If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Correspondence to: Sexual Health and Reproductive Equity Program, School of Social Welfare, University of California, Berkeley, 120 Haviland Hall MC 7400, Berkeley, CA 94720-7400.
Describe the prevalence of considering, wanting, and not obtaining a wanted abortion among a nationally-representative sample of 15-44 year olds in the United States who had ever been pregnant.
Study Design
We analyzed data from ever-pregnant respondents (unweighted n=1,789) from a larger online survey about contraceptive access using the nationally-representative AmeriSpeak panel. Among those not obtaining wanted abortions, weighted frequencies for sociodemographic characteristics and reasons for not getting the abortion are presented.
Results
Nearly 6% of the full sample reported having wanted an abortion they did not obtain. In open-ended responses, respondents most frequently reported individual reasons (43.8%) for not getting an abortion (e.g., changing their mind; personal opposition) and financial, logistical, or informational barriers (24.7%) likely related to policy. A quarter (24.1%) of the sample reported a past abortion. Among those who reported no past abortions, about one-fifth had considered abortion in the past, and 6.8% had wanted or needed one. Among those reporting no prior abortions who had considered abortion, only a third (34.3%) also report ever wanting or needing one.
Conclusions
This study begins to quantify the experience, even before the Supreme Court’s 2022 decision in Dobbs v. Jackson Women’s Health Organization, of being unable to obtain a wanted abortion. Additionally, findings suggest that people in a national sample will answer questions about whether and why they did not obtain a wanted abortion.
Implications
This study provides the first known national estimates of lifetime history of not getting a wanted abortion. Survey questions can be used for future research. Prospective and ongoing measurement of the inability to get a wanted abortion could be one part of documenting the effects of Dobbs on abortion access.
]. Such restrictions are accelerating in the wake of the Supreme Court’s June 2022 decision in Dobbs v. Jackson Women’s Health Organization, which overturned the constitutional right to abortion in the U.S. [
]. People who make it to an abortion clinic often have to travel long distances and incur high costs to receive care, as well as navigate misinformation, stigma, and harassment [
Given these barriers, there are people who want abortions who are unable to obtain them. Being unable to obtain a wanted abortion is harmful to people. Compared to people who obtained an abortion, people denied a wanted abortion experienced worse physical health, worse economic stability, and were more likely to remain connected to a violent partner [
The turnaway study: ten years, a thousand women, and the consequences of having--or being denied--an abortion [Internet]. First edition. Scribner,
2020
]. However, less is known about people who may want an abortion but never make it to a clinic. To address this gap, recent studies have examined the experiences of people who considered but did not obtain an abortion [
State abortion policies and Medicaid coverage for abortion are associated with pregnancy outcomes among individuals seeking abortion recruited using Google Ads: A national cohort study.
“It just seemed like a perfect storm”: A multi-methods feasibility study on the use of Facebook, Google Ads, and Reddit to collect data on abortion-seeking experiences from people who considered but did not obtain abortion care in the United States.
]. A mixed methods study among prenatal patients in Louisiana and Maryland found that approximately 30% had considered abortion for their current pregnancy [
]. A survey of pregnant people searching Google for information about how to access abortion found that four weeks later 32% were still seeking an abortion, and 20% decided to continue the pregnancy [
State abortion policies and Medicaid coverage for abortion are associated with pregnancy outcomes among individuals seeking abortion recruited using Google Ads: A national cohort study.
]. In these two studies, as well as a 2022 study among a sample of people who had not obtained an abortion despite having a pregnancy for which they believed abortion could have been their best option [
“It just seemed like a perfect storm”: A multi-methods feasibility study on the use of Facebook, Google Ads, and Reddit to collect data on abortion-seeking experiences from people who considered but did not obtain abortion care in the United States.
], people reported not obtaining abortions for a combination of personal and policy-related reasons.
