Translate this page into:
Implementation of a reproductive goals SmartForm to improve contraception and preconception counseling in an internal medicine resident clinic
-
Received: ,
Accepted: ,
How to cite this article: Dickens RR, Decicco E, Cardin J, Wainstein HM. Implementation of a reproductive goals SmartForm to improve contraception and preconception counseling in an internal medicine resident clinic. J Reprod Healthc Med. 2026;7:13. doi: 10.25259/JRHM_5_2026
Abstract
Objectives:
While internal medicine residents agree on the importance of reproductive counseling, time and knowledge constraints often inhibit counseling. One Key Question (OKQ) is a pregnancy intention screening tool that can facilitate counseling. We implemented a screening tool based on OKQ with the primary aim of increasing residents’ reproductive counseling; secondary outcomes included rates of contraception prescriptions and gynecology referrals.
Materials and Methods:
The quality improvement study was conducted in an internal medicine resident clinic within a safety-net health system. We included women ages 18–49 with reproductive capability. Our primary intervention was an electronic medical record (EMR) tool based on OKQ (Reproductive Goals SmartForm) and a prescribing tool (Contraception SmartSet). We conducted surveys to assess the usability of the tools. Pre- and post-intervention, we surveyed residents on their counseling habits and compared the results via Chi-square tests.
Results:
The SmartForm was completed on 27% of eligible patients. Compared to the prior 9 months, there was an increase in contraception and emergency contraception prescriptions and gynecology referrals. There was a statistically significant change in residents’ subjectively reporting routinely asking their female clinic patients of reproductive age about their reproductive goals or desire to get pregnant (p = 0.001), considering their reproductive status when starting or continuing medication (p = 0.047), and counseling on contraception (p = 0.023).
Conclusion:
To our knowledge, this is one of the first studies on implementing OKQ in an internal medicine resident clinic. The improvement in reported counseling practices suggests that a clinic-based EMR screening tool can enhance resident reproductive counseling.
Keywords
Contraception
Quality improvement
Screening tool
Women’s health
INTRODUCTION
Only 14% of outpatient visits in the United States for women of reproductive age include preconception or contraceptive counseling.[1] While over 95% of internal medicine faculty and residents agree that reproductive counseling is important, as few as 17–25% of providers report routinely providing reproductive counseling in outpatient visits, citing inadequate time and knowledge as barriers to counseling.[2,3] Internists care for a medically complex patient population who are at risk for adverse events from unplanned pregnancy and are often prescribed potentially teratogenic medications. Yet competing medical problems can cause reproductive goals counseling to be deferred to other providers.
Several interventions have been successfully implemented to address resident time and knowledge barriers, resulting in improvements in confidence, knowledge, and reproductive counseling rates.[4-8] One such intervention is employing a standardized screening and counseling tool.[9] One key question (OKQ) is a clinical screening tool used to assess the desire to become pregnant within the following year.[10] OKQ is simple to implement with little impact on workflow, is associated with high patient satisfaction, and has been shown to increase contraceptive counseling (52% vs. 76%, p = 0.040) and long-acting reversible contraceptive (LARC) counseling (10% vs. 32%, p = 0.035).[11-13]
OKQ has been implemented in various primary care clinics, including the Veterans Health Administration, federally qualified health centers, family medicine resident clinics, and OBGYN clinics; however, its implementation in an internal medicine resident clinic has not been widely documented.[11-15] We discuss the results of implementing a standardized, electronic medical record (EMR) screening tool based on OKQ to facilitate reproductive counseling in an internal medicine resident clinic.
MATERIALS AND METHODS
Methods
Our quality improvement study consisted of three phases: Pre-intervention, intervention, and post-intervention. Pre-intervention, we conducted a survey to evaluate residents’ baseline reproductive counseling practices and barriers to counseling from December 2023 to January 2024. During the 9-month intervention (April 01, 2024, to December 31, 2024), we implemented a screening tool (Reproductive Goals SmartForm) and a prescribing tool (Contraception SmartSet) and conducted a survey to assess the usability of the tools. Post-intervention, we conducted a second usability survey and a second survey on residents’ counseling practices. At the start of the intervention, residents were asked to complete a self-guided educational module on how to use the smart tools, which included a refresher on preconception and contraception counseling. Mid-intervention, residents received a 1-h didactic session on contraception counseling, which included a reminder about this project. Reminders were also sent out via email and posted in the clinic.
