When the Institute of Medicine (IOM)2 issued its 1975 report on the public health impact of legalized abortion, the scientific evidence on the safety and health effects of legal abortion services was limited. It had been only 2 years since the landmark Roe v. Wade decision had legalized abortion throughout the United States, and nationwide data collection was just under way. Today, the available evidence on abortion’s health effects is quite robust. There is a great deal of related scientific research, including well-designed randomized controlled trials, systematic reviews, and epidemiological studies examining the relative safety of abortion methods and the appropriateness of methods for different clinical circumstances. With this growing body of research, medical and surgical abortion methods have been refined or discontinued, and new techniques have been developed.
In 2016, six private foundations came together to ask the Health and Medicine Division of the National Academies of Sciences, Engineering, and Medicine to conduct a comprehensive review of the state of the science on the safety and quality of legal abortion services in the United States. The sponsors—The David and Lucile Packard Foundation, The Grove Foundation, The JPB Foundation, The Susan Thompson Buffett Foundation, Tara Health Foundation, and William and Flora Hewlett Foundation—asked
1 This summary does not include references. Relevant citations appear in subsequent chapters.
2 In March 2016, the division of the National Academies of Sciences, Engineering, and Medicine that focuses on health and medicine, previously known as the Institute of Medicine (IOM), was renamed the Health and Medicine Division.
that the review focus on the eight research questions listed in Box S-1. The Committee on Reproductive Health Services: Assessing the Safety and Quality of Abortion Care in the U.S. was appointed in December 2016 to conduct the study and prepare this report.
CONTEXT FOR THIS REPORT
What Is Quality Abortion Care?
The committee agreed that two fundamental principles would guide its work: first, that women should expect that the abortion care they receive meets well-established clinical standards for objectivity, transparency, and scientific rigor; and second, that the quality of abortion care should be assessed using the six dimensions of health care quality first described in the 2001 IOM report Crossing the Quality Chasm: A New Health System for the 21st Century (see Box S-2). These dimensions—safety, effectiveness, patient-centeredness, timeliness, efficiency, and equity—have guided public and private efforts to improve U.S. health care delivery at the local, state, and national levels for more than 15 years. Donabedian’s structure-process-outcome framework was also foundational for this report. Figure S-1 illustrates the committee’s adaptation of these concepts for assessing abortion care.
In the immediate years after national legalization, legal abortions increased steadily until peaking in the 1980s. Since then, there has been a steady decline in both the annual number and rate of abortions. Between 1980 and 2014, the abortion rate among U.S. women fell by more than half, from 29.3 to 14.6 per 1,000 women. In 2014, the aggregate number of abortions reached a low of 926,190. The reason for these declines is not fully understood, but they have been attributed to the increasing use of contraceptives, especially long-acting methods (e.g., intrauterine devices [IUDs] and implants), historic declines in the rate of unintended pregnancy, and increasing numbers of state regulations that limit the availability of otherwise legal abortion services.
Since national legalization, most abortions in the United States (91.6 percent) have been performed in early pregnancy (i.e., ≤13 weeks). With advances in technology such as highly sensitive pregnancy tests and the availability of medication abortion, abortions are being performed at increasingly earlier gestation. According to the Centers for Disease Control and Prevention, the percentage of early abortions performed at ≤6 weeks’ gestation increased by 16 percent from 2004 to 2013. In 2013, 38 percent of early abortions occurred at ≤6 weeks’ gestation. The proportion of early-gestation abortions occurring at ≤6 weeks is expected to increase even further as the use of medication abortion becomes more common.