These studies document the experience of those who consider but do not obtain abortions and provide insight into people’s reasons for not getting an abortion. However, as a field, our ability to describe and draw conclusions about the scope of this phenomenon remains limited by the limited research with small or non-representative samples. The present study aimed to describe the prevalence of reporting a history of considering, wanting, and not obtaining a wanted abortion among a national sample. This study contributes to a growing understanding of the nature and extent of not getting a wanted abortion, which will be increasingly necessary in a post-Dobbs landscape in which abortion seekers will face increased barriers to care.
2. Methods
Data came from a larger study of person-centered contraceptive need. Between January and March 2022, NORC at the University of Chicago [
] collected these data using their probability-based, nationally-representative AmeriSpeak panel, which is estimated to cover 97% of U.S. households and includes 54,001 members ages 13 and older from over 43,000 households [
Eligible panel members were between the ages of 15-44, assigned female sex at birth, not known to be sterile, and able to complete the survey in English or Spanish. NORC invited panelists to participate via email and panelists’ AmeriSpeak online account or mobile app. Respondents accessed the self-administered survey via web browser by clicking the link in the study invitation, which led to an online informed consent form. Those who provided consent proceeded to a series of screening questions, and eligible respondents continued to the survey. The survey took an average of 21 minutes. A cash equivalent of $8 was distributed to each participant in the form of NORC’s AmeriPoints [
]. The Committee for Protection of Human Subjects at the University of California, Berkeley and the NORC Institutional Review Board approved the study protocol.
2.1 Measures
This analysis focused on questions assessing the frequency and context of considering, wanting, and not getting a wanted abortion. We calculated the prevalence of four items: 1) ever considering an abortion, 2) ever wanting an abortion, 3) not obtaining a wanted abortion, 4) reasons for not obtaining a wanted abortion(s). The research team developed, selected, and refined questions based on a literature review, existing survey items, interviews with stakeholders working in abortion access, and feedback from abortion researchers. Figure 1 presents a flowchart of abortion-related questions and which respondents received them.
Fig. 1Survey questions and flow for items related to considering, wanting, and not obtaining a wanted abortion.
], which focused on pregnant individuals seeking prenatal care. We modified questions to be relevant regardless of pregnancy status. We created measures of ever considering or wanting an abortion based on two items. First, respondents who reported having ever been pregnant but did not report ever having an abortion were asked, “Have you ever considered having an abortion for any current or past pregnancy, even for just one second?” (yes, no, I’m not sure). We classified respondents who reported prior abortions as having considered abortion. Second, given that the Abortion Prenatal Study found that many people who considered abortion decided they did not want one [
], we added a second question: “Have you ever wanted or needed an abortion?” (yes, no, I’m not sure). Again, we classified respondents reporting prior abortions has having wanted or needed one. We classified respondents who reported a prior abortion as having both considered and wanted or needed an abortion to provide a holistic picture of the prevalence of these experiences among ever-pregnant individuals. This is in line with prior research showing that people who present for abortion care are highly certain of their decision (i.e., they want or need the abortion) [
We identified respondents who may have wanted an abortion but not obtained one in two ways: (1) those who responded “yes,” “I’m not sure,” or did not respond to ever wanting an abortion; and (2) respondents whose total number of pregnancies was higher than the number of abortions they reported, as they may have also wanted an abortion for a different pregnancy. We asked these respondents, “Have you ever wanted or needed an abortion that you did not get?” (yes, no, I’m not sure). Those who responded “yes” received an open-ended follow-up question: “What was the reason(s) you did not get the abortion(s)?” In these questions, we used the phrase “wanted or needed” to provide multiple entry points for responses, without making a value distinction between “wanting” or “needing” an abortion. Throughout the manuscript, we use the term “wanted” for simplicity and so as not to suggest a hierarchy of abortions, with the exception of using “wanted or needed” in the results to accurately describe responses to survey items.
