Abstract
Background
This study provides a baseline assessment of abortion incidence and service delivery prior to Roe v. Wade being overturned.
Methods
We collected information from all facilities known to have provided abortion services in the United States in 2019 and 2020. We examined abortion incidence by state, region and nationally and combined data on number of abortions with population data to estimate abortion rates. We also examined the number of abortion clinics, trends in medication abortion and service disruptions and changes in abortion protocols that occurred during the COVID‐19 pandemic. We compare these findings to those of our prior Abortion Provider Census, which collected information for 2017.
Results
We documented 930,160 abortions in 2020, an 8% increase from 2017. Between 2017 and 2020, abortion incidence increased in all four regions of the country and in a majority of states. The total number of clinics providing abortion care remained stable nationally but increased in the Midwest and the West and declined in the Northeast and South. There were 492,210 medication abortions in 2020, a 45% increase from 2017. A substantial minority of clinics adjusted protocols in response to COVID, most commonly adopting remote pre‐ and post‐abortion counseling.
Discussion
This study did not address factors behind the increase in abortion. However, this report demonstrates that the need for abortion care was growing just prior to the overturning Roe v. Wade, and the impact of this decision will be even more far‐reaching than previously expected.
INTRODUCTION
Abortion helps women, transgender men, and gender non‐binary individuals with the capacity to become pregnant to control their fertility, thereby maintaining autonomy over their bodies and the trajectory of their lives. The Supreme Court of the United States (US) established the federal constitutional right to an abortion in 1973 in its Roe v. Wade (Roe) ruling. However, on June 24, 2022 the Court overturned this decision in Dobbs v. Jackson Women's Health Organization (Dobbs) ruling that states now have authority to determine the legality of abortion and the circumstances under which it can be provided.
In removing the federal right to an abortion, the Dobbs ruling effectively established multiple state‐based legal regimes for abortion within one country. In the weeks following the decision, 16 states moved to implement near‐total or early abortion bans; as of early November 2022 bans were in effect in 14. 1 As many as 26 states are expected to ban abortion, either through laws that will go into effect in the months after the Supreme Court decision or through special legislative sessions during 2022 or 2023. 2 As of October 2022, legislatures in 17 states that are supportive of abortion rights are enacting protections for abortion providers and expanding access to care. 3
It is against this rapidly changing—and diverging—context that we present the most recent comprehensive estimates of abortion incidence for 2019 and 2020. These estimates describe the landscape of abortion provision in the US before the Dobbs decision.
Trends in abortion
After Roe was handed down in January 1973, abortion increased more or less steadily until 1990, when the number peaked at 1.6 million abortions. 4 The annual number of abortions then declined for several decades until 2017 when it reached its lowest level since 1973 at 863,320. 5 Research has not identified any definitive factor or factors responsible for the long‐term drop in abortion, but several dynamics that likely contributed include delays in sexual activity among adolescents and young adults 6 , 7 and improvements in contraceptive use—including use of long‐acting methods, 8 , 9 dual method use, 10 and improvements in use of methods such as condoms. 11 Additionally, state‐level abortion restrictions have proliferated since 2011. While there is no evidence that, these laws were the main driver behind the national decline in abortion, 12 they did prevent some individuals in some states from obtaining care. 13
Context
Between 2017 and 2020, 18 states passed 75 provisions to protect abortion rights. 14 Some of these expanded access to abortion by requiring Medicaid or private health plans to cover abortion or allowed qualified clinicians such as nurse practitioners, physician assistants, or certified nurse midwives to provide at least some abortion care. During this same period, 25 states enacted 168 abortion restrictions and bans. 14 However, some were met with legal challenges and did not go into effect. In addition, many of these new restrictions were enacted in states where clinics had already been operating in hostile environments, which may have reduced the practical impacts of the new laws.
Clinics are the main source of abortion care in the US, accounting for 95% of abortions in 2017, 5 and changes to the number of these facilities in a given state or region can impact the availability and accessibility of abortion care. Between 2014 and 2017, the total number of clinics increased from 789 to 808 (2%), but clinic numbers fell 6% in the Midwest and 9% the South, and increased in the Northeast (16%) and West (4%). 5 A recent report by researchers at Advancing New Standards in Reproductive Health (ANSIRH) suggests some of these regional patterns may have changed. 15 Between 2017 and 2020 ANSIRH researchers documented a 2% national decline in the total number of facilities that provided abortion care, though the number increased in the Midwest (4%) and decreased in the Northeast (6%). They also reported a continued decline in facilities in the South (10%) and an increase in the West (6%). Number of facilities providing abortion care is not the only, or even necessarily the most important, measure of abortion access. However, depending on the location, affordability and type of care provided by the facilities that opened—or closed—in a given state or region, they may have made abortion more—or less—accessible. 16
The COVID‐19 pandemic had a wide‐ranging impact on all aspects of society, including abortion care availability and accessibility. 17 Governors in several states, mostly in the Midwest and South, issued bans on abortion at the beginning of the pandemic, deeming it non‐essential health care. 18 The pandemic also disrupted clinic workforces and patient flow. One national study of independent abortion clinics found that 50% of facilities reported that staffs were unable to work because of childcare responsibilities and 44%–45% were sick or had to quarantine. 19 These factors may have contributed to reduced access to abortion care, particularly at the beginning of the pandemic.
At the same time, one study found that one‐third of US women wanted to delay or have fewer children in response to COVID and that a similar proportion reported delays or cancellations of contraceptive and other sexual and productive health care. 20 A study of patients at five clinics in the Washington, DC metro area found that 40% of patients calling between September 2020 and March 2021 cited COVID‐related disruptions as a reason for having an abortion. Additionally, these patients were much more likely to report financial hardship resulting from COVID than abortion patients who did not include COVID among their reasons for terminating their pregnancies. 21 These combined dynamics might have contributed to more unintended pregnancies and increased demand for abortion.
This study summarizes findings from the Guttmacher Institute's 2019–2020 Abortion Provider Census, including estimates of abortion incidence and the number of clinics offering abortion care, nationally and by state, as well as other aspects of abortion service provision. The landscape of abortion care is ever evolving and has changed considerably since these data were collected. However, the figures in this article will continue to serve as a baseline for abortion incidence prior to the implementation of state bans on abortion.
METHODS
Survey fielding
Data for this study come from the 19th iteration of the Guttmacher Institute's Abortion Provider Census, which collects data about abortion from all healthcare facilities known or suspected to provide abortion care. The most recent Abortion Provider Census collected data for 2019 and 2020 from facilities across the US and Puerto Rico. This was the first time the Abortion Provider Census included facilities in Puerto Rico and the findings will be summarized in a separate report.
The universe included facilities known to have provided abortions in 2017 (the year of the prior Abortion Provider Census), as well as facilities identified through web‐based searches, media reports, and directories of organizations and associations that work with abortion‐providing facilities. We asked all facilities to provide their 2019 and 2020 abortion caseloads and requested that clinics and physicians' offices (but not hospitals) provide additional information about their facilities' practices for providing abortion care.
In February 2021 we mailed paper questionnaires to the 2131 health facilities known or suspected to have provided abortion care for any part of 2019 and 2020. In addition to materials for mailing back the completed survey, we provided all sites with a link and unique access code so that they could complete the survey online. In March 2021, we began conducting non‐response follow‐up (NRFU), reaching out to clinics, physicians' offices, and higher volume hospital facilities* via telephone, fax, and e‐mail. Data collection and NRFU ended in May 2022 and the fielding team conducted 7131 phone calls, e‐mails, and faxes during the 15‐month period.
We also collected data about the number of abortions from state health departments in order to supplement data obtained directly from facilities and to inform estimates for sites that did not respond. We contacted health departments in 44 states, the District of Columbia, New York City, and Puerto Rico.†
During NRFU we determined that 550 health facilities previously known or suspected to provide abortions did not provide abortions in 2019 or 2020; this included low‐volume facilities from prior surveys and facilities that we contacted because we suspected they might provide abortions. Among these facilities, over three‐quarters (77%) were hospitals, many of which provide abortion occasionally and in low volume. Nearly one‐half (47%) of the 1687 facilities performing one or more abortions during the study period provided information using the questionnaire and an additional 5% offered this information by e‐mail or telephone during NRFU. We used health department data to determine the abortion caseloads of 17% of facilities and we estimated caseloads for 31%. We adopted a variety of strategies and information sources to make caseload estimates, including responses to prior surveys, key informants, media stories, on‐line reviews, and other tools. Some 80% of the facilities for which we had to make estimates were either hospitals (49%) or physicians' offices (31%) (not shown); both of these facility types typically have small abortion caseloads. In turn, 84% of abortions that we counted were based on information obtained from health care facilities: 79% were reported via completed surveys and 5% by e‐mail or telephone during NRFU– slightly lower than reported in the last Abortion Provider Census (89%). 5 We obtained an additional 4% of abortions from state health department data and we estimated the remaining 12% of abortions. This is similar to the 2017 Abortion Provider Census, which used health department data for 2% of abortions and estimated the remaining 9%. The degree to which we estimated data varied by state. We estimated more than 10% of abortions in California (11%), Maryland (14%), Hawaii (20%), New York (30%), Florida (33%) and New Jersey (40%).
