Is third‐trimester abortion exceptional? Two pathways to abortion after 24 weeks of pregnancy in the United States

Abstract

Context

In the United States, third‐trimester abortions are substantially more expensive, difficult to obtain, and stigmatized than first‐trimester abortions. However, the circumstances that lead to someone needing a third‐trimester abortion may have overlaps with the pathways to abortion at other gestations.

Methods

I interviewed 28 cisgender women who obtained an abortion after the 24th week of pregnancy using a modified timeline interview method. I coded the interviews thematically, focusing on characterizing the experience of deciding to obtain a third‐trimester abortion.

Results

I find two pathways to needing a third‐trimester abortion: new information, wherein the respondent learned new information about the pregnancy—such as of an observed serious fetal health issue or that she was pregnant—that made the pregnancy not (or no longer) one she wanted to continue; and barriers to abortion, wherein the respondent was in the third trimester by the time she was able to surmount the obstacles to abortion she faced, including cost, finding a provider, and stigmatization. These two pathways were not wholly distinct and sometimes overlapped.

Conclusions

The inherent limits of medical knowledge and the infeasibility of ensuring early pregnancy recognition in all cases illustrate the impossibility of eliminating the need for third‐trimester abortion. The similarities between respondents' experiences and that of people seeking abortion at other gestations, particularly regarding the impact of barriers to abortion, point to the value of a social conceptualization of need for abortion that eschews a trimester or gestation‐based framework and instead conceptualizes abortion as an option throughout pregnancy.

INTRODUCTION

Most abortions in the United States take place in the first trimester of pregnancy. ¹ Abortions at later gestational durations are comparatively uncommon: only 1.0% of abortions take place at or after 21 weeks after the first day of the pregnant person's last menstrual period (LMP). ² Given that difference in volume, understandably, most clinical training for, research on, and advocacy for abortion and abortion rights has focused on first‐trimester care. However, little research has examined whether and how abortions at other gestational durations are similar or different.

In this article, I examine the specific case of third‐trimester abortion, defined as abortions that take place at or after 24 weeks LMP. Third‐trimester abortion care in the United States is substantially different from first‐trimester abortion care. First, abortion in the first trimester is most available in the United States, with approximately 780 outpatient facilities providing such care, although access varies considerably by geography. ³ However, as gestation increases the number of facilities decreases. There are only four facilities that publicly advertise care after 24 weeks LMP. ⁴ This scarcity is in part an effect of state‐level gestation‐based bans. A total of 44 states generally prohibit abortion in the third trimester. ⁵ Although these bans typically have exceptions, they are so narrow that few cases fall under them. Further, only one of these four facilities is located in a major urban center. To obtain a third‐trimester abortion, then, pregnant people must travel, accruing attendant travel costs for transportation, accommodations, and food, which can represent a substantial burden. ⁶

Second, third‐trimester abortion care is distinct in its cost. In 2020, while first‐trimester abortions had an average cost of $644 for medication abortion and $715 for aspiration abortion and the average cost of a second‐trimester abortion was $1068, ⁷ third‐trimester abortions cost much more: they range in cost from a few thousand dollars to over $25,000, depending on gestation and clinical complexity. Third‐trimester abortions typically take place over 3 days and can include laboring, which contributes to their high cost. Federal and state‐level bans on public insurance coverage in 34 states ⁸ and regulation of ⁹ or high deductibles in ¹⁰ private insurance mean that most people must pay out‐of‐pocket for abortion care. Given research that finds that the out‐of‐pocket costs of a first‐trimester abortion strain the finances of many abortion patients, ¹¹ the cost of a third‐trimester abortion likely exceeds the financial capacity of most pregnant people.

