An Expanded Primary Care–Based Women’s Health Clinic to Improve Resident Education and Patient Care in Resident Continuity Clinic

Abstract

Background

Internal medicine (IM) residents are underprepared in women’s health. Lack of properly trained faculty and clinic culture limits the ability to provide bedside teaching.

Aim

Assess the impact of a primary care–based, women’s clinic on residents’ quality of care for females.

Setting

Large academic, urban primary care clinic with resident and faculty practices

Participants

PGY-2 IM and Med-Peds (MP) residents

Program Description

A weekly half-day, women’s clinic to provide expanded women’s healthcare to primary care group patients. Residents rotate through the clinic to receive bedside teaching.

Program Evaluation

Chart review was performed for a representative sample of reproductive-aged women seen in primary care before and after the establishment of the women’s clinic. A total of 666 charts were reviewed (314 pre, 352 post). Improvement was seen in residents’ rate of sexual histories (54% vs 75%, p< 0.01) with a significant decrease in women not asked about contraception (15% vs 3%, p<0.01). Overall there was a decrease in gynecology referrals (18 to 11%, p=0.02).

Discussion

After implementing the women’s health clinic, more women were asked about sexual health needs, and fewer were referred to gynecology, suggesting increased women’s healthcare provided by residents.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11606-022-07512-6.

INTRODUCTION

Internal medicine (IM) providers care for the most medically complicated reproductive-aged women yet residents often feel underprepared in gender-based care for this population.^1,2 Residents report low levels of confidence and knowledge in many core women’s health care topics including contraception, menopause, and pelvic and breast exams.^3–7 Residents report limited clinical exposure as a reason for this practice gap.⁵ This is also seen at the faculty level and leaves many training programs without the critical mass of faculty to provide in-clinic teaching and direct observation.^8–10 To bridge these gaps, many internal medicine residencies develop a referral-heavy culture that further limits gender-based teaching opportunities.

With updated cervical cancer screening guidelines, pap smears are required less often, limiting opportunities for in-clinic pelvic exam teaching.^11,12 Additionally, the American Board of Internal Medicine (ABIM) has removed the pap smear requirement and moved towards a more individualized procedure experience for residents.¹³ This change removes the incentive for programs to ensure residents receive direct observation of pelvic exams and places the burden on individual residents to seek out these educational experiences. This presents a challenge for residents to perform enough observed exams to feel confident for their future practice.

In a survey of our graduating residents from 2011 to 2016, we found 62% were uncomfortable with collecting a pap smear. Similar findings were seen in the area of contraception with 37% of residents reporting they had not received adequate training.¹⁹ To help bridge this educational gap, the expanded, primary care women’s clinic was established.

The use of primary care–based referral clinics for these less commonly encountered topics is an appropriate alternative to provide residents with in-clinic, gender-based care teaching.^14–16 This model has been widely used in the Veterans Healthcare Administration (VA) to overcome faculty development and patient demographic barriers. VA patients who were seen in a women’s clinic had higher levels of satisfaction than those seen in the traditional primary care clinic.^15,17,18 The use of gender-based clinics provides an opportunity to focus on educational resources to improve resident training.

However, evaluation of this model has been limited to patients’ and providers’ self-reported data. Prior studies have been focused on the patient and resident experience within the women’s clinic and have not evaluated the effects of these clinics on the larger primary care practice as a whole. Our study is unique in that it utilizes patient-level, clinical data and evaluates the effects of the women’s health clinic on the quality of care of women throughout the primary care practice. Through the lenses of sexual health and contraception, we investigated the effects of the women’s clinic on the care of reproductive-aged women in the resident continuity clinics.

SETTING AND PARTICIPANTS

The women’s health clinic is part of a large, academic primary care practice that is available to resident and faculty patients (see figure 2). Patients were referred to the clinic by their PCP or could self-refer. Post-graduate year 2 (PGY-2) IM and internal medicine-pediatrics (MP) residents (38 total residents per year) spend a half-day in this clinic at least once during their residency and see patients with a faculty preceptor.