2.2 Analysis
A total of 3,059 eligible individuals completed the online survey. Of these, 44.8% reported ever being pregnant and are therefore included in the analytic sample (unweighted n=1,789). We tabulated the prevalence of considering, wanting, and not getting a wanted abortion. To better understand respondents who reported not getting a wanted abortion, we present descriptive statistics about this group to describe their sociodemographic characteristics. For reasons people did not get a wanted abortion, we reviewed open-ended responses and categorized them into higher-level groupings. Categories included individual reasons (e.g., changing their mind; personal opposition); influence or pressure from others against abortion; financial, logistical, or informational barriers; and other reasons (e.g., COVID-19). We also coded responses indicating an abortion was not obtained because the pregnancy ended another way and respondents who did not provide a reason. We created binary variables to represent each category, as some respondents provided multiple reasons for not obtaining an abortion. We conducted analyses in STATA (version 17.0), using svy commands to account for weighting and complex survey design.
3. Results
3.1 Sample characteristics
The sample consists of 1,789 respondents who reported ever being pregnant (Table 1). Most respondents (82.3%) were parents, and 70.5% were married, engaged, or in a serious relationship. Nearly half (48.1%) were working full-time, and most (56.4%) had commercial health insurance.
Table 1Characteristics of a 2022 national online sample of individuals (ages 15-44) who had ever been pregnant in the U.S.
Unweighted n
Weighted %
Race/Ethnicity
White
1049
54.2
Black
234
15.5
Hispanic/Latinx
318
21.2
Asian and/or Pacific Islander
68
4.4
Multiracial
97
4.0
Other
23
0.7
Age
15-17
6
0.5
18-29
405
37.9
30-39
994
45.5
40-44
384
16.1
Parent
No
203
17.1
Yes
1580
82.3
Missing
6
0.6
Relationship status
Married, engaged, or in a serious relationship
1390
70.5
In another type of relationship
199
15.6
Not in a relationship (including separating/divorcing)
194
13.4
Missing
6
0.5
Census region
Northeast
214
12.0
Midwest
549
24.2
South
612
38.3
West
414
25.6
Urbanicity
Urban
718
36.5
Suburban
763
44.4
Rural
308
19.2
Born in the United States
Yes
1552
85.0
No
204
12.2
Prefer not to say
25
2.3
Missing
8
0.5
Current employment status
Working full time
915
48.1
Working part time
311
15.9
Not working for pay
550
34.9
Other
5
0.4
Missing
8
0.7
Unweighted n
Weighted %
Highest education completed
High school equivalent or less
319
33.3
Vocational or technical school, some college, or associates degree
706
35.7
Bachelor's degree, post graduate study, or professional degree
764
31.1
Insurance type
Commercial (e.g., employer-based, direct purchase, health insurance exchange)
1162
56.4
State Medicaid or CHIP
366
26.8
Other public insurance (including Medicare, military/VA, Indian Health Service)
3.2 Prevalence of considering, wanting, and (not) obtaining an abortion
About a quarter (24.1%) of the sample reported having had an abortion (Table 2). Nearly 40% of the sample had ever considered abortion for a pregnancy and nearly 30% had wanted or needed one. Among those who reported no past abortions, about one-fifth had considered abortion in the past, and 6.8% had wanted or needed one.
Table 2Obtaining, considering, and wanting or needing an abortion among a 2022 national sample of individuals (ages 15-44) who had ever been pregnant in the U.S.
All individuals who had ever been pregnant (full sample)*
Individuals who did not report a past abortion
Unweighted n
Weighted %
Unweighted n
Weighted %
Ever had an abortion
Yes
400
24.1
-
-
No
1374
75.1
1374
98.9
Missing
15
0.9
15
1.1
Had ever considered abortion
Yes**
653
38.5
253
19.0
No
1084
57.6
1084
75.9
Not sure
32
2.3
32
3.5
Missing
20
1.2
20
1.6
Had ever wanted or needed an abortion
Yes***
485
29.2
85
6.8
No
1243
66.4
1243
87.4
Not sure
39
3.2
39
4.2
Missing
22
1.2
22
1.6
Note: Full sample unweighted n=1789; individuals who did not report a past abortion unweighted n=1389
* Only respondents who did not report a prior abortion were asked directly if they had ever considered abortion or had ever wanted or needed an abortion (Figure 1). In this table (full sample column), those who reported a past abortion were categorized as having considered and wanted or needed an abortion.
