Physician Recommendations Against Pregnancy Among Young Women With Disabilities in the United States

We used data from four continuous waves (2011–2019) of the National Survey of Family Growth (NSFG), a nationally representative survey of the non-institutionalized U.S. population ages 15–49. Administered by the Centers for Disease Control and Prevention’s National Center for Health Statistics, the NSFG collects detailed information on reproductive health using computer-assisted personal interviewing. 33 Additional information on sampling design is available from the NSFG documentation. 33 We restricted the analytic sample to women aged 15–25 years (n = 8,242) to focus on recent experiences of reproductive counseling during a formative period when young women first establish relationships with healthcare providers. We also chose this age range to reflect a uniform exposure to a distinct policy environment under the Affordable Care Act (ACA), which allows young adults to remain on parents’ health insurance until age 26. We excluded respondents who were surgically sterile (n = 95), physically unable to become pregnant (n = 95), or missing data on the outcome or covariates (n = 34), yielding a final sample of 8,018 women ( Figure 1 ). The primary outcome was self-reported physician recommendation against pregnancy, assessed by the question: “At any time has a medical doctor ever advised you never to become pregnant?” . Among respondents who reported receiving physician recommendations against pregnancy, we examined a secondary outcome of self-reported physician reasons for the recommendation which included the following categories: (1) “dangerous for you”, (2) “dangerous for your baby”, and (3) “some other reason”. Disability status was measured using the six-item American Community Survey (ACS) disability question set: 34 (1) Are you deaf, or do you have serious difficulty hearing?; (2) Are you blind, or do you have serious difficulty seeing, even when wearing glasses?; (3) Because of a physical, mental, or emotional condition, do you have serious difficulty remembering, concentrating, or making decisions?; (4) Do you have serious difficulty walking or climbing stairs?; (5) Do you have difficulty bathing or dressing?; and (6) Because of a physical, mental, or emotional condition, do you have difficulty doing errands alone, such as visiting a doctor’s office or shopping? We created a binary variable for having any disability and coded 1 if respondents answered yes to any of these disability questions and 0 if they answered no to all of them. We also created mutually exclusive disability type variables with the following categories: 1) no disability, 2) hearing disability only, 3) vision disability only, 4) physical disability only, 5) cognitive disability only, and 6) multiple disabilities. The multiple disabilities category included respondents who answered yes to the self-care or independent living questions (difficulty bathing, dressing, or doing errands) or who reported more than one type of disability. Sociodemographic variables included race/ethnicity (Non-Hispanic (NH) White, NH Black, Hispanic, and Other), age (15–17, 18–20, 21–23, and 24–25), education (less than high school, high school diploma, more than high school), marital status (yes/no), health insurance (private, public, uninsured), employment (yes/no), household income less than 100% of the federal poverty level (yes/no), and rural versus non-rural residence. Clinical covariates included self-reported general health status as poor or fair (yes/no), history of endometriosis (yes/no), and history of menstrual or ovulation problems (yes/no). We calculated weighted prevalences and used chi-squared tests to examine differences by disability status. Poisson regressions estimated prevalence ratios (PR) and 95% confidence intervals (CI). Model 1 adjusted for sociodemographic factors. Model 2 additionally adjusted for sociodemographic and clinical characteristics. Separate regression models were run for the binary disability status variable and for the six-category disability type variable. Analyses used NSFG survey weights for the combined 2011–2019 dataset to account for the complex sampling design and to obtain nationally representative estimates. Stata version 18.0 was used to perform the analyses. The STROBE guidelines 35 were followed. This study received approval from the Brandeis University Institutional Review Board.

