Establishing Consensus on Residency Education in Women's Health

Background

Upon graduation from residency, internal medicine residents should possess the knowledge and skills to provide gender-specific care to their female patients. Specifically, the Accreditation Council for Graduate Medicine Education (ACGME) lists the care of “gender-specific diseases” among the graduation requirements for internal medicine residency. ¹

Despite calls for competency in women's healthcare, residents are still underprepared to provide comprehensive gender-specific primary care to female patients. Research demonstrates that residents infrequently provide preconception counseling to female patients, ² and several studies have found that while internal medicine residents feel that contraceptive counseling is important, only a minority of residents routinely provide contraceptive counseling to their patients. ^2,3 This may be due to a lack of knowledge as other studies have demonstrated poor contraceptive knowledge among graduating residents from primary care fields, including internal medicine. ⁴

Previous surveys of primary care residency program directors found agreement that their residents should master a range of women's health clinical topics and skills, including incontinence, polycystic ovarian syndrome (PCOS), violence, and menopausal symptoms; many feel that their residents fall short of that goal ^5,6 (Casas' et al. unpublished). In addition, internal medicine residents self-report limited clinical experience and low comfort levels with the management of a range of women's health topics, including preconception counseling, urinary incontinence, abnormal vaginal bleeding, and menopausal symptoms. ⁷

This disconnect between ACGME articulated residency training requirements and the knowledge and skills of graduating residents may be a result of a lack of specific standards for women's health education. The requirement from the ACGME is vague, stating that “residents are expected to demonstrate the ability to manage patients…in the prevention, counseling, detection, diagnosis, and treatment of gender-specific diseases,” which provide little guidance to internal medicine program directors with regard to specific topics or content areas for women's health education. The American Board of Internal Medicine (ABIM), in 1997, put forth a list of core competencies in women's health that are more specific and range from menopause to family planning to gender-related social issues. ⁸ However, the competencies are quite broad. For example, they list “common gynecological disorders: management and prevention” and “cancer screening and principles of management” among their core competencies. While these are important topics in women's health, the broad nature of these topics can make it difficult for program directors to develop a standardized women's health curriculum, and thus, difficult to assess the competencies of their residents. In addition, there are discrepancies between the ABIM's core competencies and the topics listed for the ABIM certification examination. ⁹ The ABIM certification examination lists more granular topics such as sexually transmitted infections, dysmenorrhea, and contraception. ⁹ This discrepancy may add to the confusion regarding which women's health topics should be included in residency education.

The objective of this study was to establish a consensus list of specific topics and procedural skills in women's health that every internal medicine resident should be proficient in upon graduation to ensure the ability to provide comprehensive care to women. Establishing a consensus list will provide internal medicine residency programs with a roadmap by which to fulfill the requirements regarding women's health training and assess women's health educational efforts and the competency of their residents.

Methods

We conducted a two-round Delphi of women's health experts to establish a consensus list of specific topics and procedural skills in women's health. We identified nationally recognized women's health experts from two different sources, including the 2015 Director of Residency and Fellowship Programs in Women's Health maintained by the Association of Academic Women's Health Programs and Society of General Internal Medicine's Women's Health Education Interest Group membership list. Additional women's health leaders identified by the study authors who were not members of the above groups were also invited to participate. All members of the panel had to be a physician, have a confirmed email address, and not be a member of the research team. We established a priori a desired panel size of 15–20 experts consistent with standard Delphi procedure. We identified a total of 41 individuals who were invited to be a part of the panel. All 41 individuals were sent an email invitation explaining the purpose of the study, as well as a link to the website, for the first round of the Delphi. While the research team was aware of who was participating in the study and their responses, the participants remained anonymous to one another.