Third, socially, there is starkly lower support for third‐trimester abortion than for first‐trimester abortion. Public opinion polls consistently show majority support for safe and legal abortion, an overall consensus that has remained relatively unchanged since 1973 when abortion was legalized nationally in the United States. ¹² Nonetheless, that support does not hold for abortion throughout pregnancy: only 13% of respondents in a 2018 Gallup poll believed abortion should generally be legal in the third trimester (although support was higher for specific circumstances, including life endangerment [75%], cases of rape or incest [52%], and serious fetal health issue [ranged by issue from 29% to 48%]). ¹³ As the gestational duration of the pregnancy increases, support for abortion wanes, ¹⁴ potentially because of a lack of familiarity with and empathy for third‐trimester abortion patients and providers. ¹⁵

There are thus many reasons—financial, logistical, and social—why third‐trimester abortion care is exceptional compared to first‐trimester abortion care. However, there is reason to believe that the circumstances that lead to someone needing a third‐trimester abortion are not exceptional. Several studies have highlighted the importance of the timing of pregnancy discovery, with later discovery associated with later presentation to abortion care. ¹⁶ , ¹⁷ Other research has identified how laws that complicate people's ability to access abortion, including parental involvement laws ¹⁸ and laws that contribute to the reduction of abortion clinics, ¹⁹ , ²⁰ are associated with later presentation to abortion care for patients. Broadly, the literature on delay has tended to focus on abortion in the later second trimester, not specifically on contributors to needing a third‐trimester abortion. To begin to fill this gap, I draw on interviews with 28 cisgender women who obtained an abortion after 24 weeks LMP to examine the factors that contribute to obtaining abortion care in the third trimester.

METHODOLOGY

Recruitment

I describe the study methods in depth elsewhere. ²¹ Briefly, I recruited people who had obtained an abortion after 24 weeks LMP. The recruitment flyer described the project as an interview study to understand the experiences of seeking and obtaining an abortion in the third trimester. It included a toll‐free phone number to call for more information and enroll in the study. Starting in September 2017, clinic staff at an outpatient facility that offers abortion into the third trimester distributed the study flyer to patients, along with other paperwork, as they were leaving the facility to return home after the procedure was complete. In January 2018, I engaged a local research assistant to present the flyer to prospective participants on‐site and in‐person. Also in January 2018, I asked people from the general public who had contacted me based on my previous published research who had personal experience of third‐trimester abortion to share the flyer with their relevant networks, which included online support groups for people who had obtained abortions.

Data collection

I screened prospective participants who called the study phone number for eligibility. To be eligible, participants had to be over 18 years of age, be comfortable speaking in English, and have obtained an abortion after 24 weeks of pregnancy. If the potential participant was eligible and interested, I completed an oral consenting process with the caller, collected contact information in a secure Salesforce database, and scheduled a phone interview for a future time that was convenient to the participant and was at least 3 weeks after the abortion, with the intention of allowing participants time to reflect on their experience. To encourage prospective participants to call the study phone, I offered all callers a $10 gift card incentive, regardless of eligibility determination.

Interviews were semi‐structured, guided by open‐ended questions that enabled both the interviewer and the respondent to introduce new ideas, build rapport, and follow the flow of the conversation. After collecting initial demographic information, I employed a modified version of the timeline interview methodology, ²² starting out by asking participants to talk about the points in their life that they thought of as consequential to their third‐trimester abortion experience and then following up to understand their experiences and the significance of these time points. This open structure is based on the recognition in the timeline methodology that people's memories do not always operate linearly, and that the meaning of certain life experiences is dependent on understanding other life experiences that both preceded and followed the focal experience. ²² By the end of the interview, I aimed to capture the participant's experience of discovering pregnancy, deciding on abortion, seeking abortion care, receiving abortion care, and returning home after the abortion. With permission from participants, I recorded all interviews. I offered all participants a $50 gift card to a major national retailer to remunerate them for their time. The interviews ranged in length, averaging about 1 h and 45 min. The Institutional Review Board at the University of California, San Francisco approved all study protocols.

