Pregnancy care-seeking stress in the USA: a prospective cohort study Authors: M Antonia Biggs, Lauren Ralph, Diana Greene Foster, Stephanie Arteaga, Corinne H Rocca Published: 2026 ABSTRACT AbstractObjectivesTo prospectively assess pregnancy-related care sought and obstacles and stress experienced by newly pregnant people.DesignThe ADAPT Study, a longitudinal cohort study, followed 2015 non-pregnant participants aged 15–34 years for a year. Those with confirmed incident pregnancies were followed through their pregnancies and for 3 years.SettingWe recruited participants from 23 reproductive and primary care facilities located in five southwestern states with different sociopolitical reproductive health contexts (restrictive, Arizona and West Texas; protective, southeastern California, Nevada and New Mexico).Participants334 people reported a new pregnancy within 1 year of enrolment; 324 with outcome data are included in this analysis.Primary and secondary outcome measuresTypes of pregnancy care sought (‘Have you looked into where or how you could get (prenatal care, abortion care or adoption services)?’) and care-seeking stress (‘How stressful was it to find (prenatal, abortion or adoption) care for this pregnancy?’).ResultsMost participants (83%, 270/324) sought prenatal care; 43% (138/324) sought abortion care; and 5% (17/324) sought adoption services. Overall, 17%, 29% and 23%, respectively, reported that care-seeking was extremely/quite a bit stressful. Abortion care-seeking was associated with significantly more stress than seeking prenatal care in the ordinal (adjusted odds ratio (aOR 1.70, 95% CI 1.10 to 2.62) but not logistic (aOR 1.33, 95% CI 0.74 to 2.38) model. Adoption care-seeking stress did not differ from prenatal care-seeking stress in either model. Participants who experienced any type of abortion care-seeking obstacle and those recruited in a state with a restrictive policy environment (aOR 2.72, 95% CI 1.09 to 6.80) reported more care-seeking stress than their counterparts.ConclusionsPeople who seek pregnancy care often experience some care-seeking stress, regardless of the type of care they seek. Findings point to the need to reduce the burden of the pregnancy care-seeking process across all types of pregnancy care.Trial registration numberNCT03888404.Keywords: Pregnancy; Stress, Psychological; OBSTETRICS; PUBLIC HEALTH STRENGTHS AND LIMITATIONS OF THIS STUDY This cohort study followed participants prospectively from before pregnancy through their care-seeking process, capturing the types of pregnancy care sought, associated factors and care-seeking stress. This cohort study followed participants prospectively from before pregnancy through their care-seeking process, capturing the types of pregnancy care sought, associated factors and care-seeking stress. This prospective study design allowed us to robustly capture people’s care-seeking as they became pregnant and throughout their care-seeking process. This prospective study design allowed us to robustly capture people’s care-seeking as they became pregnant and throughout their care-seeking process. With a sample of 324 pregnancies, analyses may have been underpowered to detect meaningful differences in care-seeking stress by type of care sought, particularly among those seeking adoption care. With a sample of 324 pregnancies, analyses may have been underpowered to detect meaningful differences in care-seeking stress by type of care sought, particularly among those seeking adoption care. While we captured the obstacles faced by people seeking abortion and their relationship with care-seeking stress, we did not capture obstacles faced by those who sought prenatal or adoption care. While we captured the obstacles faced by people seeking abortion and their relationship with care-seeking stress, we did not capture obstacles faced by those who sought prenatal or adoption care. Introduction Pregnancy is a time when people face an array of decisions, including navigating whether and how to seek pregnancy-related care such as prenatal, abortion and/or adoption services. The existing literature has explored some of the challenges associated with each pregnancy-care pathway. While each of these care options involve considerations and stressors, access to abortion care may be particularly difficult due to stigma and policies that restrict access to abortion.1,6 In the USA, facility-based abortion care is unevenly distributed, with about one in four states lacking a brick-and-mortar abortion facility due to state-level abortion bans.7 8 Even states that have maintained access to abortion since the 2022 overturning of Roe vs Wade, impose restrictions such as gestational limits, mandatory waiting periods, two-visit requirements, parental involvement requirements and lack of insurance coverage for abortion. These restrictions add logistical and financial burdens, forcing some people to travel long distances, causing delays and potentially preventing people from even considering or seeking abortion care.9 10 Few studies on abortion care-seeking have examined whether experiencing obstacles to abortion care is associated with increased stress, anxiety and depression.11,14 Similarly, research on prenatal care has shown that personal and structural factors as well as having an unplanned pregnancy prevented people from timely access to prenatal care services.15,19 In particular, groups who may fear criminalisation, such as people who are undocumented or have substance use disorders, report fears of being reported to government authorities as a reason to avoid accessing prenatal care.20,22 However, the extent to which seeking prenatal care is stressful has not been well explored. Finally, studies examining the barriers to and stress of seeking adoption services are limited; research suggests that adoption tends to be the least preferred pregnancy option.23 This literature has mostly examined the coercive practices and emotional complexities of relinquishment for birth parents.23,28 There remains a notable gap in the literature in terms of examining people’s stress as they pursue different pregnancy care-seeking pathways, including the unique structural and societal obstacles faced by those seeking abortion. By better understanding how stress differs across the types of care sought and identifying the obstacles that contribute to care-seeking stress, we can better develop