The Reproductive Goals SmartForm we developed is based on OKQ: Would you like to become pregnant in the next year? The SmartForm directs residents to the Contraception SmartSet to simplify prescribing contraception and placing gynecology referrals. A non-interruptive best practice alert (BPA) reminds residents to complete the SmartForm during clinic visits for eligible patients at least once per year.
Our SmartForm first confirms that a woman has reproductive capability, and then asks the woman’s desire to become pregnant in the next 12 months. Based on the patient’s answer of yes, no, or unsure, additional prompts cascade to guide counseling on preconception, contraception, or a combination of both. If patients desired pregnancy in the next year, residents were reminded to discuss health and medication optimization, prenatal vitamin or folate supplementation, and short-term contraception options if applicable. If patients did not desire pregnancy in the next year, residents were prompted to ask about their current contraception method, offer a discussion of contraception options, and offer a prescription for emergency contraception. If patients were unsure of their pregnancy goals, residents were reminded to complete a combination of both: Discussing preconception health optimization and offering contraception, including emergency contraception.
The Contraception SmartSet contains 4 sub-sections: Combined estrogen/progesterone, progesterone only, nonhormonal, and emergency contraception. The combined sub-section includes a single mid- and low-dose estrogen pill option. The SmartSet includes gynecology referrals for LARC placement and tubal ligation, with a reminder to have a short-term contraception plan while the patient is waiting to see gynecology. The SmartSet also includes patient instructions in English and Spanish.
Our primary outcome was to increase completion of the Reproductive Goals SmartForm from 0 to 50% for eligible patients seen in the clinic during the 9-months period of the intervention. Secondary outcomes included (1) improving resident reproductive counseling practices and reducing barriers to counseling based on pre-and post-survey responses and (2) increasing rates of residents ordering contraception, emergency contraception, and gynecology referrals pre-and post-intervention. We defined pre-intervention as the 9 months before the intervention (July 01, 2023–March 31, 2024). This 9-months period was the comparison for contraception prescriptions and gynecology referrals.
To gain feedback on the smart tools, surveys on the usability of the SmartForm and SmartSet were sent out from August to September 2024 (mid-intervention) and from January to February 2025 (post-intervention). Finally, a post-survey was sent out from January to February 2025 to re-evaluate residents’ counseling practices and barriers to counseling. We compared the pre- and post-survey results with Chi-square tests, as the survey results were anonymous.
Setting and participants
Our intervention was conducted in an internal medicine resident clinic. The clinic is located within an integrated, safety-net health system that utilizes Epic as its EMR software. The clinic is a primary care teaching clinic with 98 residents. The project was approved as an exempt quality improvement project by both Institutional Review Boards associated with the clinic.
Since we are a resident clinic, there was resident turnover at the end of the academic year. Twenty-seven Post-Graduate Year 3 (PGY3) residents graduated in June 2024, 7 PGY1 and PGY2 residents left the clinic, and 34 PGY1 residents matriculated into the program in July 2024. We sent pre- and post-intervention surveys to all current residents, so the resident groups surveyed were not completely comprised of the same residents. Sixty-four residents were sent both surveys. While there were differences between the two groups surveyed, we continued to screen from the same clinic.
Our screening tool targeted women ages 18–49 with reproductive capability, meaning that they had not yet undergone menopause or permanent sterilization (bilateral tubal ligation or hysterectomy).
RESULTS
Pre-intervention survey results
Fifty-four residents out of 98 (43% PGY1, 57% PGY2 or 3, 61% female) completed the pre-survey [Table 1]. Only 17% (9) of the residents reported often (50–79%) or routinely (>80%) asking their female clinic patients of reproductive age about their reproductive goals or desire to get pregnant. Forty-six percent (25) of residents often or routinely consider their female clinic patients’ reproductive status when starting or continuing a medication. Thirty-seven percent (20) of residents reported often or routinely counseling their female clinic patients of reproductive age on contraception; only 2% (1) of residents often or routinely counsel on emergency contraception. The predominant barriers to counseling reported by residents were limited time and uncertainty about available contraception options and LARC referrals [Table 2].