The most recent Abortion Provider Census was the first iteration of the survey to have an online option. We used Qualtrics, a secure online survey platform, and we allowed respondents the option of answering the survey in English or Spanish. One‐third (33%) of the 1687 facilities that provided data chose to submit their data online, and an additional 34% submitted paper or PDF surveys. The remaining survey respondents (33%) used an Excel spreadsheet adaptation of the survey, which was made available to respondents providing data for multiple sites.
The Guttmacher Institute's federally registered institutional review board deemed the study exempt.
Analysis
We categorized facilities into four types: hospitals, physicians' offices, specialized abortion clinics, and nonspecialized clinics. Hospitals are sites that typically have operating rooms, emergency departments, and labor and delivery and maternal‐fetal departments. We classified hospital‐affiliated clinics as specialized or non‐specialized clinics. We define physicians' offices as facilities offering fewer than 400 abortions per year whose names suggested they were a private practice. We categorized physicians' offices that provided more than 400 abortions per year as clinics.
We classified clinic facilities where abortion services account for half or more of all patient visits as specialized abortion clinics. Nonspecialized clinics are facilities where fewer than half of all patient visits were for abortion services.
We used data on numbers of women aged 15–44 on July 1, 2019 and July 1, 2020 from the Census Bureau 22 to calculate abortion rates.‡ We used our abortion incidence data in combination with US birth data for the 1‐year periods following July 1, 2019 and 2020 from the National Center for Health Statistics 23 to calculate the national abortion ratio, or the proportion of pregnancies, excluding miscarriages that ended in abortion.
We also examine incidence of medication abortion. More than two‐thirds (67%) of nonhospital facilities provided annual counts of medication abortions including those using mifepristone with misoprostol, methotrexate, or misoprostol alone. More than 98% of medication abortions reported to us were provided with mifepristone and misoprostol. The response rate on this measure varied by facility type and caseload and we constructed weights to account for these differences. When available, we used state health department figures on medication abortion to cross‐check the medication abortion figures we generated.
In 2020, COVID disrupted health care systems across the US (and the globe) and we examined several survey items intended to capture this impact. We first asked clinics “At any point in 2020, did this facility stop or ‘pause’ providing the below types of abortion care.” Types of care included medication abortions, first‐trimester procedural abortions, and second or later trimester procedural abortions. Some 625 clinic facilities answered this question and our analysis of disruptions around COVID focuses on these facilities. Notably, the item did not specify whether care was paused in response to COVID and some clinics may have reported pauses in abortion care in 2020 for other reasons. For example 4% of non‐hospital facilities reported being unable to provide care on one or more days in 2014 in order to be in compliance with local or state regulations. 24
We also adapted items from a prior study 25 to assess whether clinics had adopted five potential changes to their abortion protocols: remote pre‐abortion visits, remote post‐abortion visits, quick pick‐up of mifepristone,§ mailing medications for medication abortion, and using an online pharmacy for prescribing abortion medications. For each item we asked if the service was “offered pre‐COVID (before March 2020),” “began offering in response to COVID,” “offering now,” or “never offered.” Between 554 and 573 clinics (69%–71%) answered each item. Some 61 facilities that responded to the items about disruptions (Question 19) did not answer any items about changes in protocols (Question 20) even though the two were next to each other at the end of the survey. Handwritten notes on two paper surveys next to Question 20 indicated that they were not responding to the item(s) because they had never adopted any of these practices. We considered this as a reason that all 61 facilities did not respond to these items and, in turn, our estimates of the extent to which each protocol was adopted are conservative or “lower bound” estimates.
Compared to clinics that responded to the items about service disruptions and changes in COVID protocols, a larger proportion of those that did not (n = 182) were located in the Northeast (34% vs 26%) and the South (33% vs. 21%), were categorized as specialized abortion clinics (34% vs. 26%), and in the lowest volume category (14% vs. 5% with caseloads <30). Together, facilities that did not respond to the question about service disruptions or COVID protocols accounted for 20% of abortions in 2020.
RESULTS
Abortion incidence
The total number of abortions, the abortion rate, and the abortion ratio in the US all increased between 2017 and 2020 (Table 1). In 2020, 930,160 abortions were provided in clinical settings, an 8% increase from 2017. The 2020 abortion rate of 14.4 abortions per 1000 women aged 15–44 represented a 7% increase from 2017. The abortion ratio also increased and slightly more than one in five pregnancies (births and abortions), 20.6%, ended in abortion in 2020, up from 18.4% in 2017.
TABLE 1.
Number of reported abortions, abortion rate, and abortion ratio, United States, 1973–2020
| Year | Number | Rate a | Ratio b |
|---|---|---|---|
| 1973 | 744,610 | 16.3 | 19.3 |
| 1974 | 898,570 | 19.3 | 22 |
| 1975 | 1,034,170 | 21.7 | 24.9 |
| 1976 | 1,179,300 | 24.2 | 26.5 |
| 1977 | 1,316,700 | 26.4 | 28.6 |
| 1978 | 1,409,600 | 27.7 | 29.2 |
| 1979 | 1,497,670 | 28.8 | 29.6 |
| 1980 | 1,553,890 | 29.3 | 30 |
| 1981 | 1,577,340 | 29.3 | 30.1 |
| 1982 | 1,573,920 | 28.8 | 30 |
| 1983 | 1,575,000 | 28.5 | 30.4 c |
| 1984 | 1,577,180 | 28.1 | 29.7 |
| 1985 | 1,588,550 | 28 | 29.7 |
| 1986 | 1,574,000 | 27.4 | 29.4 c |
| 1987 | 1,559,110 | 26.9 | 28.8 |
| 1988 | 1,590,750 | 27.3 | 28.6 |
| 1989 | 1,566,900 | 26.8 | 27.5 c |
| 1990 | 1,608,600 | 27.4 | 28 c |
| 1991 | 1,556,510 | 26.3 | 27.4 |
| 1992 | 1,528,930 | 25.7 | 27.5 |
| 1993 | 1,495,000 | 25 | 27.4 c |
| 1994 | 1,423,000 | 23.7 | 26.6 c |
| 1995 | 1,359,440 | 22.5 | 25.9 |
| 1996 | 1,360,160 | 22.4 | 25.9 |
| 1997 | 1,335,000 | 21.9 | 25.5 c |
| 1998 | 1,319,000 | 21.5 | 25.1 c |
| 1999 | 1,314,780 | 21.4 | 24.6 |
| 2000 | 1,312,990 | 21.3 | 24.5 |
| 2001 | 1,291,000 | 20.9 | 24.4 c |
| 2002 | 1,269,000 | 20.5 | 23.8 c |
| 2003 | 1,250,000 | 20.2 | 23.3 c |
| 2004 | 1,222,100 | 19.7 | 22.9 |
| 2005 | 1,206,200 | 19.4 | 22.4 |
| 2006 | 1,242,000 | 19.9 | 22.9 c |
| 2007 | 1,209,640 | 19.5 | 21.9 |
| 2008 | 1,212,350 | 19.4 | 22.5 |
| 2009 | 1,151,600 | 18.5 | 22.2 c |
| 2010 | 1,102,670 | 17.7 | 21.7 |
| 2011 | 1,058,490 | 16.9 | 21.2 |
| 2012 | 1,011,000 | 16.1 | 20.4 c |
| 2013 | 958,700 | 15.2 | 19.4 |
| 2014 | 926,190 | 14.6 | 18.8 |
| 2015 | 899,500 | 14.2 | 18.5 c |
| 2016 | 874,080 | 13.7 | 18.3 |
| 2017 | 862,320 | 13.5 | 18.4 |
| 2018 | 885,800 | 13.8 | 18.4 c |
| 2019 | 916,460 | 14.2 | 19.8 |
| 2020 | 930,160 | 14.4 | 20.6 |
Abortions per 1000 women aged 15–44 as of July 1 of each year.