and support approaches tailored to reduce the stress faced by pregnant individuals. This analysis is part of the Attitudes and Decision-Making After Pregnancy Testing (ADAPT) study, which followed non-pregnant people who became pregnant within 1 year of recruitment and continued to follow them through their care-seeking process. This paper aims to assess the types of pregnancy care people seek, experiences of care-seeking stress, and whether care-seeking stress differs by type of care sought. We also describe the obstacles individuals face when trying to find abortion care and whether these obstacles are associated with care-seeking stress. Unlike retrospective measurements of people’s care-seeking experiences which can lead to underreporting in the types of care sought, we assess the care-seeking experiences of people who were followed prospectively through their pregnancy care-seeking processes. Methods We analysed data from the ADAPT study, a longitudinal cohort study which aims to assess the factors associated with people’s pregnancy and childbearing preferences, their pregnancy decision-making and care-seeking, and the effects of experiencing pregnancy and birth on people’s health and well-being. A detailed cohort profile and methods of the ADAPT study have been previously published.29 Briefly, we recruited study participants from 23 reproductive and primary care facilities in five states in the Southwestern USA, Arizona, southeastern California, Nevada, New Mexico and West Texas, with the aim of including safety-net facilities that varied in patient volume and urbanicity. We enrolled and prospectively followed an underlying cohort (UC) of participants who were not pregnant (or were having an abortion) at screening. Participants who experienced new pregnancies during the 1-year UC were transferred into a new pregnancy cohort, where we continued to follow them over 3 years and documented their pregnancy experiences. This analysis examines the participants who experienced and reported incident pregnancies. This study follows Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines for cohort studies. All participant recruitment procedures were approved by the University of California, San Francisco (UCSF) Institutional Review Board (#17-23592). Participant eligibility and recruitment Participant eligibility criteria were designed to include people likely to experience an incident pregnancy across a spectrum of pregnancy preferences. To participate, individuals had to be 15–34 years of age given that people outside of this age range have lower pregnancy rates;30 have a uterus; be sexually active in the past 3 months with someone who has sperm; and not be using a long-acting or permanent contraceptive and not pregnant. Individuals needed to speak and read English or Spanish, reside in a study state or a bordering state, have access to a phone, smartphone or the internet and be willing to be contacted by the research team over 1–4 years. Participants were recruited from March 2019 to October 2022. Research assistants (RAs) approached patients in the facility waiting room, handed a study flyer to those eligible and interested in participating in a study and reviewed an informed consent form, which the participant signed electronically on a tablet. Data collection procedures Participants completed a baseline survey and quarterly follow-up surveys (3, 6, 9, 12 months) about participants’ sociodemographic characteristics, pregnancy preferences, new pregnancies. They completed interim pregnancy check-in surveys (1.5, 4.5, 7.5, 10.5, 13.5 months) about incident pregnancies. Participants received a gift card after completing the baseline (US$50) and each UC follow-up (US$20) and check-in (US$5) survey. Participants who reported an incident pregnancy (confirmed via urine pregnancy test or healthcare provider) were then followed for 3 years and completed surveys at pregnancy report; 6 weeks; and 3, 6, 9, 12, 18, 24, 30 and 36 months. Participants were surveyed prospectively about their pregnancy circumstances, decisions, care-seeking, and health and well-being outcomes. Questions were repeated retrospectively after the end of their pregnancy. Participants received a US$50 gift card for the first and last surveys related to the incident pregnancy and a US$35 gift card for those in between. Outcome variables In this analysis, we had two primary outcome variables of interest. Type of pregnancy care sought was measured by asking participants separately about each type of pregnancy care (prenatal, abortion and adoption) ‘At any point in this pregnancy, have you looked into where or how you could get [prenatal care, abortion care or adoption services]?’ (yes/no). Each question was followed by examples (eg, looking for clinics, healthcare providers or information on the internet or asking friends, family or someone else; looking for agencies or services or for another person to place the baby with permanently). These questions were intended to capture information-seeking and care-seeking behaviours and not whether the participant actually received care. Using these responses, we created three dichotomous variables since people could seek more than one type of care: prenatal (yes/no), abortion (yes/no) and/or adoption (yes/no) care. We also examined care-seeking stress by asking participants separately about each type of care sought, ‘How stressful was it to find (prenatal care, abortion care or adoption services) for this pregnancy?’ with the following Likert-scaled options: not at all stressful, a little bit stressful, moderately stressful, quite a bit stressful, extremely stressful and don’t know. For some analyses, we dichotomised responses: 1 (extremely stressful/quite a bit stressful) and 0 (not at all stressful/a little bit/moderately stressful/don’t know). These data were collected prospectively: we surveyed participants about their pregnancy care seeking and care-seeking stress once they reported a new incident pregnancy and then repeatedly over time, including once after they reported a pregnancy outcome. For participants with multiple measurements, we included their highest response (most stressful). Pregnancy outcome categories included gave birth, had an abortion and miscarried. For those who gave birth, we also examined whether they placed the baby for adoption. Pregnancy-care visits and sources of information about pregnancy care-seeking served as secondary outcomes. To identify prenatal care visits, we asked ‘At any point in this pregnancy, have you seen a doctor, midwife or nurse for prenatal care?’; for people who sought abortion care, we asked ‘Have you visited a clinic or health facility for an abortion during this pregnancy?’; and for those who sought adoption we asked ‘Have you visited any adoption agencies or services?’ (yes/no). To assess sources of information, we asked ‘Where did you look for information (on where to get prenatal care, on where or how to get an abortion, about adoption)?’, followed by a ‘select all that apply’ list of response options (internet; a clinic or health facility; asked a doctor, midwife or nurse; a pregnancy resource centre (sometimes known as a crisis pregnancy centre); asked friends, asked family, a church and other). Independent variables Type of pregnancy care sought (ie, ‘looked into where or how you could get (prenatal care, abortion care or adoption services)’) also served as an independent variable: prenatal (yes/no), abortion (yes/no) and/or adoption (yes/no) care. We assessed the outcomes of each participant’s pregnancy (birth and raise, abortion, miscarriage, birth and adoption). We also examined abortion care-seeking obstacles which included 15 obstacles derived from the published literature31 as well as the state policy context of the recruitment site. We grouped obstacles according to three thematic content areas: personal obstacles (four items: not sure I wanted abortion, and family, partner or friends not wanting me to have an abortion); practical obstacles (six items: difficulty arranging childcare, getting time off work, raising money to pay for the abortion or travel and getting insurance to cover the abortion, not being able to find a clinic nearby and not knowing where to go); and clinic or legal obstacles (five items: not wanting to pass clinic protestors, parental involvement or waiting period requirement, and waiting for the first or second abortion appointment). For each type (personal, practical, or clinic or legal) of obstacle and all 15 obstacles combined, we created a categorical variable describing the number of abortion care-seeking obstacles (ie, ‘zero’, ‘one’, ‘two or more obstacles’) experienced. To account for the different sociopolitical contexts and structural barriers to reproductive healthcare services in which participants might have sought care, including contraception, abortion, and pregnancy care, we included the policy environment of the state of the recruitment site categorised as restrictive (Arizona and Texas) or protective (California, Nevada, and New Mexico) access to reproductive health services.32 Covariates For covariates, we included participant sociodemographic characteristics, mental health, social support and pregnancy history, which were measured at baseline, and pregnancy preferences, as reported at the quarterly survey prior to the incident pregnancy. Sociodemographic characteristics included age; self-reported race and ethnicity; employment status; experienced food insecurity in the past month (how often worried whether food would run out before got money to buy more); and difficulty paying mortgage, rent or utility bills in the past year. We measured preferences about pregnancy and childbearing before the incident pregnancy using the 14-item Desire to Avoid Pregnancy (DAP) scale.33 DAP scores range from 0 to 4, with higher scores representing greater desire to avoid pregnancy. We included state policy environment of the recruitment site (restricted vs protected). We also accounted for people’s history of depression or anxiety, history of adverse childhood experiences (ACEs) and instrumental social support, because these have been shown to be associated with subsequent perceived stress.12 34 35 ACEs included 9 of the 10 ACEs questionnaire questions, which measure participants’ exposure to childhood (1) neglect, (2) physical, (3) sexual or (4) emotional abuse; household adversity, living with a household member with (5) substance use issues, (6) mental illness or (7) who has been incarcerated, and (8) parents fought physically; and (9) witnessed violence in the neighbourhood before the age of 18.36 Participants <18 years were only asked questions four through nine. Instrumental social support was measured with a modified version of the two-Way Social Support Scale, assessing the degree to which individuals feel they have received support or concrete forms of assistance (ie, needed financial assistance, help with responsibilities, etc).37 Responses ranged from 5 ‘always’ to 0 ‘never’. Statistical analyses We conducted descriptive analyses of participants’ characteristics, the types of pregnancy care they sought and their pregnancy outcomes. To assess differences in care-seeking stress experienced seeking each kind of pregnancy-related care, we conducted bivariable and multivariable mixed-effects ordinal (to examine the ordered, 5-point Likert scaled stress variable) and logistic regression (for the dichotomous stress variable) analyses. Models accounted for clustering by site and individual, given that people could seek more than one type of pregnancy care. To explore whether participant or pregnancy characteristics might confound the relationship between type of care sought and care-seeking stress, we conducted three multivariable mixed-effects logistic regression analyses assessing whether any covariates that were associated with seeking abortion or adoption care were also associated with care-seeking stress. Among people who sought abortion care, we conducted bivariable mixed-effects logistic regression analyses to examine the personal, practical, and clinic and legal abortion-seeking obstacles experienced as well as state policy environment of recruitment site by whether it was extremely/quite a bit stressful to try to seek abortion care. To assess whether experiencing abortion care-seeking obstacles or state policy environment were associated with abortion care-seeking stress, we conducted a series of multivariable and bivariable mixed-effects logistic regression analyses with the number or type of abortion care-seeking obstacles and policy environment as independent variables, with and without adjusting for model covariates. All statistical analyses accounted for clustering by including a random intercept for recruitment