| Reproductive goals counseling | Pre-survey n(%) | Post-survey n(%) | p-value |
|---|---|---|---|
| Number of residents | 54 | 54 | |
| Sex | |||
| Female | 33 (61.1) | 30 (55.6) | 0.558 |
| Male | 21 (38.9) | 24 (44.4) | |
| PGY level | |||
| PGY1 | 23 (42.6) | 20 (37.0) | 0.792 |
| PGY2 | 17 (31.5) | 17 (31.5) | |
| PGY3 | 14 (25.9) | 17 (31.5) | |
| I ask my female clinic patients of reproductive age (18–49 years old) about their reproductive goals or desire to get pregnant | |||
| Rarely or never (0–20%) | 21 (38.9) | 8 (14.8) | 0.001 |
| Sometimes (21–49%) | 24 (44.4) | 19 (35.2) | |
| Often (50–79%) | 7 (13.0) | 18 (33.3) | |
| Routinely (>80%) | 2 (3.7) | 9 (16.7) | |
| I consider my female clinic patients’ reproductive status when starting or continuing a medication | |||
| Rarely or never (0–20%) | 12 (22.2) | 6 (11.1) | 0.047 |
| Sometimes (21–49%) | 22 (40.7) | 23 (42.6) | |
| Often (50–79%) | 17 (31.5) | 13 (24.1) | |
| Routinely (>80%) | 3 (5.6) | 12 (22.2) | |
| I counsel my female clinic patients of reproductive age on contraception | |||
| Rarely or never (0–20%) | 17 (31.5) | 6 (11.1) | 0.023 |
| Sometimes (21–49%) | 17 (31.5) | 24 (44.4) | |
| Often (50–79%) | 17 (31.5) | 15 (27.8) | |
| Routinely (>80%) | 3 (5.6) | 9 (16.7) | |
| I counsel my female clinic patients of reproductive age on emergency contraception | |||
| Rarely or never (0–20%) | 40 (74.1) | 31 (57.4) | 0.129 |
| Sometimes (21–49%) | 13 (24.1) | 17 (31.5) | |
| Often (50–79%) | 0 (0.0) | 3 (5.6) | |
| Routinely (>80%) | 1 (1.8) | 3 (5.6) | |
A p value significant level p< 0.05. PGY: Postgraduate year
| Barriers to counseling | Pre-survey n(%) | Post-survey n(%) | p-value |
|---|---|---|---|
| Number of residents | 54 | 54 | |
| Barriers | |||
| Uncertainty about available contraceptive options and/or LARC referrals | 77 | 76 | 0.315 |
| Limited time | 41 | 41 | |
| Inadequate knowledge or training | 29 | 16 | |
| Sensitivity of the subject, personal discomfort, or cultural and/or language barriers | 29 | 22 | |
| Implicit bias, outside the scope of practice, few patients of reproductive age, or discussion never occurred | 21 | 12 | |
| Total barriers selecteda | 197 | 167 |
Intervention results
During the intervention, 311 women (mean age 36 ± 8 years, 64.0% uninsured, 50.8% Hispanic/Latino, and 69.5% with English as their primary language) were eligible for screening [Table 3]. During our post-analysis, we excluded patients with a prior history of bilateral tubal ligation; these women were not eligible for screening as they were not capable of becoming pregnant but were initially included due to a BPA error. In total, the SmartForm was completed on 27% of eligible patients (83 women) [Table 3]. 83% (69) of women did not desire pregnancy in the next 12 months. Seven percent (6) were unsure, and 10% (8) desired pregnancy in the next year. The most common contraceptive methods before the visit were using condoms (19%) or practicing abstinence (13%). Compared to pre-intervention, there was an increase in contraception prescriptions (16 vs. 21), emergency contraception prescriptions (3 vs. 11), tubal ligation referrals (0 vs. 2), and gynecology referrals during the study period (104 vs. 113). Due to limitations of our EMR, we were unable to determine if the gynecology referral was for LARC placement.
