Unmasking Barriers in the Delivery of Preconception Counseling and Contraception Provision for Patients With Type 1 or Type 2 Diabetes

This is a pre- and post-intervention cohort chart review. Biologic females between the ages of 18 to 45 years with type 1 or type 2 diabetes who were seen in person by an endocrinologist for an outpatient visit at our tertiary academic center between 2020 and 2021 were included. Patients who had previously undergone sterilization or hysterectomy or who were currently pregnant were excluded. The same patient was not included more than once. If a patient had more than one visit in the study period, all of the visits were reviewed.

Because of the comprehensive health care system at our center and insurance that favors care within the same institution, the vast majority of patients are seen by other subspecialty or primary care providers (PCPs) within our system and can be captured. If care with an obstetrician/gynecologist was obtained within our system, those records were also reviewed to verify applicable clinical information. A preselected target of 40 patients pre-intervention and 48 patients post-intervention was based on 80% power to detect an increase in preconception counseling from 0 to 9% and contraception from 25 to 54% with a significance level of P <0.05.

Between the pre- and post-intervention periods, we performed a series of QI initiatives for endocrinologists. These included distribution of a series of distilled electronic messages with education about diabetes in pregnancy and contraceptive options (Supplementary Figure S1) and development of teaching materials about diabetes in pregnancy and contraception options that could be easily distributed to patients (Supplementary Figure S2). We also educated providers on the use of International Classification of Diseases, 10th Revision, code Z31.69 (preconception counseling) for billing purposes. Additionally, providers were given a list of oral contraceptive pill options that are safe for women with diabetes (Supplementary Figure S3), and the EMR system processes for ordering contraception or consultations with obstetricians/gynecologists were simplified. Finally, several EMR record tools were developed to serve as reminders to discuss contraception or preconception counseling and to simplify documentation for providers, including note templates for obtaining history, developing an assessment and plan, and providing patient information on preconception counseling (Supplementary Figure S4). These tools were highlighted in monthly endocrinology conferences and disseminated to endocrinologists who provided diabetes care.

After our intervention and post-interventional analysis, we conducted a confidential survey of all endocrinology providers at our center to better assess persistent barriers to the provision of contraception and preconception counseling. This survey was distributed electronically and consisted of several choices of barriers, of which any number could be selected. A free text box allowed providers to fill in their additional perspectives on persistent difficulties implementing contraception and preconception counseling for this population (Supplementary Figure S5).

All aspects of this research were approved by the University of Colorado Institutional Review Board.

Fifty patients were included in the pre-intervention cohort, and another 49 were included in the post-intervention cohort. There were no differences in clinical characteristics between these cohorts (Table 1).

Prior to our QI intervention, only 4% of patients had contraception specifically documented by their endocrinologist, although 62% were documented to be using contraception around the time of the visit based on our chart review. Table 2 summarizes the types of contraception patients were using. Endocrinologists did not specifically provide preconception counseling to any of these patients, although 14% of the patients had previously undergone counseling with a maternal-fetal medicine provider based on additional chart review. No referrals were placed to an obstetrician/gynecologist during the pre-intervention period.

After the QI intervention, contraception was specifically addressed by endocrinologists with 18% of patients, representing a significant increase from the pre-intervention rate of 4% (P = 0.03). Only one patient (2%) received a prescription for contraception. Additionally, 52% of patients were elsewhere documented to be on contraception around the time of their endocrinology visit. Compared with 0% before the intervention, 6% underwent preconception counseling with their endocrinologist (Table 2). Despite the provision of EMR links for referrals, there were no referrals placed to an obstetrician/gynecologist for contraceptive or preconception needs, which was similar to the pre-intervention period.

We were frankly surprised by the modest improvement in preconception counseling and contraception provision after what we viewed to be a comprehensive intervention focused on education and EMR tools designed to directly facilitate these practices in a time-efficient manner. As a result, we designed a provider survey to interrogate perceived barriers.

