Introduction

Abortion stigma is one of a broader range of reproductive stigmas that affects most people seeking abortion [1, 2]. A national study of over 4,000 U.S. abortion patients indicated that nearly two-thirds reported that people would look down on them if they knew they had an abortion [3]. Kumar, Hessini and Mitchell have defined abortion stigma as “a negative attribute ascribed to women who seek to terminate a pregnancy that marks them, internally or externally, as inferior to ideals of womanhood” [4]. By this definition, abortion stigma is specific to women seeking an abortion, whereby they feel devalued for not meeting societal expectations of motherhood. Others have conceptualized abortion stigma more broadly to also include providers and other individuals involved in abortion [5, 6]. Findings from cross-sectional studies have linked perceptions of abortion stigma with pre-abortion [7] and post-abortion psychological health [8].

Stigma is a complex concept that involves a number of interrelated factors. Pescosolido and Martin conceptualize stigma as a system or “stigma complex” which they define as “the set of interrelated, heterogeneous system structures, from the individual to the society,… that constructs, labels, and translates difference into marks. In turn, reactions … to a cultural bundle of prejudice… and discrimination… are produced” [9]. Thus, according to this and other definitions, abortion stigma is a product of the larger social, cultural and community contexts in which abortion seeking occurs [6, 10–12]. The contextual factors that define one’s community—race, ethnicity, religion, religiosity, class, age, and social support—are known to be associated with perceived abortion stigma, although the relationships are interrelated and complex [3, 6, 9, 13]. The experience of reproduction, including abortion and abortion stigma, differ depending on one’s group identity and community context [1, 9, 14]. In the U.S., a long history of White supremacy, racism and social inequalities based on race and class likely lead to differences in perceptions and experiences of abortion stigma. The largest U.S. study examining abortion stigma in the United States demonstrated that across ethnic/racial groups, African American/Black women reported lower levels of perceived and internalized abortion stigma, White women scored highest on perceptions of stigma from the general public, and Latina women scored highest in perceptions of stigma from family and friends [3]. Another national study surveying women seeking reproductive health services found that religiosity was strongly associated with overall perceived abortion stigma, but race was not [15]. However, when further examining relationships among different dimensions of stigma, African American/Black women scored lower on worries about judgment and higher on feelings of isolation, when compared to White women [15]. Being raised in a family with anti-abortion attitudes and lack of partner support have also been found to be associated with greater perceptions of abortion stigma [3, 13]. In this study, we explore the longitudinal relationships between these contextual factors, including race/ethnicity, religion and social support, on perceived abortion stigma and psychological well-being.

While perceived abortion stigma refers to people’s perceptions of how others judge them for seeking or obtaining an abortion, internalized stigma is when these views have been incorporated into one’s sense of self in the form of shame, guilt, or secrecy. People who have had an abortion and hold anti-abortion attitudes may have internalized stigma. Several participants in one qualitative study in the U.S. who had sought an abortion described anti-abortion attitudes [16]. Similarly, some people seeking abortion in England and Wales described negative views about abortion and believed they had done something morally wrong, views that were accompanied by feelings of guilt and shame [17]. While participants attributed their feelings of shame to their rejection of motherhood, those who resisted abortion stigma, justified their abortions by claiming they were being responsible mothers, in that they chose abortion in order to focus on their existing children. Participants in both studies reconciled their internalized anti-abortion attitudes with their desire to have an abortion by distinguishing their personal circumstances around their abortions from those of other people’s abortions [16, 17].

People may conceal their abortions in order to cope with their negative emotions around the abortion or to prevent negative reactions from others [3, 15, 18, 19]. A small qualitative study of women who sought an abortion in the UK, found that all women described the abortion as socially unacceptable, which motived their desire to keep it secret [19]. A national study that analyzed a subsample of women and men in the United States who had a personal experience with abortion found that over one-quarter of abortion disclosures received a negative, stigmatizing reaction [20]. However, the real-world figure is likely higher, as the study only captured disclosures among those willing to disclose their abortions as part of the study, an inherent challenge in studying abortion, as stigmatized events are notoriously underreported [21]. Similarly, several respondents in a qualitative study that specifically sampled people who had emotional difficulties related to the abortion reported that their friends and family were not supportive and some even lost relationships due to the abortion [22].