By June 2018, I had interviewed 28 participants and experienced the interviews to be very rich, reaching thematic saturation on key research questions. Additionally, I experienced the interviews collectively to be emotionally intense, which is consistent with research on the emotion work required in conducting qualitative research on sensitive topics. ²³ , ²⁴ On occasion, I found it difficult to manage my emotions during interviews, leading me to be concerned about whether and how an increase of such occasions would affect the quality of the interviews. I did not want to unintentionally convey to an interviewee that a particular emotional response to what they were saying was expected. Further, I recognized that the expression of spontaneous emotions by an interviewer represents an inappropriate experience for the interviewees, who may feel compelled to offer emotional support. I therefore judged it appropriate to close recruitment.

Analysis

A professional transcription company transcribed the interviews verbatim. I analyzed the transcripts in ATLAS.ti 7 using thematic coding. For the purposes of this analysis, coding focused on characterizing the broad experience of deciding to obtain a third‐trimester abortion, with attention to variations across the respondents. Recognizing the limits of understanding seeking an abortion with an exclusively individual‐level focus, ²⁵ I explicitly attuned coding to capture participants' social circumstances and context. Through memoing, I grouped the thematic codes into higher‐level categories, identifying two emergent pathways to needing a third‐trimester abortion among interviewees. I then returned to the transcripts to characterize each respondent's experience according to the identified pathways, fleshing out the contours of each pathway, sequencing each respondent's experience, and seeking evidence of additional pathways. I did not find evidence for additional pathways but did identify some respondents whose experiences were characterized by both pathways.

All names used below are pseudonyms. Reported ages are age at time of the interview, which may differ from age at the time of the pregnancy and abortion.

RESULTS

Sample characteristics

I interviewed 28 cisgender women, who ranged in age from 18 to 46. A total of 19 identified as white, five as Hispanic or Latina, two as Black or African American, one as Asian, and one as biracial (Asian and white). In terms of highest educational attainment, two had not completed high school, five had completed high school, six had completed some college, six had graduated from a four‐year college, and nine held advanced degrees. Although most respondents described themselves as able to meet their basic financial needs, eight were unemployed at the time of their abortion, including one who was homeless. A total of 22 participants had obtained their abortion within the year prior to the interview and six had obtained their abortion more than 1 year prior, with 8 years the longest time reported since abortion. At the time of their abortion, participants' pregnancy duration ranged from 24 to 35 weeks LMP.

New information about the pregnancy

In one path by which participants came to need a third‐trimester abortion, they learned a new piece of information in or nearly in their third trimester of pregnancy that made them realize that this was not, or was no longer, a pregnancy they wanted to carry to term. Because this determination happened later in pregnancy, they needed abortion care in the third trimester. The new information that changed their determination about the pregnancy was often about the health of their fetus.

Rachel, a 46‐year‐old white woman in the Northeast, is one example. Rachel's pregnancy was going as expected. At the 20‐week diagnostic scan, her prenatal care team noticed nothing irregular. In fact, Rachel recalled, “I specifically remember at my 20‐week checkup, the doctors saying, don't worry, everything's great.” Because of a complication in her previous pregnancy, her doctors scheduled her for another diagnostic scan at 29 weeks. During this scan, they observed problems with her fetus's brain. Initially, her prenatal care team encouraged her not to worry too much. As she related, “It wasn't a ‘yeah, everything's great,’ but she [my doctor] was very much, you know, ‘Wait to see more. It might not be a big deal. Don't read on the Internet.’” Of the additional testing her doctor wanted, Rachel recalled thinking, “This is a waste of time.”

That changed after the testing, which included capturing an image of her fetus's brain. Rachel described it as “Just pieces missing, concaved. And when you see a normal brain and you see that, [it] doesn't even matter what the technical terms are, it's clear it was not right.” After meetings with specialists, Rachel and her husband determined that “There was nothing to say this child has any possibility of having even the capability of existing as a baby, as a child, as a person.” That night, she said, “We both agreed that terminating was the right decision for all involved. For our [two‐year‐old] son, for us, our marriage, us as individuals, and for the baby we were carrying.” Importantly, the information that led Rachel and her husband to decide to have an abortion was not available earlier in her pregnancy. Indeed, the diagnostic scan at 20 weeks identified no problems.