We reviewed the literature on residency education in women's health, including the ABIM core competencies, and developed an initial list of topics in women's health. This list was then reviewed by five local women's health experts to produce a final list of 54 topics. The first round of the Delphi process involved four different components. First, the list of topics was presented to the participants and they were asked to rank each topic, using a 1–5 Likert scale with 1 indicating “not important” and 5 indicating “essential,” based on how important they thought each topic was for every internal medicine resident to be proficient in upon graduation to be able to provide high quality care to women. Second, they were asked using the same list of topics to identify which items, if any, they felt were “critical” for a standard women's health curriculum if time for such a curriculum was limited. Third, participants were given the opportunity to list any additional topics they felt were critical for a women's health curriculum in a free text response section. Fourth, participants were asked to complete a six question demographic survey. Participants had 2 weeks to complete the first round.

After the first round, we calculated the mean importance rating and standard deviation for each topic. These data were given to participants in the second round. We also calculated the percentage of participants that identified a topic as “critical” for a women's health curriculum. Topics that were identified as critical by at least 50% of the participants were retained for the second round.

Only those who completed the first round were invited to participate in the second round of the study, which involved two parts. For the first part, participants were presented with the list of 54 topics, as well as two additional topics (diagnosis of mental health disorders and treatment of mental health disorders), identified in the free response section of the first round for a total of 56 topics. They were again asked to rate each topic using the same 1–5 Likert scale used in the first round based on how important they felt each topic was, taking into consideration the mean rating and standard deviation from the first round. Second, participants were presented with a list of 40 topics identified as critical for a standard women's health curriculum and asked to indicate whether or not they agreed with the designation as a critical topic using a 1–5 Likert scale with 1 indicating strongly disagree and 5 indicating strongly agree.

At the end of the second round, mean importance rating for all topics and mean agreement rating for the critical topics were calculated. The list of critical topics mirrored the list of important topics; therefore, our consensus list included any topic that received a mean importance rating of ≥4. Stata SE was used for all calculations. The University of Pittsburgh IRB approved our study.

Results

Of the 41 individuals we invited to participate in our Delphi, 19/41 (46%) agreed to participate with 100% (19/19) completion in the second round. Our experts represented 17 different programs from across the country, including both Veteran's Administration (VA) and non-VA programs. Participant demographic data are shown in Table 1. The majority of our experts were well-established clinicians who had been in practice for over 10 years and approximately one-third had obtained the rank of full professor. In addition, 32% directed a women's health residency or fellowship and 37% had completed a women's health fellowship. There were no statistically significant differences in association with VA programs, membership to the Society of General Internal Medicine's Women's Health Education Interest Group, or overseeing a women's health residency/fellowship between those that completed our Delphi and those that did not.

Of the 56 topics that our experts were asked to evaluate, 35 topics (62.5%) received an importance rating of ≥4, indicating that they were “very important” or “essential” to women's health education. Table 2 divides these 56 topics into “more important” or “less important” topics, using the mean importance rating of 4 as the cutoff between groups. The mean importance rating is given for each topic, as well as the standard deviation. The topics are categorized into larger clinical domains of women's health; these are displayed to the far left. It can be appreciated that even within a clinical domain, there are topics that are considered “more important” and topics that are considered “less important.” Regarding the important topics, there was an emphasis on the diagnosis of various conditions, such as PCOS, abnormal uterine bleeding, and pelvic pain. Many of the topics that were deemed less important were related to management (as opposed to diagnosis) of various conditions, including: chronic pelvic pain, endometriosis, sexual dysfunction, eating disorders, and PCOS.

Discussion

Our study generated a consensus list of 35 core women's heath topics and procedural skills to help guide internal medicine residency women's health education. Women's healthcare is rapidly evolving as is the demand for physicians who specialize in the care of women. ¹⁰ Internists have been called upon by the American College of Physicians to provide comprehensive care to women. ¹¹ Moreover, data from the VA demonstrates that patients are more likely to receive female-specific cancer screening, ¹² contraception, ¹³ and have a better overall experience when they see a dedicated women's health primary care provider with established competencies in gender-specific care. ¹⁴ Despite the need for internists who are well-trained in women's health, recent work by Casas et al. demonstrates that there is still room for improvement when it comes to women's health education. They found that almost half of program directors are unfamiliar with the current competencies put forth by the ABIM and that many core topics are still not being addressed in residency education.