On the other hand, a substantial body of evidence outside the field of abortion indicates that concealing a stigmatized behavior, event or identity can have negative psychological consequences [23]. While there is ample evidence indicating that women keep their abortions secret [4, 5], research on the psychological effects of this secrecy is scarce. Major and Gramzow (1999) examined the effects of stigma and secrecy on experiencing intrusive thoughts and psychological distress [24]. They found that perceived abortion stigma was associated with keeping the abortion secret, that secrecy was associated with increases in thought suppression and intrusive thoughts about the abortion, and that both experiencing intrusive thoughts and thought suppression were associated with greater psychological distress. Due to simultaneous measurement of factors, all at two years post-abortion seeking, they were unable to establish causality or the sequential nature of these relationships. Findings from qualitative studies suggest that secret keeping can be experienced as a psychological burden and contribute to anxiety [19, 22]. This study will explore the relationship between abortion disclosure at the time of seeking an abortion and psychological distress up to five years later.

Enacted stigma refers to experiences of being negatively judged by family, friends, the community, or culture, whereas structural stigma, refers to discrimination from governments, organizations, institutions, and/or policies [9]. Primary examples of structural stigma in abortion care include the myriad of state-level policies that intentionally restrict people’s access to care, whereas enacted stigma may come in the form of judgment from health care providers, anti-abortion clinic protesters, and loved ones. Both enacted and structural abortion stigma can contribute to the process of stigmatization, which is the social process by which people are affected by stigma [9]. Thus, people’s interactions with medical institutions, health care providers, and policies may further contribute to their perceptions of abortion stigma and the process of stigmatization [6, 13, 19, 25]. However, these experiences are not necessarily stigmatizing for all people. Beynon-Jones (2017) described how some people navigate experiences of enacted abortion stigma by critiquing or rejecting judgment from others if they felt they were infringing on an individual’s experience [26]. Whereas, a qualitative study of people accessing abortion in the U.S. found that the presence of clinic protesters, elaborate clinic security, and the formality of clinic procedures contributed to perceived stigma [27]. Research on abortion stigma up to date, has primarily relied on cross-sectional studies, which has impeded our ability to adequately capture the social process of stigmatization.

The current U.S. political climate that has led to decreased access to abortion care represents a form of structural stigma for people needing care. Some people may experience these legal and clinic restrictions to abortion care as enacted stigma, since they may result in people feeling that they are not morally deserving of an abortion. Using Turnaway Study data, this study seeks to explore whether abortion denial due to facility gestational age limits is associated with perceived abortion stigma at the time of abortion seeking. The Turnaway Study, a longitudinal study that followed four groups of nearly 1,000 people seeking abortion, many of whom were denied the service, found evidence of internalized stigma. Approximately one in five people seeking abortion thought that abortion was morally wrong or should be illegal [28]. At six months post-abortion seeking, those denied an abortion and who carried to term (Turnaway-births) were the least likely to support the legal right to abortion, followed by those who obtained a Near-limit or First-trimester abortion; people who were denied an abortion and later miscarried or sought an abortion elsewhere (Turnaway-no-births) were most likely to support the legal right to an abortion. Further, people’s perceptions of stigma were significantly associated with reporting negative emotions (regret, anger, guilt, sadness) about their abortion years later [5, 12]. Left unanswered by the Turnaway Study is whether abortion denial affects perceptions of abortion stigma, whether perceptions of abortion stigma change over time, and whether disclosing abortion-seeking and abortion denial are associated with adverse mental health outcomes. The current analysis uses Turnaway Study data to answer these questions and to specifically assess whether abortion denial and other contextual factors are associated with perceived abortion stigma over five years post-abortion seeking and whether perceived abortion stigma and telling others about abortion seeking are associated with psychological health years later. As part of this latter research question, we tested whether race/ethnicity moderates the relationship between perceived abortion stigma and psychological health. We hypothesized that the structural stigma of abortion denial, is associated with perceptions of abortion stigma, and that perceived abortion stigma and abortion denial are associated with psychological distress years later. For all analyses we were specifically interested in assessing differences between people obtaining a Near-limit abortion and the three other Turnaway Study groups (Turnaway-births, Turnaway-no-births, and First-trimester abortion).

