Provider Bias in Long-Acting Reversible Contraception (LARC) Promotion and Removal: Perceptions of Young Adult Women

Abstract

Long-acting reversible contraception (LARC) is effective and acceptable. However, concern exists about potential provider bias in LARC promotion. No study has documented contraceptive users’ attitudes toward or experiences with provider influence and bias regarding LARC.

We collected qualitative data in 2014 to address this gap. Participants were 50 young adult women with any history of contraceptive use (including LARC) in Dane County, Wisconsin. Women often described providers as a trusted source of contraceptive information. However, several women reported that their preferences regarding contraceptive selection or removal were not honored. Furthermore, many participants believed that providers recommend LARC disproportionately to socially marginalized women.

We encourage contraceptive counseling and removal protocols that directly address historical reproductive injustices and that honor patients’ wishes.

One of the most significant reproductive health developments of the past decade is the rise in use of long-acting reversible contraception (LARC). For at least 3 reasons, LARC methods, which include intrauterine devices (IUDs) and implants,¹ can be welcome options for many people who wish to prevent pregnancy. First, people who use contraception tend to prefer IUDs and implants at higher rates than they do other methods.² Second, IUDs and implants are much more effective than other contraceptive methods.³ Third, LARC methods are also cost-effective.⁴ These benefits mean that increasing knowledge of and access to these methods is both a public health and a social justice imperative. Across age groups, racial/ethnic groups, and social classes, many people still do not know about or have access to these methods.^5–8

However, it is critical to take a user-centered approach to increasing LARC knowledge and access. Reproductive justice supporters have described a variety of ways LARC methods might be promoted or practiced in socially unjust ways, particularly among poor women of color. Some argue that policymakers’ enthusiasm about LARC may pertain more to lowering certain groups’ birth rates than to improving women’s lives.^9,10 Others fear that promoting LARC over all other methods could threaten the reproductive autonomy of the most socially marginalized women.¹¹ (A recent analysis by Kavanaugh et al. showed that 11% of Whites, 9% of Blacks, 15% of Hispanics, and 11% of women of other races were using LARC as their current contraceptive method; in multivariate models, Black women were significantly less likely to have used LARC than White women.¹² However, the authors did not have sufficient power to run separate regressions by race. More complicated racial patterns may emerge when race is treated not as a control variable, but as a potential context in which people use or do not use LARC.)

Although outright coercion may be unlikely, subtler biases shape the ways in which race and class influence women’s contraceptive decision-making and patient–provider interactions.¹³ For example, evidence suggests that providers recommend IUDs and implants more to poor women of color than to poor White women and more to poor White women than middle-class women.¹⁴ Finally, historical reproductive injustices—from forced sterilizations to Norplant insertion in exchange for welfare benefits—could lead communities of color to perceive well-intended IUD and implant programs as engaging in racial targeting.^15,16

Despite these concerns, no study to our knowledge has documented contraceptive users’ perspectives on LARC-related provider influence and potential bias. To bridge this gap, we use qualitative data to describe 50 contraceptive users’ perceptions of provider influence and bias, mostly in relationship to LARC but also (when relevant) to other contraceptive experiences.

METHODS

We derived our data from a qualitative study of IUD and implant use among 18- to 29-year-old women³ in Dane County, Wisconsin, a semiurban area of approximately 500 000 inhabitants and home to the University of Wisconsin–Madison. (Although all of the participants in our study identified as women, we recognize that not all contraceptive users do so. We encourage the development of LARC programs and services that are inclusive of transgender and gender-queer individuals.) Approximately 13% of the county’s residents live below the federal poverty level (compared with 15% nationally), and 19% are people of color (compared with 23% nationally).¹⁷ The purpose of the larger study was to assess barriers to and facilitators of LARC use among young adult women in Dane County and at the university. Because the role of providers emerged so strongly over the course of the study, we conducted a closer analysis of the data pertaining specifically to provider influence and bias.

In phase 1 of the larger study, investigators conducted focus groups with women who had any history of contraceptive use. These focus groups were designed to explore young adult women’s LARC-related knowledge and attitudes, as well as various factors associated with LARC acceptability and access.¹⁸ We also conducted 12 one-on-one interviews with former or current LARC users to more deeply explore personal experiences with these methods (phase 2).