Kara, a 36‐year‐old biracial woman in the Midwest, too, obtained new information in her third trimester of pregnancy that made it one she no longer wanted to continue. Her experience was different from Rachel's, however, in that a scan at 18 weeks gestation revealed impairments in her fetus, but their severity was unknown. A follow‐up scan at 22 weeks revealed additional impairments. However, as Kara explained, even though it was now clear that her fetus had serious health issues, what that meant for its capacity to live was unclear. Her doctors told her that “Based on what they could see at that point, there was a 70% chance the baby would be either totally fine or maybe have some mild to moderate cognitive impairment. Perhaps some mild disabilities, but the kind of things that some therapy basically resolves. […] 30% chance of more severe problems or death.” To Kara, this seemed like good odds that her baby would be healthy. Although her doctors offered her the opportunity to have an abortion at that point, she declined. Instead, she opted for a series of additional tests and scans. Six weeks later, an ultrasound scan revealed that her fetus was in the 30% and “It would definitely carry a very bad prognosis.” Kara and her husband decided to seek an abortion: giving birth to a baby with these conditions, she said, “Was just beyond what we could in good conscience proceed with.” Notably, the information that informed that decision, even as they knew there were some observed fetal health issues, was unavailable earlier in the pregnancy. As Kara explained, “Brain development happens so much in the last second trimester and early third trimester that they really could not confidently tell us more [at those earlier scans].” Simply put, there were limits on what information Kara could obtain earlier in pregnancy.

The new information respondents received that led to their decision to obtain an abortion was not exclusively related to fetal health. For some respondents, the new information they obtained was that they were pregnant. Autumn, a 22‐year‐old white woman in the West, was having a regular period but felt a bit “off,” as she put it. She stopped by the local health clinic and took a pregnancy test, which came back positive. She and her husband discussed the pregnancy and, she said, “We both decided to get an abortion.” She made an appointment at a nearby abortion clinic. The ultrasound worker at the clinic thought she was early in pregnancy, opting to conduct a transvaginal ultrasound, which is preferred for diagnosing and dating early pregnancies. Then, Autumn explained, the ultrasound worker “Kind of got like a confused face and she was like stuttering and she was sounded very like worried.” Autumn was not early in pregnancy. Based on the subsequent abdominal ultrasound the clinic worker conducted, she was 26 weeks into her pregnancy. Autumn was shocked and confused. She said, “I immediately burst into tears “cause I was like, “How is this possible?” Autumn sought an abortion in the third trimester because she did not know she needed one until then.

Veronica, a 21‐year‐old Latina woman in the South, also did not realize she was pregnant until she was in the third trimester of pregnancy. Veronica was dating someone new and wanted to get tested for sexually transmitted infections before commencing a sexual relationship with this man. The clinic also ran a pregnancy test, which was positive. Veronica was shocked. She explained that she had no recognizable pregnancy symptoms and had been having a regular period: “It seemed to me like regular periods because it lasted the same amount of time that they would usually last […] and I never got morning sickness. I wasn't lethargic.” Veronica was immediately clear that she did not want to continue the pregnancy and took the first available abortion appointment at the clinic. When Veronica presented for her abortion appointment, the ultrasound worker determined that she was 25 weeks pregnant. Veronica needed an abortion in the third trimester because the fact that she was pregnant was new information to her when she was already 25 weeks pregnant.

These four women's accounts are representative of the experience of how learning new information in the third trimester of pregnancy—including that one is pregnant—led to the need for a third‐trimester abortion. They were seeking a third‐trimester abortion because they did not have the information they needed to make the abortion decision earlier. Further, that information often could not be available earlier, as in the case of impairments diagnosable only at a certain stage of development, or would be counterintuitive to seek earlier, as in the case of a pregnancy for someone who is having a regular period, thereby evidencing the impossibility of ever eliminating the need for third‐trimester abortion.