Materials and methods

Statistical analyses

We used mixed effects regression analyses, accounting for clustering by recruitment site, to assess differences in demographic, pregnancy and other participant characteristics by study group at baseline (Table 1). For categorical variables with more than two categories (i.e. race/ethnicity and marital status), we used an omnibus post-estimation test to accommodate multiple category associations. For multivariable analyses with perceived abortion stigma as our outcome, we used mixed effects linear regression models to assess whether perceived abortion stigma trajectories differed by study group and changed over time, and accounted for clustering by site and individual. We assessed whether trajectories were curvilinear by testing whether including quadratic terms for time improved the model fit. All of the models with stigma as an outcome required quadratic terms for time as indicated by a significant (p < .05) likelihood ratio test. To model psychological distress, we used mixed effects logistic regression analyses to predict any psychological distress from six months through five years post-abortion seeking, adjusting for baseline factors and accounting for clustering by site and individual. To assess whether race/ethnicity modifies the relationship between perceived abortion stigma and psychological distress we included an interaction term by race/ethnicity and baseline perceived stigma. To ease interpretation of the results of the interaction term used in this model, we also used a post-estimation command to generate marginal predictive values of each categorical independent variable and to assess statistically significant differences between these variables. Because the interaction term is included in this model, one should interpret the race/ethnicity main effects as from people without any baseline perceived stigma. We conducted all analyses in STATA 15.0.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of women who sought and either received or were denied abortion, at the time they sought an abortion (N = 928).

https://doi.org/10.1371/journal.pone.0226417.t001

Results

Of the 1,132 eligible people recruited, 956 people (84.5%) completed baseline interviews. Participation and attrition rates by study group have been published elsewhere [35]. We removed three people who changed their mind about wanting an abortion and 25 people who did not respond to the stigma items at baseline, leaving a final analysis sample of 928 (435 Near-limits, 159 Turnaway-births, 70 Turnaway-no-births, and 264 First-trimesters).

Baseline perceived abortion stigma

We present participant baseline characteristics by study group in Table 1 and their baseline perceptions of abortion stigma in Table 2. According to unadjusted analyses, at approximately one-week after being denied an abortion, over half of people seeking abortion reported that people in their community (60%) or people close to them (56%) would look down on them if they knew they had sought an abortion (Table 2). People in the Turnaway-birth group scored significantly lower on overall and individual perceived abortion stigma items, than people in the Near-limit group (Table 2). African American/Black people scored lower on both stigma items when compared to White people. People who reported that the man involved in the pregnancy was not involved in the abortion decision had higher levels of perceived abortion stigma than people who reported that the man involved wanted to carry the pregnancy to term. When compared to people who did not tell anyone other than the man involved in the pregnancy that they were seeking an abortion, those who told two or more people reported significantly lower perceived abortion stigma overall (1.6 vs 1.3, p < .05) and from their community (1.6 vs 1.2, p < .05). People who were stopped by clinic protestors reported significantly higher perceptions of abortion stigma (1.7), than people who did not see protestors (1.3, p < .05). There were no significant differences in baseline perceived abortion stigma and being lost at the last interview wave five years later.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Baseline perceived abortion stigma by certain participant descriptive characteristics, unadjusted (n = 928).

https://doi.org/10.1371/journal.pone.0226417.t002

Five-year perceived abortion stigma trajectories

According to adjusted mixed effects linear regression analyses, when compared to Near-limits, people in the Turnaway-birth group reported significantly lower baseline perceived stigma from people in their community (-0.30; 95% CI, -0.52, -0.08), from people close to them (-0.38; 95% CI, -0.59, -0.16) and overall (-0.34; 95% CI, -0.54, -0.14, Table 3). People in the Turnaway-no-birth group were significantly more likely to perceive community abortion stigma than people in the Near-limit group (0.34; 95% CI, 0.03, 0.64). From one week to five years post-abortion seeking, overall perceived abortion stigma declined significantly (p < .001) for all study groups (Fig 1). These study group differences were no longer statistically significant (p>0.05) by two years post-abortion seeking (not shown).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Perceived abortion stigma marginal predicted probabilities over time by study group, using adjusted mixed effects linear regression.