Among 8,018 women in our sample, representing 21.4 million women nationally, 21.2% indicated having a disability (see Table 1 ). Young women with disabilities differed from those without disabilities across most characteristics. They were more likely to be younger, have less than a high school diploma, be unmarried, receive public health insurance, be unemployed, live in households with income below the federal poverty level, and reside in rural areas as compared to those without disabilities. They were also more likely to report worse health status, experience menstrual/ovulation problems, and have endometriosis. Significant differences were observed across disability types. Among young women with disabilities, those with multiple disabilities had the highest prevalence of having less than a high school diploma, followed by those with cognitive disabilities. Those with cognitive disabilities only and those with multiple disabilities were the least likely to be employed among disability subgroups. In contrast, those with vision disabilities had the highest proportion employed, which also exceeded that of peers without disabilities (68.9% vs. 58.7%). Among all subgroups, those with physical disabilities had the highest prevalence of endometriosis, followed by those with vision disabilities. Young women with multiple disabilities also had the highest prevalence of menstrual or ovulation problems and self-reported poor or fair health status compared to other subgroups. Women with disabilities were substantially more likely to report having received a physician recommendation never to become pregnant: 2.6% of women with disabilities versus 1.1% of nondisabled women (see Table 2 ). Physician recommendations against pregnancy varied substantially by disability type. Women with physical disabilities had the highest prevalence ( 5.4%), followed by women with hearing disabilities ( 4.1%), vision disabilities ( 3.9%), multiple disabilities (3.4%), and cognitive disabilities ( 1.8%). Figure 2 presents the reasons physicians provided to 122 women who received recommendations against pregnancy, stratified by disability status. The most frequently cited reasons for women with and without disabilities were: “dangerous for you” (48% and 43%, respectively), followed by “some other reason” (44% and 40%), and “dangerous for your baby” (17% and 8%). In unadjusted analyses, women with disabilities were 2.55 times (95% CI: 1.45 – 4.47) more likely to receive recommendations against pregnancy compared with those without disabilities (see Table 3 ). After adjusting for sociodemographic factors (Model 1), the association remained significant (adjusted PR=2.15; 95% CI: 1.30–3.54). Further adjustment for clinical factors (Model 2) attenuated but did not fully explain the association (adjusted PR=1.91; 95% CI: 1.10–3.29). In the fully adjusted models (Model 2), prevalence ratios were 5.73 times (95% CI: 1.79–18.29) higher for women with physical disabilities, 4.24 times (95% CI: 1.49–12.07) higher for those with hearing disabilities, and 3.99 times (95% CI: 1.60–9.99) higher for those with vision disabilities compared with women without disabilities. Women with multiple disabilities also had an elevated prevalence ratio (adjusted PR, 2.45; 95% CI: 0.80–7.53), but this did not reach statistical significance. The association for women with cognitive disabilities alone was not significant in either model. Given wide confidence intervals, likely due to the relatively small numbers of respondents in each disability subgroup who reported receiving recommendations against pregnancy, these estimates should be interpreted with caution.