** This 38.5% who had ever considered an abortion includes 14.4% who did not report a prior abortion and the 24.1% who reported having had an abortion.
*** This 29.2% who had ever wanted an abortion includes 5.1% who did not report a prior abortion and the 24.1% who reported having had an abortion.
Table 3 presents the prevalence of not obtaining a wanted abortion among the overall sample and the subgroup of respondents who may have wanted an abortion but did not obtain one. In the full analytic sample, including those who reported prior abortions, 5.8% had wanted or needed an abortion that they did not get during their lifetimes, and 1.2% were unsure. The subgroup of respondents who may have wanted an abortion but did not obtain one comprises 25.2% of the sample (unweighted n=452) and includes those who (1) said they had wanted or needed an abortion in the past but reported no abortions and/or (2) reported more pregnancies than abortions (see Methods and Figure 1). Among this subset, 22.6% reported wanting or needing an abortion they did not get, and 4.8% were not sure. Of note, not everyone who reported considering abortion also reported wanting an abortion: among those reporting no prior abortion who had considered abortion, 34.3% also reported having wanted or needed one, while 8.3% were unsure if they had ever wanted or needed an abortion (not shown).
Table 3Not obtaining a wanted or needed abortion among a 2022 national sample and subgroup of individuals (ages 15-44) who had ever been pregnant in the U.S.
Full sample
Subgroup
15-44 year olds who had ever been pregnant
Had a past pregnancy for which they may have wanted an abortion and not obtained one*
Unweighted n
Weighted %
Unweighted n
Weighted %
Have you ever wanted or needed an abortion that you did not get?
Yes
96
5.8
96
22.6
No
322
17.4
322
67.9
Not sure
16
1.2
16
4.8
Missing
18
1.2
18
4.6
N/A
1337
74.3
Total
1789
452
Notes:
Full sample unweighted n=1789; subgroup unweighted n=452
See Figure 1 for a full depiction of which respondents were asked which questions
* Subgroup includes respondents who reported more pregnancies than past abortions, as they could have wanted an abortion for a different pregnancy and not obtained one. It also includes respondents who answered yes, not sure, or skipped the question asking if they had ever wanted or needed an abortion.
3.3 Respondents who reported not getting a wanted abortion
Among the subset of respondents who reported not getting a wanted abortion (unweighted n=96), 39.4% identified as white, 33.2% as Hispanic/Latinx, and 15.7% as Black (Table 4). Nearly 45% each were between 18-29 and 30-39. The majority were parents (90.5%). Over a quarter (27.6%) reported having had at least one abortion. Nearly half (45.1%) lived in the South. Similarly, approximately half had public insurance. Most reported some current financial strain, with two-thirds (65.4%) indicating it was difficult to live on their current household income. Nearly three-quarters (73.1%) reported they could not currently afford an emergency expense of $400 or more.
Table 4Characteristics of respondents who reported not obtaining a wanted or needed abortion among a 2022 national sample of individuals (ages 15-44) in the U.S.