site and were conducted in STATA 18. Significance was reported at p <0.05. To handle missing covariate data, which was very low (<1.3%), we employed casewise deletion methods. Patient and public involvement statement Patients and the public were not involved in the development of the research questions and other aspects of the surveys. Fifteen former ADAPT study participants served as part of two participant advisory groups and provided input on study processes and experiences to improve protocols and dissemination strategies. Study updates and select initial results are disseminated through electronic newsletters to active participants and those who exited the study and expressed interest in learning about study results. A stakeholder advisory group is being assembled to provide input on interpretation and dissemination of study results. Results A total of 2015 participants completed the baseline survey, of which 334 reported and confirmed a new pregnancy (incident pregnancy) during the year of follow-up. Six were lost to follow-up after reporting the pregnancy, and four had ambiguous data regarding the ongoing status of their confirmed pregnancy, leaving 324 as our analytic sample. Most participants were in their 20s (37% were ages 20–24 years, 35% were ages 25–29), identified as Latine ethnicity (59%), and as employed (62%, table 1). About 46% were nulliparous. Average DAP scale score prior to the pregnancy was 1.7 (range: 0–4). Table 1Participants’ pre-pregnancy demographic and health historyDemographic characteristics measured at baselineN (%)Total324 (100%)Age, mean years (SD)25.5 (4.4)Age group, years 15–175 (2%) 18–1933 (10%) 20–24119 (37%) 25–29112 (35%) 30–3455 (17%)Race/ethnicity Latine191 (59%) White58 (18%) Black30 (9%) Multiple or another race46 (14%)Parity Nulliparous148 (46%) Parous176 (54%)Employed No118 (36%) Yes202 (62%) Missing4 (1%)Food insecure, past month No192 (59%) Yes128 (40%) Missing4 (1%)Ever unable to pay rent, mortgage or utilities, past year No158 (49%) Yes162 (50%) Missing4 (1%)State policy environment of recruitment site Restrictive policies (Arizona and Texas)161 (49.7%) Protective policies (California, New Mexico and Nevada)163 (50.3%)Mental health history and social support measured at baselineEver diagnosed with depression or anxiety No190 (59%) Yes131 (40%) Missing3 (1%)Adverse childhood experiences range 0–9, mean (SD)2.4 (2.4) None96 (30%) One56 (17%) Two45 (14%) Three to four59 (18%) Five to six40 (12%) Seven to nine25 (8%) Missing3 (1%)Instrumental Social Support, range (0–5), mean (SD)3.8 (1.2)Pregnancy and childbearing preference measured at survey prior to incident pregnancy Desire to Avoid Pregnancy scale score, range 0–4, mean (SD)1.7 (1.1) Participants’ pre-pregnancy demographic and health history Approximately 83% sought prenatal care for the incident pregnancy, 43% sought abortion care and 5% sought adoption services (table 2). About 69% of people who sought prenatal care gave birth; 58% of those who sought abortion care had an abortion; and no one who sought adoption care placed their child for adoption. While most participants sought only one type of pregnancy care (66%, 214/324) for the incident pregnancy, over one-quarter sought two types of care (27%, 89/324), 3% (11/324) sought all three and 3% (10/324) did not seek or did not report seeking care for the pregnancy (not shown). Table 2Pregnancy outcomes and pregnancy care-seeking stress by type of pregnancy care sought (n=324)Type of pregnancy care soughtPrenatal careAbortion careAdoptionNoneTotal, row %270 (83%)138 (43%)17 (5%)10 (3%)Pregnancy outcome, col % Gave birth185 (69%)45 (33%)10 (59%)1 (10%) Had an abortion44 (17%)80 (58%)3 (18%)1 (10%) Had a miscarriage34 (13%)10 (7%)4 (24%)6 (60%) Unknown outcome*7 (3%)3 (2%)0 (0%)2 (20%) Placed for adoption0 (0%)0 (0%)0 (0%)0 (0%)How stressful has it been to find the (prenatal care, abortion care, adoption services) you have wanted?, col %NA Not at all stressful114 (42%)37 (28%)4 (24%) A little bit stressful68 (25%)20 (15%)2 (12%) Moderately stressful36 (13%)18 (13%)2 (12%) Quite a bit stressful21 (8%)12 (9%)2 (12%) Extremely stressful24 (9%)26 (19%)2 (12%) Don't know7 (3%)19 (14%)5 (29%) Missing0 (0%)2 (1.5%)0 (0%)Care-seeking stress (dichotomous), col % Not at all/a little/moderately/don't know225 (83%)94 (71%)13 (77%) Extremely/quite a bit stressful45 (17%)38 (29%)4 (23%)Note: People may have sought more than one type of pregnancy care.*Nine people who reported an incident pregnancy were still pregnant at the last observation before they were lost to follow-up.NA, Not asked. Pregnancy outcomes and pregnancy care-seeking stress by type of pregnancy care sought (n=324) Note: People may have sought more than one type of pregnancy care. Nine people who reported an incident pregnancy were still pregnant at the last observation before they were lost to follow-up. NA, Not asked. Nearly one in five (17%) of those who sought prenatal care reported that it was extremely (9%) or quite a bit (8%) stressful to find the care they wanted, compared with 29% and 23% of those who sought abortion care or adoption services, respectively (table 2). In ordinal regression analyses, participants seeking abortion care were more likely than those seeking prenatal care to find it stressful to find the care they wanted in unadjusted (OR, 2.37, 95% CI 1.33 to 4.23) and adjusted models (aOR 1.70, 95% CI 1.10 to 2.62). However, in logistic regression models, this association was statistically significant in unadjusted (2.17, 95% CI 1.14 to 4.12) but not adjusted (aOR 1.33, 95% CI 0.74 to 2.38) analyses (table 3). The odds of experiencing care-seeking stress did not differ significantly between people seeking adoption versus prenatal care or abortion care in unadjusted or adjusted ordinal or logistic regression analyses (table 3). Table 3Association between type of pregnancy care sought and pregnancy care-seeking stress according to ordinal and logistic regression analysesComparison of care-seeking stress with by type of care soughtOrdinal regressionOdds ratio (95% CI)Logistic regressionOdds ratio (95% CI)UnadjustedAdjustedUnadjustedAdjustedAbortion vs prenatal care (ref.) 2.37* (1.33 to 4.23) 1.70* (1.10 to 2.62) 2.17* (1.14 to 4.12) 1.33 (0.74 to 2.38)Adoption vs prenatal care (ref.)1.88 (0.88 to 4.02)1.67 (0.64 to 4.36)1.41 (0.45 to 4.48)1.30 (0.34 to 4.92)Adoption vs abortion care (ref.)0.79 (0.36 to 1.75)0.98 (0.36 to 2.65)0.65 (0.27 to 1.54)0.98 (0.25 to 3.81)Adjusted models account for age group, race/ethnicity, employment status, experienced food insecurity in the past month, inability to pay for housing or utilities in the past year, parity, history of depression or anxiety, history of adverse childhood experiences (ACEs), instrumental social support, state policy environment and the desire to avoid pregnancy. Logistic regression modelled care-seeking stress as a dichotomous outcome (extremely/quite a bit stressful vs not at all/a little/moderately stressful/don’t know). Ordinal regression modelled care-seeking stress as an ordered, 5-point Likert-scaled outcome (ranging from ‘not at all stressful’ to ‘extremely stressful’, with ‘don’t know’ responses coded as ‘moderately’).