "State Abortion Policy, Geographic Access to Abortion Providers and Changing Family Formation." Family Planning Perspectives 30:281–287.

National Abortion Federation 1999 "Clinic Support Update." Washington, D.C.: NAF (April).

National Center for Health Statistics 1999 Births: FinalData for 1997. Atlanta, Ga.: Center for Disease Control and Prevention.

O'Connor, M. L. 1998 "Home Nurse Visits from Pregnancy until Child's Second Birthday Have Sustained Benefits for Mother and Child." Family PlanningPerspectives 30:47–48.

Paul, Maureen, E. Steve Lichtenberg, Lynn Borgatta, David A. Grimes, and Phillip G. Stubblefield 1999 AClinician's Guide to Medical and Surgical Abortion. New York: Churchill-Livingston.

Population Reports 1997 Caring for Postabortion Complications: Saving Women's Lives. Baltimore, Md.: Population Information Center (September).

Reproductive Freedom News 1999 Anti-Abortion Legislature Swamps Choice Activists. New York: Center for Law and Reproductive Policy (May).

Rooks, Judith S. 1997 Midwifery and Childbirth in America. Philadelphia, Pa.: Temple University Press.

Rosen, Mortimer G., and Janet C. Dickinson 1992 "The Incidence of Cerebral Palsy." American Journal ofObstetrics and Gynecology 167:417–423.

Rosenberg, Irwin H. 1992 "Folic Acid and Neutral-Tube Defects—Time for Action?" New England Journal ofMedicine 327:1875–1877.

Swiss, S. 1993 "Rape as a Crime of War: A Medical Perspective." Journal of the American Medical Association 270:612–615.

Taylor, Paul (ed.) 1980 "Parent-Infant Relationships." New York: Grune and Stratton.

Wallace, H. M., R. P. Nelson, and P. J. Sweeney 1994 Maternal and Child Health Practices, 4th ed. Oakland, Calif.: Third Party.

Winikoff, Beverly 1994 Acceptability of First TrimesterMedical Abortion. New York: Population Council.

World Health Organization, Division of Reproductive Health 1998a Postpartum Care of the Mother and Newborn: A Practical Guide. Geneva, Switzerland: WHO.

——1998b Unsafe Abortion, 3rd ed. Geneva, Switzerland: WHO.

Lynn Borgotta