To ensure socioeconomic diversity among participants, we designed a stratified sampling frame: one third of focus groups and interviews were conducted with current university students, and two thirds were conducted with women from the community who were currently receiving at least 1 form of public assistance. We were interested in lower-income individuals given their increased likelihood of unintended pregnancy,¹⁹ whereas university students were of interest for a related research project. We selected a participant age range of 18 to 29 years given this group’s disproportionate burden of unintended pregnancy¹⁹ and comparatively low likelihood of using LARC.²⁰ Although race and ethnicity were not part of our sampling frame, we strove for racial/ethnic diversity among both the university and community respondents.

Recruitment and Data Collection

The study’s data collection and recruitment procedures have been described in detail elsewhere²¹; what follows is a brief summary. To recruit participants, study team members posted and distributed flyers in university buildings, public libraries, health clinics, bus shelters, and Job Corps offices. Recruitment e-mails were circulated to public health departments and other pertinent health and social organizations. Information about the study also appeared in the Craigslist community volunteer and “etc.” jobs sections. Some participants were referred by friends or family members who qualified for the study. All of the participants were screened via telephone.

Data collection took place between January and June 2014. We conducted 6 focus groups with 40 women who had any history of contraceptive use; of these women, 8 had used an IUD and 2 had used an implant. Focus groups included 4 to 10 participants and were between 1.5 and 2.5 hours in duration. Interviews, which were conducted with women who had any history of LARC use, were between 25 and 55 minutes in duration, with a mean length of about 50 minutes. At the conclusion of the focus group or interview, university participants received $20 gift cards and community participants received $30 gift cards. All focus group sessions and interviews were audio-recorded and transcribed.

Analysis

We used an inductive, modified grounded theory approach in analyzing the data, meaning that we drew on preexisting themes from the literature and research questions as well as themes arising from the data. For the purposes of our analysis, study team members applied a “provider” code to all sections of the transcripts pertaining to health care professionals who provide contraceptive services. The first and second authors reviewed the coding reports and met to compare and confirm a list of subthemes, which became the basis of the results described subsequently.

To conduct descriptive and analytic cross-case analyses,²² the second author reviewed the coding reports a second time and created a theme-based matrix to assess differences, if any, across the following groups: LARC ever users versus never users, White respondents versus respondents of color, and university students versus women in the community currently receiving public assistance. Because we found few differences between groups, we largely describe commonalities across respondents. We draw distinctions between racial groups only in 2 relevant cases (under the first and third themes described subsequently).

RESULTS

Table 1 provides an overview of the 50 women who participated in the study: 23 current university students and 27 women from the community who were currently receiving at least 1 form of public assistance. About two fifths (n = 21) had had any prior experience with LARC methods. About three fifths (n = 32) identified as White alone, and 18 women identified as Black, Latina, Asian, Native American, or biracial.

TABLE 1—

Overview of Participant Characteristics: 18–29-Year-Old Women in Dane County, Wisconsin, With a History of Contraceptive Use, 2014

Characteristic	Focus Group Participants (n = 40), No.	Interview Participants (n = 12), No.	Totala (n = 50), No.
Status
Current university student	19	4	23
Community resident receiving at least 1 form of public assistance	21	8	27
Race/ethnicityb
White	22	10	32
Black	5	1	5
Latina	6	. . .	6
Asian	3	. . .	3
Native American	2	. . .	2
Biracial	3	1	3
Highest level of education attainedc
High school	2	. . .	2
Some college	24	6	29
College or more	12	4	15
Any history of intrauterine device or implant use	9	12	21

^aTwo focus group participants also participated in interviews, and thus the total sample size is 50 rather than 52.

^bTwo participants self-identified as being of more than 1 race.

^cFour participants omitted educational information.

In the sections to follow, we present 4 themes related to provider influence and bias that emerged from the analyses. Although all of the data derived from the study focused specifically on LARC, some participants also shared relevant stories related to experiences with other patient–provider interactions associated with contraceptive use. Quotations from focus groups are not fully comparable to quotations from interviews as units of analysis, given the inherently different dynamics of these two data collection mechanisms. However, given the exploratory nature of our study, as well as the fact that focus group participants did share both personal and anecdotal stories (as opposed to merely attitudes and larger social norms), we mix both interviewee and focus group data in our presentation of results. All participant names are pseudonyms.

Providers as Trusted Source of Information

Many women, especially White respondents, described health care providers as a trusted source of information regarding contraceptives and LARC methods in particular. For example, when asked “Whose opinion matters most to women as they choose contraceptive methods, including IUDs and implants?” most focus group participants mentioned providers first:

Certainly my doctor I would trust.