| Patient characteristics | Patients eligible n (%) | Patients sampled n (%) |
|---|---|---|
| Total participants, n | 311 | 83 |
| Race/Ethnicity | ||
| White, non-Hispanic | 47 (15.1) | 18 (21.7) |
| Black, non-Hispanic | 87 (28.0) | 19 (22.9) |
| Hispanic/Latino | 158 (50.8) | 41 (49.4) |
| Asian or unknown | 19 (6.1) | 5 (6.0) |
| Language | ||
| Spanish language | 90 (28.9) | 21 (25.3) |
| English language | 216 (69.5) | 62 (74.7) |
| Other language* | 5 (1.6) | 0 (0.0) |
| Insurance type | ||
| Uninsured | 199 (64.0) | 49 (59.0) |
| Medicaid | 66 (21.2) | 20 (24.1) |
| Medicare | 15 (4.8) | 4 (4.8) |
| Commercial | 31 (10.0) | 10 (12.1) |
| Urbanization | ||
| RUCA 1† | 302 (97.1) | 76 (91.6) |
| RUCA other‡§‖ | 9 (2.9) | 7 (8.4) |
| Marital status | ||
| Single | 171 (55.0) | 46 (55.4) |
| Married | 96 (30.9) | 29 (35.0) |
| Other¶ | 44 (14.1) | 8 (9.6) |
| Co-morbidities | ||
| Hypertension | 89 (28.6) | 20 (24.1) |
| Diabetic | 58 (18.6) | 12 (14.5) |
| BMI | ||
| Underweight or healthy weight (<25) | 65 (20.9) | 12 (14.5) |
| Overweight (>25–30) | 93 (29.9) | 23 (27.7) |
| Obese>30 | 153 (49.2) | 48 (57.8) |
Usability survey results
20% of residents (20 out of 98) completed the initial usability survey. Twelve (60%) had never completed the SmartForm, citing barriers of limited time, more pressing issues, few eligible patients, and forgetting or not realizing it existed. Of the 8 residents who had completed the SmartForm at least once, 7 (88%) were satisfied with the ease of completing it, 6 (75%) were satisfied with the time it took to complete, and 5 (63%) felt they could easily integrate the SmartForm into clinic visits. Only 2 of the 8 residents had utilized the Contraception SmartSet, with the others stating they did not need it or were unaware of its existence.
Post-intervention, 46% of residents completed the second usability survey (45 of 98). Of the 26 residents who had completed the SmartForm at least once, 25 (96%) were satisfied with the ease of completing it, 25 (96%) were satisfied with the time it took to complete, and 21 (81%) felt they could easily integrate the SmartForm into clinic visits [Table 4]. Eleven of the 26 residents (42%) had utilized the Contraception SmartSet. Of these, 9 (82%) were satisfied with the ease of use of the SmartSet, and 10 (91%) were satisfied with the support information they had when prescribing contraception using the SmartSet.
| Usability | n, (%) |
|---|---|
| Residents who completed the usability survey | 45 |
| Residents who used the SmartForm | 26 |
| Residents who used the SmartSet | 11 |
| How many times have you completed the Reproductive Goals SmartForm? | |
| 1–2 | 20 (76.9) |
| 3 or more | 6 (23.1) |
| I am satisfied with the ease of completing the SmartForm | |
| Strongly agree | 9 (34.6) |
| Agree | 16 (61.5) |
| Neutral | 1 (3.9) |
| Disagree or strongly disagree | 0 |
| I am satisfied with the amount of time it took to complete the SmartForm | |
| Strongly agree | 11 (42.3) |
| Agree | 14 (53.8) |
| Neutral | 1 (3.9) |
| Disagree or strongly disagree | 0 |
| I could easily integrate the SmartForm into my clinic visits | |
| Strongly agree | 9 (34.6) |
| Agree | 12 (46.1) |
| Neutral | 4 (15.4) |
| Disagree | 1 (3.9) |
| Strongly disagree | 0 |
| Did you utilize the contraception SmartSet for orders and/or patient instructions? | |
| Yes | 11 (42.3) |
| No | 15 (57.7) |
| I am satisfied with the ease of using the SmartSet | |
| Strongly agree | 4 (36.4) |
| Agree | 5 (45.5) |
| Neutral | 2 (18.2) |
| Disagree or strongly disagree | 0 |
| I am satisfied with the support information I had when prescribing contraception using the SmartSet | |
| Strongly agree | 4 (36.4) |
| Agree | 6 (54.5) |
| Neutral | 1 (9.1) |
| Disagree or strongly disagree | 0 |
Post-intervention survey results
Post-intervention, 54 residents out of 98 (37% PGY1, 63% PGY2 or 3, 56% female) completed the survey. Now, 51% (27) of the residents reported often or routinely asking their female clinic patients of reproductive age about their reproductive goals or desire to get pregnant (p = 0.001) [Table 1]. Forty-six percent (25) of residents often or routinely consider their female clinic patients’ reproductive status when starting or continuing a medication (p = 0.047). Forty-five percent (24) of residents reported often or routinely counseling their female clinic patients of reproductive age on contraception (p = 0.023); only 11% (6) of residents often or routinely counsel on emergency contraception (p = 0.129). There was no statistically significant difference in reported barriers to counseling (p = 0.315) [Table 2].