Abortion methods have evolved and improved in the decades since national legalization. Four legal abortion methods—medication,3 aspiration, dilation and evacuation (D&E), and induction—are used in the United States. Today, aspiration is the most common abortion method used in the United States, accounting for almost 68 percent of abortions performed overall in 2013. Its use, however, is likely to decline as the use of medication abortion increases. The percentage of total abortions by the medication method rose from 10.6 to 22.3 percent between 2004 and 2013. In 2014, approximately 45 percent of abortions up to 9 weeks’ gestation were medication abortions, up from 36 percent in 2011. Fewer than 9 percent of abortions are performed after 13 weeks’ gestation—typically by D&E. In 2013, approximately 2 percent of U.S. abortions at 14 weeks’ gestation or later were induction procedures.
3 The terms “medication abortion” and “medical abortion” are used interchangeably in the literature. This report uses “medication abortion” to describe the U.S. Food and Drug Administration (FDA)-approved prescription drug regimen used up to 10 weeks’ gestation.
In 2014, the vast majority of abortions were performed in nonhospital settings: either an abortion clinic (59 percent) or a clinic offering a variety of medical services (36 percent). Fewer than 5 percent of abortions were provided in hospitals.
The overall number of facilities providing abortions—especially specialty abortion clinics—is declining. The greatest proportional decline is in states that have enacted abortion-specific regulations. In 2014, there were 272 abortion clinics in the United States—17 percent fewer than in 2011—and 39 percent of women of reproductive age resided in a county without an abortion provider. Twenty-five states have five or fewer abortion clinics; five states have only one abortion clinic. An estimated 17 percent of women travel more than 50 miles to obtain an abortion.
Women Who Have Abortions
Most women who have abortions are under age 30 (72 percent), are unmarried (86 percent), and are poor or low income (75 percent). Women who have abortions are also more likely to be women of color4 (61.0 percent); half of all women who have abortions are black (24.8 percent) or Hispanic (24.5 percent). This distribution is similar to the racial and ethnic distribution of women with household incomes below 200 percent of the federal poverty level (FPL). Poor women and women of color are also more likely than others to experience an unintended pregnancy.
Unique Regulatory Environment
Abortion is among the most regulated medical procedures in the nation. While a comprehensive legal analysis of abortion regulation is beyond the scope of this report, the committee agreed that it should consider how abortion’s unique regulatory environment relates to the safety and quality of abortion care. Federal restrictions on the distribution of mifepristone (one of the drugs used in medication abortion) also merit attention given its increasing use and the extensive body of research demonstrating its safety and effectiveness.
States play an essential role in ensuring the safety of health care services, especially through their licensure of clinicians and health care facilities. In
4 Includes all nonwhite races and ethnicities.
every state, clinicians and inpatient facilities (e.g., hospitals, rehabilitation centers) must be licensed by a state board or agency to provide health care services legally. When states regulate specific office-based health care procedures, the requirements are usually triggered by the level of sedation that the facility offers. Abortion services are an exception. A wide variety of state regulations affect abortion care, including the type of clinician permitted to perform an abortion, independently of the relevant scope of practice laws (e.g., qualified advanced practice clinicians [APCs] or physicians without hospital privileges may be barred from performing abortions); health insurance coverage (e.g., Medicaid or private insurance plans may be prohibited from paying for abortions); how the informed consent process is conducted (e.g., providers may be required to inform women that abortion increases their risk of breast cancer or mental illness, despite the absence of valid scientific evidence); the abortion method that is used (e.g., D&Es may be banned); the timing and scheduling of procedures (e.g., women may have to wait 18 to 72 hours after a counseling appointment); and the physical attributes of the clinical setting (e.g., procedure room size, corridor width). In most states, the regulations apply to all abortion methods regardless of weeks’ gestation, the use of sedation, or the invasiveness of the procedure.
U.S. Food and Drug Administration’s (FDA’s) Risk Evaluation and Mitigation Strategy (REMS) Program
The distribution and use of mifepristone has been restricted under the requirements of the FDA’s REMS program since 2011. The FDA-approved protocol for medication abortion was updated in 2016 based on extensive clinical research demonstrating the safety of the revised regimen. The revised REMS continues to limit the distribution of Mifeprex (the brand name for mifepristone) to patients in clinics, hospitals, or medical offices under the supervision of a certified prescriber and cannot be sold in retail pharmacies. The committee could not find evidence on how this restriction impacts the safety or quality of abortions.