—Christy, White, never user, community focus group participant)

I feel like I would talk to a doctor… . I’d rather trust somebody who had a lot of experience.

—Suzanne, White, never user, university focus group participant

Trust could strongly influence women’s willingness to try IUDs and implants and could ameliorate concerns about these methods for potential users. For example, Madeline, a White LARC user and community interviewee, said that she initiated IUD use because her provider suggested it to her, even though she at first was wary about certain aspects of the device. “I trust him,” she said. “My provider is awesome.”

A minority of participants indicated that they were reluctant to trust health care providers regarding LARC recommendations. These participants were disproportionately women of color. For example, Sandra, a Latina university focus group participant who had not used an LARC method, said, “I actually really don’t trust providers. . . . So I go through other means of finding out what [contraceptive method] I want to be on.”

Removal and Other Patient Preferences Sometimes Unheeded

Despite the trust described by a number of participants, women also reported that their own preferences could be undervalued by providers when it came to contraception. Here we highlight 2 issues in particular: contraceptive selection and provider minimization of side effects, especially in relationship to desired LARC removal. The former could pertain to contraceptive methods more broadly, whereas the latter was more specific to LARC methods.

In terms of the contraceptive decision-making process, some women said that providers’ preferences for particular methods could outweigh patients’ desires. For example, Josie, a White university interviewee and current IUD user, reported,

I was pressured to use NuvaRing a long time ago. And I hated it. And I knew I was going to hate it and I told them [my providers] I was going to hate it. But they were like “No, it’s the greatest thing ever. You’re never ever going to have a problem with it.” And I used it for a week and I absolutely hated it and took it out.

In Josie’s case, the stage may have been set for dissatisfaction given her provider’s singular promotion of a method she was not interested in. Such an experience could also potentially undermine future trust in her care providers or lead to contraceptive decisions independent of a provider. (Josie made her own decision to initiate an IUD prior to any interaction with a care provider: “I Googled … kinds of birth control and then narrowed from there because I wanted to look into everything before I decided.” In a qualitative study of approaches to contraceptive counseling, Dehlendorf et al. found that women aged 25 years or younger were more likely than women aged 35 years or older to receive such “foreclosed” contraceptive counseling; that is, the provider discussing only those methods brought up by the patient and leaving decision-making to the patient.²³)

Some women reported feeling disrespected or patronized during provider–patient interactions regarding contraception. For example, Latina focus group participant Sandra, whose distrust of providers was just described, reported the following patient–provider interaction:

I went to the doctor and I didn’t even bring up the pill, and she was like “You know, you have to take it at the same time every day,” and I said “yeah.” “You have to take it at the same time every day at the same time, like not an hour after.” And I was like “yeah.” And she was like “same time every day.” Like she told me four times and I got so upset that I wanted to walk out. I was like “Are you kidding me? I’m not stupid. Stop it.”

Such experiences could undermine provider–patient trust and decrease women’s receptiveness to LARC recommendations from providers.

A substantial number of women reported provider minimization of side effects such as heavy cramping and bleeding, particularly if those side effects led women to request the removal of the LARC device. For example, Dawnesha, a Black community interviewee and current implant user, reported several months’ worth of difficult bleeding and cramping with her implant. She said,

I was telling the nurse how I been on my period for like 3 weeks now, and I’m having bad cramps, and I’m even having them in my back, which I never had before. And she was saying, “Just give it another month or so and see how it goes.” . . . I was mad, but then I’m like, I’m just going to give it another try. They know best because they go to school for this stuff.

At the time of her interview, Dawnesha was still hoping to have her implant removed if her care providers would agree to it.

Heather, a White community interviewee who was a satisfied IUD user at the time of the interview, said that she had faced provider resistance when she wanted her IUD removed a month or slightly more after its insertion. She said,

I told them that I wanted it out and they said that it’s really expensive and that the IUD’s the best option. I got some resistance there. . . . I was a little emotional at the time and she [the provider] didn’t even care, it seemed.

On one hand, Heather was a satisfied IUD user at the time of the study, so her symptoms had improved; on the other hand, she felt frustrated and disrespected during that particular patient–provider interaction.

Elizabeth, a White community interviewee, encountered similar resistance when seeking the removal of her first IUD, ParaGard, after experiencing significant bleeding and cramping. She reported,

My provider was really hesitant to remove the ParaGard. She kept telling me, “Well, we should wait 3 months and see if your symptoms have worsened.” And I waited 3 months and she’s like “Well, you should wait some more.” And I’m like “No. So take it out or I’m going to a different doctor. Those are your options.”