DISCUSSION
To our knowledge, this is one of the first quality improvement studies to implement OKQ in an internal medicine resident clinic. We achieved 27% completion of the SmartForm for eligible patients seen in the clinic during the 9-months intervention. Our utilization rate is similar to that of another OKQ study over 3 years, which reported 39% utilization.[16] Our relatively low utilization could be due to a number of factors, including the high number of residents and attendings in our clinic, making widespread adoption challenging; resident turnover during the study; and competing medical problems to address. Although uptake of the tools was limited, the tools were well received by the subgroup of residents who utilized them.
There was a statistically significant change in residents reporting routinely asking their female clinic patients of reproductive age about their reproductive goals or desire to get pregnant, considering their reproductive status when starting or continuing medication, and counseling on contraception. This suggests that clinic-based implementation of a screening EMR tool can change reproductive counseling practices, as prior research has shown.[13,17] However, we were unable to compare residents directly pre- and post-intervention, and we acknowledge that the pre- and post-survey cohorts were partially independent groups due to resident turnover between the surveys.
The low emergency contraception counseling we saw may be in part due to low awareness, misinformation, and the current politicization of emergency contraception.[18] We did not see a statistically significant change in reported barriers to counseling, possibly due to small sample size, resident turnover, and limited utilization of the tools. We did demonstrate an increase in contraceptive prescriptions, though it was not statistically significant, which is similar to other implementations of OKQ in non-resident clinics.[13]
We hypothesize that the relatively low SmartForm completion rate and small increase in contraception prescriptions and gynecology referrals in contrast to residents’ significant self-reported improvements in counseling habits, may be the result of recall bias where the residents overconfidently rated their counseling. This discrepancy in perceived improvement versus objective data could in part reflect increased comfort in counseling due to education and some utilization of the smart tools, as well as more informal conversations occurring about reproductive goals without SmartForm completion (as this was a process measure rather than proof of behavior change).
The implementation of OKQ in a safety-net internal medicine resident clinic is novel and material to underserved women’s reproductive health. In the United States, 45% of pregnancies were unintended and disproportionately affect women from negative social determinant backgrounds, including those below the federal poverty line.[19,20] Women who have low socioeconomic status and are uninsured are more likely to receive reproductive health services from primary care providers, versus dedicated women’s health specialists.[21] Internal medicine residents care for medically complex patients who are at risk for adverse events from unplanned pregnancy, yet competing medical problems can cause reproductive goals counseling to be deferred. Thus, our Reproductive Goal SmartForm shows promise, as it helped to mitigate barriers to reproductive counseling for this vulnerable population.
Our study has various limitations. Despite the large size of our clinic, the number of residents who completed the usability surveys and had utilized the tools was relatively small. Changes in counseling practices must be interpreted cautiously due to resident turnover and the inability to track residents over time, and do not prove causality. Moreover, the study was conducted in a single safety-net clinic with Epic as the EMR and may not be broadly applicable to other settings and patient populations. We also did not assess patient satisfaction and recommend that future researchers include patient-centered outcomes. Finally, there may be a selection bias, as the residents determined who to screen. Thus, the residents could have screened patients who were more straightforward due to lower medical complexity, already utilizing contraception, or not currently engaging in sexual activity.
Further research is needed to investigate how to overcome barriers to implementing and sustaining this type of intervention in a high-volume resident clinic. The clinic stakeholders’ goals must align with OKQ, and additional approaches to provider engagement should be considered, such as an interruptive BPA, meaning it requires provider interaction.[11] Implementing similar tools in other primary care settings and specialty clinics can also be explored.
CONCLUSION
In conclusion, the Reproductive Goals SmartForm shows promise in improving the reproductive counseling habits of internal medicine residents. Future quality improvement projects should consider implementing tools based on OKQ to increase internal medicine residents’ reproductive counseling and help patients plan for pregnancy.