This report provides a comprehensive review of the state of the science on the safety and quality of abortion services in the United States. As noted earlier (see Box S-1), the committee was charged with answering eight specific research questions. The committee’s conclusions regarding each question appear below. The committee was also asked to offer recommendations regarding the eight questions. However, the committee decided that its conclusions regarding the safety and quality of U.S. abortion care responded comprehensively to the scope of this study. Therefore,
the committee does not offer recommendations for specific actions to be taken by policy makers, health care providers, and others.
The Research Questions
1. What types of legal abortion services are available in the United States? What is the evidence regarding which services are appropriate under different clinical circumstances (e.g., based on patient medical conditions such as previous cesarean section, obesity, gestational age)?
As noted above, four legal abortion methods—medication, aspiration, D&E, and induction—are used in the United States. Length of gestation—measured as the amount of time since the first day of the last menstrual period—is the primary factor in deciding which abortion procedure is the most appropriate. Both medication and aspiration abortions are used up to 10 weeks’ gestation. Aspiration procedures may be used up to 14 to 16 weeks’ gestation.
Mifepristone, which, as noted above, is sold under the brand name Mifeprex, is the only medication specifically approved by the FDA for use in medication abortion. As discussed earlier, the drug’s distribution has been restricted under the requirements of the FDA REMS program since 2011—it may be dispensed only to patients in clinics, hospitals, or medical offices under the supervision of a certified prescriber. To become a certified prescriber, eligible clinicians must register with the drug’s distributor, Danco Laboratories, and meet certain requirements. Retail pharmacies are prohibited from distributing the drug.
When abortion by aspiration is no longer feasible, D&E and induction methods are used. D&E is the superior method; in comparison, inductions are more painful for women, take significantly more time, and are more costly. However, D&Es are not always available to women. The procedure is illegal in Mississippi and West Virginia.5 Elsewhere, access to the procedure is limited because many obstetrician/gynecologists (OB/GYNs) and other physicians lack the requisite training to perform D&Es. Physicians’ access to D&E training is very limited or nonexistent in many areas of the country.
Few women are medically ineligible for abortion. There are, however, specific contraindications to using mifepristone for a medication abortion or induction. The drug should not be used for women with confirmed or suspected ectopic pregnancy or undiagnosed adnexal mass; an IUD in place; chronic adrenal failure; concurrent long-term systemic corticosteroid
5 Both states allow exceptions in cases of life endangerment or severe physical health risk to the woman.
therapy; hemorrhagic disorders or concurrent anticoagulant therapy; allergy to mifepristone, misoprostol, or other prostaglandins; or inherited porphyrias.
Obesity is not a risk factor for women who undergo medication or aspiration abortions (including with the use of moderate intravenous sedation). Research on the association between obesity and complications during a D&E abortion is less certain—particularly for women with Class III obesity (body mass index ≥40) after 14 weeks’ gestation.
A history of a prior cesarean delivery is not a risk factor for women undergoing medication or aspiration abortions, but it may be associated with an increased risk of complications during D&E abortions, particularly for women with multiple cesarean deliveries. Because induction abortions are so rare, it is difficult to determine definitively whether a prior cesarean delivery increases the risk of complications. The available research suggests no association.
2. What is the evidence on the physical and mental health risks of these different abortion interventions?
Abortion has been investigated for its potential long-term effects on future childbearing and pregnancy outcomes, risk of breast cancer, mental health disorders, and premature death. The committee found that much of the published literature on these topics does not meet scientific standards for rigorous, unbiased research. Reliable research uses documented records of a prior abortion, analyzes comparable study and control groups, and controls for confounding variables shown to affect the outcome of interest.