Barriers to abortion before the third trimester

The second path by which participants came to need an abortion in the third trimester was characterized by barriers to abortion before the third trimester. These women decided they wanted an abortion before the 24th week of pregnancy but were delayed into the third trimester by obstacles to abortion care. Most but not all of the barriers they encountered were policy related.

Monique, a 30‐year‐old Black woman in the South, is one example. Monique knew she was pregnant as soon as she missed her period. She had morning sickness and felt tired. She took a store‐bought pregnancy test and confirmed that she was pregnant. She did not want to be, describing learning she was pregnant as “Shocking, surprising, [and] devastating.” Pregnancy is high risk for her health, and she did not feel emotionally ready to have another child. She and her boyfriend agreed that she would obtain an abortion. They found a clinic and made an appointment but faced financial obstacles to paying for care. The federal Hyde amendment and Monique's state prohibit public insurance coverage for abortion, which meant Monique, who relied on public insurance, had to pay out‐of‐pocket for care. Monique's contract job had just ended, and her boyfriend had lost his job. She did not have any of the money she needed to pay for an abortion. She said, “I was more focused on worrying about bills.” She sought help from abortion funds—non‐profit organizations that offer financial and practical support to abortion patients—but still could not secure enough money to cover the cost. Abortion funds, it bears noting, are often unable to meet client demand for funding and practical support. And so, Monique resigned herself to continuing the pregnancy, although, she said, “I can't say like I ever truly grasped it or fully accepted it.” Monique mostly ignored her pregnancy. As she said, “Honestly, I tried not to think about it.” She did not tell any friends or family about it. At about 18 weeks gestation, she started prenatal care. Just over a month later, her (now) ex‐boyfriend got a new job that came with a signing bonus, enabling them to afford an abortion, even at the higher costs associated with later gestations. Monique was able to obtain the abortion she “still always felt was the right choice.” To sum up, Monique had a third‐trimester abortion because she could not overcome state policy‐created financial barriers to abortion until then.

Victoria, a 26‐year‐old white woman in the South, too, faced substantial barriers to abortion when she first decided she wanted one. When she discovered her pregnancy, Victoria and her boyfriend spent a week considering their options and decided abortion was the right decision for them. They found trying to implement this decision extremely difficult, however. The local abortion clinic had closed a few years prior following a series of state laws that made it no longer feasible to keep the clinic open. The closest abortion clinic was an hour's drive away and, due to her state's two‐visit requirement, Victoria would have to travel there twice. Victoria and her boyfriend were homeless and did not have a car; they had no way to travel to the clinic. She also did not know how she could afford the abortion, since, like Monique, her public insurance coverage did not cover abortion care due to state policy. Any money Victoria and her boyfriend had went toward meeting their basic needs. As she explained, “My boyfriend was working odd jobs, and I don't have a job, and we were homeless on the street. So, I mean, we would have definitely tried, and we, you know, did try, but there was no way that we could get a couple hundred [dollars], let alone a grand, if not more.” She tried calling multiple abortion funds but, “Seems like every time I would call, they would call back saying that they were already, you know, out of funds.” By the time Victoria was able to raise enough money (from abortion funds) and figure out transportation, she was in the third trimester of pregnancy. When I asked whether she would have had an abortion before the third trimester, had she had the opportunity, she replied without hesitating, “Yes, ma'am.” For Victoria and several other interviewees, bans on public insurance coverage of abortion and difficulty finding a provider intersected with the structural barrier of poverty to make abortion before the third trimester impossible to obtain despite respondents' desire.