PROGRAM DESCRIPTION

The women’s health clinic is a consulting clinic for providers in the primary care group. The patients are seen within the primary care clinic. This replaced a pre-existing “pap clinic” model that had varying different faculty involvement but without a consistent preceptor. The revised model included a faculty preceptor who completed a women’s health-focused, general medicine fellowship. This clinic provides expanded services including cancer screenings, Long-Acting Reversible Contraception (LARC) placement, and contraception management, menopause treatment, breast and vaginal health, STI screening, and prevention, and eating disorder care, etc. (see figure 1). On average, 4–6 pap smears were performed per weekly session and 3–4 LARC/contraception visits per month.

All patients are seen by a resident and the supervising faculty preceptor. Residents are provided with direct observation of all exams and procedures. This ensures that all residents receive feedback on pelvic examination skills. Visits were billed by the faculty preceptor. If the faculty preceptor was not available, the residents did not attend the clinic that week and the patients were redirected to the nurse practitioner clinic.

PROGRAM EVALUATION

Sample Size and Chart Review

A retrospective chart review was performed for female patients of reproductive age (15–46 years old) seen in the primary care practice before and after the addition of the expanded women’s clinic. The expanded clinic began seeing patients in September 2017. Visits from July 1, 2015–June 30, 2016, were considered pre-intervention, and those from January 3, 2018–January 4, 2019, were post-intervention (see figure 1). The first visit to the primary care clinic within these time frames was reviewed for each patient. Visits to the women’s clinic were not included. The visits were identified using the data provided by the EPIC Analytics Request System and the Slicer Dicer tool.

A sample size calculation was performed to identify the size necessary to produce a representative population with a 95% confidence level and 5% error. Charts were randomly selected to be included as representative samples for the pre- and post-intervention groups. Since a major addition to the Expanded Women’s Clinic was the addition of LARC placement, we chose contraception use as a primary endpoint. Secondary endpoints of sexual history recording and referrals to gynecology were selected to measure a contraception counseling pre-skill and teaching-case retention.

As contraception was a primary outcome, visits were excluded if the patient had a hysterectomy, was currently pregnant, was seen in urgent care, managing infertility, not sexually active or post-menopausal. Women’s health clinic visits were not included since these patients were referred from the primary care group and this would lead to significant overlap and double counting of patients. Also, the goal of this study was to assess the effects of a women’s clinic on the quality of care for female patients in the resident continuity clinics outside of the women’s clinic.

The note from the first visit within the pre- or post-intervention time frame was reviewed for record of sexual history, age, current most effective method of contraception, and referral to gynecology. The note and embedded social history section in EPIC were both reviewed on the day of the visit for a record of sexual history. Patients who did not have a recorded method of contraception were listed as “none” if the option of “none” was entered in the embedded social history section or a discussion of the patient declining contraception was recorded in the note. If there was no record of a contraception discussion in the note or embedded social history, this was recorded as “not asked.” For a patient with multiple contraception methods, the most effective options were listed based on the Centers for Disease Control and Prevention tired decisions tool.¹⁷ LARC was defined as implant and Intrauterine Device (IUD). Surgical contraception was tubal ligation or partner vasectomy (Fig. 2).

Statistical Analysis

Using STATA SE 17, we created descriptive statistics to summarize patient outcomes in the pre- and post-intervention groups. An unpaired t-test was used for comparing age groups. Using a chi-squared test, we examined differences between the pre- and post-intervention groups in regards to rates of sexual history recording, contraception use, and referrals to gynecology. The project was formally determined to be of quality improvement by the Institutional Review Board, per institutional policy at the University of Chicago.

RESULTS

3720 pre-intervention and 4084 post-intervention visits were identified. A total of 666 charts were reviewed with 314 pre and 352 post-intervention. Demographic data was assembled (Table 1). The groups were similar in regard to mean age, age group distribution, and PCP type (resident or faculty). During the intervention period from September 2017 to January 2019, 65 women’s clinic sessions were completed with 45 residents attending (76 total eligible residents, 59% participation) (6 residents attending multiple women’s health clinics).

Table 1.

