|Home | About | Journals | Submit | Contact Us | Français|
Although three quarters of reproductive-age women see a health provider annually, less than half receive recommended contraceptive counseling services. We sought to explore providers' perspectives on the challenges to contraceptive counseling in primary care clinics to develop strategies to improve counseling services.
A qualitative, focus group (n=8) study was conducted in November and December 2007; 48 of 90 providers practicing in four primary care clinics at the University of Pittsburgh Medical Center participated. Providers included physicians, nurses, and pharmacists working in these clinics' multidisciplinary teams. Discussions explored perceived barriers to the provision of counseling services. All groups were audiorecorded, transcribed, and entered into Atlas.Ti, a qualitative data management software. The data were analyzed using a grounded theory approach to content analysis.
Perceived patient, provider, and health system barriers to contraceptive counseling were identified. Perceived patient barriers included infrequent sexual activity, familiarity with a limited number of methods, desire for pregnancy despite medical contraindications, and religious beliefs. Provider barriers included lack of knowledge, training, and comfort; assumptions about patient pregnancy risk; negative beliefs about contraceptive methods; reliance on patients to initiate discussions; and limited communication between primary care providers (PCPs) and subspecialists. Health system barriers included limited time and competing medical priorities.
PCPs vary widely in their knowledge, perceived competence, and comfort in providing contraceptive counseling. General efforts to improve integration of contraceptive counseling into primary care services in addition to electronic reminders and efficient delivery of contraceptive information are needed.
Nearly half of U.S. pregnancies are unintended.1–3 Although three quarters of reproductive-age women see a health provider annually, less than half receive contraceptive or other family planning services4–6 despite the fact that contraceptive and family planning counseling are considered core competencies for primary care physicians (PCPs).7 The provision of contraceptive services by PCPs is meant to minimize fragmentation of women's healthcare and maximize opportunities for comprehensive primary and preventive care at each clinical encounter. Because women receive the majority of preventive care from nongynecological providers,7,8 PCPs are in a unique position to offer contraceptive counseling services, potentially reducing unintended pregnancies. However, PCPs are less likely to provide contraceptive services than are obstetrician/gynecologists, even when these services are medically indicated.9–13 To improve delivery of contraceptive counseling services, it is important to identify factors that limit provision of these services in primary care practices.
We explored opinions about barriers to contraceptive counseling among providers in primary care settings. This study was conducted among PCPs in a large, urban health system as part of a larger effort to develop system-based strategies to improve contraceptive counseling.
We conducted eight focus groups with 48 providers in November and December of 2007. Participants were recruited from four academic and community-based primary care practices affiliated with the University of Pittsburgh Medical Center (UPMC) in Pittsburgh, Pennsylvania. Participants provided written consent. The study was approved by the university's Institutional Review Board.
The four targeted clinics all provide comprehensive women's healthcare (e.g., Pap smears, sexually transmitted disease [STD] testing, contraceptive counseling and prescribing, pregnancy testing), although individual providers in each clinic vary in their scope of practice. The targeted practices had 90 eligible providers, and 92% of patients are white, 4% black, and 3% Asian.
The UPMC health system has a unique arrangement in the targeted clinics. These clinics have integrated, multidisciplinary teams that include physicians, nurses, and pharmacists who work in teams to provide coordinated care for patients. We refer collectively to these three provider types as “providers” throughout this article. To adequately assess barriers to contraceptive counseling services within our health system, we included all three provider types. We recognize that similar practice structures that involve pharmacists as an integral part of service delivery do not exist at many other institutions. However, we included pharmacists in our sampling frame because they are part of the local process of providing medication counseling and prescribing in our health system. The clinical care teams generally comprise more nurses at various skill levels than physicians or pharmacists. Thus, there was a larger pool of nurses than physicians or pharmacists from which to recruit. Each clinic generally has only one pharmacist as part of its team; thus, the pool of pharmacists to recruit from was quite small.
Eligible providers had to work in clinics that provide outpatient primary care services for reproductive-age women. We purposefully sampled physician providers from three nongynecological primary care specialties (internal medicine, family medicine, and adolescent medicine) as well as nurses and pharmacists working as part of the multidisciplinary teams. Participants were recruited using mail and e-mail invitations and fliers posted in staff rooms.
Each focus group had 6–10 participants of mixed gender and lasted approximate 2 hours. Groups were stratified by provider type: four were conducted with physician/pharmacists and four with nurses of varying levels. Physicians and pharmacists were included in the same focus groups because the pool of pharmacists to recruit from was small and because physicians and pharmacists have a similar scope of work regarding contraceptive counseling and prescribing privileges, whereas not all nurses have the same scope of work. Participants completed a brief demographic questionnaire that assessed sociodemographic (age, gender, and race/ethnicity) and individual providers' clinical practice characteristics (number of hours per week spent providing direct patient care, number of reproductive-age women seen weekly, and whether participants prescribed contraceptives). All participants received a cash incentive of $100.