Elizabeth was finally able to get her ParaGard removed, and she went on to get a Mirena, with which she was satisfied at the time of the interview. However, she reflected that many patients may not be able to harness as much agency or “feistiness” when pushing back on providers’ recommendations. She also reflected on institutional or peer pressures that might shape providers’ LARC practices. She said,

I don’t know if it makes them [providers] look bad if you have an IUD removed and they’re the one who placed it, or I don’t know if they have some stat chart somewhere, like a contest board in the breakroom.

By contrast, some women received assurances from their provider about LARC removal. For example, Kelli, a White IUD user and community interviewee, reported that “I remember my providers saying, ‘You know, if you do want it out any time before the 5 years, just make an appointment and we can take it out.’ ” Such assurances from providers appeared to increase women’s willingness to try LARC methods. They could also help women “stick it out” with side effects such as bleeding and cramping, which are likely to decrease over time.

Prior Reproductive Injustices' Role

Women shared their personal experiences with provider pressure regarding LARC but also identified the potential for provider bias based on social and reproductive factors. When asked “Do you think providers are more likely to recommend IUDs and implants to some groups more than others?” a number of participants, especially White participants, mentioned clinical or reproductive characteristics, particularly multiparity (i.e., a history of 2 or more full-term pregnancies). However, participants across racial groups also cited the potential for racial and socioeconomic bias in provider recommendations. Women expected that providers would be more likely to recommend IUDs and implants to women of color, poor women, and women deemed uneducated or unintelligent by providers. For example:

Young African American women are more pressured [to use LARC] from my point of view. Even if they’re in a responsible relationship or state of mind or set of circumstances, or even if they’re just going in for education, providers can be very judgmental.

—Loretta, African American, IUD user, community focus group participant

I can definitely see providers maybe pushing for the long-term method more with poorer women.

—Kelli, White, IUD user, community interviewee

I think probably minorities and lower-income people could be more likely to be pressured [to use LARC].

—Marissa, White, IUD user, community interviewee

Women across racial groups, both from the community and from the university, linked these potential biases to historical reproductive injustices such as forced sterilizations and eugenic social policies. For example, when asked about whether providers might recommend LARC methods more to some women than others, White community interviewee Elizabeth responded,

Historically there have been government efforts to actually sterilize Native Americans and Black women because they didn’t want those populations growing. There’s this sort of idea, “because you’re poor we can’t trust you to make good decisions about birth control and so we’re going to make that decision for you.”

One participant linked historical injustices to her own unwillingness to try a particular LARC method:

In school, we learned a lot about the Norplant implant and how women of color were specifically targeted for that. I don’t know if that’s still the case, or what happened with that, but because of that I’m really anti-implant.

—Heidi, White, never user, community focus group participant

As these examples suggest, White women cited historical reproductive injustices and then linked these injustices to LARC promotion. However, women of color were more likely to describe prior injustices in a way that personally affected them and their communities and that increased their wariness of LARC recommendations as a result. For example, African American focus group participant Loretta, quoted earlier, described how she had seen cousins and nieces “lectured to” about birth control by providers “even when they were responsible.” And Sandra (the Latina university focus group participant quoted earlier), who had never used an LARC method, reported,

Birth control in general makes me really wary because it was meant to keep people like me from procreating and having more of us, right? . . . I don’t really trust doctors because I don’t know what subconscious things are going on when certain methods are being recommended to me.

Larger Influences on Providers

A final salient theme pertained to participants’ identification of larger influences that may shape providers’ contraceptive recommendations, including those relating to LARC. Some respondents contextualized providers within larger institutional cultures, often sympathetically. For example, rather than singling out providers for being uniquely biased, several women argued that everyone in American society is affected by racial and social class biases. Some used the term “unconscious.” As Mary, a White focus group participant who had never used LARC, said,

I don’t think doctors are really exempt from being prejudiced. I mean, we’re all still human, so I think it’s silly to say definitely that providers will recommend LARC methods in an unbiased way.

A few participants seemed sympathetic to the stresses placed on health care providers, particularly in resource-deprived settings. For example, Elizabeth reported,

I mean [providers] see a huge volume of people, and after you see your thousandth pregnant 14-year-old, you’re probably like “This is ridiculous. These people are clearly not smart enough to handle this. So [LARC] is what my policy is going to be moving forward.”