*Indicates p<0.05 and presented in bold.ACEs, adverse childhood experiences. Association between type of pregnancy care sought and pregnancy care-seeking stress according to ordinal and logistic regression analyses Adjusted models account for age group, race/ethnicity, employment status, experienced food insecurity in the past month, inability to pay for housing or utilities in the past year, parity, history of depression or anxiety, history of adverse childhood experiences (ACEs), instrumental social support, state policy environment and the desire to avoid pregnancy. Logistic regression modelled care-seeking stress as a dichotomous outcome (extremely/quite a bit stressful vs not at all/a little/moderately stressful/don’t know). Ordinal regression modelled care-seeking stress as an ordered, 5-point Likert-scaled outcome (ranging from ‘not at all stressful’ to ‘extremely stressful’, with ‘don’t know’ responses coded as ‘moderately’). Indicates p<0.05 and presented in bold. ACEs, adverse childhood experiences. When we explored whether participant background and pregnancy characteristics were associated with seeking abortion care or experiencing care-seeking stress, only DAP scores (desire to avoid pregnancy) were significantly and positively associated with both seeking abortion care (aOR 3.08, 95% CI 2.26 to 4.22) and care-seeking stress (aOR 1.59, 95% CI 1.20 to 2.11, table 4). No participant characteristics were significantly associated with both seeking adoption care and care-seeking stress (table 4). Table 4Factors associated with seeking abortion and adoption care and factors associated with care-seeking stressParticipant characteristics measured at baselineSought abortion care (vs all other)(n=319)Sought adoption care (vs all other)(n=319)Care-seeking stress(n=430)*aOR (95% CI)aOR (95% CI)aOR (95% CI)Age group, years 15–190.92 (0.35 to 2.44)5.30† (1.09 to 25.76)1.03 (0.43 to 2.46) 20–24Ref.Ref.Ref. 25–291.89 (0.91 to 3.94)1.09 (0.21 to 5.75)0.74 (0.35 to 1.53) 30–342.46† (1.00 to 6.03)5.63† (1.00 to 31.58)1.06 (0.43 to 2.61)Race/ethnicity LatineRef.Ref.Ref. White0.63 (0.27 to 1.48)0.70 (0.12 to 3.95)1.25 (0.56 to 2.80) Black0.38 (0.13 to 1.09)0.56 (0.13 to 2.44)0.58 (0.18 to 1.85) Multiple or another race0.72 (0.30 to 1.71) ‡ 1.70 (0.74 to 3.86)Parity ParousRef.Ref.Ref. Nulliparous1.26 (0.64 to 2.48)2.13 (0.54 to 8.38)1.99 (0.97 to 4.10)Employed NoRef.Ref.Ref. Yes0.65 (0.35 to 1.21)1.55 (0.48 to 5.06)0.69 (0.38 to 1.26)Food insecure, past month NoRefRef.Ref. Yes0.86 (0.44 to 1.68)1.46 (0.40 to 5.32)1.69 (0.86 to 3.31)Ever unable to pay rent, mortgage or utilities, past year NoRef.Ref.Ref. Yes1.33 (0.67 to 2.62)0.85 (0.23 to 3.08)0.96 (0.49 to 1.86)State policy environment of recruitment site Restrictive policies (Arizona and Texas)0.32† (0.14 to 0.75)0.86 (0.21 to 3.47)1.39 (0.66 to 2.90) Protective policies (California, New Mexico and Nevada)Ref.Ref.Ref.Mental health history and social support measured at baselineEver diagnosed with depression or anxiety NoRef.Ref.Ref. Yes1.27 (0.69 to 2.32)0.54 (0.17 to 1.77)1.61 (0.88 to 2.94)Number of adverse childhood experiences (0–9)1.12 (0.99 to 1.28)0.81 (0.61 to 1.08)1.04 (0.93 to 1.18)Instrumental social support, range (0–5)1.00 (0.77 to 1.31)0.74 (0.43 to 1.27)0.90 (0.69 to 1.17)Pregnancy and childbearing preference measured at survey prior to incident pregnancy Desire to Avoid Pregnancy scale score prior to reporting incident pregnancy (range 0–4)3.08† (2.26 to 4.22)1.24 (0.71 to 2.15)1.59† (1.20 to 2.11)*Includes multiple observations per individual given than people could seek more than one type of care.† p<0.05 and presented in bold text.‡Due to small cell sizes, we combined ‘Black’ and ‘Multiple or another race’ race categories.aOR, adjusted OR; Ref, Reference. Factors associated with seeking abortion and adoption care and factors associated with care-seeking stress Includes multiple observations per individual given than people could seek more than one type of care. p<0.05 and presented in bold text. Due to small cell sizes, we combined ‘Black’ and ‘Multiple or another race’ race categories. aOR, adjusted OR; Ref, Reference. Among those who looked into where or how to get abortion care (n=132), personal care-seeking obstacles (67%) were the most common type of obstacle reported, followed by clinic or legal (56%) and practical (56%) obstacles; 33% were recruited in a restrictive policy environment (table 5). Those who reported that the care-seeking process was quite a bit or extremely stressful were significantly more likely to report that they were slowed down from getting the abortion because they were not sure they wanted an abortion (71% vs 52%, p=0.049), had difficulty getting time off work (47% vs 18%, p=0.001), had difficulty raising money for the abortion or travel (61% vs 31%, p=0.002) and not knowing where to go (42% vs 11.7%, p<0.001) than those who said it was not at all or moderately stressful to seek care. The requirement to wait for an abortion appointment (55% vs 26%, p=0.002) and not wanting to pass by clinic protestors (50% vs 31%, p=0.044) were also reported as obstacles more often among those who found it extremely or quite a bit stressful to seek abortion care. Participants who reported it was quite a bit or extremely stressful to seek care reported nearly double the obstacles (6.0 obstacles), compared with those who found it moderately or not at all stressful (3.4 obstacles, p<0.001) and were more likely to have been recruited from a state with a restrictive policy context (47% vs 27%, p=0.049). Table 5Abortion care-seeking obstacles by abortion seeking