Unweighted n
Weighted %
Race/Ethnicity
White
36
39.4
Black
19
15.7
Hispanic/Latinx
24
33.2
Asian and/or Pacific Islander
4
4.6
Multiracial
10
6.0
Other
3
1.1
Age
15-17
0
0.0
18-29
30
44.1
30-39
49
44.9
40-44
17
11.0
Parent
No
7
9.5
Yes
89
90.5
Ever had an abortion
No
67
72.4
Yes
29
27.6
In a serious relationship
Married, engaged, or in a serious relationship
59
60.6
In another type of relationship
18
19.3
Not in a relationship (including separating/divorcing)
19
20.1
Census region
Northeast
11
9.8
Midwest
26
18.3
South
38
45.1
West
21
26.8
Urbanicity
Urban
38
35.9
Suburban
42
42.5
Rural
16
21.6
Born in the United States
Yes
86
90.8
No
10
9.2
Unweighted n
Weighted %
Current employment status
Working full time
42
45.3
Working part time
18
17.2
Not working for pay
36
37.5
Highest education completed
High school equivalent or less
24
39.7
Vocational or technical school, some college, or associates degree
47
40.0
Bachelor's degree, post graduate study, or professional degree
25
20.4
Insurance type
Commercial (e.g., employer-based, direct purchase, health insurance exchange)
37
36.7
State Medicaid or CHIP
38
47.5
Other public insurance (including Medicare, military/VA, IHS)
Most respondents (91.2%) provided the reason(s) they did not obtain a wanted abortion in the open-ended follow-up question. The majority (95.9%) of responses addressed a single category (Table 5). Notably, 11.4% reported that their pregnancy ended another way (miscarriage, stillbirth, ended up not being pregnant). Most commonly, respondents reported what we classified as an individual reason for not getting an abortion (43.8%), such as not being able to “go through with it.” Some specified they worried about experiencing difficult emotions if they had an abortion, such as guilt or regret. Other common individual reasons included respondents reporting they changed their minds, decided to continue the pregnancy, or personally opposed abortion for moral or religious reasons. Beyond individual reasons, some respondents reported influence or pressure from others (including influence, pressure, or threats from their partner at the time) as the reason they did not obtain an abortion.
Table 5Categorization of open-ended responses of reasons respondents reported not getting a wanted or needed abortion among a 2022 national sample in the U.S.
Unweighted n
Weighted %
Individual reasons
41
43.8
Influence or pressure from others against abortion
A quarter (24.7%) of respondents who did not obtain a wanted abortion also reported financial, logistical, and informational barriers. Some said they did not get the abortion because the pregnancy was “too far along.” Elaborations on this phrase referred to state-level gestational duration bans, facility duration limits, and respondents’ assessment of a point in pregnancy after which they would not choose abortion. About 1 in 10 reported high out-of-pocket financial costs as a reason they did not obtain an abortion.
4. Discussion
This study describes the prevalence of considering, wanting, and not obtaining an abortion in a nationally-representative sample of 15-44-year-olds who had ever been pregnant. Nearly 6% of respondents reported having wanted an abortion they did not get. To our knowledge, no similar data exist from a national population-based sample.
In addition to asking respondents if they had ever considered abortion, as prior studies have [
“It just seemed like a perfect storm”: A multi-methods feasibility study on the use of Facebook, Google Ads, and Reddit to collect data on abortion-seeking experiences from people who considered but did not obtain abortion care in the United States.
], we also asked whether they had ever wanted an abortion. While there was overlap, only about one-third of respondents who had not had an abortion but had considered abortion also said they had wanted one, and 8% were unsure. Findings highlight the conceptual importance of distinguishing between considering and wanting an abortion in studies that aim to characterize pregnancy decision making. After all, the true measure of reproductive autonomy is not having an abortion; it is having the ability to consider all pregnancy outcomes and choose and enact one’s preferred outcome (including both abortion and continuation) without constraints [
]. To the extent that researchers and advocates do not recognize this distinction, they may misunderstand the role of abortion in people’s lives as well as barriers and facilitators to ensuring people’s abortion needs are met, which could result in misdirecting attention and resources.
These data highlight the importance of a nuanced approach to investigating wanting an abortion. In terms of why people did not get a wanted abortion, respondents most commonly reported what we labeled “individual reasons,” which have also been reported in other studies [
“It just seemed like a perfect storm”: A multi-methods feasibility study on the use of Facebook, Google Ads, and Reddit to collect data on abortion-seeking experiences from people who considered but did not obtain abortion care in the United States.
]. These data support the idea that, just as with other reproductive experiences, it is common for people to weigh abortion as an option for a pregnancy, even one they ultimately decide to continue. It may be more accurate to describe respondents’ experience as not getting an abortion that was wanted at one point and distinguishing them from people seeking abortions who could not obtain them due to structural barriers. For example, as in prior studies [
“It just seemed like a perfect storm”: A multi-methods feasibility study on the use of Facebook, Google Ads, and Reddit to collect data on abortion-seeking experiences from people who considered but did not obtain abortion care in the United States.