Physical health effects The committee identified high-quality research on numerous outcomes of interest and concludes that having an abortion does not increase a woman’s risk of secondary infertility, pregnancy-related hypertensive disorders, abnormal placentation (after a D&E abortion), preterm birth, or breast cancer. Although rare, the risk of very preterm birth (<28 weeks’ gestation) in a woman’s first birth was found to be associated with having two or more prior aspiration abortions compared with first births among women with no abortion history; the risk appears to be associated with the number of prior abortions. Preterm birth is associated with pregnancy spacing after an abortion: it is more likely if the interval between abortion and conception is less than 6 months (this is also true of pregnancy spacing in general). The committee did not find well-designed research on abortion’s association with future ectopic pregnancy, miscarriage or stillbirth, or long-term mortality. Findings on hemorrhage during a subsequent pregnancy are inconclusive.
Mental health effects The committee identified a wide array of research on whether abortion increases women’s risk of depression, anxiety, and/or posttraumatic stress disorder and concludes that having an abortion does not increase a woman’s risk of these mental health disorders.
3. What is the evidence on the safety and quality of medical and surgical abortion care?
Safety The clinical evidence clearly shows that legal abortions in the United States—whether by medication, aspiration, D&E, or induction—are safe and effective. Serious complications are rare. But the risk of a serious complication increases with weeks’ gestation. As the number of weeks increases, the invasiveness of the required procedure and the need for deeper levels of sedation also increase.
Quality Health care quality is a multidimensional concept. As noted above, six attributes of health care quality—safety, effectiveness, patient-centeredness, timeliness, efficiency, and equity—were central to the committee’s review of the quality of abortion care. Table S-1 details the committee’s conclusions regarding each of these quality attributes. Overall, the committee concludes that the quality of abortion care depends to a great extent on where women live. In many parts of the country, state regulations have created barriers to optimizing each dimension of quality care. The quality of care is optimal when the care is based on current evidence and when trained clinicians are available to provide abortion services.
4. What is the evidence on the minimum characteristics of clinical facilities necessary to effectively and safely provide the different types of abortion interventions?
Most abortions can be provided safely in office-based settings. No special equipment or emergency arrangements are required for medication abortions. For other abortion methods, the minimum facility characteristics depend on the level of sedation that is used. Aspiration abortions are performed safely in office and clinic settings. If moderate sedation is used, the facility should have emergency resuscitation equipment and an emergency transfer plan, as well as equipment to monitor oxygen saturation, heart rate, and blood pressure. For D&Es that involve deep sedation or general anesthesia, the facility should be similarly equipped and also have equipment to provide general anesthesia and monitor ventilation.
Women with severe systemic disease require special measures if they desire or need deep sedation or general anesthesia. These women require
|Quality Attributea||Definition||Committee’s Conclusions|
|Safety||Avoiding injuries to patients from the care that is intended to help them.||Legal abortions—whether by medication, aspiration, D&E, or induction—are safe. Serious complications are rare and occur far less frequently than during childbirth. Safety is enhanced when the abortion is performed as early in pregnancy as possible.|
|Effectivenessb||Providing services based on scientific knowledge to all who could benefit and refraining from providing services to those not likely to benefit (avoiding underuse and overuse, respectively).||
Legal abortions—whether by medication, aspiration, D&E, or induction—are effective. The likelihood that women will receive the type of abortion services that best meet their needs varies considerably depending on where they live. In many parts of the country, abortion-specific regulations on the site and nature of care, provider type, provider training, and public funding diminish this dimension of quality care. The regulations may limit the number of available providers, misinform women of the risks of the procedures they are considering, overrule women’s and clinician’s medical decision making, or require medically unnecessary services and delays in care. These include policies that
|Quality Attributea||Definition||Committee’s Conclusions|
|Patient-Centeredness||Providing care that is respectful of and responsive to individual patient preferences, needs, and values and ensuring that patient values guide all clinical decisions.||
Patients’ personal circumstances and individual preferences (including preferred abortion method), needs, and values may be disregarded depending on where they live (as noted above). The high state-to-state variability regarding the specifics of abortion care may be difficult for patients to understand and navigate. Patients’ ability to be adequately informed in order to make sound medical decisions is impeded when state regulations require that
|Timeliness||Reducing waits and sometimes harmful delays for both those who receive and those who give care.||
The timeliness of an abortion depends on a variety of local factors, such as the availability of care, affordability, distance from the provider, and state requirements for an in-person counseling appointment and waiting periods (18 to 72 hours) between counseling and the abortion.