Abortions per 100 pregnancies ending in an abortion or live birth; for each year, the ratio is based on births occurring during the 12‐month period starting in July 1 of that year.
Figures are estimated by interpolation of numbers of abortions counted by Guttmacher and adjustments made to Centers for Disease Control and Prevention Abortion Surveillance Reports.
It is worth noting that patterns in abortion incidence were not uniform across the time periods of 2017–2019 and 2019–2020 (Table 2). Nationally, most of the increase in abortion numbers took place between 2017 and 2019, and only increased 1% between 2019 and 2020.
TABLE 2.
Abortion incidence, abortion rates, and change in rates, nationally, by region and by state, 2017, 2019 and 2020
| Number of abortions | Abortion rate | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| % Change | % Change | |||||||||||
| 2017 | 2019 | 2020 | 17–19 | 19–20 | 17–20 | 2017 | 2019 | 2020 | 17–19 | 19–20 | 17–20 | |
| United States | 862,320 | 916,460 | 930,160 | 6 | 1 | 8 | 13.5 | 14.2 | 14.4 | 5 | 1 | 7 |
| Northeast | 224,310 | 235,920 | 228,100 | 5 | −3 | 2 | 20.5 | 21.8 | 21.2 | 6 | −3 | 3 |
| Connecticut | 11,910 | 11,990 | 11,170 | 1 | −7 | −6 | 17.7 | 17.9 | 16.7 | 1 | −7 | −6 |
| Maine | 2040 | 2100 | 2370 | 3 | 13 | 16 | 8.8 | 9.0 | 10.1 | 2 | 12 | 15 |
| Massachusetts | 18,590 | 19,050 | 17,060 | 2 | −10 | −8 | 13.5 | 13.6 | 12.2 | 1 | −10 | −10 |
| New Hampshire | 2210 | 2090 | 2050 | −5 | −2 | −7 | 9.2 | 8.5 | 8.3 | −8 | −2 | −10 |
| New Jersey | 48,110 | 48,280 | 48,830 | 0 | 1 | 1 | 28.0 | 28.8 | 29.2 | 3 | 1 | 4 |
| New York | 105,380 | 117,140 | 110,360 | 11 | −6 | 5 | 26.3 | 30.2 | 28.8 | 15 | −5 | 10 |
| Pennsylvania | 31,260 | 31,250 | 32,260 | 0 | 3 | 3 | 13.1 | 13.1 | 13.6 | 0 | 4 | 4 |
| Rhode Island | 3500 | 2840 | 2760 | −19 | −3 | −21 | 16.7 | 13.6 | 13.3 | −19 | −2 | −20 |
| Vermont | 1300 | 1190 | 1230 | −8 | 3 | −5 | 11.4 | 10.4 | 10.7 | −9 | 3 | −6 |
| Midwest | 133,120 | 143,360 | 146,540 | 8 | 2 | 10 | 10.2 | 10.9 | 11.2 | 7 | 3 | 10 |
| Illinois | 42,080 | 52,220 | 52,780 | 24 | 1 | 25 | 16.6 | 20.9 | 21.3 | 26 | 2 | 28 |
| Indiana | 7710 | 7720 | 7880 | 0 | 2 | 2 | 5.9 | 5.9 | 6.0 | −1 | 2 | 2 |
| Iowa | 3760 | 3470 | 3510 | −8 | 1 | −7 | 6.3 | 5.8 | 5.9 | −8 | 2 | −6 |
| Kansas | 6830 | 6740 | 8180 | −1 | 21 | 20 | 12.2 | 12.0 | 14.5 | −2 | 21 | 19 |
| Michigan | 26,630 | 29,160 | 31,500 | 10 | 8 | 18 | 14.2 | 15.5 | 16.8 | 9 | 8 | 18 |
| Minnesota | 10,740 | 11,190 | 11,060 | 4 | −1 | 3 | 10.1 | 10.3 | 10.2 | 2 | −1 | 1 |
| Missouri | 4710 | 1520 | 170 | −68 | −89 | −96 | 4.0 | 1.3 | 0.1 | −68 | −92 | −98 |
| Nebraska | 2020 | 2150 | 2200 | 6 | 2 | 9 | 5.5 | 5.8 | 5.9 | 5 | 2 | 7 |
| North Dakota | 1160 | 1120 | 1170 | −3 | 4 | 1 | 7.9 | 7.5 | 7.8 | −5 | 4 | −1 |
| Ohio | 20,630 | 20,400 | 20,990 | −1 | 3 | 2 | 9.4 | 9.2 | 9.5 | −2 | 3 | 1 |
| South Dakota | 500 | 420 | 130 | −16 | −68 | −74 | 3.1 | 2.5 | 0.8 | −19 | −68 | −74 |
| Wisconsin | 6360 | 7260 | 6960 | 14 | −4 | 9 | 5.9 | 6.6 | 6.4 | 12 | −3 | 8 |
| South | 295,290 | 308,280 | 320,410 | 4 | 4 | 9 | 12.1 | 12.5 | 12.9 | 3 | 3 | 7 |
| Alabama | 6110 | 5910 | 5700 | −3 | −4 | −7 | 6.4 | 6.2 | 6.0 | −3 | −3 | −6 |
| Arkansas | 3200 | 2920 | 3250 | −9 | 11 | 2 | 5.5 | 5.0 | 5.6 | −9 | 12 | 2 |
| Delaware | 1900 | 2040 | 1830 | 7 | −10 | −4 | 10.5 | 11.3 | 10.0 | 8 | −12 | −5 |
| District of Columbia | 5630 | 9900 | 9410 | 76 | −5 | 67 | 30.2 | 51.8 | 48.9 | 72 | −6 | 62 |
| Florida | 71,050 | 72,210 | 77,400 | 2 | 7 | 9 | 18.6 | 18.5 | 19.7 | −1 | 6 | 6 |
| Georgia | 36,330 | 39,980 | 41,620 | 10 | 4 | 15 | 16.9 | 18.3 | 18.9 | 8 | 3 | 12 |
| Kentucky | 3200 | 3660 | 4080 | 14 | 11 | 28 | 3.8 | 4.3 | 4.8 | 13 | 12 | 26 |
| Louisiana | 9920 | 8150 | 7360 | −18 | −10 | −26 | 10.6 | 8.8 | 8.0 | −17 | −9 | −25 |
| Maryland | 29,800 | 30,030 | 30,750 | 1 | 2 | 3 | 25.0 | 25.3 | 25.9 | 1 | 2 | 4 |
| Mississippi | 2550 | 3190 | 3560 | 25 | 11 | 40 | 4.3 | 5.5 | 6.1 | 28 | 11 | 42 |
| North Carolina | 29,500 | 29,320 | 31,850 | −1 | 9 | 8 | 14.6 | 14.2 | 15.3 | −3 | 8 | 5 |
| Oklahoma | 4780 | 9070 | 9690 | 90 | 7 | 103 | 6.2 | 11.7 | 12.4 | 89 | 6 | 100 |
| South Carolina | 5120 | 5000 | 5300 | −2 | 6 | 4 | 5.3 | 5.1 | 5.3 | −4 | 4 | 0 |
| Tennessee | 12,140 | 9970 | 10,850 | −18 | 9 | −11 | 9.2 | 7.5 | 8.1 | −18 | 8 | −12 |
| Texas | 55,440 | 59,280 | 58,020 | 7 | −2 | 5 | 9.4 | 9.8 | 9.5 | 4 | −3 | 1 |
| Virginia | 17,210 | 16,470 | 18,740 | −4 | 14 | 9 | 10.2 | 9.7 | 11.0 | −5 | 13 | 8 |
| West Virginia | 1430 | 1170 | 990 | −18 | −15 | −31 | 4.4 | 3.7 | 3.1 | −16 | −16 | −30 |
| West | 209,600 | 228,900 | 235,120 | 9 | 3 | 12 | 13.5 | 14.5 | 14.9 | 7 | 3 | 10 |
| Alaska | 1260 | 1320 | 1240 | 5 | −6 | −2 | 8.6 | 9.1 | 8.6 | 6 | −5 | 0 |
| Arizona | 12,400 | 13,020 | 13,320 | 5 | 2 | 7 | 9.2 | 9.3 | 9.3 | 1 | 0 | 1 |
| California | 132,680 | 150,660 | 154,060 | 14 | 2 | 16 | 16.4 | 18.7 | 19.2 | 14 | 3 | 17 |
| Colorado | 12,390 | 12,410 | 13,420 | 0 | 8 | 8 | 10.9 | 10.5 | 11.2 | −4 | 7 | 3 |
| Hawaii | 3200 | 3150 | 3130 | −2 | −1 | −2 | 12.0 | 12.1 | 12.1 | 1 | 0 | 1 |
| Idaho | 1290 | 1520 | 1690 | 18 | 11 | 31 | 3.9 | 4.4 | 4.8 | 13 | 9 | 23 |
| Montana | 1580 | 1600 | 1630 | 1 | 2 | 3 | 8.3 | 8.2 | 8.2 | −1 | 0 | −1 |
| Nevada | 9690 | 9920 | 11,010 | 2 | 11 | 14 | 16.4 | 16.4 | 17.9 | 0 | 9 | 9 |
| New Mexico | 4620 | 4470 | 5880 | −3 | 32 | 27 | 11.7 | 11.2 | 14.7 | −4 | 31 | 26 |
| Oregon | 9640 | 9130 | 8560 | −5 | −6 | −11 | 11.9 | 11.0 | 10.3 | −8 | −6 | −13 |
| Utah | 2990 | 3030 | 3120 | 1 | 3 | 4 | 4.4 | 4.3 | 4.4 | −2 | 2 | 0 |
| Washington | 17,740 | 18,570 | 17,980 | 5 | −3 | 1 | 12.1 | 12.2 | 11.7 | 1 | −4 | −3 |
| Wyoming | 140 | 90 | 100 | −36 | 14 | −29 | 1.3 | 0.8 | 0.9 | −38 | 12 | −31 |
Note: State‐based rates use the population of women of reproductive age (15–44) living in the state as the denominator, however all patients (including those traveling from out of state) are considered part of the numerator.