Ethics approval:
The research/study approved by the Institutional Review Board at University of Texas Southwestern and Parkland, number Y2-23-0366, dated 20th October 2023.
Declaration of patient consent:
Patient’s consent is not required as there are no patients in this study.
Conflicts of interest:
There are no conflicts of interest.
Use of artificial intelligence (AI)-assisted technology for manuscript preparation:
The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.
Financial support and sponsorship: Nil.
References
- Trends in contraceptive and preconception care in United States ambulatory practices. Fam Med. 2015;47:264-71.
- [Google Scholar]
- Contraceptive counseling by general internal medicine faculty and residents. J Womens Health (Larchmt). 2014;23:707-13.
- [CrossRef] [PubMed] [Google Scholar]
- Provision of contraceptive counseling by internal medicine residents. J Womens Health (Larchmt). 2009;18:127-31.
- [CrossRef] [PubMed] [Google Scholar]
- Using video modules and simulation learning to improve IUD counseling among internal medicine residents-A randomized controlled educational trial. J Gen Intern Med. 2021;36:1446-7.
- [CrossRef] [PubMed] [Google Scholar]
- Case-based curriculum with integrated smartphone applications improves internal medicine resident knowledge of contraceptive care. Adv Med Educ Pract. 2019;10:971-7.
- [CrossRef] [PubMed] [Google Scholar]
- A new contraception curriculum in the internal medicine residency primary care clinic. Mil Med. 2019;184:e61-4.
- [CrossRef] [PubMed] [Google Scholar]
- An expanded primary care-based women's health clinic to improve resident education and patient care in resident continuity clinic. J Gen Intern Med. 2022;37:2314-7.
- [CrossRef] [PubMed] [Google Scholar]
- Standardized contraceptive handout facilitates contraceptive counseling. Fam Med. 2018;50:146-8.
- [CrossRef] [PubMed] [Google Scholar]
- Preconception, interconception, and reproductive health screening tools: A systematic review. Health Serv Res. 2023;58:458-88.
- [CrossRef] [PubMed] [Google Scholar]
- One key question®: Preventive reproductive health is part of high quality primary care. Contraception. 2013;88:3-6.
- [CrossRef] [PubMed] [Google Scholar]
- Facilitators of and barriers to successful implementation of the one key question® pregnancy intention screening tool. Womens Health Rep (New Rochelle). 2022;3:326-34.
- [CrossRef] [PubMed] [Google Scholar]
- Effects of clinic-level implementation of one key question® on reproductive health counseling and patient satisfaction. Contraception. 2021;103:6-12.
- [CrossRef] [PubMed] [Google Scholar]
- Increase in contraceptive counseling by primary care clinicians after implementation of one key question® at an urban community health center. Matern Child Health J. 2019;23:996-1002.
- [CrossRef] [PubMed] [Google Scholar]
- Pregnancy risk screening and counseling for women veterans: Piloting the one key question in the veterans healthcare administration. South Med J. 2021;114:150-5.
- [CrossRef] [PubMed] [Google Scholar]
- Routine screening for pregnancy intention to address unmet reproductive health needs in two urban federally qualified health centers. J Health Care Poor Underserved. 2017;28:1477-86.
- [CrossRef] [PubMed] [Google Scholar]
- Clinic factors associated with utilization of a pregnancy-intention screening tool in community health centers. Contraception. 2021;103:336-41.
- [CrossRef] [PubMed] [Google Scholar]
- Improving contraceptive services in primary care: A quality improvement collaborative pilot. Contraception. 2015;92:364.
- [CrossRef] [Google Scholar]
- Emergency contraception: Access and challenges at times of uncertainty. Am J Ther. 2022;29:e553-67.
- [CrossRef] [PubMed] [Google Scholar]
- Declines in unintended pregnancy in the United States, 2008-2011. N Engl J Med. 2016;374:843-52.
- [CrossRef] [PubMed] [Google Scholar]
- Disparities in rates of unintended pregnancy in the United States, 1994 and 2001. Perspect Sex Reprod Health. 2006;38:90-6.
- [CrossRef] [Google Scholar]
- A population-based study of US women's preferred versus usual sources of reproductive health care. Am J Obstet Gynecol. 2015;213:352.e1-14.
- [CrossRef] [PubMed] [Google Scholar]