stress among people seeking abortion careDid any of the following slow you down from getting the abortion?Abortion care-seeking stressP value*Totaln (col%)Not at all to moderately stressful or don't known (col %)Quite a bit to extremely stressfuln (col %)Total94 (71%)38 (29%)132 (100%)Personal abortion care-seeking obstacles Not sure I wanted abortion49 (52%)27 (71%) 0.049 76 (58%) Family not wanting me to have an abortion27 (29%)13 (34%)0.53540 (30%) Partner not wanting me to have an abortion31 (33%)13 (34%)0.89244 (33%) Friends not wanting me to have an abortion17 (18%)7 (18%)0.96424 (18%) Number of personal obstacles, mean (SD)1.32 (1.3)1.58 (1.2)0.2991.4 (1.3) None35 (37%)8 (21%)Ref.43 (33%) One23 (25%)11 (29%)0.16934 (26%) Two to four36 (38%)19 (50%)0.08455 (42%)Practical abortion care-seeking obstacles Difficulty arranging childcare17 (18%)6 (16%)0.75323 (17%) Difficulty getting time off work17 (18%)18 (47%) 0.001 35 (26.5%) Difficulty raising money to pay for abortion or travel29 (31%)23 (61%) 0.002 52 (39%) Not being able to find a clinic near me13 (14%)15 (40%) 0.002 28 (21%) Not knowing where to go11 (12%)16 (42%)p<0.00127 (21%) Trying to get insurance to cover the abortion22 (23%)18 (47%) 0.008 40 (30%) Number of practical obstacles, mean (SD)1.16 (1.7)2.53 (1.9)p<0.0011.6 (1.8) None51 (54%)8 (21%)Ref.59 (44%) One13 (14%)5 (13%)0.16718 (14%) Two to six30 (32%)25 (66%)p<0.00155 (42%)Clinic or legal abortion care-seeking obstacles Not wanting to pass by clinic protestors29 (31%)19 (50%) 0.044 48 (36%) Requirement that a parent or guardian be notified/consent5 (5%)6 (15%)0.06711 (8%) Requirement to wait to get the abortion19 (20%)16 (42%) 0.018 35 (27%) Waiting for a first abortion appointment24 (26%)21 (55%) 0.002 47 (35%) Waiting for a second abortion appointment13 (14%)10 (26%)0.09223 (17%) Number of clinic or legal obstacles, mean (SD)1.0 (1.3)1.9 (1.4) 0.001 1.23 (1.4) None51 (54%)7 (18%)Ref.58 (44%) One17 (18%)8 (21%) 0.038 25 (19%) Two to five26 (28%)23 (61%)p<0.00149 (37%)Overall total abortion care-seeking obstaclesNumber of obstacles, mean (SD)3.4 (3.2)6.0 (3.2)p<0.0014.2 (3.5) Zero to one35 (37%)3 (8%)Ref.38 (29%) Two to three19 (20%)6 (16%)0.08725 (19%) Four to five19 (20%)9 (24%) 0.018 28 (21%) Six to fifteen21 (22%)20 (53%)p<0.00141 (31%)State policy environment of recruitment site Restrictive policies (Arizona and Texas)25 (27%)18 (47%) 0.049 43 (33%) Protective policies (California, New Mexico, and Nevada)69 (73%)20 (53%)Ref.89 (67%)Only 132 out of the 138 people who sought abortion care reported on care-seeking stress.Bold text indicates p<0.05. *Based on mixed-effects logistic regression analyses accounting for clustering by recruitment site. Abortion care-seeking obstacles by abortion seeking stress among people seeking abortion care Only 132 out of the 138 people who sought abortion care reported on care-seeking stress. Bold text indicates p<0.05. Based on mixed-effects logistic regression analyses accounting for clustering by recruitment site. In the adjusted models examining the association between abortion care-seeking obstacles and policy environment with abortion care-seeking stress, experiencing any personal (aOR 3.58, 95% CI 1.22 to 10.49), practical (aOR 4.67, 95% CI 1.69 to 12.87) or clinical or legal abortion care-seeking obstacles (aOR 4.66, 95% CI 1.62 to 13.41) was associated with greater odds of reporting abortion care-seeking stress (table 6). Individuals who reported experiencing four to five obstacles (aOR 4.80, 95% CI 1.02 to 22.65) or six or more care-seeking obstacles had significantly higher odds (aOR 13.68, 95% CI 3.07 to 60.90) of reporting it was extremely or quite a bit stressful to seek abortion care. People recruited in a state with a restricted policy environment reported more care-seeking stress (aOR 2.72, 95% CI 1.09 to 6.80) than those recruited in a protected access state (table 6). Table 6Associations between care-seeking obstacles and policy environment and finding it extremely or quite a bit stressful to obtain abortion care according to bivariable and multivariate logistic regression analyses (n=132)Unadjusted models*Fully adjusted models†OR95% CIOR95% CIAny personal abortion care-seeking obstacles NoRef.Ref. Yes2.23(0.91 to 5.49) 3.58* (1.22 to 10.49)Any practical abortion care-seeking obstacles NoRef.Ref. Yes 4.45* (1.85 to 10.71) 4.67* (1.69 to 12.87)Any clinical or legal abortion care-seeking obstacles NoRef.Ref. Yes 5.51* (2.09 to 14.47) 4.66* (1.62 to 13.41)Number of abortion care-seeking obstacles 0-1 obstaclesRef.Ref. 2-3 obstacles3.68(0.83 to 16.41)3.46(0.68 to 17.70) 4-5 obstacles 5.53* (1.33 to 22.88) 4.80* (1.02 to 22.65) 6-15 obstacles 11.11* (2.94 to 41.95) 13.68* (3.07 to 60.90)State policy environment of recruitment site Restricted policies (Arizona and West Texas) 2.48* (1.13 to 5.44) 2.72* (1.09 to 6.80) Protected policies (California, New Mexico and Nevada)Ref.Ref.*Indicates p<0.05 and presented in bold text. All models are based on mixed-effects logistic regression accounting for clustering by recruitment site.*Unadjusted models include only number or type of care-seeking obstaclesenvironment and policy as independent variables, each set run separately.†Fully adjusted models include each independent variable (obstacle or policy environment of recruitment site), and also adjust for age group, race/ethnicity, employment status, experienced food insecurity past month, inability to pay for utilities or housing in the past year, parity, history of depression or anxiety, history of adverse childhood experiences, instrumental social support and the desire to avoid pregnancy. Associations between care-seeking obstacles and policy environment and finding it extremely or quite a bit stressful to obtain abortion care according to bivariable and multivariate logistic regression analyses (n=132) *Indicates p<0.05 and presented in bold text. All models are based on mixed-effects logistic regression accounting for clustering by recruitment site. Unadjusted models include only number or type of care-seeking obstaclesenvironment and policy as independent variables, each set run separately. Fully adjusted models include each independent variable (obstacle or policy environment of recruitment site), and also adjust for age group, race/ethnicity, employment status, experienced food insecurity past month, inability to pay for utilities or housing in