Each discussion was facilitated by two moderators: one led the discussion, and the other took detailed notes about the mood during discussions, nonverbal communication, and group affirmations/rejections of stated opinions. This information was incorporated into the written transcripts to aid with analysis. A pool of four individuals experienced in qualitative and health services research moderated: two physicians (an obstetrician/gynecologist and an adolescent medicine specialist), one Ph.D. psychologist, and one Masters in Public Health trained researcher. None shared clinical responsibilities with participants. Moderators used a standardized, semistructured interview guide developed using a conceptual framework that explored primary care clinicians' perceived knowledge, self-efficacy, barriers/facilitators, and personal role in contraceptive counseling. Recognizing that women's contraceptive needs may change throughout the year, we asked providers about contraceptive counseling provided in general primary care practice and did not limit our questions to counseling provided solely during annual examinations. Sample questions include: Which types of birth control do you particularly encourage your female patients to consider and why? In what situations do you recommend that women use contraception? What tends to prompt you to talk with a patient about her need for contraception? After each focus groups, moderators and the principal investigator (PI) debriefed about the discussion content and compared emergent themes with previous focus groups. Focus groups were held until thematic saturation was reached.
The data were cleaned and analyzed between December 2007 and September 2008. For questionnaire data, Excel was used to calculate frequencies for categorical variables and means (or medians) for continuous variables.
All focus groups were audiorecorded and transcribed, and the transcripts were entered into ATLAS.ti, version 5.2, a qualitative data management program. We used a grounded theory approach to content analysis.14,15 To develop thematic codes, transcripts were systematically and independently reviewed by two coders using an iterative process referred to as open coding. Both coders had moderated or taken notes during focus groups and were, therefore, familiar with the data. One coder was an adolescent medicine specialist, and the other was an M.P.H. researcher. The coders and PI synthesized open codes into broad themes, producing a final codebook containing four themes: perceived barriers, discussion prompts, discussion content (e.g., types of contraception), and contraceptive knowledge. Coders independently recoded all transcripts using the codebook. In 85% of cases, there was complete agreement in coding. Consensus was reached about the remaining 15% of passages via brief discussion in the presence of a third party (PI). The goal of this process was to triangulate the perspectives of multiple individuals to promote fuller understanding of the data.16 We report themes that were consistent and common across all focus groups and participant characteristics (e.g., provider type).
Sociodemographic data were available for 41 of the 48 providers (Table 1). Fourteen providers were physicians (8 general internal medicine faculty, 1 internal medicine resident, 1 adolescent medicine faculty, and 4 family medicine faculty), 26 were nurses (16 registered nurses, 4 license practical nurses, 6 nurse practitioners), and 1 was a pharmacist. Of the 6 nurse practitioners, 4 were from family medicine, 1 was from internal medicine, and 1 did not specify the primary specialty. The general distribution of the three provider types reflects the general distribution of physicians, nurses, and pharmacists within the multidisciplinary teams across the four clinics sampled. Whereas the majority of all participants prescribed contraception, 27% preferred to refer women elsewhere for contraception services. Participating physicians and pharmacists all reported prescribing contraception.
Providers identified a number of issues they believe patients bring to clinical encounters that hinder providers from engaging patients in contraceptive counseling discussions. Providers defined contraceptive counseling as encompassing initial discussions to start a method as well as engaging a woman in follow-up discussions about the adequacy of the woman's current contraception method. The latter was believed to be important, as a woman's medical comorbidities, medication use, or pregnancy intentions may change and necessitate a change in her contraceptive method. Barriers to both types of conversations were discussed.
The major themes and representative quotes related to perceived patient barriers are shown in Table 2. The most common perceived patient barrier was patient preference for particular contraceptive methods. Method preference was thought to be primarily influenced by prior experience with or side effects from a method or dislike of a method's mode of administration (e.g., injectables). Method preference was also perceived to be influenced by family, peers, or drug marketing. Providers thought these preferences limited women's willingness to consider all available contraceptive options.
A lot of times I notice my college-age [patients] already coming in knowing what they want because their friends are all on that.…Discussions with alternatives are actually quite limited.…I think they choose it more just because it's all they've known.
Some providers believed current contraceptive use also made women less willing to hear about alternative or even more effective options.
A second challenge was effectively conveying the health risks associated with pregnancy. One provider described managing a liver transplant patient through two pregnancies.