In this nationally representative study, young women with disabilities were nearly twice as likely to report receiving physician recommendations never to become pregnant compared to their nondisabled peers, even after adjusting for sociodemographic and clinical factors. This finding confirms our hypothesis that disability status is independently associated with such recommendations, i.e., the association persists after statistically controlling for potential confounding variables. These disparities were particularly pronounced among women with physical, hearing, and vision disabilities, who were four to six times more likely to receive such recommendations. Notably, the associations persisted after adjusting for health conditions that could plausibly inform physician counseling, such as endometriosis, ovulation disorders, and self-reported general health. This suggests that disability status itself, rather than underlying clinical factors, may be driving differential physician recommendations. These findings raise concerns that physician judgments may be influenced by ableist assumptions and pejorative attitudes about disability and parenting capacity rather than evidence-based medical considerations. The differences in rationale cited by physicians provide essential insights into potential bias mechanisms. Compared to women without disabilities, women with disabilities were slightly more likely to receive maternal safety concerns (“dangerous for you”: 48% vs. 43%) and substantially less likely to receive fetal safety concerns (“dangerous for your baby”: 8% vs. 17%). This pattern suggests that clinicians may disproportionately focus on the perceived risks of pregnancy for the disabled person, potentially reflecting different standards of counseling for disabled compared to nondisabled women. Most concerning is the proportion of recommendations attributed to “some other reason” (44% for women with disabilities and 40% for women without disabilities), suggesting that physicians may base reproductive recommendations on factors outside established clinical guidelines. Additionally, these findings point to a broader issue of the lack of sufficient standardization and objectivity in current reproductive counseling practices which may allow subjective judgments and implicit biases to influence clinical recommendations. Our findings extend previous qualitative research documenting biased or coercive reproductive counseling for women with disabilities. 4 , 11 , 12 , 32 The magnitude of observed disparities suggests that physicians may operate based on implicit or explicit biases about disability and parenting capability rather than objective assessments of medical risk. Research demonstrates that women with disabilities can have healthy pregnancies with appropriate prenatal care and accommodations. 28 , 29 , 36 The differential reproductive counseling patterns observed in this study contradict evidence-based practice and may deny women with disabilities their reproductive autonomy. The disparity in counseling by disability type may also reveal how physician bias operates. Women with visible disabilities, such as physical, hearing, or vision disabilities, were more likely to receive recommendations against pregnancy than those with less visible disabilities, such as cognitive disabilities. This pattern may suggest that visual cues may trigger physicians’ assumptions about patients’ fitness for pregnancy and parenting, rather than individualized assessment of health risks. This interpretation assumes that cognitive disability status is not readily apparent to clinicians, as the cognitive disability measure used in this study—based on self-reported difficulty with remembering, concentrating, or making decisions—may not capture many individuals with apparent and visible intellectual and developmental disabilities (IDD). Furthermore, women with IDD might not have participated in the survey at all, as this survey does not allow proxies and requires self-reporting. Previous qualitative research has documented such biased attitudes among healthcare providers. Studies have found that physicians routinely question the appropriateness of pregnancy for women with disabilities and their ability to care for their children. 4 , 11 , 12 These attitudes may translate into paternalistic counseling practices that prioritize provider assumptions rather than patients’ preferences and reproductive autonomy -- the power to decide whether or not to become pregnant, free from coercion or discrimination. 3 Our findings indicate that for women with disabilities, this autonomy is routinely undermined in clinical encounters. These disparities represent significant concern in ensuring delivery of quality and equitable reproductive healthcare. These disparities take on heightened urgency in the post-Dobbs legal landscape, where access to abortion and other reproductive health services is increasingly restricted. 37 , 38 Women with disabilities already face significant barriers to reproductive healthcare, including provider bias, economic hardship, and inaccessible facilities. Restrictive abortion laws may further reinforce paternalistic counseling and reduce options for those discouraged from pregnancy. Our study has several limitations. First, the cross-sectional design precludes the determination of causal associations between disability status and reproductive counseling. Second, disability status and physician counseling were self-reported and may be subject to recall or misclassification bias. Third, the NSFG survey excludes institutionalized populations, potentially underrepresenting women with severe disabilities. Fourth, the relatively small number of respondents who reported receiving recommendations against pregnancy (n=122) limits the statistical power for subgroup analyses and may result in unstable prevalence ratios, particularly for specific disability types, as evidenced by the wide confidence intervals in Table 3 . These findings have important implications for clinical healthcare practice and policy. Healthcare institutions must ensure physical accessibility and compliance with the Americans with Disabilities Act, as many providers remain unfamiliar of their legal obligations under the law. 39 Quality improvement initiatives should include standardized reproductive counseling protocols and establish accountability mechanisms through regular monitoring of care quality across diverse patient populations. Indicators should include assessment of equitable care delivery to patients with disabilities. Medical education should incorporate disability competency training, including attention to reproductive rights and autonomy. Clinical guidelines must emphasize individualized, evidence-based counseling and caution against reliance on subjective assessments that may introduce bias into clinical practice. Future research priorities should investigate how provider characteristics, training, and institutional cultures influence reproductive counseling patterns. Implementation research is needed to evaluate the effectiveness of interventions aimed at reducing provider biases. Additionally, qualitative research should examine the perspectives of disabled women who receive recommendations against pregnancy from physicians, which could inform patient-centered improvement strategies. Young women with disabilities face systematic bias during reproductive healthcare counseling, with nearly double the likelihood of receiving recommendations against pregnancy compared to nondisabled women. The high proportion of recommendations under “some other reason” category beyond maternal or fetal health concerns suggests that current reproductive counseling practices lack sufficient standardization and objectivity, which may potentially allow subjective judgments or biases to influence clinical recommendations. These findings underscore the urgent need for changes in reproductive healthcare delivery, including evidence-based clinical guidelines and ongoing monitoring to ensure that all women receive respectful, individualized, and medically appropriate reproductive healthcare counseling regardless of disability status.