Between 2017 and 2020, abortion incidence increased in all four regions of the country (Figure 1) and was largest in the Midwest (10%) and the West (12%). Abortion incidence increased in 33 states and DC and decreased in 17 states. The degree of change varied substantially across states, even within regions. Between 2017 and 2020 abortion incidence increased by more than 10% in 12 states, including large ones such as California, Georgia, Illinois, and Michigan. States where the number of abortions declined by more than 10% included Louisiana, Missouri, Oregon, Rhode Island, South Dakota, Tennessee, West Virginia, and Wyoming.
FIGURE 1.
United States Census Regions
Changes in state abortion rates between 2017 and 2020 tracked with changes in abortion incidence,¶ with a few exceptions. In Montana, North Dakota, and Washington, the number of abortions increased 1%–5% between 2017 and 2020 while abortion rates decreased by 1%–4%. These changes are due to increases in the population of women of reproductive age. Even though there were more abortions, there were also more women 15–44 in 2020 than in 2017. In Utah, abortion incidence increased 4% while the rate remained stable.
Areas with the highest abortion rates in 2020 were the District of Columbia, Florida, Illinois, Maryland, New Jersey, and New York. Rates were lowest in Missouri, South Dakota, Utah, West Virginia, and Wyoming. Notably, our study measures abortion by state of occurrence and does not account for individuals crossing state lines for abortion care. With the exception of Utah, states with the lowest abortion rates in 2020 had anywhere from 52% (West Virginia) to 99% (Missouri) of residents obtaining abortions out of state. 26
Abortion facilities
We identified 1603 health care facilities that provided abortions in 2020 (Table 3); in 2017 the comparable figure was 1587. Table A1 shows the number of all abortion providing facilities, including hospitals and physicians' offices, by state. While hospitals accounted for one‐third of facilities providing abortion care in 2020, they only accounted for 3% of all abortions. This is because 71% of hospitals provide fewer than 30 abortions per year (not shown), and these are often restricted to high‐risk pregnancies and/or those for fetal indications. 27
TABLE 3.
Number and percentage distribution of abortion providers and of abortions, by provider type, and caseload, 2017 and 2020
| Facilities | Abortions | |||||||
|---|---|---|---|---|---|---|---|---|
| Number | % | Number | % | |||||
| 2017 | 2020 | 2017 | 2020 | 2017 | 2020 | 2017 | 2020 | |
| Total | 1587 | 1603 | 100 | 100 | 862,320 | 930,160 | 100 | 100 |
| Facility type | ||||||||
| Hospital | 518 | 530 | 33 | 33 | 28,760 | 28,010 | 3 | 3 |
| Abortion clinic | 253 | 227 | 16 | 14 | 519,180 | 499,420 | 60 | 54 |
| Nonspecialized clinic | 555 | 580 | 35 | 36 | 302,860 | 395,550 | 35 | 43 |
| Physicians' offices a | 261 | 266 | 16 | 17 | 11,510 | 7190 | 1 | 1 |
| Facility caseload | ||||||||
| 1–29 | 609 | 636 | 38 | 40 | 5310 | 5610 | 1 | 1 |
| 30–399 | 474 | 432 | 30 | 27 | 75,280 | 67,220 | 9 | 7 |
| 400–999 | 230 | 241 | 14 | 15 | 148,140 | 163,600 | 17 | 18 |
| 1000–4999 | 255 | 275 | 16 | 17 | 499,010 | 555,650 | 58 | 60 |
| ≥5000 | 19 | 19 | 1 | 1 | 134,580 | 138,080 | 16 | 15 |
Physicians' offices reporting 400 or more abortions a year are classified as clinics.
Some 50% of facilities providing abortion care were clinics, but they accounted for 96% of all abortions. While there were more non‐specialized (n = 580) than specialized clinics (n = 227) in 2020, clinics specializing in abortion care provided the majority of abortions (54%). Still, the share of abortions provided by non‐specialized facilities increased from 35% in 2017 to 43% in 2020.
Abortion clinics
The total number of clinics was virtually unchanged between 2017 and 2020, at 808 and 807, respectively, although there were differences by region and state (Table 4). The number of clinics increased in the Midwest (11%) and in the West (6%) and decreased in the Northeast (9%) and the South (3%). The Midwest had 10 more clinics in 2020 than in 2017 and increases were largest in Illinois (5 additional clinics), Michigan (3) and Minnesota (3). The West had 18 more clinics in 2020 than in 2017 with an additional 12 clinics in California and five in Colorado. In the Northeast, New York had nine fewer clinics and, in the South, Florida had seven fewer. The loss of only one or two clinics can have significant impacts on states that had few to begin with. For example, the loss of two clinics in Iowa resulted in 25% fewer clinics in that state and the loss of one clinic had a substantial impact on some Southern states like Arkansas, Louisiana, and South Carolina. In 2020 six states had only one clinic: Mississippi, Missouri, North Dakota, Rhode Island, South Dakota, and West Virginia.
TABLE 4.