We have one patient with a liver transplant who was cautioned not to get pregnant.…The patient decided herself that she wanted to be pregnant and did get through the first pregnancy fine, but then we always caution again, “You really should be on the pill.” But then [she] got pregnant again and now is having a lot of medical problems….
Additional perceived patient barriers included religious prohibitions against contraceptive use, patient discomfort discussing contraception, and perceived lack of need for contraception. Providers thought that discomfort discussing or using contraception was particularly salient for minors. Parental presence during visits and fear that parents would discover contraceptive packaging or learn of contraceptive use through insurance statements were believed to limit adolescent comfort discussing contraception.
Provider barriers generated the greatest amount of discussion (Table 3). Six themes arose, including pregnancy risk classification; lack of knowledge, training, or comfort; beliefs about certain methods; perceived patient responsibility for initiating discussions; need for skilled personnel for certain methods; and lack of communication with subspecialists.
The most common provider barrier was providers' use of informal risk classification schemes to determine which patients needed contraceptive counseling. For example, older women of reproductive age who had previously had children were assumed to be “responsible enough not to get pregnant.” Women who reported not currently being sexually active were thought not to need contraceptive counseling, despite the reality that some could initiate sexual activity before their next clinic visit. Need for contraceptive counseling was often prompted by issues that arose during a visit, such as prescribing a teratogenic medication or diagnosing a medical condition for which pregnancy was contraindicated. Providers universally agreed teenagers and college students should be routinely offered contraceptive counseling. It is important to note that not all providers used subjective risk classification to determine women's contraceptive needs. Some reported offering contraceptives to all reproductive-age women.
Providers varied in their self-assessments of their contraceptive knowledge. Condoms and oral contraceptives were the methods providers cited being most familiar with and prescribing most often. Most stated they lacked knowledge about other contraceptive options and contraceptive counseling training. The main challenge providers reported facing was keeping abreast of new contraceptive options. One provider lamented, “There are so many out there.” Providers articulated that lack of knowledge and training worked in tandem to limit their comfort with and likelihood of providing contraceptive counseling. As a consequence, a number of participants preferred to refer patients women's health specialists rather than engage in contraceptive counseling discussions.
Provider beliefs, particularly inaccurate beliefs about contraceptive methods, influenced their contraceptive counseling practices. Providers described a number of beliefs about the appropriateness of contraceptive options for certain types of patients as influencing the content of their discussions: “We discourage the patch because of the excessive hormones that the students really don't need.”
Some providers stated that patients, not providers, are responsible for initiating contraceptive discussions. One provider stated, “If they don't bring it up, I often don't.” Another explained that it required very unusual circumstances for him to feel an obligation to initiate a discussion of contraception, “I had a patient once who almost died during pregnancy and then sort of felt like I needed to talk to her about it.”
The need for skilled personnel to initiate certain methods, such as intrauterine devices (IUDs) or contraceptive implants (Implanon, Organon International) was another important barrier. Some providers articulated concerns that referring patients to family planning specialists could hurt their business. As one said, “As soon as you recommend the IUD, the woman is no longer our patient.” Another offered, “Because I have a [provider] who puts in IUDs in my office, I talk about IUDs.”
Finally, providers complained that lack of communication between providers and subspecialists adversely affects contraceptive counseling. Providers complained that subspecialists often initiate medications that affect women's fertility or are teratogenic without providing contraceptive counseling necessitating follow-up by PCPs. One shared an anecdote:
She called about whether or not she was pregnant, and I asked her a couple questions.…She wasn't aware that she was more likely to get pregnant once they got her polycystic ovarian syndrome (PCOS) under control. That was, I thought, unfortunate. Turns out she was not pregnant, which was a good thing, and then of course there was a follow-up conversation about, well now you need to make sure you don't get pregnant.
On the other hand, some providers refused to provide contraceptive counseling to women being treated for medical conditions, believing that it was the responsibility of the woman's subspecialists.
Providers described a number of perceived system-level challenges to contraceptive counseling (Table 4). Lack of insurance coverage for contraceptives was most commonly cited, particularly for adolescents and young adults. As one provider offered, “‘It's a high copay.…Money is a barrier.”
Lack of time and the need to address competing medical priorities also limited provision of contraceptive counseling. One provider stated, “If they're coming in with a cold, it's not going to come up.” Work setting and case mix also play a role. Providers practicing in environments with access to providers trained in women's health or family planning reported being more likely to routinely provide comprehensive contraceptive counseling. Similarly, those who saw a sizable number of young, reproductive-age patients were more likely to provide contraceptive counseling routinely.