We sent this survey out to the 43 providers in our adult endocrinology clinic (24 physicians, 11 advanced practice providers, and 8 fellows), of whom 31 see patients with diabetes. We received a total of 25 unique responses, all of which were providers who cared for patients with diabetes (81% response rate). Providers could select more than one barrier or write in their own rationale.

Among the choices of reasons why providers do not perform this care, an overwhelming 80% stated that there was inadequate time during visits. Thirty-two percent identified a lack of proper knowledge to provide this care, 24% noted that they could not easily navigate the EMR tools we provided, and 8% said that they forgot to bring up these issues during visits. When queried about ways in which to improve contraception provision and preconception counseling, several respondents indicated that features of the EMR could be better used to this end. For example, an automatic advisory that requires relevant questions to be addressed before closing the EMR encounter was suggested. Other recommendations included education or links in the materials provided to patients at the conclusion of the visit and posted signs or provision of informational flyers in the clinic. Interestingly, some endocrinologists recommended that a dedicated telehealth clinic be created for this purpose and that it was more appropriate for diabetes educators, PCPs, and obstetrician/gynecologists to improve these aspects of care. Of all respondents after the intervention, only 8% stated that they faced no barriers to successfully providing contraception and preconception counseling to this population.

We demonstrate extremely low rates of documentation of preconception counseling and contraception provision by endocrinologists for women of reproductive age with diabetes that improved only modestly after a series of multitargeted QI initiatives. Provider surveys indicated that the overwhelming barrier was lack of time during visits. Despite our multimodal attempts to disseminate key information, one-third of the providers responded that inadequate knowledge was still a major reason behind their inability to implement preconception and contraception counseling for these patients.

In our pre-intervention cohort, only 4% of patients had contraception specifically documented by their endocrinologist, similar to findings previously reported. Schwarz et al. (14) found that only 4% of U.S. ambulatory visits with women with diabetes included contraception counseling, which increased minimally to 6% after adjusting for time spent with these patients. When specifically looking at primary care practices, Magdaleno et al. (15) reported that 19% of women with diabetes received preconception counseling over a 3-year period. Our study resonated with similarly identified barriers—predominantly a lack of time and knowledge.

It is unclear whether preconception counseling rates are low in this population or whether reported rates are related to deficiencies in documentation. Recent data from a single institution showed that 25% of endocrinologists reported offering preconception counseling, although 55% of women with diabetes reported receiving preconception counseling, possibly from other sources (16). Interestingly, among the visits in which preconception counseling was self-reported by an individual provider, this care was documented only 67% of the time (16). Similarly, Magdaleno et al. (15) showed that 39% of PCPs reported regularly providing preconception counseling, despite a low rate of documentation of only 19%. The low rate of preconception counseling did not appear to be the result of a lack of assigned value given that 89% of providers indicated that preconception counseling is important. These findings indicate a disconnect between provider priorities and values, implementation of practices, and documentation of practices.

Pregnant women with preexisting diabetes face increasing maternal, fetal, and neonatal morbidity and mortality. Women who met the inclusion criteria for our study had a mean A1C of 7.8% and 7.9% in the pre- and post-intervention cohorts, respectively, which is well above the upper threshold of 6.5% recommended by both the ADA and ACOG (7,8). This level of glycemic control indicates a substantially increased risk for major malformations and poor gestational outcomes and the need for contraception until adequate glycemic control is achieved. Only 62% and 53% of women who met inclusion criteria pre- and post-intervention, respectively, were using contraception. If those women not currently on effective contraception were to become pregnant, they would be at a markedly increased risk of poor, largely preventable outcomes.