Total number of clinics providing abortion care, 2017 and 2020; percentage of counties without a clinic and percentage of women living in those counties; all by region and by state
| Region and state | Number of clinics | Percentage of counties without a clinic, 2020 | Percentage of women in counties with no clinic, 2020 | ||
|---|---|---|---|---|---|
| 2017 | 2020 | Percentage change, 2017–2020 | |||
| United States total | 808 | 807 | −0.1 | 89 | 38 |
| Northeast | 245 | 223 | −9 | 52 | 18 |
| Connecticut | 26 | 20 | −23 | 0 | 0 |
| Maine | 16 | 17 | 6 | 13 | 16 |
| Massachusetts | 19 | 17 | −11 | 50 | 14 |
| New Hampshire | 4 | 4 | 0 | 60 | 30 |
| New Jersey | 41 | 37 | −10 | 29 | 21 |
| New York | 113 | 104 | −8 | 37 | 6 |
| Pennsylvania | 18 | 17 | −6 | 82 | 38 |
| Rhode Island | 2 | 1 | −50 | 80 | 36 |
| Vermont | 6 | 6 | 0 | 64 | 38 |
| Midwest | 91 | 101 | 11 | 95 | 54 |
| Illinois | 25 | 30 | 20 | 89 | 32 |
| Indiana | 6 | 7 | 17 | 95 | 66 |
| Iowa | 8 | 6 | −25 | 95 | 65 |
| Kansas | 4 | 4 | 0 | 98 | 60 |
| Michigan | 21 | 24 | 14 | 86 | 34 |
| Minnesota | 7 | 10 | 43 | 95 | 58 |
| Missouri | 3 | 1 | −67 | 99 | 94 |
| Nebraska | 3 | 3 | 0 | 97 | 39 |
| North Dakota | 1 | 1 | 0 | 98 | 72 |
| Ohio | 9 | 10 | 11 | 93 | 55 |
| South Dakota | 1 | 1 | 0 | 98 | 76 |
| Wisconsin | 3 | 4 | 33 | 96 | 68 |
| South | 195 | 189 | −3 | 94 | 52 |
| Alabama | 5 | 5 | 0 | 93 | 59 |
| Arkansas | 3 | 2 | −33 | 99 | 86 |
| Delaware | 4 | 3 | −25 | 33 | 19 |
| District of Columbia a | 5 | 5 | 0 | 0 | 0 |
| Florida | 65 | 58 | −11 | 75 | 25 |
| Georgia | 15 | 14 | −7 | 95 | 55 |
| Kentucky | 1 | 2 | 100 | 99 | 82 |
| Louisiana | 4 | 3 | −25 | 95 | 75 |
| Maryland | 25 | 23 | −8 | 63 | 23 |
| Mississippi | 1 | 1 | 0 | 99 | 92 |
| North Carolina | 14 | 16 | 14 | 91 | 53 |
| Oklahoma | 4 | 5 | 25 | 96 | 53 |
| South Carolina | 4 | 3 | −25 | 93 | 71 |
| Tennessee | 8 | 7 | −13 | 95 | 61 |
| Texas | 21 | 24 | 14 | 96 | 44 |
| Virginia | 16 | 17 | 6 | 92 | 79 |
| West Virginia | 1 | 1 | 0 | 98 | 90 |
| West | 276 | 294 | 6 | 78 | 15 |
| Alaska | 4 | 4 | 0 | 87 | 33 |
| Arizona | 8 | 8 | 0 | 80 | 18 |
| California | 161 | 173 | 7 | 38 | 3 |
| Colorado | 18 | 23 | 28 | 77 | 26 |
| Hawaii | 4 | 4 | 0 | 40 | 5 |
| Idaho | 3 | 3 | 0 | 95 | 67 |
| Montana | 5 | 6 | 20 | 91 | 47 |
| Nevada | 7 | 9 | 29 | 88 | 9 |
| New Mexico | 6 | 6 | 0 | 91 | 48 |
| Oregon | 16 | 17 | 6 | 75 | 22 |
| Utah b | 2 | 2 | 0 | 97 | 64 |
| Washington | 40 | 37 | −8 | 56 | 9 |
| Wyoming | 2 | 2 | 0 | 96 | 96 |
2017 differs from published figure (of 4) as we determined during the most recent round of data collection that a facility had been misclassified in the prior round.
2017 differs from published figure (of 3) as we determined during the most recent round of data collection that a facility had been misclassified as a clinic in the prior round.
The negligible change in the number of clinics nationwide masks turnover among existing clinics. Of the 807 clinic facilities providing abortions in 2020, 95 had not been providing this care in 2017 (not shown). Similarly, of the 808 clinic facilities providing abortions in 2017, 86 had stopped doing so or had closed by 2020.**
In 2020, 89% of US counties did not have a clinic facility that provided abortion care and 38% of women aged 15–44 lived in these counties (Table 4); these figures are identical to those in 2017. 5 In six states, fewer than 10% of women lived in a county without a clinic facility: California, Connecticut, Hawaii, Nevada, New York, and Washington. In Mississippi, Missouri, West Virginia, and Wyoming, at least 90% of women lived in a county without such a clinic.
Medication abortion
In 2020, 492,210 medication abortions were provided in nonhospital facilities (Table 5), a 45% increase from 2017. Medication abortion accounted for the majority of all abortions in 2020 (53%), compared to 39% in 2017. 5 It is worth noting that there were 412,130 medication abortions in 2019 (not shown). The relative increase between 2019 and 2020 (19%) was only slightly lower than the one for the two‐year period of 2017 and 2019 (21%).
TABLE 5.
Number of medication abortions and share of all abortions provided by medication, by facility type and facility caseload, 2017 and 2020
| 2017 | 2020 | |||
|---|---|---|---|---|
| N | % | N | % | |
| United States total | 339,650 | 39 | 492,210 | 53 |
| Facility type | ||||
| Abortion clinics | 181,280 | 35 | 232,240 | 47 |
| Other clinics | 153,350 | 51 | 254,870 | 64 |
| Physicians' offices | 5020 | 44 | 4960 | 70 |
| Facility caseload (excluding hospitals) | ||||
| 1–29 | 1230 | 50 | 1510 | 55 |
| 30–399 | 39,880 | 69 | 39,950 | 83 |
| 400–999 | 84,330 | 60 | 109,330 | 70 |
| ≥1000 | 214,190 | 34 | 341,270 | 49 |
Note: 2017 Data from Jones et al. 5
The share of medication abortions increased for all facility types and caseloads. In 2020, the majority of abortions provided by physicians' offices (70%) and non‐specialized clinics (64%) were medication abortions, up from 44% and 51%, respectively, in 2017. Among abortion clinics, the number and proportion of abortions that were done with medication increased from 35% in 2017 to 47% in 2020 though in both years the majority of abortions were procedural. Similarly, between 2017 and 2020 the number and share of abortions provided with medication(s) increased across all facility caseloads, but in the largest facilities the majority of abortions in both years were procedural (66% and 51%). In 2020, at least 26% of all nonhospital facilities (including physicians' offices) and 31% of clinics provided only medication abortion (not shown); these figures were very similar to 2017 (25% and 30%, respectively).
Changes around COVID
Among clinics that responded to the items assessing abortion protocols in relation to the pandemic, less than 10% had any of the five practices in place prior to COVID (Table 6). Most commonly, 8% indicated that they had provided remote post‐abortion visits prior to COVID and 1%–3% had provided any of the other services.
TABLE 6.
Clinic changes to abortion protocols in relation to COVID
| N | % | ||||
|---|---|---|---|---|---|
| Abortion protocol | Adopted prior to COVID | Adopted in response to COVID | Currently offer | Never offered | |
| Remote pre‐abortion visit | 625 | 3 | 34 | 22 | 63 |
| Remote post‐abortion visit | 625 | 8 | 42 | 30 | 53 |
| Quick pick‐up of mifepristone | 625 | 3 | 16 | 11 | 80 |
| Mailing medications for medication abortion | 625 | 3 | 5 | 5 | 91 |
| Online pharmacy for medication abortion | 625 | 1 | 3 | 4 | 95 |
The most common changes in protocol in response to COVID applied to remote visits: 42% of responding clinics adopted remote post‐abortion visits and 34% adopted remote pre‐abortion visits. Just under one in five responding clinics, 16%, allowed for people to drop into the clinic to get medication abortion drugs (“quick pick up”) and substantially lower proportions reported mailing abortion drugs (5%) or using an online pharmacy to mail medications to patients (3%).
Not all clinics that adopted these protocols were still using them when they filled out the survey (2021 for most of the clinics). Remote post‐abortion visits were still offered by 30% and remote pre‐abortion visits by 22%. Only 11% were still offering quick pick‐up of medication abortion drugs. Four to 5% were mailing medications and/or using an online pharmacy.
Of the 625 clinics that responded to the item about pausing abortion care, 15% reported service disruptions during COVID (not shown). Disruptions were reported by half or more of all clinics in Alabama (5), Arkansas (2), Oklahoma (5), South Dakota (1 clinic in the state), Tennessee (7), West Virginia (1), and Wyoming (2). Among the responding clinics, disruptions were most common in the South (25%) and Midwest (22%) and disruptions were also reported by 16% of clinics in the Northeast (including 35% of clinics in Connecticut and 13% in New York). Some 7% of clinics providing abortions in the West reported service disruptions around the time of COVID.
DISCUSSION
Our findings demonstrate that the long‐term decline in abortion in the US had reversed. The number of abortions increased 8% between 2017 and 2020 and the abortion rate increased 7%. The nationwide increase was not uniform and we found heterogeneity across states and across time periods within states: some states saw steady increases between 2017 and 2020 while others saw increases between 2017 and 2019 but decreases between 2019 and 2020, or vice‐versa. Below, we provide some contextual considerations and highlight findings that merit further inquiry.
Clinics play an important role in abortion access and, as in the past, we have found that the overwhelming majority of abortions (96%) were provided by these facilities. Notably, the total number of clinics providing this care nationally was stable between 2017 and 2020. While there were changes in clinic numbers for many states, they were not always in ways to suggest increased access to abortion. For example, the number of clinics fell in eight states where the number of abortions increased: Arkansas, Florida, Georgia, Maryland, New Jersey, New York, South Carolina, and Washington. In Oregon the number of clinics increased while the number of abortions declined. Indicators such as percentage of counties without a clinic and percent of women living in them remained stable nationally and for most states. Still, it is potentially meaningful that the regions that had the biggest increase in abortion incidence—the Midwest and the West—also had increases in number of clinics.