Finally, the lack of clinical care systems to remind providers to provide contraceptive counseling or to identify patients at risk of pregnancy was cited as problematic. Providers in practices with standardized annual examination and new patient history forms reported that these forms prompted them to at least inquire about contraceptive use. Similarly, providers in practices with electronic medical record (EMR) systems said that electronic prompts at annual examinations were helpful. Providers indicated that it would be helpful if the EMRs could quickly link them with up-to-date information about available contraceptives to aid with patient counseling and prescribing.
We explored the perspectives of providers working in four multidisciplinary primary care clinics at a single health system regarding challenges to contraceptive counseling in primary care. This study was conducted as part of a larger effort to develop system-based strategies to improve contraceptive counseling within our health system. Providers cited multiple challenges to the provision of contraceptive counseling services in their practices. Similar to studies on barriers to the provision of general preventive services17 and STD screening18 in primary care, providers reported challenges at the provider, perceived patient, and health system levels. Our results suggest that no single approach to improving provision of contraceptive counseling services will likely suffice. Rather, a variety of strategies is needed to effectively improve the provision of these services.
Many reasons have been suggested to explain why primary care providers do not offer recommended prevention services to patients, including lack of reimbursement, patient refusal, and lack of time. Studies demonstrate that the large number of screening and counseling recommendations make provision of prevention services, including contraceptive services, challenging.12,19–21 Hence, providers often feel forced to decide whom to provide services to on a case-by-case basis. Our finding that providers' use of a variety of techniques to identify women to whom they will provide contraceptive counseling reflects this. Although we acknowledge the existence of financial, time, and system-based challenges that can be cited to explain the low rates of delivery of contraceptive counseling services in primary care, the reality is that existing practices that restrict a woman's access to complete information or referral services have been construed by some as unethical. The provision of basic contraceptive counseling is part of the scope of basic practice for PCPs,7 and many providers are not meeting this requirement. Certainly, referring patients with complicated medical histories for contraceptive counseling with a contraceptive specialist makes sense; however, the vast majority of reproductive-age women do not fall into this category and would benefit from brief, targeted contraceptive counseling delivered in the primary care setting. Providers need to strike a careful balance between addressing patients' acute care and prevention needs, maximizing reimbursement and adhering to time constraints. Given the low reimbursement rates for contraceptive and family planning services, cost-effective solutions are needed to help providers appropriately identify patients in need of contraceptive services.22 In the meantime, policy advocates should continue to lobby to obtain adequate reimbursement for contraceptive and other family planning services.
Providers overwhelmingly thought that they lacked the knowledge and skills to provide contraceptive counseling. Over the last three decades, studies have consistently reported that contraceptive training for PCPs is insufficient.23–32 Opportunities for practicing providers to obtain continuing medical education (CME) regarding advances in contraceptive methods also are limited.24,33 Thus, our finding that providers' report an overwhelming lack of knowledge, comfort, and self-efficacy for providing contraceptive counseling services comes as no surprise; however, it is disconcerting that this problem persists. Developing standardized curricula in contraception counseling, adding contraceptive questions to provider certification examinations, offering CME opportunities, and increasing the availability of electronic counseling resources represent just a few tools that have been proposed to address the problem.33 Success depends on training programs and health systems assuming responsibility for making these things happen.