Although not strictly rooted in policies, the stigmatization of abortion also served as a barrier to abortion before the third trimester for some respondents, as did a history of sexual trauma. Cristina, an 18‐year‐old Latina woman in the South, experienced both. A few years prior to our interview, Cristina was raped. She became pregnant, and the pregnancy ended in a miscarriage. It took her a long time to trust a romantic partner again, but she had come to trust her boyfriend and they had sex once. When she learned that her contraception had failed, she said, “I was devastated, pretty much. I was really mad, and I was really upset because I did not want a baby.” As clear as her desire to terminate this pregnancy was, her history of sexual trauma and her fear that her parents would disapprove of her pregnancy and her abortion decision meant that Cristina took no immediate action toward obtaining an abortion. She said, “I cried so hard. I didn't want to be touched by anybody. I didn't want to be talked to by anybody. […] I felt like it was a dream.”

When she eventually came to terms with her pregnancy, Cristina found a clinic 2 h from her home. She chose that distance because “I did not want to know anybody there.” She was also trying to find a place where she would not have to experience pregnancy counseling. She explained, “I did not want counseling. I did not want to talk about it. Those were the kinds of things that I was trying to avoid [in choosing a clinic].” At the clinic, she learned that she was farther along than she thought and there was no provider in the state who would provide abortion care for her at that gestation. As she said, “The door had just been shut for me.” Desperate, she confided in her brother, who helped her find an out‐of‐state abortion provider. By the time Cristina found the provider, raised the money for the procedure and travel, and was able to get to the out‐of‐state clinic, she was in the third trimester of pregnancy. Her history of sexual trauma, her fear of others' judgment, and the paucity of abortion facilities in her state made it impossible for Cristina to implement her abortion decision until she was in the third trimester of pregnancy.

These respondents' experiences illustrate how policies and fear of opposition to abortion can cause substantial delay in people's ability to obtain abortion care. Pointedly, these barriers to abortion did not result in continuation of pregnancy. Rather, they resulted in abortions later in pregnancy than the respondent initially wanted, prolonging their experience of pregnancy and of seeking abortion. It also bears noting that most women on this second pathway were struggling financially (like Monique and Victoria) or did not have their own income (as in Cristina's case). These statuses compounded the effects of low service availability and policies regulating abortion care.

New information compounded by barriers

Even as these two paths characterized distinct ways that respondents came to need an abortion in the third trimester, there was no bright line between them. Several respondents described obtaining new information that changed their desired pregnancy outcome in the second trimester, but then faced barriers to abortion that delayed them into the third trimester. For example, Carrie, a 33‐year‐old white woman in the South, desired her pregnancy and initially intended to continue it. That changed when she was 21 weeks pregnant: her medical team identified several serious fetal health issues. She said, “I think the deciding factor for us was the last symptom [the sonographer] found […] I think that's when we came home, and we said, ‘There's nothing really left to fight for.’ So, then it became a decision [to seek abortion].” However, her current health providers declined to provide abortion care and she was forced to travel out‐of‐state to obtain an abortion. It took her and her husband 3 weeks to find a provider, pull together money to pay for the procedure and travel, find childcare for their three‐year‐old son, get sufficient time off from work, and drive the 9 h to the clinic. Carrie was over 24 weeks pregnant at the time of her abortion.

Several other respondents shared similar accounts of learning a new piece of information that made this a pregnancy they did not want to continue before the 24th week of pregnancy and then experiencing obstacles to obtaining a timely abortion that meant they obtained care after the 24th week of pregnancy. Most commonly, as for Carrie, state‐level gestational duration bans complicated respondents' ability to obtain an abortion when they first wanted one, delaying them into the third trimester. As another example, Isabella, a 24‐year‐old Latina woman in the South, did not recognize that she was pregnant until she was in the second trimester of pregnancy. She had difficulty locating an abortion provider, finally finding a clinic when she estimated she was 21 weeks into pregnancy. Because of her state's ban on abortions after 22 weeks gestation, the distance she would have to travel to get to the clinic, and the state's two‐visit requirement that meant added time until she could have an abortion, the clinic advised her she was too close to the limit and they could not care for her. She was in the third trimester of pregnancy by the time she found and traveled to a clinic that could legally provide her with abortion care.