|Efficiency||Avoiding waste, including waste of equipment, supplies, ideas, and energy.||An extensive body of clinical research has led to important refinements and improvements in the procedures, techniques, and methods for performing abortions. The extent to which abortion care is delivered efficiently depends, in part, on the alignment of state regulations with current evidence on best practices. Regulations that require medically unnecessary equipment, services, and/or additional patient visits increase cost, and thus decrease efficiency.|
|Quality Attributea||Definition||Committee’s Conclusions|
|Equity||Providing care that does not vary in quality because of personal characteristics such as gender, ethnicity, geographic location, and socioeconomic status.||
State-level abortion regulations are likely to affect women differently based on their geographic location and socioeconomic status. Barriers (lack of insurance coverage, waiting periods, limits on qualified providers, and requirements for multiple appointments) are more burdensome for women who reside far from providers and/or have limited resources.
a These attributes of quality health care were first proposed by the Institute of Medicine’s Committee on Quality of Health Care in America in the 2001 report Crossing the Quality Chasm: A New Health System for the 21st Century.
b Elsewhere in this report, effectiveness refers to the successful completion of the abortion without the need for a follow-up aspiration.
further clinical assessment and should have their abortion in either an accredited ambulatory surgery center or hospital.
5. What is the evidence on what clinical skills are necessary for health care providers to safely perform the various components of abortion care, including pregnancy determination, counseling, gestational age assessment, medication dispensing, procedure performance, patient monitoring, and follow-up assessment and care?
Required skills All abortion procedures require competent providers skilled in patient preparation (education, counseling, and informed consent); clinical assessment (confirming intrauterine pregnancy, determining gestation, taking a relevant medical history, and physical examination); pain management; identification and management of expected side effects and serious complications; and contraceptive counseling and provision. To provide medication abortions, the clinician should be skilled in all these areas. To provide aspiration abortions, the clinician should also be skilled in the technical aspects of an aspiration procedure. To provide D&E abortions,
the clinician needs the relevant surgical expertise and sufficient caseload to maintain the requisite surgical skills. To provide induction abortions, the clinician requires the skills needed for managing labor and delivery.
Clinicians that have the necessary competencies Both trained physicians (OB/GYNs, family medicine physicians, and other physicians) and APCs (physician assistants, certified nurse-midwives, and nurse practitioners) can provide medication and aspiration abortions safely and effectively. OB/GYNs, family medicine physicians, and other physicians with appropriate training and experience can provide D&E abortions. Induction abortions can be provided by clinicians (OB/GYNs, family medicine physicians, and certified nurse-midwives) with training in managing labor and delivery.
The extensive body of research documenting the safety of abortion care in the United States reflects the outcomes of abortions provided by thousands of individual clinicians. The use of sedation and anesthesia may require special expertise. If moderate sedation is used, it is essential to have a nurse or other qualified clinical staff—in addition to the person performing the abortion—available to monitor the patient, as is the case for any other medical procedure. Deep sedation and general anesthesia require the expertise of an anesthesiologist or certified registered nurse anesthetist to ensure patient safety.