Given the critical importance of glycemic control periconceptually and particularly during organogenesis, endocrinologists and PCPs have more power to reduce preventable adverse pregnancy outcomes in women with diabetes—especially major malformations—than do obstetricians, who often do not see these women until after 8 weeks of pregnancy, when organogenesis is completed. Given that approximately 50% of pregnancies are unplanned, ensuring adequate contraception until metabolic control and optimal treatment of underlying comorbidities is optimized is fundamental to improving pregnancy outcomes (17).

There is not only a continued need to raise awareness of the importance of preconception counseling in this population, but also a need to design better initiatives to improve competency through resident and provider professional training opportunities. A serious and substantial increase in public health funding for these important initiatives should drive future clinical research efforts.

Limitations of this study include that it was performed at a single subspecialty practice and center. The interventions were designed before assessing providers’ fundamental knowledge, acknowledgment of the problem, and readiness for change. Insight into provider attitudes and barriers toward preconception counseling and contraception provision was informally interrogated before the intervention and more formally afterward, through the post-implementation survey. Additionally, patients’ knowledge, practices, and outcomes were not evaluated. Strengths of the study include its adequate power to detect an improvement in outcomes, the multifaceted approach of patient and provider education, provider outreach, and the provision of resources, including EMR tools to directly facilitate prescriptions, referrals, and documentation.

Future initiatives should focus on how to mitigate barriers to the provision of counseling and contraception, and particularly the lack of time and knowledge. These initiatives should include a multidisciplinary approach during medical training and in health care professionals in practice, including PCPs, obstetrician/gynecologists, and endocrinologists, as well as a wide array of other types of providers (e.g., advanced practice practitioners, nurses, certified diabetes care and education specialists, pharmacists, and community counselors) to assuage the time burden on one specialty or provider. More EMR tools built with input from all stakeholders will also improve the efficiency of counseling and facilitate documentation, prescriptions, and referrals. Informing providers that preconception counseling is a billable code can provide incentive as well. Public health efforts at educating patients through different materials such as digital platforms and social networks may also empower them to ask questions of their providers and advocate for their reproductive needs and metabolic health.

Clearly, adequate resources directed to design innovative, multifaceted approaches are needed, particularly when taking into account the time constraints and baseline knowledge of providers in this arena. Figure 1 summarizes various interventions within the Institute of Healthcare Improvement QI impact-effort matrix (18) that might improve such efforts.

Our study shows that educational and documentation tools afford modest improvements in preconception counseling and contraception provision to reproductive-aged women with diabetes. The data and QI initiative presented here were from the endocrinology clinic at a single tertiary academic center, and one might assume a greater rate of adoption of professional organization guidelines within academic practices. However, our findings clearly showcase the global phenomenon of lack of preconception counseling and contraception provision for these patients, and the barriers endocrinology providers continue to face in this arena.

Our study serves as a call to action and incorporates elements that can serve as a road map for future, more effective efforts. Without renewed momentum, this critical, yet persistently neglected, public health priority for women with diabetes will fail to avert the rise in adverse outcomes in this rapidly growing population of these women and their offspring.

The authors thank the patients on whom data were collected; Drs. Michael McDermott, Cecilia Low Wang, and David Saxon of the University of Colorado Anschutz Medical Campus, in Aurora, CO, for their help in disseminating provider surveys and educational materials; and the endocrinology and obstetrics/gynecology providers at the University of Colorado Anschutz Medical Campus who take care of our high-risk patients of reproductive age with type 1 or type 2 diabetes.

This research was supported by American Diabetes Association Cardiovascular-Metabolic Fellowship Award #1-21-CMF-003 (L.A.A.) and National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases grant F32 DK123878 (A.Z.).

No potential conflicts of interest relevant to this article were reported.

L.A.A., L.A.B., and A.Z. conceived the idea for this QI initiative. L.A.A., C.N., and A.Z. executed the initial interventions. L.S. and R.J. performed pre- and post-intervention chart review and analysis. L.A.A., L.S., L.A.B., and A.Z. wrote the manuscript. All authors reviewed and edited the manuscript and approved the final version for submission. L.S. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.