Note: Values are based on post-estimation estimates following an adjusted mixed effects linear regression analysis controlling for race/ethnicity, age, education, marital status, parity, difficulty deciding to have an abortion, history of abortion, region of residence, religion, exposure to abortion protesters, emotional social support, and history of depression or anxiety; Differences in perceived abortion stigma between the Turnaway-birth and Near-limit abortion groups are statistically significant between one week and 1.5 years post-abortion seeking.

https://doi.org/10.1371/journal.pone.0226417.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Predictors of perceived abortion stigma over five years, using adjusted mixed effects linear regression .

https://doi.org/10.1371/journal.pone.0226417.t003

With the exception of education level, nulliparity and exposure to anti-abortion protesters, all covariates were significantly associated with perceived abortion stigma. When compared to White people, African American/Black people were significantly less likely to perceive abortion stigma by their community (-0.45; 95% CI, -0.63, -0.27, Table 3), people close to them (-0.29; 95% CI, -0.47, -0.12) and overall (-0.37; 95% CI, -0.54, -0.20). Married people were significantly more likely than never married people, people who identified as Protestant or Catholic religion were more likely than people who identified as no religion, and people living in the South were more likely than people living in the West to perceive abortion stigma from their community, people close to them and overall. People with no prior history of abortion and who reported difficulty deciding to have the abortion were significantly more likely than their counterparts to perceive abortion stigma from their community, from people close to them, and overall. Higher emotional social support was associated with lower perceived abortion stigma. People with a history of depression or anxiety were significantly more likely than people without such a history to perceive abortion stigma from their community and overall. When compared to people ages 19 and under, those ages 20 and over were significantly less likely to perceive abortion stigma overall and from their community, whereas people ages 25–29 were significantly less likely to perceive abortion stigma from people close to them, than people ages 19 and under.

Perceived abortion stigma, abortion disclosure, race/ethnicity and psychological distress

In adjusted multivariable analyses, people who reported high overall perceived abortion stigma at baseline had significantly higher odds of reporting psychological distress years later (aOR; 3.98, 95% CI, 1.39, 11.37, Table 4 and Fig 2). People who told one (aOR; 2.22, 95% CI, 1.13, 4.37) or more (aOR; 2.21, 95% CI, 1.16, 4.21) people that they had sought an abortion (other than the man involved) also had higher odds of reporting psychological distress years later than people who told no one. According to the main effects for race/ethnicity, among those without any baseline perceived abortion stigma, African American/Black people had significantly lower odds (aOR; 0.13, 95% CI, 0.03, 0.61) as non-Hispanic White people to experience psychological distress. Significant race/ethnicity by abortion stigma interaction terms indicated that race/ethnicity modified the relationship between perceptions of abortion stigma and psychological distress (Table 4 and Fig 2). As seen in Fig 2 and Table 5, among people without any reported baseline overall abortion stigma, African American/Black people had significantly lower psychological distress than non-Hispanic White people (1.3% vs 5.1%, p < .01), whereas there were no statistically significant differences in psychological distress by race/ethnicity among people with medium or high baseline perceived abortion stigma. However, while not statistically significant, among those with high perceived abortion stigma, Hispanic/Latina people were less likely to report psychological distress than non-Hispanic White people (8% vs. 13%, p = 0.056).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Marginal predicted probabilities of experiencing psychological distress from six months to five years post-abortion seeking by race/ethnicity and baseline perceived abortion stigma, using adjusted mixed effects logistic regression.

Note: Values are based on post-estimation estimates following an adjusted mixed effects logistic regression analysis controlling for study group, number of abortion disclosures, age, education, marital status, parity, difficulty deciding to have an abortion, emotional social support, pregnancy preference of the man involved in the pregnancy, and history of depression or anxiety.

https://doi.org/10.1371/journal.pone.0226417.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Predictors of psychological distress over five years according to mixed effects logistic regression analyses.

https://doi.org/10.1371/journal.pone.0226417.t004

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Predictive margins of experiencing any psychological distress six months to five years post-abortion seeking.

https://doi.org/10.1371/journal.pone.0226417.t005

Discussion

In this study we found that over half of people seeking an abortion think that people in their community and people close to them would look down on them if they knew they had sought an abortion, a proportion similar to that reported in other studies [3]. With one in four people having an abortion in their lifetime, we find that despite its commonness, abortion is a highly stigmatized experience [36] with potential implications for future psychological health.