6. What safeguards are necessary to manage medical emergencies arising from abortion interventions?
The key safeguards—for abortions and all outpatient procedures—are whether the facility has the appropriate equipment, personnel, and emergency transfer plan to address any complications that might occur. No special equipment or emergency arrangements are required for medication abortions; however, clinics should provide a 24-hour clinician-staffed telephone line and have a plan to provide emergency care to patients after hours. If moderate sedation is used during an aspiration abortion, the facility should have emergency resuscitation equipment and an emergency transfer plan, as well as equipment to monitor oxygen saturation, heart rate, and blood pressure. D&Es that involve deep sedation or general anesthesia should be provided in similarly equipped facilities that also have equipment to monitor ventilation.
The committee found no evidence indicating that clinicians that perform abortions require hospital privileges to ensure a safe outcome for the patient. Providers should, however, be able to provide or arrange for patient access or transfer to medical facilities equipped to provide blood transfusions, surgical intervention, and resuscitation, if necessary.
7. What is the evidence on the safe provision of pain management for abortion care?
Nonsteroidal anti-inflammatory drugs (NSAIDs) are recommended to reduce the discomfort of pain and cramping during a medication abortion. Some women still report high levels of pain, and researchers are exploring new ways to provide prophylactic pain management for medication abortion. The pharmaceutical options for pain management during aspiration, D&E, and induction abortions range from local anesthesia, to minimal sedation/anxiolysis, to moderate sedation/analgesia, to deep sedation/analgesia, to general anesthesia. Along this continuum, the physiological effects of sedation have increasing clinical implications and, depending on the depth of sedation, may require special equipment and personnel to ensure the patient’s safety. The greatest risk of using sedative agents is respiratory depression. The vast majority of abortion patients are healthy and medically eligible for all levels of sedation in office-based settings. As noted above (see Questions 4 and 6), if sedation is used, the facility should be appropriately equipped and staffed.
8. What are the research gaps associated with the provision of safe, high-quality care from pre- to postabortion?
The committee’s overarching task was to assess the safety and quality of abortion care in the United States. The committee decided that its findings and conclusions fully respond to this charge. The committee concludes that legal abortions are safe and effective. Safety and quality are optimized when the abortion is performed as early in pregnancy as possible. Quality requires that care be respectful of individual patient preferences, needs, and values so that patient values guide all clinical decisions.
The committee did not identify gaps in research that raise concerns about these conclusions and does not offer recommendations for specific actions to be taken by policy makers, health care providers, and others.
The following are the committee’s observations about questions that merit further investigation.
Limitation of Mifepristone distribution Mifepristone (Mifeprex) is the only medication approved by the FDA for use in medication abortion. Extensive clinical research has demonstrated its safety and effectiveness using the FDA-recommended regimen. Furthermore, few women have contraindications to medication abortion. Nevertheless, as noted earlier, the FDA REMS restricts the distribution of mifepristone. Research is needed on how the limited distribution of mifepristone under the REMS process impacts dimensions of quality, including timeliness, patient-centeredness,
and equity. In addition, little is known about pharmacist and patient perspectives on pharmacy dispensing of mifepristone and the potential for direct-to-patient models through telemedicine.
Pain management There is insufficient evidence to identify the optimal approach to minimizing the pain women experience during an aspiration procedure without sedation. Paracervical blocks are effective in decreasing procedural pain, but the administration of the block itself is painful, and even with the block, women report experiencing moderate to significant pain. More research is needed to learn how best to reduce the pain women experience during abortion procedures.
Research on prophylactic pain management for women undergoing medication abortions is also needed. Although NSAIDs reduce the pain of cramping, women still report high levels of pain.
Availability of providers APCs can provide medication and aspiration abortions safely and effectively, but the committee did not find research assessing whether APCs can also be trained to perform D&Es.
Addressing the needs of women of lower income Women who have abortions are disproportionately poor and at risk for interpersonal and other types of violence. Yet little is known about the extent to which they receive needed social and psychological supports when seeking abortion care or how best to meet those needs. More research is needed to assess the need for support services and to define best clinical practice for providing those services.