The strength of our list is that it offers very specific topics, such as the diagnosis and management of osteoporosis, on which to build curricula, compared to the vague nature of the current ABIM guidelines. ⁸ While the majority of our important topics would fit under the ABIM core competencies, there are some differences. Notably, there was no mention of lesbian, gay, bisexual, and transgender (LGBT) health in the ABIM competencies ⁸ ; however, our expert panel felt that lesbian/bisexual health was important. This likely reflects the fact that the ABIM competencies were written nearly two decades ago and since then there have been significant changes in societal norms with an accompanying realization of the important healthcare needs of the LGBT community.

There are some discrepancies in the opinions of our experts and the ABIM certification examination. For instance, pregnancy-related topics such as hypertension and diabetes during pregnancy and peripartum cardiomyopathy are topics on which internists may be questioned on the ABIM national board examination, ⁹ yet pregnancy-related topics were not identified by our expert panel as important for internists to provide high-quality clinical care. This may be reflective of the increasing specialization of medicine, in which these women are most likely to be seen by specialists in high-risk maternal care. One survey of internal medicine residents found that 95% of residents reported rarely, if ever, providing clinical care related to pregnancy. ¹⁵ Of note, there is no specific mention of pregnancy-related health topics in the ABIM competencies, unless they are defined under “family planning and reproductive health” or “common gynecological disorders.” ⁸ This disconnect poses a challenge both for residents preparing for the examination and for test question developers.

For many of the topics, such as osteoporosis, sexually transmitted infections, and vaginitis, our experts felt that internists should be able to both screen for and manage these medical conditions. However, for other topics, our expert panel placed more emphasis on diagnosis rather than management. For example, the diagnosis of PCOS, abnormal uterine bleeding, and sexual dysfunction was all rated as important by our expert panel, but the management of these conditions did not make our consensus list. The emphasis on diagnosis over management by our expert panel was not surprising to us. It is not expected that every categorical internal medicine resident should graduate as an expert in the comprehensive primary care of women. However, they should all be able to recognize and diagnose common women's health concerns and be able to make appropriate referrals. For example, our expert panel felt that it is important for all graduating residents to be able to screen for breast cancer and assess breast cancer risk, but prescribing chemoprophylaxis was not seen as being important. This is consistent with current literature, which shows that very few internists have ever recommended or prescribed chemoprophylaxis. ¹⁶ In addition, it is important to note that just because a topic was rated as less important does not mean that it is actually unimportant, but rather reflects the fact that curricular time is limited and topics need to be prioritized for residency education in a diverse group of residents pursuing both subspecialty and primary care careers.

Finally, there is a growing understanding that sex and gender impact all aspects of medicine. We expect internists to be aware of sex and gender differences in the conditions that fall within standard internal medicine domains, such as cardiovascular disease. Our project sought to establish the parameters specific to women's health within internal medicine so as to better organize women's health curricula. Therefore, we did not include topics like cardiovascular disease in our topic list. In addition, we acknowledge that gender-specific care is not just limited to women's health, but encompasses men's health and LGBT health as well. Future work will be needed to help shape the content of residency education within these domains.

There are some limitations to our study. Our results only represent the opinions of our experts, all of which come from large academic centers and may not be reflective of the opinions of other internists. Approximately, one-third of the participants in our Delphi panel spent less than 25% of their time in direct clinical settings, which may give them a different perspective regarding residency education compared to physicians who spend more time doing direct clinical work. These individuals, however, were senior physicians and their current clinical effort does not necessarily represent their lifetime experience; thus, we felt that their views were particularly valuable to the development of residency guidelines. Also, our study only represents one time point. Opinions of the importance of clinical topics are likely to change over time as the practice of medicine changes. Finally, the Delphi technique favors consensus, which may overshadow minority opinions.

Given the growing awareness that sex and gender impact all domains of care, internal medicine residents need to be well trained in women's healthcare. Our list will allow internal medicine residencies to ensure that their graduates live up to the ACGME requirements of being able to provide “gender-specific care” by incorporating the 35 consensus topics into their residency curricula.