While we hypothesized that people denied an abortion (Turnaway-births) would report higher levels of perceived abortion stigma given that people might interpret the denial of services as a stigmatizing experience, we instead found that they reported less perceived abortion stigma than their counterparts, a difference that remained statistically significant for nearly two years. Presumably the lower levels of perceived abortion stigma among people denied an abortion and who carried to term, stems from their not engaging in the stigmatized behavior (obtaining an abortion), thus enabling them to ward off any feelings of shame or guilt. Interestingly, people who were denied an abortion and who later miscarried or had an abortion elsewhere (Turnaway-no-births) reported the highest levels of perceived community abortion stigma across study groups. These people experienced not only the structural stigma of being denied an abortion but they also were exposed to the potentially stigmatizing experience of abortion or miscarriage. This group of people was also the most likely to experience subsequent psychological distress. Their higher levels of perceived abortion stigma supports our hypothesis that abortion denial is stigmatizing. Further, it suggests that people who carry a pregnancy to term are able to avert abortion stigma by meeting the cultural ideals of motherhood by having a baby, as suggested by Kumar and colleagues [4].

The emerging public discourse condemning abortions later in pregnancy through the onslaught of legal restrictions to abortion may affect how people and their communities view abortions depending on their gestational ages. Thus, while we may have expected that people obtaining abortions earlier in pregnancy to be less likely to perceive abortion stigma than those obtaining abortions at later gestational ages, we found no differences between these two groups of people. The perceived stigma of having an abortion beyond the first trimester may change if efforts to set gestational age limits earlier in pregnancy continue to increase and to change people’s moral views about abortion at differing gestational ages.

The higher levels of abortion stigma among people who obtained an abortion persisted for nearly two years, demonstrating the long-standing effects of stigma. Participants’ abortion stigma trajectories may reflect how perceptions of stigma move through stages, according to how others’ react to the stigma and as the process of disclosing the abortion becomes easier over time [37]. The sharp decline in perceived stigma over time, not only among people who carried to term but also among people who had an abortion, suggests that many people were eventually able to liberate themselves of feelings of stigma by potentially becoming more accepting of the abortion, forgetting about the abortion, and/or engaging in experiences and social interactions that reduced perceptions of stigma. The decline in perceptions of stigma coincide with a previous Turnaway Study analysis indicating that people’s negative and positive emotions about the abortion, including feelings of guilt, and thoughts about the abortion, decreased significantly over time [12].

While previous Turnaway Study findings have demonstrated that people who had abortions were not more likely to experience adverse mental health outcomes than those denied an abortion and carried their pregnancies to term, it also found that for all groups of people, psychological distress was highest at the time of seeking abortion when compared to years later [35]. In the current analysis, we found that in addition to history of mental health problems and other covariates, the higher levels of perceived abortion stigma at the time of seeking an abortion, also contribute to people’s experiences of psychological distress. Perceived abortion stigma was strongly associated with experiences of psychological distress years later, both among people who had abortions and people who were denied abortions. This association suggests that people may have internalized these perceptions of stigma, with long-term consequences for their mental health. Another interpretation to this finding is that, people may have experienced enacted stigma rather than merely perceptions of stigma, for example if they received negative reactions from others post-abortion seeking and these experiences may have led to subsequent negative mental health outcomes.

Consistent with the published literature, we found that people’s social and cultural contexts, including race/ethnicity, age, religion, and where they lived, were significantly associated with perceived abortion stigma [3, 29, 36]. One of the strengths of our study design lies that we collected data from people seeking abortions in 21 states, representing every region of the United States, which allowed us to capture the experience of a diverse range of individuals. Perceptions of stigma were higher in certain parts of the country where abortion is more restricted and among people who identified as Protestant or Catholic. Married people reported higher levels of perceived stigma than never married people. This is perhaps a product of the social pressures and expectations that married people may feel to have children, whereas unmarried people may feel more social pressure to avoid childbearing.