Changes in pregnancy prevention strategies can also impact need for abortion. The most recent available data do not suggest substantial changes in types of contraceptives used. The National Survey of Family Growth for 2015–2017 28 and 2017–2019 29 reports that 65% of women 15–49 were using contraception at the time they were interviewed and similar proportions during both time periods were using a method of long‐acting reversible contraceptives (10%), male or female sterilization (25% and 24%), or oral contraceptive pills (13% and 14%). 30 It is possible that individuals and couples used methods less consistently or had more sex more frequently, but these are unlikely to explain the scale of the increase in abortion documented in this study. Along with the increase in abortion, there were 241,600 fewer births in 2020 than 2017 and the birth rate decreased 6%. 23 These trends mean that compared to 2017, fewer people were getting pregnant in 2020 and, among those who did, a larger proportion were having abortions. Abortion is only one piece of a larger, still unclear, story about what drives changes in fertility in the US.
Medication abortion provided in a clinical setting increased substantially during the study period and, for the first time, in 2020 accounted for the majority of abortions. While mifepristone may have made abortion more accessible—for example, allowing abortion to be provided via telemedicine 31 , 32 —it is unlikely to have been the primary driver behind the increase in abortion during the time period covered by this study. Medication abortion has been increasing since it was introduced in 2000, 33 even while abortion was declining nationally; greater use of medication abortion does not automatically increase abortion incidence. Additionally, the main increase in abortion incidence nationally, and in many states, was concentrated between 2017 and 2019, but reliance on medication abortion increased about the same over the two periods.
It is likely that the larger increase in medication abortion between 2019 and 2020 was due, at least in part, to COVID. Compared to procedural abortions, medication abortions require less, or even no, face‐to‐face contact between patient and provider, 34 and the drugs can be obtained via quick pick‐up or the mail. 25 Indeed, we found that a substantial minority of clinics adopted these strategies in response to COVID and some were still offering them in 2021. We also found that the number and share of clinics offering only medication abortion remained stable, suggesting changes in the availability of medication relative to procedural abortions was not a primary driver behind the increased reliance on medication abortion.
COVID also impacted abortion care in other ways. States were affected by the pandemic at different times and state government responses varied widely. Some states, mostly in the Midwest and South, implemented temporary bans on abortion by deeming it “nonessential.” While some bans were only in effect for several days or weeks, they created confusion and disruptions to care. They may have contributed to the decreases in abortion between 2019 and 2020 in Alabama, Alaska, Louisiana, and Texas. 18 , 35 It is worth noting that some state's COVID abortion bans only applied to procedural abortions; 18 limitations on procedural, but not medication, abortions could have contributed to the increase in medication abortions.
It is likely that many facilities—including those in states with abortion bans—still had to reduce services because clinic staffs were dealing with illness, quarantines, remote schooling, and childcare. 19 , 36 For example, COVID had intense and early impacts on New York state. At least 13% of clinics in New York reported disruptions and these dynamics might explain why abortion declined 6% between 2019 and 2020 after an 11% increase between 2017 and 2019. Similarly, the one clinic in South Dakota that provided abortion care had to halt services for 7 months in 2020 in response to COVID 35 and the number of abortions declined 69%.
In states where abortion care was not severely disrupted by COVID, clinics may have served more people traveling from states where abortion was less accessible. These dynamics may have contributed to increases in abortion between 2019 and 2020 in Colorado, Florida, Kansas, New Mexico, and Oklahoma. 37 , 38
Other factors that may have impacted abortion incidence and provision were unrelated to COVID. In Missouri the number of abortions decreased 96%, from 4710 in 2017 to 170 in 2020. Even in 2017, the majority of Missouri residents went out of state for abortion care. 39 This pattern was magnified between 2017 and 2020 in response to intrusive regulation by state health officials and the number of Missouri residents that obtained abortions in Illinois, in particular, increased substantially. 26
The overwhelming majority of people with Medicaid coverage use it to pay for their abortions when they live in states with state funding for abortion care. 40 Changes in Medicaid policy in two states may have made abortion more accessible to residents. In January 2018, Illinois allowed the use of state Medicaid funds to pay for abortion care and the number of abortions rose 24% between 2017 and 2019. Illinois also had five more clinics in 2020 as compared to 2017. Similarly, in January 2020, Maine allowed the use of state Medicaid funds to pay for, and required private health insurance plans to cover, abortion care; this coincided with a 13% increase in abortion between 2019 and 2020 compared to 3% between 2017 and 2019. On a more concerning note, West Virginia stopped allowing Medicaid coverage of abortion in 2018 and this may have contributed to the decline in abortion incidence in the state, 18% between 2017 and 2019 and 31% overall.
Slightly less than half of abortions occur in states that do not have Medicaid funding for abortion. 41 In 2014, one in five people obtaining abortions in these states relied on financial assistance to cover at least some of the costs of this care. 40 Grassroots organizations and local abortion funds have existed for decades and one large, national fund subsidized more than 10% of abortions that occurred in 2020. 42 There is some evidence that money distributed by grassroots organizations and local abortion funds, and the number of people who received it, increased substantially between 2017 and 2020. 43 If this same dynamic applied to large national abortion funds, this could have allowed more people to access abortions, especially in states where Medicaid could not be used to pay for this care.
LIMITATIONS
This study had a number of limitations. Although 84% of the abortions we counted in 2020 were based on information provided by individuals working at or for the facilities from which the data were collected, we estimated caseloads at facilities that accounted for 12% of abortions and used state health department data for the remaining 4% of abortions. This problem was particularly pronounced in six states, including larger ones such as New York (30%), Florida (33%), and New Jersey (40%). Readers and investigators should keep in mind that abortion estimates are therefore less precise in such states.
Our study underestimates the true amount of the increase in abortion since we only measure abortions that occurred in clinical settings. Research documented that more than 55,000 US residents requested abortion drugs from one online provider during the study period 44 and many of these requests undoubtedly resulted in self‐managed abortions. US residents may also have obtained abortion drugs for self‐managed abortion through other websites. 45
We are aware that our study did not capture some facilities that provide a small number of abortions per year—especially in hospitals and private physicians' offices. This, too, means that our study undercounts that actual number of abortions that occurred.
Our findings on disruptions to abortion services and changes to abortion provision around the time of COVID were obtained using data from the 77% of clinics. Facilities in the South were over‐represented among those that did not respond to items about disruptions, but among those that did, clinics in this region reported the highest levels of disruptions. The opposite dynamic applies to clinics in the West. It is possible, then, that the prevalence of disruption in abortion provision is actually higher than our estimates suggest.
These analyses focus on abortion by state of occurrence and, for some states, do not reflect use of abortion by residents. In 2020, 9% of individuals obtaining abortions crossed state lines to obtain care. For example, recently published data show that 26% of abortions reported to have occurred to residents of Idaho and 88% of those to residents of Wyoming were obtained in other states. 26 Similarly, although the District of Columbia had the highest abortion rate in the country in 2019 and 2020, the majority of abortions provided in the District of Columbia in 2020 were for nonresidents, most commonly individuals from Maryland or Virginia. 26
CONCLUSIONS
This study demonstrates that the 30‐year decline in abortion had reversed, underscoring that the need for abortion care had grown just as the US Supreme Court overturned Roe v. Wade. State bans on abortion care may decrease the number of abortions that occur but will not eliminate, or even reduce, need or desire for abortion. Further, following the Supreme Court's Dobbs ruling the number of clinics (and other health care facilities) offering abortion care undoubtedly dropped and the number of women of reproductive age living in a county without a clinic increased. To compensate for these restrictions, as well as the rapidly changing landscape, patients may travel to other states for services (which imposes additional cost barriers, already known as a hinderance for accessing care), become increasingly reliant on self‐managed care, or forgo care altogether. While abortion can be safely self‐managed with medications, 46 patients deserve access to knowledgeable, non‐judgmental, and culturally competent healthcare should they have questions or concerns about their abortion.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
ACKNOWLEDGMENTS
The authors are grateful to the clinicians and clinic staff who provided data, without whom this study would not have been possible. We thank our fielding staff for their tireless data collection efforts: Lilian Ha, Madeleine Haas, Audrey Maynard, Rayan Sadeldin Bashir Mohamed and Parisa Thepmankorn. We gratefully acknowledge the critical feedback and contributions from the following Guttmacher colleagues: Joerg Dreweke, Kathryn Kost, Isaac Maddow‐Zimet and Elizabeth Nash.
Biographies
Rachel K. Jones is a Principal Research Scientist at the Guttmacher Institute. Her research focuses on abortion in the United States, and she has conducted studies on contraceptive use and health insurance.