the past year, parity, history of depression or anxiety, history of adverse childhood experiences, instrumental social support and the desire to avoid pregnancy. When compared with participants who ever looked into where or how to get prenatal care (91%), significantly fewer of those who sought an abortion (29%) or adoption services (7%) ended up visiting a clinic, facility or adoption agency for that care (p<0.05, table 7). Individuals who looked into prenatal care were significantly more likely to seek information from a clinic/health facility (88%), doctor, midwife or nurse (87%), friends (67%) and family (69%) compared with those seeking abortion (63%, 40%, 36%, 28%, respectively) or adoption (38%, 25%, 33%, 13%, respectively). While the internet was the most common source of information for those seeking abortion (92%) and adoption care (72%), a clinic or healthcare facility was the most common for those seeking prenatal care (88%). Pregnancy resource centres were a source of information for about half of those seeking prenatal care (51%) and 38% to 44% of those seeking abortion or adoption, with no statistically significant differences between the groups. Table 7Pregnancy care visits and sources of information by type of pregnancy care soughtType of pregnancy care sought(‘looked into where or how you could get (prenatal, abortion or adoption) care’)Prenatal (Reference, n=270)Abortion(n=138)Adoption(n=17)At any point in the pregnancy, visited a clinic or health facility for an abortion or prenatal care or visited an adoption agency or services221 (91%) 26 (29%)* 1 (7%)* Where sought information about (prenatal, abortion or adoption care) Internet209 (80%) 118 (92%)* 12 (71%) A clinic/health facility225 (88%) 69 (63%)* 6 (38%)* Asked doctor, midwife or nurse220 (87%) 37 (40%)* 4 (25%)* A pregnancy resource centre (sometimes known as crisis pregnancy centres)126 (51%)42 (44%)6 (38%) Asked friends166 (67%) 34 (36%)* 5 (33%) Asked family171 (69%) 26 (28%)* 2 (13%)* A churchNANA3 (19%) Other16 (11%)7 (13%)0 (0%)*P<.05 based on mixed-effects logistic regression analyses accounting for clustering by site and individual. Pregnancy care visits and sources of information by type of pregnancy care sought P<.05 based on mixed-effects logistic regression analyses accounting for clustering by site and individual. Discussion In this prospective study investigating the types of pregnancy care newly pregnant individuals sought, experiences of care-seeking stress were relatively common across all types of care. Over two-thirds found it at least a little bit stressful seeking the pregnancy care they wanted, a finding that is significant to highlight given that stress may contribute to barriers accessing timely pregnancy care and individuals’ mental health outcomes. While nearly 1-in-6 people seeking prenatal care found it extremely or quite a bit stressful to find the care they wanted, those seeking abortion care reported more care-seeking stress in unadjusted and adjusted ordinal regression analyses. However, this association was only statistically significant in unadjusted but not adjusted logistic regression models. This attenuated relationship may be due to underlying contextual and social circumstances that could be associated with choosing a type of care and care-seeking stress. Notably, we found that having a stronger pre-pregnancy desire to avoid pregnancy and childbearing, as indicated by a higher DAP score, was significantly associated with both seeking abortion care and experiencing care-seeking stress. This suggests that experiencing an undesired pregnancy may drive both the decision to seek abortion and the stress experienced during the care-seeking process. Thus, having a pregnancy that is highly undesired may amplify people’s emotional burden and stress at the moment they discover they are pregnant, accompanied by an urgency to seek timely care, thereby increasing care-seeking stress. Although we adjusted for several contextual variables, other unmeasured factors may have also contributed to differences in care-seeking experiences. These findings underscore the need for comprehensive and integrated pregnancy care options that reduce structural barriers and the burden of care-seeking. Notably, we found that while most newly pregnant people looked into only one type of pregnancy care, over one-quarter sought multiple types of pregnancy care. Most people who sought prenatal or abortion care ultimately gave birth or had an abortion; however, none placed their child for adoption. This finding is consistent with national estimates that very few people consider adoption and only about 0.5% of births nationally are placed for adoption.23 38 These results suggest that for some people, the care-seeking process may be an integral part of their decision-making, as they may still be considering and gathering information about their choices as they navigate the pregnancy care landscape. We explored the types of obstacles pregnant people faced when seeking abortion care, including the contribution of the broader sociopolitical context and their association with people’s experiences with care-seeking stress. More than two-thirds of people reported facing two or more obstacles, of all types. Consistent with research showing that experiencing delay-causing obstacles to care is associated with experiencing more stress, anxiety and depression,11 we also found that experiencing any type of obstacle, but in particular practical or clinic/legal obstacles, was associated with significantly higher odds of reporting care-seeking stress. In addition, participants recruited from states with restrictive abortion policy environments reported greater care-seeking stress than those in protective policy environments, suggesting that broader policy contexts may independently contribute to the burden of seeking care. These results suggest that structural barriers, particularly those embedded in public and institutional policies and practices, may be more psychologically taxing than internal or personal care-seeking obstacles, such as those related to pregnancy decision-making or social support. Thus, legislative decisions which exacerbate structural barriers to care may negatively impact people’s well-being, despite their claim that they aim to protect the mental health of people seeking abortion.39 This study contributes empirical evidence of the