], our respondents also named gestational duration and financial cost as reasons for not obtaining an abortion. These reasons may be attributed to policies that set gestational limits and limit insurance coverage, which are widely known to act as barriers to abortion care [
], the number of people who cannot obtain a wanted abortion for structural reasons may also increase.
Importantly, individual reasons always exist within broader context and may be influenced by the norms, values, and input of important others and community as well as people’s ability to implement their decision. People live within contexts and structures that can make abortion unfeasible, unobtainable, or unchooseable [
]. To the extent those influences are anti-abortion, proffered reasons that we categorized as individual reasons could be underpinned by interpersonal, social, and structural factors. It is possible that these were responses to real assessments of how (im)possible obtaining an abortion was—geographically, logistically, financially, and socially.
This study makes methodological contributions and has implications for future research. We asked respondents two original questions, which can be used and adapted for future research: whether they had ever wanted or needed an abortion and whether they had ever wanted or needed an abortion they did not get. The vast majority of respondents who reported not getting a wanted abortion responded to an open-ended question about why, demonstrating the feasibility of asking a national sample to disclose information about abortions they wanted at (at least) one point but did not obtain.
Numerous limitations of this study must be acknowledged. Given limited space in the extant survey, we asked abortion-related questions in terms of lifetime history. We therefore do not know for which or how many pregnancies respondents considered, wanted, or did not obtain a wanted abortion. We also do not know when or where respondents did not get their wanted abortion(s), or other information to contextualize their lives at the time. To better understand health equity implications of these data, future research should explore patterns in the sociodemographic contexts of not obtaining a wanted abortion. We relied on retrospective report across a lifetime; respondents’ recall may decline or shift over time, which may explain why some respondents were unsure whether they ever wanted an abortion they did not get. Additionally, research among people denied an abortion shows that as time passes, reports of the intendedness of the pregnancy shifts [
]; similarly, assessment of wanting an abortion in the past could evolve. Respondents may have been reluctant to report that they wanted an abortion if they went on to parent the resulting child. Abortion is highly stigmatized and often underreported in research [
]; data likely underestimate how many people had, considered, wanted, and did not obtain a wanted abortion. That a quarter of the sample reported a past abortion suggests a willingness to disclose. Finally, while adapted and new survey questions were reviewed by subject experts, we did not conduct cognitive interviews [
], and so cannot confirm that respondents understood the questions as intended. Future research should investigate how people understand terms such as considering, wanting, and/or needing an abortion.
In the wake of the 2022 Dobbs Supreme Court decision and state-level restrictions that are proliferating in response [
], it will likely become increasingly common that people do not obtain abortions they want. Prospective and ongoing measurement of the prevalence of not obtaining a wanted abortion could be one part of documenting the effects of Dobbs on abortion access. This study contributes to a growing and timely body of research about the frequency and context of not obtaining wanted abortions in the U.S.
Acknowledgements
The authors thank Dr. Jennet Arcara for her role in designing the survey for the parent study, reviewing questions, and providing support and consultation on data analysis. We also thank Dr. Corinne Rocca for reviewing survey questions.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declarations of interest
None
References
Nash E.
Cappello O.
Naide S.
Mohammed L.
Ansari-Thomas Z.
Radical Attempts to Ban Abortion Dominate State Policy Trends in the First Quarter of 2019 [Internet].
The turnaway study: ten years, a thousand women, and the consequences of having--or being denied--an abortion [Internet]. First edition. Scribner,
2020 (Available from)
State abortion policies and Medicaid coverage for abortion are associated with pregnancy outcomes among individuals seeking abortion recruited using Google Ads: A national cohort study.
“It just seemed like a perfect storm”: A multi-methods feasibility study on the use of Facebook, Google Ads, and Reddit to collect data on abortion-seeking experiences from people who considered but did not obtain abortion care in the United States.
00069-0&id=gr1.jpg){kind=link}