Marielle Kirstein is a Senior Research Assistant at the Guttmacher Institute. Ms. Kirstein's research focuses on access to abortion and contraceptive services in the United States.
Jesse Philbin is a Senior Research Associate at the Guttmacher Institute. Ms. Philbin's research focuses on measurement of abortion incidence and abortion service quality in legally restrictive settings.
APPENDIX A.
TABLE A1.
Total number of abortion‐providing facilities, 2017 and 2020, and percentage change, by region and by state
| Region and state | 2017 | 2020 | % Change |
|---|---|---|---|
| United States total | 1587 | 1603 | 1 |
| Northeast | 518 | 545 | 5 |
| Connecticut | 54 | 78 | 44 |
| Maine | 21 | 24 | 14 |
| Massachusetts | 47 | 44 | −6 |
| New Hampshire | 12 | 12 | 0 |
| New Jersey | 76 | 75 | −1 |
| New York | 252 | 262 | 4 |
| Pennsylvania | 43 | 36 | −16 |
| Rhode Island | 3 | 4 | 33 |
| Vermont | 10 | 10 | 0 |
| Midwest | 137 | 150 | 9 |
| Illinois | 40 | 45 | 13 |
| Indiana | 9 | 12 | 33 |
| Iowa | 9 | 7 | −22 |
| Kansas | 4 | 4 | 0 |
| Michigan | 30 | 33 | 10 |
| Minnesota | 11 | 14 | 27 |
| Missouri | 4 | 2 | −50 |
| Nebraska | 7 | 6 | −14 |
| North Dakota | 1 | 2 | 100 |
| Ohio | 14 | 15 | 7 |
| South Dakota | 2 | 2 | 0 |
| Wisconsin | 6 | 8 | 33 |
| South | 314 | 291 | −7 |
| Alabama | 7 | 5 | −29 |
| Arkansas | 4 | 2 | −50 |
| Delaware | 6 | 5 | −17 |
| District of Columbia | 8 | 8 | 0 |
| Florida | 85 | 76 | −11 |
| Georgia | 26 | 24 | −8 |
| Kentucky | 3 | 6 | 100 |
| Louisiana | 4 | 3 | −25 |
| Maryland | 44 | 38 | −14 |
| Mississippi | 3 | 3 | 0 |
| North Carolina | 26 | 23 | −12 |
| Oklahoma | 6 | 7 | 17 |
| South Carolina | 10 | 6 | −40 |
| Tennessee | 12 | 11 | −8 |
| Texas | 35 | 37 | 6 |
| Virginia | 32 | 31 | −3 |
| West Virginia | 3 | 6 | 100 |
| West | 618 | 617 | 0 |
| Alaska | 6 | 5 | −17 |
| Arizona | 11 | 16 | 45 |
| California | 419 | 418 | 0 |
| Colorado | 32 | 35 | 9 |
| Hawaii | 28 | 25 | −11 |
| Idaho | 5 | 5 | 0 |
| Montana | 5 | 6 | 20 |
| Nevada | 11 | 11 | 0 |
| New Mexico | 7 | 8 | 14 |
| Oregon | 29 | 29 | 0 |
| Utah | 12 | 8 | −33 |
| Washington | 51 | 49 | −4 |
| Wyoming | 2 | 2 | 0 |
Jones RK, Kirstein M, Philbin J. Abortion incidence and service availability in the United States, 2020. Perspect Sex Reprod Health. 2022;54(4):128‐141. doi: 10.1363/psrh.12215
Endnotes
Our designation of high volume depended on the total number of abortions in the state but was typically more than 200 abortions per year as recorded in the 2017 Abortion Provider Census. 5
We did not contact the Colorado, Kansas, Maryland, New Hampshire, New York state, or South Dakota state health departments, as they either do not collect these data, provide detailed reports online, or have delayed release of data.
While our figures include abortions provided to women, transgender men, and non‐binary and other individuals with the capacity to become pregnant, the Census Bureau data count women. We adopt the terminology of “women” when we discuss abortion rates but recognize that this denominator does represent all individuals who obtain abortions.
While we did not provide a definition, we intended “quick pick up” to refer to situations where an individual could go to the facility only to pick up medication abortion drugs, without receiving in‐person counseling or examination.
Our analysis places more emphasis on abortion incidence as opposed to rates because abortion rates for many states, and perhaps even the nation, will change, at least slightly, when population figures from the Census 2020 become available in late 2022 or early 2023.
The figures of 95 and 86 do not perfectly align with the overall change of one less clinic in 2020 than 2017 due to change in classification of some facilities. Between 2017 and 2020, 22 facilities classified as clinics in 2017 were reclassified as physicians' offices (n = 13) or hospitals (n = 9) in 2020 and 12 facilities that were clinics in 2020 had been classified as physicians' offices (n = 10) or hospitals (n = 2) in 2017. Reclassification of facilities between surveys is standard; it is based on changes in abortion caseloads and, in the case of most hospitals, new information acquired during fielding.
REFERENCES
- 1. Guttmacher Institute . Interactive Map: US abortion policies and access after Roe. 2022. Accessed September 9, 2022. https://states.guttmacher.org/policies/.
- 2. Guttmacher Institute . 26 states are certain or likely to ban abortion without roe: here's which ones and why. 2021. Accessed July 8, 2022. https://www.guttmacher.org/article/2021/10/26-states-are-certain-or-likely-ban-abortion-without-roe-heres-which-ones-and-why.
- 3. Personal communication with Elizabeth Nash. 2022.
- 4. Jones RK, Zolna MRS, Henshaw SK, Finer LB. Abortion in the United States: incidence and access to services, 2005. Perspect Sex Reprod Health. 2008;40(1):6‐16. doi: 10.1363/4000608 [DOI] [PubMed] [Google Scholar]
- 5. Jones RK, Witwer E, Jerman J. Abortion incidence and service availability in the United States, 2017. 2019. Accessed January 25, 2022. https://www.guttmacher.org/report/abortion-incidence-service-availability-us-2017. doi: 10.1363/2019.30760 [DOI]
- 6. Lindberg LD, Santelli JS, Desai S. Changing patterns of contraceptive use and the decline in rates of pregnancy and birth among U.S. adolescents, 2007‐2014. J Adolesc Health. 2018;63(2):253‐256. doi: 10.1016/j.jadohealth.2018.05.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Santelli JS, Morrow B, Anderson JE, Lindberg LD. Contraceptive use and pregnancy risk among U.S. high school students, 1991‐2003. Perspect Sex Reprod Health. 2006;38(2):106‐111. doi: 10.1363/psrh.38.106.06 [DOI] [PubMed] [Google Scholar]
- 8. Kavanaugh ML, Jerman J. Contraceptive method use in the United States: trends and characteristics between 2008, 2012 and 2014. Contraception. 2018;97(1):14‐21. doi: 10.1016/j.contraception.2017.10.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Kavanaugh ML, Jerman J, Finer LB. Changes in use of long‐acting reversible contraceptive methods among U.S. women, 2009‐2012. Obstet Gynecol. 2015;126(5):917‐927. doi: 10.1097/AOG.0000000000001094 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kavanaugh ML, Pliskin E, Jerman J. Use of concurrent multiple methods of contraception in the United States, 2008 to 2015. Contraception: X. 2021;3:100060. doi: 10.1016/j.conx.2021.100060 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Sundaram A, Vaughan B, Kost K, et al. Contraceptive failure in the United States: estimates from the 2006‐2010 National Survey of family growth. Perspect Sex Reprod Health. 2017;49(1):7‐16. doi: 10.1363/psrh.12017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Austin N, Harper S. Quantifying the impact of targeted regulation of abortion provider laws on US abortion rates: a multi‐state assessment. Contraception. 2019;100:374‐379. doi: 10.1016/j.contraception.2019.06.003 [DOI] [PubMed] [Google Scholar]
- 13. Grossman D, White K, Hopkins K, Potter JE. Change in distance to nearest facility and abortion in Texas, 2012 to 2014. JAMA. 2017;317(4):437‐439. doi: 10.1001/jama.2016.17026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Jones RK, Philbin J, Kirstein M, Nash E, Lufkin K. Long‐term decline in US abortions reverses, showing rising need for abortion as supreme Couty is posed to overturn roe v. Wade. 2022. https://www.guttmacher.org/article/2022/06/long-term-decline-us-abortions-reverses-showing-rising-need-abortion-supreme-court.
