Timing of pregnancy discovery among women seeking abortion^,

Not recognizing pregnancy in the first few weeks increases the likelihood of needing an abortion in the second trimester, which can exacerbate barriers to care. Most obviously, one cannot get an abortion in the first trimester if one only realizes that one is pregnant in the second. But the delays associated with later recognition are not limited to just the time it takes to suspect and test for pregnancy. When a pregnant person is past the first trimester, the logistical barriers to getting an abortion—the cost of the procedure and travel, the time off required, the need for transportation and finding childcare for existing children—increase []. The further along they are in pregnancy, the higher the cost of an abortion and the farther they must travel to find one of the few existing clinics able to provide one at their stage in pregnancy. In the Turnaway Study, we found that later recognition of pregnancy, beyond 10 weeks, sets off a cascade of delays [].

Delays in recognition can have important effects on the ability to receive care, especially given that restrictions on abortion can delay care and even lead to denial of services, particularly where gestational age limits establish a cut off for abortion services across a whole state and in states with mandatory waiting period laws. Furthermore, seeking abortion early in pregnancy allows people the option of a medication abortion (often only available through 11 weeks of pregnancy) for those who prefer it over an in-clinic procedure []. First trimester abortion is also substantially less expensive than second trimester abortions []. Given the restrictions on funding, abortion, particularly later procedures, can be a catastrophic health expenditure for many pregnant people [].

Yet, few studies have explored the factors that influence the timing of pregnancy recognition, especially in pregnancies ending in abortion. A study of women seeking abortion in California found that among those in the second trimester, one fifth suspected they were pregnant and over half (58%) confirmed they were pregnant in the second trimester []. The study found that pregnancy symptoms, such as nausea and vomiting, shortened the time to recognize pregnancy while not keeping track of one's menstrual period was associated with a longer time to recognize pregnancy. Another study among women in Utah similarly found that women who did not track their menstrual periods, as well as those who frequently used drugs, had higher odds of discovering their pregnancies later [].

The Turnaway Study was a longitudinal study of nearly 1000 women in the United States who either received or were denied an abortion because they were just under or just over the gestational limit at the clinic in which they sought care. This study provides a unique sample of people seeking abortion later in pregnancy. Previous analyses from the US Turnaway Study found that many women, including those seeking abortion in the first trimester (104/273 38%), just under the gestational limit (194/452 43%) and just over the gestational limit (111/231 48%), report that not realizing that they were pregnant was a cause of delay []. However, the point in pregnancy at which they discovered pregnancy was very different for those receiving an abortion within the first 14 weeks and those receiving an abortion after 20 weeks (5 weeks vs 12 weeks). Understanding the prevalence and predictors of later recognition of pregnancy sheds light on the impact of policies which restrict access to abortion based on gestational age.

The Turnaway Study was a 5-year longitudinal study of women who presented for abortion care at 1 of 30 facilities throughout the United States between 2008 and 2010. Gestational limits at the study facilities ranged from the end of the first trimester to the end of the second. Each facility had the latest gestation age limit of any provider within 150 miles. Study participants were pregnant with no known fetal anomalies or demise who spoke English or Spanish and were aged 15 years or older. Participants were enrolled into 3 study groups in a two-to-one-to-one ratio on the basis of ultrasound dating of gestational age relative to each facility's limit: (1) Near Limits presented for abortion up to 2 weeks under the facility's gestational age limit and obtained wanted abortions, (2) Turnaways presented for abortion up to 3 weeks over a facility's limit and were denied abortions, and (3) First Trimesters received abortions at gestations up to 14 weeks. Study participants completed a baseline telephone interview 1 week after either receiving or being denied an abortion and telephone interviews every 6 months.

The Turnaway Study also included semistructured in-depth qualitative telephone interviews with 31 participants who had completed the 5 years of interviews; all respondents had discovered their index pregnancy at least 5 years before the interview. Twenty-eight were randomly selected from the pool of participants who had completed the 5-year semi-annual interviews by October 2014 and had agreed to future contact by the researchers. An additional 3 participants who did the interviews in Spanish or who had voluntarily placed a baby for adoption were also interviewed. The qualitative interviews took place between December 2015 and March 2015 and ranged from 30 minutes to 2 hours in length. The interview guide was designed to elicit open ended responses about participants’ reflections about their pregnancy experiences. Questions explored participants’ pregnancy-related experiences including when they first discovered they were pregnant, when they sought pregnancy-related care (abortion, prenatal care, birth), as well as significant life changes that occurred in the intervening years related to work, school, family, and relationships. Data were coded using deductive methods using the interview guide and inductively through open coding in Atlas.ti7. We developed thematic summaries and, for this paper, focused in on the experiences of pregnancy discovery.