While most instances of new information intersecting with barriers to abortion followed the pattern of learning the new information before experiencing barriers to abortion, Tonya, a 37‐year‐old white woman in the South, experienced a different ordering. Tonya and her husband were happy to be pregnant. Because of two unexpected bleeds early in her second trimester, Tonya's prenatal team scheduled her for weekly ultrasound scans. These scans identified physical impairments. As Tonya recounted, “every week, we went to a [new specialist], we would see something new.” As each impairment was identified, she reported, “My OB [obstetrician] that I was assigned [would say], ‘it's going to be no problem.’” But as the weeks passed, Tonya said, “we noticed something about his [the fetus's] hands, that they were always held in a certain way. But nobody ever noticed it or commented, and so we didn't either.” At her ultrasound scan at about 20 weeks gestation, Tonya asked the sonographer about what she had observed. The sonographer showed her several issues that the scan revealed, including describing how the team had observed aspects of the impairments grow more severe over time. Tonya was shocked: “That growth hadn't been told to us.” Her physician, she realized, had withheld information on the severity of the fetal impairments. When I asked whether she thought that was because he believed she would seek an abortion if she knew that information, she said, “Yes, I do. I think he was prolonging.” Tonya and her husband experienced a barrier to receiving significant information about the health of their fetus because of their doctor's opposition to abortion.

Tonya and her husband sought a second opinion at a different hospital, which confirmed the severity of the fetal health issue. Tonya explained that “The prognosis for him [the fetus] was either going to be this kind of slow death in‐utero—and I kept having the bleeds and things like that that were not good—or, if he did make it to birth, it was going to be a lot of surgeries and we did not know what his quality of life would be at that point.” Now at 22 weeks gestation, she and her husband decided abortion was the right choice. It was here that they faced another healthcare worker‐based barrier to abortion: the on‐call doctor at the second hospital refused to grant permission for the abortion. He insisted that, Tonya explained, “He had had experience delivering a 22‐week baby that was viable—and I know that they are, but that was a baby that didn't have any abnormalities, it was simply born early. So, we were refused treatment.” Tonya was in the third trimester of pregnancy by the time she could locate and travel to an out‐of‐state abortion provider. She needed a third‐trimester abortion because, first, her prenatal team conspired to withhold information from her about her fetus that was material for her decision and, second, a different healthcare provider refused to approve her for an on‐site abortion.

DISCUSSION

I find two pathways by which people come to need a third‐trimester abortion: new information and barriers to abortion before the third trimester. These findings, drawn from women's lived experiences, enrich our understanding of why people seek third‐trimester abortion care. Pointedly, findings illustrate the importance of attention to pregnant people's clinical, legal, and social circumstances—and not just their individual preferences or proffered “reasons” for abortion.

In addition, these findings, coupled with existing research on people's experience of pregnancy and seeking abortion, demonstrate two emergent truths about third‐trimester abortion. First, they demonstrate the impossibility of eliminating the need for third‐trimester abortion care. In a small but persistent number of pregnancies, new information acquired in the third trimester of pregnancy will make the pregnancy not one, or no longer one, the pregnant person wants to continue. Clinical research demonstrates that some serious fetal health issues simply are not observable until the third trimester of pregnancy. ²⁶ Research has also found that a small subset of people do not recognize they are pregnant until the third trimester of pregnancy. ²⁷ , ²⁸ While scholars have explored efforts to reduce “late” recognition of pregnancy, ²⁹ there is no feasible way to ensure “early” recognition of all pregnancies. Centrally, this means that some people will only possess the information they need to choose abortion in the third trimester of pregnancy.