Demographics of Pre- and Post-Intervention Groups

Variable	Pre (n=314)	Post (n=352)	P value
Age, mean (SD)	33.6 (8.3)	33.9 (8.2)	0.6
Age, n (%)
15–24 years	55 (17.5)	61 (17.3)	0.95
25–34 years	95 (30.3)	114 (32.4)	0.35
35–46 years	164 (52.2)	177 (50.3)	0.62
Type of PCP, n (%)			0.87
Faculty	173 (55.1)	172 (48.9)	0.12
Resident	141 (44.9)	180 (51.1)	0.12

Results were stratified by provider type. In the residents’ patients, improvement was seen in the rate of recorded sexual histories (54% vs 75%, p< 0.01). While there was a significant decrease in the proportion of patients not asked about contraception (15 to 3%, p<0.01), there was a significant increase in those not using contraception (“none” pre 21% vs post 27%, p<0.01). Rates of any contraception use and LARC increased but did not reach statistical significance (5.5% increase, p=0.3, and 5.6% increase, p=0.28, respectively) (see Table 2).

Table 2.

Contraception Use and Type, Pre- vs Post-Intervention, Resident Patients Only

Type of contraception	Pre, n (%) (total pt 141)	Post, n (%) (total pt 179)	P value
Overall	90 (63.8%)	124 (69.3%)	0.30
LARC	15 (10.6%)	29 (16.2%)	0.28
Pill/patch/ring	32 (22.6%)	43(24%)	0.58
Shot	2 (1.4%)	6 (3.3%)	0.42
Surgical	15 (10.6%)	16 (8.9%)	0.44
Condom	26 (18.4%)	30 (16.7%)	0.44
None	30 (21.2%)	49 (27.3%)	<0.01
Not asked	21 (14.9%)	6 (3.3%)	<0.01

As this clinic was available to all patients, the data was evaluated for the practice as a whole. Overall, there were similar findings with sexual history recordings and contracepting us to the resident-only data. A significant decrease was seen in referrals to gynecology with the addition of the women’s care clinic (18 to 11%, p=0.02).

DISCUSSION

This quality improvement project suggests that the addition of a primary care, women’s health clinic improved the frequency of sexual history taking and discussing contraception. Prior studies showed that residents who regularly take a sexual history are more likely to also provide contraceptive counseling.^9,20 Our study supports the concept that sexual history recording is an essential pre-skill in proving reproductive health care to patients.

We were able to demonstrate a reduction in gynecology referrals suggesting that more routine office-based gynecology is being provided within primary care. Residents frequently report a lack of clinical exposure to key gender-based care topics as a barrier to their education.^5,9,20 By providing an internal referral option, these key teaching moments remain in primary care. This also suggests a change in the scope of care and clinical culture provided in the primary care group away from low-value referrals and towards more comprehensive gender-based care.

More women were asked about contraception after the addition of the expanded women’s clinic as seen in the decrease of “not asked” visits. Though we were not able to measure contraception counseling, this trend does suggest that since the addition of our expanded women’s care clinic patients in the primary care group are receiving contraception counseling. By asking more women about contraception, we identified more women who were not using contraception in the post-group. A portion of these patients likely are ambivalent about their family planning which depending on their goals not using contraception may be an appropriate option.²¹ This study did not evaluate why the women in the “none” category were not using contraception which may be of interest in future iterations of this work.

While an increase in contraception and LARC use was seen, it did not reach statistical significance. The study was likely underpowered to reach statistical significance for these variables. Since the addition of LARC services to the women’s clinic in 2017, 46 IUDs and 34 implants had been placed or removed. This is a similar rate of LARC visits as reported by other primary care–based LARC clinics.²² This suggests that improved access to these LARC services has contributed to the increase of LARC utilization by our patients.

There are several limitations to this study. While we were able to show improvement in key clinical endpoints, we are not able to show a direct causal relationship between this and the addition of the women’s care clinic. Other factors such as the addition of new faculty to the primary care group and revisions to the resident ambulatory curriculum may have contributed to the improvement in the quality of care for female patients. Next, while patients seen on the urgent care schedule were excluded, those who were presenting to their PCP with an urgent chief complaint were included. This may result in visits during which taking a sexual history or providing contraception counseling were not appropriate to be included in our chart review. Also, this study focused on cisgender women, and like many in the realm of women’s health, we are working towards a more gender-inclusive practice and resident education model.

In summary, this study demonstrates improvement in the quality of women’s healthcare provided in our resident continuity clinics using patient-level clinical data. After the addition of an expanded primary care women’s health clinic, there was a significant improvement in the rate of recorded sexual histories and women asked about contraception with a decrease in gynecology referrals.

Declarations

Conflict of interest

The authors declare that they do not have a conflict of interest.