Using the baseline quantitative surveys and in-depth qualitative interview Turnaway study data, this current analysis explores the reasons for delayed recognition of pregnancy among people seeking abortion. To measure timing of pregnancy recognition, we asked all participants at the baseline interview, “When did you find out that you were pregnant?” We recorded answers as each woman reported it: a precise date, weeks of gestation, or elapsed time. We converted all answers to weeks of pregnancy from last menstrual period. We asked about the number of previous births and date of the most recent birth. We asked the woman's household income in the past month and the number of people who shared this income to calculate the ratio of the household's income to the federal poverty threshold. We asked the woman's height and weight and calculated each woman's body mass index. We asked about the woman's physical health prior to pregnancy recognition on a 5-point scale from very good to very poor. We asked whether the woman had ever received a diagnosis of depression (includes depression, dysthymia or bipolar disorder) or a diagnosis of an anxiety disorder like panic, obsessive-compulsive, anxiety, or post-traumatic stress disorder. We asked about recreational drug use in the month prior to pregnancy recognition and whether she experienced problem alcohol symptoms including drinking more than 5 drinks on an occasion, having a drink first thing in the morning or having been unable to remember what happened the night before due to drinking. We asked whether the woman had been using a method of contraception in the month she became pregnant and what type of method she had used. We present bivariate analyses where the test for significance adjusts for clustered sites of recruitment. We use a multivariate model, also adjusted for clustered recruitment, to predict factors associated with discovery of pregnancy in the second trimester, at 14 weeks and beyond. We use a multivariate model, adjusted for clustered recruitment, to predict factors associated with discovery of pregnancy in the second trimester, at 14 weeks and beyond. In this model, we included all variables hypothesized to be associated with late discovery, based on our interviews and the literature, that were contained in our quantitative data collection. These include demographics (age, race/ethnicity, household income relative to the federal poverty threshold, and education), health variables (self-reported physical health prior to pregnancy recognition, body mass index prior to pregnancy, diagnosis of depression or anxiety, and drug use or problem alcohol symptoms prior to discovering pregnancy) as well as 2 reproductive health variables (contraceptive use in the month of conception, and when and whether she had given birth previously).

The Turnaway Study is unique in its large sample of women seeking abortions later in pregnancy, enabling us to look at how delays in recognition of pregnancy contributes to abortion seeking later in pregnancy. We found that among the almost one thousand US women who participated in the study, those seeking second trimester abortions were more often nulliparous and using hormonal contraception when they became pregnant than those seeking earlier abortions. In our in-depth interviews, we learned that some women never experienced pregnancy symptoms. Some women experienced severe nausea and others experienced none. Some have an immediate stop to their menstrual periods and others experience spotting for many months. Having few pregnancy symptoms is part of the large range of normal physiological responses to pregnancy [].

Our finding that people who had not previously carried a pregnancy to term were more likely to be delayed in recognizing pregnancy may have several explanations. A lack of familiarity with symptoms may have resulted in delayed recognition. Nulliparity is also associated with younger age and, therefore, irregular periods and a lower chance in noticing a missed period []. In this study, hormonal contraceptive use was also associated with later recognition—some methods stop menstrual periods altogether and so a woman has one less pregnancy symptom to observe. Effective contraceptive use may also have delayed recognition if the woman believed she was protected from the chance of conception or if symptoms of pregnancy were mistaken for side effects of contraception.

Although a small number of women in the Turnaway Study reported that they were “in denial" about the pregnancy, this does not seem to explain much of the observed delay in recognizing pregnancy. More often, women did not have clear symptoms that indicated that they were pregnant. Several previous studies have shown that unplanned pregnancies are discovered later than intended ones [,]. When a person is not expecting to be pregnant, symptoms stemming from other causes, such as a chronic health condition and having recently delivered a child, were not attributed to a new pregnancy. Previous studies have also found that drug use and smoking are associated with later onset of symptoms or pregnancy discovery [,,. We did not find that these factors were associated with later discovery of pregnancy in our quantitative analyses or in the study as a whole, although they may have been important factors for individual women.

We were not able to measure all the factors that may have slowed women down from recognizing pregnancy earlier; the study was not primarily designed to evaluate the causes of delay in discovery and so did not contain questions on experience of pregnancy symptoms or history of tracking menstrual periods, limiting our ability to evaluate some of the known contributors to later discovery. Further research should focus on how the experience of symptoms affects the timing of pregnancy recognition.

Late detection of pregnancy can cause a cascade of problems for people who do not want to carry a pregnancy to term. Previous work from the Turnaway study has shown that initial delays lead to greater burdens of travel, cost and logistics []. As McCarthy and colleagues point out in their study of women in Utah, “Early pregnancy discovery is particularly important for women who may face additional constraints in obtaining abortion care, such as those living in states with mandated waiting periods, gestational limits for abortion, limited availability of second-trimester abortion services, and who live a considerable distance from an abortion provider.”[]

Access to early abortion may become even more important with the possibility of the Supreme Court allowing pre-viability bans on abortion (Dobbs vs Jackson Women's Health to be decided by the US Supreme Court in 2022). Laws imposing gestational limits make abortion unavailable to people who discover pregnancy after the limit. Such laws do not consider the varied and understandable reasons people have for seeking abortion care later in pregnancy. Our findings suggest that such bans will disproportionately affect people who were using contraceptive methods to prevent pregnancy and people who have never given birth before. Backers of gestational limits may assume that people are delaying their search for abortion services for reasons that are in their control. Yet we find that many people who seek abortion services later in pregnancy seek care as expeditiously as they can after discovering that they are pregnant []. In the current policy environment, delay in discovery of pregnancy results in greater costs, farther travel and often precludes the option of abortion altogether. Removing gestational age limits would help to ensure that people who discover their pregnancies at later gestational ages are not excluded from abortion services.

The authors declare no conflict of interest.

This study was supported by research and institutional grants from the Wallace Alexander Gerbode Foundation, The David and Lucile Packard Foundation, The William and Flora Hewlett Foundation, and an anonymous foundation.

The authors thank Rana Barar and Sandy Stonesifer for study coordination and management; Mattie Boehler- Tatman, Janine Carpenter, Undine Darney, Ivette Gomez, Selena Phipps, Brenly Rowland, Claire Schreiber and Danielle Sinkford for administering surveys; Michaela Ferrari, Debbie Nguyen, Jasmine Powell, and Elisette Weiss for project support; and all the participating providers for their assistance with recruitment.