cumulative burden of logistical and legal barriers and their potentially negative impact on individuals’ experiences with stress. Addressing these obstacles, particularly those related to clinical or legal obstacles and the broader policy environment, could mitigate stress and improve access to care for pregnant individuals. We found that people rely on very different sources of information for their care-seeking depending on the type of care sought. While prenatal care information was more often sought through healthcare professionals and close social networks, people seeking information about abortion and adoption care relied heavily on the internet and were less likely to involve healthcare professionals or their social network. The reliance on self-directed information gathering for people with unwanted pregnancies may be due to the pervasive stigma surrounding becoming pregnant with unwanted pregnancy or seeking abortion,40 thereby motivating people to conceal the pregnancy from others and avoid a negative stigmatising reaction.2 41 42 This study has several limitations. Our prospective study design, which followed a cohort of individuals from before and throughout pregnancy, while resource intensive, is an important study strength. Nonetheless, our analyses among 324 pregnancies may have been underpowered to detect differences in care-seeking stress by type of care sought, particularly among those seeking adoption care. Furthermore, because the care-seeking process is usually longer for those seeking prenatal and adoption care than for those seeking abortion care, a substantial proportion of people had their abortions by the time they reported the pregnancy on a survey.43 Thus, we might not have adequately captured their care-seeking stress. Another limitation is that we only captured the obstacles faced by people seeking abortion and not of those who sought prenatal care or adoption services. Future research should explore the sources of stress in seeking prenatal care and adoption services, whether stress is largely driven by personal or structural factors and how obstacles to care differ based on the type of care sought. Furthermore, participants in this study faced multiple layers of vulnerability, with many reporting experiences of food insecurity and a history of childhood trauma and mental health conditions that likely impacted their care-seeking experiences. Consistent with the literature on pregnancy care-seeking,9 it is important to acknowledge that people’s pregnancy decision-making and care-seeking occurred within these constrained circumstances. Future research should further examine the structural, financial and informational barriers that influence people’s ability to obtain their preferred reproductive health and pregnancy-related care. Finally, our study focused on individuals seeking care from safety net healthcare facilities in states near the US-Mexico border. Results may not be generalisable to populations with different demographic characteristics (eg, advanced maternal age, lower proportions Latine) or seeking care in different US regions. While a large body of research has highlighted the wide array of personal and structural barriers people experience trying to access prenatal and abortion care,1831 44,47 none that we are aware of have directly compared the stress of care-seeking by type of pregnancy care sought among a sample that is followed prospectively. By following newly pregnant individuals prospectively as they became pregnant and then throughout their care-seeking process, we were able to capture the range of experiences and types of care sought and provide a more nuanced understanding of the stress experienced from the pregnancy care-seeking process. Findings point to a need to identify the sources of burden so as to reduce the stress of the pregnancy care-seeking process across all types of pregnancy care. Acknowledgements The authors are grateful to the healthcare facilities that served as recruitment facilities for this study, as well as the people who chose to participate, with special thanks to those who served as participant advisors. We thank Heather Gould for project direction; Rana Barar for early infrastructure support; Micah Perlin and Keith Jackman for database support; Irene Orellana Bonilla, Jasmin Sanchez and Sabrina Serrano for recruitment and data collection; Jasmine Powell and Angela Shin for project coordination; and Molly Battistelli and Michele Benjamin for grants management. Footnotes Funding: This study was funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (R01-HD108643) and the Susan Thompson Buffett Foundation. These entities had no role in the study design; collection, analysis and interpretation of the data; writing of the report; or in the decision to submit the article for publication. Prepublication history for this paper is available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2026-117096). Data availability free text: Data are available upon reasonable request. The ADAPT study dataset is managed and maintained by UCSF investigators. Investigators and trainees at other academic institutions may obtain a deidentified dataset and collaborate on data analyses after presenting analysis plans to the UCSF investigators; obtaining IRB approval (or documentation of exemption of approval) from relevant academic institutions; signing a data use agreement with the UCSF investigators; and being approved by the UCSF IRB. Data for this study are not currently being made publicly available due to the vulnerable study population (minors, pregnant people) and sensitive topics of investigation and because we did not obtain consent from participants to share their data in this way. Patient consent for publication: Not applicable. Ethics approval: This study involved human participants and was approved by the UCSF Institutional Review Board (#17-23592). Participants gave informed consent to participate in the study before taking part. Provenance and peer review: Not commissioned; externally peer-reviewed. Patient and public involvement: Patients and/or the public were involved in the design, conduct, reporting or dissemination plans of this research. Refer to the Methods section for further details. Data availability statement Data are available upon reasonable request. References Data Availability Statement Data are available upon reasonable request.