This respondent reflected that she, too, could make internal judgments about how other women should use contraception:

If someone’s not capable of … remembering to take her pill or insisting on using a condom or whatever, then it would be in her best interest to use something that she can’t mess up or forget or lose or break.

In summary, some respondents admitted that they, too, thought that they knew best when it came to (some) others’ contraceptive use, even as they criticized the same paternalism exhibited by providers.

DISCUSSION

Despite public health enthusiasm about LARC, a number of reproductive justice proponents have been concerned about how LARC methods might be promoted by health practitioners.^9–11 One central worry has been the role of bias in shaping contraceptive recommendations,¹³ a concern upheld in a trial documenting that providers were in fact more likely to recommend IUDs to poor women of color than to poor White women.¹⁴ In this exploratory study of 50 contraceptive users, we found that patients have the same expectations for provider bias in terms of LARC recommendations. We also found evidence of provider resistance to removal. Although women of all races reported prior reproductive injustices, women of color were especially likely to experience LARC promotion as racialized and to express a personal connection to such injustices. By contrast, at least some contraceptive users expressed sympathy toward the larger forces that may shape providers’ LARC recommendations, from deep-set racism and cultural bias to high-pressure health care systems.

On the basis of these findings and the growing momentum of reproductive justice approaches to LARC, we propose several strategies that might improve future provider–patient interactions. We champion contraceptive counseling and practice protocols that support both patients and providers in offering patient-centered care.

Professionals who offer contraceptive care are strongly encouraged to educate themselves about prior reproductive abuses in socially marginalized communities (if they are not deeply aware already). (A variety of helpful resources have been compiled at the University of California’s Reproductive Justice Virtual Library.²⁴) These histories are known and remembered by contraceptive users and patients; they should be on professionals’ minds as well. Even well-intentioned practitioners may hold deep-set judgments about whether (more) children are a good idea for certain women. Rather than pretending such implicit biases do not exist, we would be better served by acknowledging them and identifying techniques to actively challenge and undermine dominant stereotypes. Patient-centered practices such as values clarification²⁵ and cultural humility^26,27 have been encouraged in related health care domains, and they could be helpful in relationship to LARC as well. Providers need and deserve tools to better cope and respond when patients make different choices than they might want them to.

Aligning with exciting new work in this area,^11,28–30 we also encourage contraceptive counseling and marketing that adopt a user-centered framework that supports clients in identifying their family planning priorities. As argued by Gomez et al.,¹¹ we need to attend to users’ individual preferences and circumstances, particularly in the case of groups whose fertility has been historically devalued. We also need to temper enthusiasm that LARC is the best option for all contraceptive users.¹¹ Even briefly acknowledging historical racial injustices during contraceptive counseling sessions may be important as well. For example, counselors may wish to say something along the lines of

I want you to know that I recommend these methods to all of my patients, regardless of their race, social class, or number of children; however, these methods might not be right for everyone, and I want to make sure we find the one that works best for you.

Patients should also be offered thorough information about potential side effects so that they are empowered to better manage these effects if they occur. Skillful “expectation management” can ease women’s insertion fears and help them better weather unpleasant bleeding and cramping. In addition, practitioners should emphasize to contraceptive patients that they will support them whenever they decide to discontinue an LARC method. Finally, providers may wish to consider a developmental approach to these issues, as patients’ ages and life stages are likely to influence their willingness to consider LARC as well as the kind of reproductive health care they receive.²³

Limitations

Our findings should be considered in the light of the study’s limitations. Most centrally, the original study goals were to assess general barriers to and facilitators of increased LARC use among young adults, not specifically the issue of provider influence. Although race could be an important factor, the sampling frame was not perfectly designed to assess racial differences. For example, we would have likely received different input from a focus group composed completely of African American women than from a group including 6 White women and 2 women of color.

Furthermore, our sample contained a considerable number of relatively socially advantaged women. For example, the 12 interviewees (all former or current LARC users) included only 2 non-White women, and all had at least a high school diploma. LARC trajectories among women of color with comparatively less formal schooling may differ with regard to provider influence. Such women’s perspectives and experiences are invaluable and should be included in future research in this area. We may have also garnered different results if we had conducted a study in a large, urban area with more socioeconomic and racial diversity.

Conclusions

A user-centered approach to LARC could serve to increase people’s access to LARC methods if they wish to use them, could improve their ability to have LARC devices removed if they so choose, and could help them feel respected and cared for by their providers. System- and provider-level changes can help facilitate access to unbiased and noncoercive information through patient-centered contraceptive counseling.