Of note, was the significant interaction we found by race/ethnicity and perceived abortion stigma on psychological distress. Consistent with the literature [1, 3], African American/Black people were less likely to report perceived abortion stigma compared to non-Hispanic White people. The concept of stratified reproduction, which describes the notion that society places differential valuations on reproduction, fertility, and parenthood based on race, provides some context to this finding [1, 14]. Historical injustices, including forced sterilization among mostly low-income people and people of color, and current societal practices, including preferentially encouraging long-acting reversible contraception among communities of color and low-income people, are just two examples of policies and practices that exemplify how reproduction is stratified in the United States with a devaluation of the fertility and parenthood of African American/Black people [14, 38]. In the context of abortion stigma, White people who have an abortion may experience stigma for not conforming to these societal ideals that place a greater value on their birth and parenthood, while African American/Black people may perceive less stigma from a society that devalues their childbearing [14, 39]. Our finding that race/ethnicity modifies the relationship between perceived abortion stigma and psychological distress, is new to the literature. Differences in perceived stigma levels and in the effect of stigma on psychological distress may reflect the unique types of racialized abortion stigma that people of different races face [14]. The anti-choice movement has developed targeted anti-abortion messaging that takes advantage of historical and current reproductive injustices and stratified reproduction to place blame and shame on individual African American/Black people who choose to have abortions [40]. If African American/Black people who do perceive higher baseline abortion stigma internalize racialized abortion stigma, this may carry more weight and affect psychological distress related to their abortion.

In contrast to previous studies suggesting that concealing abortion may result in negative psychological consequences, we found that telling others about one’s abortion-seeking behavior increased people’s risk for experiencing adverse psychological outcomes, while having emotional social support, as measured by the perceived availability of friends or family to talk about problems according to a validated scale [33], was protective. This finding suggests that people may have told others who reacted negatively or who were not supportive of their pregnancy decisions, thereby affecting their mental health. Another explanation for this finding is that people who felt psychologically distressed may have been more likely to disclose their abortion to others as a way to cope with their distress. Consistent with this hypothesis, a previous study found that people who experienced intrusive thoughts about the abortion were more likely to disclose their emotions about the abortion to others, which was associated with psychological distress [24]. Unfortunately our measure of telling others about the abortion is an imperfect proxy for abortion concealment, as it does not allow us to assess the extent to which people’s abortion disclosures were by necessity or by choice or how many people they may have kept the abortion secret from. We also do not know people’s reasons for disclosing and how others reacted to their disclosures.

This study has a number of limitations. Our measure of abortion disclosure was limited, in that it captured the number but not the context or consequence of disclosures. A better measure of abortion disclosure would have also revealed people’s reason for disclosing, whether they wanted to disclose or felt forced to disclose by circumstances, as well as people’s reactions to the disclosure. Thus, our findings on the relationship between abortion disclosure and psychological health, should be viewed cautiously. While our measure on perceived abortion stigma was drawn from a previously used measure [24], it was adapted to also work with people who had sought, but not necessarily obtained an abortion, limiting our ability to compare our results across studies. Neither the original nor our adapted measure have been validated, and this measure does not capture the full range of dimensions that have been described in other measures of abortion stigma [41]. However, a notable strength of our study is its longitudinal design and ability to examine the perceived abortion stigma trajectories and the process of stigmatization and de-stigmatization over five years. We are unaware of any other study that has examined perceived abortion stigma over time. The longitudinal design offers a new contribution to the literature by measuring perceived abortion stigma before our measure of psychological distress, which is an essential first step to establishing causation. Another important study strength is that we were able to compare the experiences of abortion stigma among people seeking abortion, denied an abortion, and among people seeking abortion earlier and later in pregnancy. Taken together, this study finds that people who have obtained or have considered abortion are at risk of perceiving abortion stigma which is influenced by an array of contextual factors such as race/ethnicity, religion, and social support, all of which may have consequences for their psychological well-being years later.

Acknowledgments

The authors thank Rana Barar and Sandy Stonesifer for study coordination and management; Mattie Boehler-Tatman, Janine Carpenter, Undine Darney, Ivette Gomez, Selena Phipps, Brenly Rowland, Claire Schreiber, Danielle Sinkford, and C. Emily Hendrick for conducting interviews; Michaela Ferrari, Debbie Nguyen and Elisette Weiss for project support; Jay Fraser, John Neuhaus, Corinne Rocca, and Charles E. McCulloch for statistical and database assistance; and all the participating providers for their assistance with recruitment.