Providers described a number of barriers they believe patients bring to contraceptive counseling encounters that limit effective counseling interactions. These include preexisting patient preferences for particular methods, patient desire for pregnancy or pregnancy ambivalence despite medical contraindications to pregnancy, and religious prohibitions, to name a few. Providers' perspectives of patient-level barriers are surprisingly similar to women's own reports of what limits their use of effective contraceptive options.34 These challenges represent areas of conflict during patient-provider communication about contraception that may cause reduced contraceptive counseling because of patient refusal to engage in counseling discussions (or lack of follow-through with contraceptive recommendations), as opposed to provider noncompliance with counseling guidelines. To develop effective approaches for increasing the provision of contraceptive counseling, it is critically important that we understand the barriers that patients and providers bring to the counseling process. Few studies have assessed optimal strategies for counseling women about unintended pregnancy and contraceptive use.34 Client-centered counseling techniques, such as motivational interviewing approaches, have been proposed.35,36 However, more rigorous randomized controlled trials of such interventions are needed to test the efficacy of these approaches for assessment of women's contraceptive needs.37
The major barrier to contraceptive counseling will always vary between individual providers. Thus, an effective strategy for increasing contraceptive counseling may be to make screening for pregnancy intentions and assessing a woman's contraceptive need routine for all reproductive-age women. As with pain assessments, all reproductive-age women could be asked about their pregnancy intentions and current contraceptive use at every visit, not just at annual examinations or when patients have specific reproductive complaints. This brief screening would remind providers to address both contraception and preconception counseling. Although one may think contraception might not come up during an acute care visit for an infection, it often should. If, for example, during an acute care visit, an antibiotic or other medication is prescribed whose metabolic pathways interfere with contraceptive effectiveness, the provider must be attuned to the potential medication interaction and recommend that the woman use a backup contraceptive method temporarily. Similarly, if during any visit, a new medical diagnosis is made for which pregnancy is contraindicated or increases a women's risk of complications for herself or her fetus, the provider should recognize this, address it, and offer contraception, if appropriate. Integrating routine screening into clinic routines would also help patients and providers to view this topic as a normal, expected part of primary care. As found with pain screening, however, mandatory screening will not ensure that providers respond appropriately to women's need for information, especially if providers lack confidence about their contraceptive knowledge.38
This study was motivated by our interest in identifying system-based strategies to improve contraceptive counseling within our health system. Although no single intervention is likely to address all barriers to contraceptive counseling in primary care, EMR systems have been proposed as an important systems-based tool for identifying patients in need of contraceptive counseling.39 Based on our results, two functionalities of EMRs were suggested: routine prompts for providers to provide contraceptive screening and counseling and links to up-to-date information about contraceptive methods and medical eligibility criteria, especially for medically complicated patients.40 For example, if a patient has a medical condition and her provider is unsure if a particular method is appropriate to use, hyperlinks within the EMR could rapidly give the provider access to online resources, such as Micromedex or the World Health Organization's Medical Eligibility Criteria. EMR systems have not yet been universally implemented in the United States and will not address all barriers, such as patient-level barriers, time constraints, or the need for skilled personnel to initiate the most effective and least user-dependent contraceptive methods (e.g., IUDs, Implanon). Developing creative ways to provide contraceptive services outside of the traditional physician-patient face-to-face encounter is warranted. Contraceptive counseling offered via group visits with physicians, physician extenders, or health educators has been proposed. Social marketing campaigns using print or electronic media to educate women about their contraceptive choices represent another option.21,33 Given the complexity of factors limiting the provision of contraceptive counseling in the primary care setting, practices that employ several strategies will likely be most effective.
There are several important limitations to mention. First, we examined a heterogeneous group of provider types. Different provider types (pharmacist, nurse practitioner, physician) may have differing levels of knowledge, training, or comfort regarding contraceptive counseling. Similarly, they are subject to different professional norms that influence their counseling practices.22 This study was unable to tease out these differences. Second, our sample lacked gender and racial/ethnic diversity. Although findings were consistent between focus groups, suggesting their validity, we cannot exclude the possibility that a different sampling framework that stratified by gender or race/ethnicity might have identified different or additional themes. Third, the qualitative nature of this study allowed us to identify the range of barriers consistently cited by providers within our system; however, our approach did not allow us to rank these items or determine which are the primary barriers operating at each level.41 Fourth, this set of barriers is not exhaustive. We identified barriers that pertain to the provider community; additional challenges to the promotion of contraceptive counseling certainly exist. Finally, we present only provider perspectives. Patients may perceive a different or additional set of barriers. To effectively improve contraceptive counseling services, the challenges of both must be understood and addressed.
Providers in primary care settings face multiple barriers to the provision of contraceptive counseling services. Challenges exist at the patient, provider, and health system levels. Given the complexity of factors limiting the provision of contraceptive counseling in primary care settings, no single strategy will suffice. Clinical practices and health systems should seek to employ a variety of approaches to increase provider education and service delivery. Given the lack of progress in improving contraceptive counseling rates over the past few decades and the persistently high rates of unintended pregnancy, interventions at all levels of the healthcare system are needed.
Assistance with coding was provided by Cara Nikolajski, M.P.H. This study was funded by an R18 grant from the Agency for Health Care Research and Quality (AHRQ) grant 1R18HS017093-01. E.B.S. was supported by a K23 grant from the NICHD, grant K23HD051585. A.Y.A. was supported by grant KL2 RR024154-03 from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH), and NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NCRR or NIH. Information on NCRR is available at www.ncrr.nih.gov/. Information on Re-engineering the Clinical Research Enterprise can be obtained at nihroadmap.nih.gov/clinicalresearch/overview-translational.asp.
These results were presented as a poster at Reproductive Health 2008, the joint annual meeting of the Association of Reproductive Health Professionals, the Society for Family Planning, and the Planned Parenthood Federation of America in Washington, DC, September 19, 2008.
The authors have no conflicts of interest to report.