Women with disabilities have the same desire to become pregnant and exercise reproductive autonomy as women without disabilities. 1 , 2 Reproductive autonomy is the ability to make informed decisions about contraception, pregnancy, and childbirth while recognizing how contextual factors, including race, culture, poverty, immigration status, disability, and access to healthcare, shape these choices. 3 Despite this, disabled women face persistent barriers to reproductive healthcare access, including harmful stereotypes, 4 limited contraceptive access, 4 , 5 , 6 communication barriers, 7 , 8 , 9 economic insecurity, 10 structural discrimination, 4 , 11 , 12 , 13 and coercive practices related to sterilization and childbearing. 14 , 15 These challenges are often compounded by intersecting factors such as race, ethnicity, poverty, immigration status, and gender identity. Historically, reproductive oppression of disabled people has been rooted in eugenics. 16 The 1927 Buck v. Bell Supreme Court decision upheld the constitutionality of compulsory sterilization, with Justice Holmes writing, “Three generations of imbeciles are enough.” 16 This precedent has never been explicitly overturned. While forced sterilization laws have been repealed, sterilization remains legally permissible for people with disabilities deemed unable to consent. 17 – 19 Studies consistently demonstrate that disabled women face disproportionately higher rates of sterilizations compared to nondisabled women. 15 , 20 Ableism, the systemic discrimination against people with disabilities, is well documented in the healthcare system. 21 , 22 Surveys of physicians reveal widespread negative attitudes toward patients with disabilities: more than 80% believe people with disabilities have a lower quality of life than nondisabled people, and only 40% report high confidence in providing equal care. 23 Moreover, fewer than 60% of physicians strongly agree that they welcome patients with disabilities into their practice. 23 , 24 Additional research highlights providers’ limited training in disability care, inadequate knowledge about reproductive healthcare for women with disabilities, and insufficient confidence in providing care. 4 , 11 , 25 Reproductive healthcare counseling should be patient-centered, free from biases based on disability status, and tailored to meet the unique needs of patients. 26 , 27 Research shows that women with disabilities can have healthy pregnancies with appropriate prenatal care and accommodations. 28 – 30 However, existing clinical guidelines on reproductive counseling are not tailored towards meeting the unique needs of women with diverse disabilities, 28 , 31 which may contribute to inconsistent counseling practices influenced by provider attitudes rather than medical evidence. Qualitative research has shown that physicians sometimes recommend sterilization or termination of pregnancy to women with disabilities, even in the absence of clear medical indications. 12 , 32 However, existing studies rely heavily on small or convenience samples, and no prior research has used nationally representative data to quantify physician recommendations against pregnancy for women with disabilities. This study addresses that gap. Using data from the National Survey of Family Growth (NSFG), we examine whether young women with disabilities are more likely than their nondisabled peers to report having received physician recommendations never to become pregnant. We hypothesize that disability status will be independently associated with such recommendations, even after adjusting for clinical and sociodemographic characteristics.