Second, these findings reveal similarities in respondents' experiences of seeking third‐trimester abortion to those of people who obtain abortions in the first and second trimesters. Information about the pregnancy, including that the person is pregnant, is consequential to choosing abortion at any point in pregnancy. Time at recognition of pregnancy is consistently identified as related to abortion timing, with recognition of pregnancy at later gestations associated with presenting for abortion at later gestations and more common among specific populations (e.g., young women, women who have recently given birth, and women who use drugs). ¹⁶ , ¹⁷ And the literature comprehensively documents how policy‐related barriers to abortion contribute to delays in pregnant people's ability to obtain care, ¹⁸ , ¹⁹ , ²⁰ including making abortion impossible to obtain. ³⁰ , ³¹ This analysis extends that literature on the timing of presentation for abortion, highlighting how pregnant people find themselves in need of third‐trimester abortion care because of institutional, governmental, and societal failures—particularly failures to protect the most marginalized—and, sometimes, because of bad luck. They echo, in other words, what we already know about pathways to abortion: that they are enabled and constrained by institutions, policies, culture, and personal experience. ³¹ Thus, despite the enumerated ways that third‐trimester abortion differs from earlier abortion—in clinical complexity, cost, logistics, and social support—there are clear consistencies in why people seek abortion care, regardless of gestation.

As such, there is value to a social conceptualization of abortion not by gestation or trimesters but, rather, as care that can be needed throughout pregnancy. In this way, I join other critiques of the social and legal organization of abortion care by gestation. ³² , ³³ , ³⁴ Conceptualizing abortion throughout pregnancy can facilitate framings that center pregnant people instead of external measurements of pregnancy (e.g., gestational duration). In so doing, it can shift discussion away from normative expectations about when people should seek abortion and why to, instead, highlight how and when pregnant people's reproductive and bodily autonomy is supported and compromised.

While I have identified two pathways, there may be other pathways to needing a third‐trimester abortion that were not represented in this sample. There may also be additional contours to these pathways that are not reflected in these interviewees' experiences. For example, there may be other forms of new information (e.g., about the pregnant person's health, about the pregnant person's financial, housing, and/or relationship stability) that emerge in the late second trimester or third trimester of pregnancy that change pregnant people's desired outcome for a pregnancy. Similarly, there may be other barriers to obtaining an abortion before the third trimester, including barriers that are not policy‐related, such as experiences of abuse, violence, and/or kidnapping, and/or are rooted in the history of medical mistreatment of Black women and other women of color ³⁵ , ³⁶ , ³⁷ and ensuing medical mistrust. ³⁸ Nonetheless, a strength of the two‐pathway conceptualization is that it can accommodate these additional contours.

There are limits to the transferability of these findings: this sample includes only people who were able to obtain third‐trimester care. These findings cannot describe the experience of people who needed a third‐trimester abortion but were unable to obtain one. It may be that the pathways I delineate here are better identified as pathways to needing and obtaining third‐trimester abortion care. Still, research that includes people denied abortion care examining who is likely to seek an abortion at or after 20 weeks LMP suggests the utility of this two‐pathway framework: late recognition of pregnancy (i.e., new information) and obstacles to abortion are associated with seeking abortion later in pregnancy. ¹⁶

While third‐trimester abortion is relatively uncommon in the United States compared to first‐trimester abortion, ² investigation into how some pregnant people come to need an abortion after 24 weeks of pregnancy reveals the irreducibility of that need as well as key similarities to the experience of needing a abortion in the first or second trimester. These findings make a strong case for a social conceptualization of abortion not in terms of trimesters but, instead, as an option throughout pregnancy.

Biography

Katrina Kimport, PhD, is Associate Professor in the Department of Obstetrics, Gynecology & Reproductive Sciences and a research sociologist in the Advancing New Standards in Reproductive Health (ANSIRH) program at the University of California, San Francisco. Her research focuses on gender, health, and reproduction. She is the author of No Real Choice: How Culture and Politics Matter for Reproductive Autonomy (Rutgers, 2022).

Kimport K. Is third‐trimester abortion exceptional? Two pathways to abortion after 24 weeks of pregnancy in the United States. Perspect Sex Reprod Health. 2022;54(2):38‐45. doi: 10.1363/psrh.12190