- 15. Schroeder R, Muñoz I, Kaller S, Berglas N, Stewart C, Upadhyay U. A report from the Abortion Facility Database Project, Advancing New Standards in Reproductive Health (ANSIRH), University of California San Francisco: 37.
- 16. McGowan ML, Norris AH, Bessett D. Care churn—why keeping clinic doors open isn't enough to ensure access to abortion. N Engl J Med. 2020;383(6):508‐510. doi: 10.1056/NEJMp2013466 [DOI] [PubMed] [Google Scholar]
- 17. Roberts SCM, Berglas NF, Schroeder R, Lingwall M, Grossman D, White K. Disruptions to abortion care in Louisiana during early months of the COVID‐19 pandemic. Am J Public Health. 2021;111(8):1504‐1512. doi: 10.2105/AJPH.2021.306284 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Cappello, O . Surveying state executive orders impacting reproductive health during the COVID‐19 pandemic. Guttmacher Institute. 2020. Accessed January 25, 2022. https://www.guttmacher.org/article/2020/07/surveying-state-executive-orders-impacting-reproductive-health-during-covid-19.
- 19. Roberts SCM, Schroeder R, Joffe C. COVID‐19 and independent abortion providers: findings from a rapid‐response survey. Perspect Sex Reprod Health. 2020;52(4):217‐225. doi: 10.1363/psrh.12163 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Lindberg LD, VandeVusse A, Mueller J, Kirstein M. Early impacts of the COVID‐19 pandemic: findings from the 2020 Guttmacher survey of reproductive health experiences. Guttmacher Institute. 2020. Accessed November 15, 2022. https://www.guttmacher.org/report/early-impacts-covid-19-pandemic-findings-2020-guttmacher-survey-reproductive-health. doi: 10.1363/2020.31482 [DOI]
- 21. Dozier JL, Sufrin C, Berger BO, Burke AE, Bell SO. COVID‐19 impacts on abortion care‐seeking experiences in the Washington, DC, Maryland, and Virginia regions of the United States. Perspect Sex Reprod Health. 2022;54:68‐79. doi: 10.1363/psrh.12202 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. National Center for Health Statistics . Vintage 2019 postcensal estimates of the resident population of the United States. Accessed July 11, 2022. https://www.census.gov/programs-surveys/popest/data/data-sets.html.
- 23. Osterman M, Hamilton B, Martin J, Driscoll A, Valenzuela C. Births: Final Data for 2020. National Center for Health Statistics (U.S.); 2021. doi: 10.15620/cdc:112078 [DOI] [PubMed] [Google Scholar]
- 24. Jones RK, Jerman J. Abortion incidence and service availability in the United States, 2014. Perspect Sex Reprod Health. 2017;49(1):17‐27. doi: 10.1363/psrh.12015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Tschann M, Ly ES, Hilliard S, Lange HLH. Changes to medication abortion clinical practices in response to the COVID‐19 pandemic. Contraception. 2021;104(1):77‐81. doi: 10.1016/j.contraception.2021.04.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Maddow‐Zimet I, Kost K. Even before Roe Was Overturned, Nearly One in 10 People Obtaining an Abortion Traveled across State Lines for Care. Guttmacher Institute; 2022. [Google Scholar]
- 27. Zeldovich VB, Rocca CH, Langton C, Landy U, Ly ES, Freedman LR. Abortion policies in U.S. teaching hospitals: formal and informal parameters beyond the law. Obstet Gynecol. 2020;135(6):1296‐1305. doi: 10.1097/AOG.0000000000003876 [DOI] [PubMed] [Google Scholar]
- 28. Daniels K, Abma JC. Current Contraceptive Status among Women Aged 15–49: United States, 2015–2017. National Center for Health Statistics; 2018:8. [Google Scholar]
- 29. Daniels K, Abma JC. Current Contraceptive Status among Women Aged 15–49: United States, 2017–2019. National Center for Health Statistics; 2020:8. [PubMed] [Google Scholar]
- 30. National Survey of Family Growth . Key statistics from the National Survey of Family Growth: C listing. 2022. Accessed June 30, 2022. https://www.cdc.gov/nchs/nsfg/key_statistics/c-keystat.htm.
- 31. Kohn JE, Snow JL, Simons HR, Seymour JW, Thompson T‐A, Grossman D. Medication abortion provided through telemedicine in four U.S. states. Obstet Gynecol. 2019;134(2):343‐350. doi: 10.1097/AOG.0000000000003357 [DOI] [PubMed] [Google Scholar]
- 32. Beardsworth KM, Doshi U, Raymond E, Baldwin MK. Miles and days until medical abortion via TelAbortion versus clinic in Oregon and Washington, USA. BMJ Sex Reprod Health. 2022;48(e1):e38‐e43. doi: 10.1136/bmjsrh-2020-200972 [DOI] [PubMed] [Google Scholar]
- 33. Jones RK, Nash E, Cross L, Philbin J, Kirstein M. Medication Abortion Now Accounts for More than Half of all US Abortions. Guttmacher Institute; 2022. [Google Scholar]
- 34. Raymond EG, Grossman D, Mark A, et al. Commentary: no‐test medication abortion: a sample protocol for increasing access during a pandemic and beyond. Contraception. 2020;101(6):361‐366. doi: 10.1016/j.contraception.2020.04.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Sobel L, Ramaswamy A, Frederiksen B, Salganicoff A. State action to limit abortion access during the COVID‐19 pandemic. Kaiser Family Foundation. 2020. Accessed June 30, 2022. https://www.kff.org/coronavirus‐covid‐19/issue‐brief/state‐action‐to‐limit‐abortion‐access‐during‐the‐covid‐19‐pandemic/.
- 36. Wolfe T, van der Meulen RY. Abortion during the COVID‐19 pandemic: racial disparities and barriers to care in the USA. Sex Res Soc Policy. 2022;19(2):541‐548. doi: 10.1007/s13178-021-00569-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Cleveland C. Colorado's abortion providers see more out‐of‐state patients during the coronavirus pandemic. Colorado Public Radio. Accessed July 8, 2022. https://www.cpr.org/2020/04/15/colorado‐abortion‐providers‐are‐seeing‐more‐out‐of‐state‐patients‐during‐the‐coronavirus‐pandemic/.
- 38. White K, Kumar B, Goyal V, Wallace R, Roberts SCM, Grossman D. Changes in abortion in Texas following an executive order ban during the coronavirus pandemic. JAMA. 2021;325(7):691‐693. doi: 10.1001/jama.2020.24096 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Smith MH, Muzyczka Z, Chakraborty P, et al. Abortion travel within the United States: an observational study of cross‐state movement to obtain abortion care in 2017. Lancet Regional Health. 2022;10:100214. doi: 10.1016/j.lana.2022.100214 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Jerman J, Jones RK, Onda T. Characteristics of U.S. Abortion Patients in 2014 and Changes since 2008. Guttmacher Institute; 2016:29. [Google Scholar]
- 41. Guttmacher Institute . State funding of abortion under Medicaid. 2016. Accessed July 6, 2022. https://www.guttmacher.org/state-policy/explore/state-funding-abortion-under-medicaid.
- 42. Tan R. Demand for abortion subsidies surges in the D.C. area as funding declines. Washington Post. 2021. Accessed August 26, 2022. https://www.washingtonpost.com/local/dc‐abortion‐funding‐crisis/2021/06/09/1fee804c‐c7a2‐11eb‐a11b‐6c6191ccd599_story.html.
- 43. National Network of Abortion Funds . Seeds & roots: fortifying network infrastructure and leading with values. 2018. Accessed November 15, 2022. https://media.abortionfunds.org/cms/assets/uploads/2021/07/29145901/NNAF-Annual-Report-%E2%80%93-FY-2019-FY-2020.pdf
- 44. Aiken ARA, Starling JE, Gomperts R. Factors associated with use of an online telemedicine service to access self‐managed medical abortion in the US. JAMA Netw Open. 2021;4(5):e2111852. doi: 10.1001/jamanetworkopen.2021.11852 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Murtagh C, Wells E, Raymond EG, Coeytaux F, Winikoff B. Exploring the feasibility of obtaining mifepristone and misoprostol from the internet. Contraception. 2018;97(4):287‐291. doi: 10.1016/j.contraception.2017.09.016 [DOI] [PubMed] [Google Scholar]
- 46. Moseson H, Herold S, Filippa S, Barr‐Walker J, Baum SE, Gerdts C. Self‐managed abortion: a systematic scoping review. Best Pract Res Clin Obstet Gynaecol. 2020;63:87‐110. doi: 10.1016/j.bpobgyn.2019.08.002 [DOI] [PubMed] [Google Scholar]