Diagnosis and Treatment of Gestational Diabetes Mellitus: A National Survey of Physician Practices

Abstract

Aims:

We aimed to identify changes in United States practice patterns in gestational diabetes mellitus (GDM) diagnosis and treatment following publication of the 2008 Hyperglycemia and Adverse Pregnancy Outcomes (HAPO) study that supported transition toward a 2-hour oral glucose tolerance test.

Methods:

A total of 1,030 U.S. obstetric providers were surveyed in 2021 about GDM screening, diagnosis, and treatment, as well as perceptions surrounding preparation for the 1-hour, 50-g glucose loading test (GLT). Data were compared with data from a similar 2003 survey. The study was reviewed by the Institutional Review Board at Stanford University and was determined to be exempt.

Results:

Of 1,030 providers surveyed, 304 (30%) responded. Most respondents continued using the two-step screening method (95.0% versus 95.2% in 2003, p = 0.18). Fewer providers used insulin as a first-line medication (64.1% in 2021 versus 82.3% in 2003, p < 0.001). However, providers practicing for 0–10 years often used insulin as first-line compared with providers practicing for over 10 years (79% versus 55%, p < 0.001). Of 2021 respondents, 39.3% believed that fasting before the 1-hour GLT lowers the glucose result, 34.3% believed it increases the result, and 26.4% believed it would have no effect.

Conclusions:

Despite data from the HAPO trial, the majority of providers surveyed still use the two-step method for GDM screening. There is wide variability in perceptions and counseling regarding preparation for the 1-hour GLT.

Introduction

Gestational diabetes mellitus (GDM) complicates approximately 8% of pregnancies in the United States and is associated with significant adverse pregnancy outcomes, as well as an increased lifetime risk of type 2 diabetes mellitus (T2DM).^1,2 Adverse maternal outcomes include cesarean birth and preeclampsia, whereas adverse fetal outcomes include large for gestational age infants, birth trauma, neonatal hypoglycemia, and neonatal hyperbilirubinemia.^3

–7 Given this, the American College of Obstetricians and Gynecologists (ACOG) and the U.S. Preventative Task Force recommend universal screening for GDM after 24 weeks gestation as treatment of GDM in pregnancy has shown to improve outcomes.^8,9

Strategies for the diagnosis and management of GDM have evolved over time, and clinical recommendations are conflicting. GDM is commonly diagnosed using one of two methodologies. The “two-step method” begins with a 1-hour, 50 g oral glucose loading test (GLT). Those screening positive then undergo a diagnostic 3-hour oral glucose tolerance test (OGTT) following a 100 g glucose load. Another diagnostic modality is the “one-step method,” in which a 75 g glucose load is administered after a fasting blood draw, with subsequent blood draws 1 and 2 hours later. In 2008, the Hyperglycemia and Adverse Pregnancy Outcomes (HAPO) study demonstrated a continuous and linear association between increasing levels of maternal blood glucose and adverse pregnancy outcomes when the one-step 75 g OGTT was administered during pregnancy.¹⁰ This resulted in the International Association of Diabetes and Pregnancy Study Group’s (IADPSG) recommendation of the one-step 75 g OGTT utilizing a predefined odds ratio of 1.75 for adverse outcomes to define blood glucose thresholds for the diagnosis of GDM. This recommendation was anticipated to result in a two- to threefold increase in the prevalence of GDM with conflicting data regarding improvement in outcomes.^10
–12

Despite these findings, ACOG continued to recommend the two-step method in the 2013 GDM Practice Bulletin. This decision was due to the lack of evidence that using the new criteria would lead to clinically significant improvements in perinatal outcomes.¹³ Surveys in the United States have shown that the vast majority of obstetricians use the two-step method, with a survey in 2014 of Society of Maternal-Fetal Medicine members reporting 90% using the two-step method.¹⁴ In 2018, ACOG updated its Practice Bulletin for GDM diagnosis and management recommendations.⁹ The updated recommendations continued to support the two-step method, highlighted the recommended transition toward using insulin as a first-line medication, and provided more specific antenatal testing recommendations, prompting an opportunity for a more updated survey of practice patterns among obstetricians in the United States. We therefore conducted a national survey of obstetric care providers in the United States to evaluate the latest practice patterns in the diagnosis and treatment of GDM, compared with those published in a 2003 survey study by Gabbe et al., as well as to assess provider perceptions and patient advice regarding preparation for the 1-hour, 50-g GLT.

Materials and Methods

Survey design

An electronic provider survey was developed using Qualtrics (Qualtrics, Provo, UT). The survey queried provider demographic data in addition to GDM screening practices and therapy choices (Supplementary Appendix). Survey questions were developed to mirror those asked by Gabbe and colleagues in 2003, to determine the potential effect of practice changes since the HAPO trial.¹⁵ Additional questions surveyed provider perceptions about the effect of oral intake before the 1-hour, 50-g GLT and recommendations provided to patients before this screen. The study was reviewed by the Institutional Review Board at Stanford University and was determined to be exempt.

Survey distribution

Anonymous surveys were emailed to 1,030 U.S. obstetric providers, including obstetrics and gynecology residents, maternal-fetal medicine fellows, practicing obstetricians/gynecologists, advanced practice providers (which included nurse practitioners, physician assistants and certified nurse midwives), and family medicine physicians who care for pregnant patients. A total of 546 of those surveyed are members of the Pregnancy-Related Care Research Network (PRCRN), a national collective of women’s health care clinicians who are recruited to participate in research studies that assess clinical care patterns and provider education needs to support efforts to improve the delivery of health care to mothers and their children.¹⁶ The PRCRN has been the source of over 200 peer-reviewed publications, including the survey study by Gabbe et al. published in 2004.¹⁵

Participants were contacted by email with basic information about the study and an electronic link to access the survey. Survey recipients were asked if they currently diagnose or treat patients with GDM—if their response was “No,” the survey ended and their responses were excluded. There was no minimum number of questions answered to be included in the survey results. One reminder email was sent approximately 1 month following the initial email request.

Data analysis

Surveys were emailed between April and May of 2021. Survey respondent demographics are presented as count and percent. The survey responses of 2021 were then compared with those conducted in 2003 when applicable.¹⁵ The 2021 survey responses were also compared by providers who had up to 10 years of experience versus those who had 11 or more years. Chi-squared tests or, in the presence of cells ≤5, Fisher’s exact tests were performed using SAS 9.4 (Cary, NC).

Results

Of 1,030 providers emailed, 319 responded (31%). Ten providers responded that they did not diagnose or treat patients with GDM so were excluded, and an additional five providers were excluded given that they did not complete any survey questions after the first yes/no screening question (see Supplementary Appendix)—thus, these 15 responses were excluded from the analysis for a total of 304 responses (30%). Of the remaining 304 responses included in the analysis, 165 (54.5%) were generalist obstetrician/gynecologists, 62 (20.5%) were maternal-fetal medicine specialists, 53 (17.5%) were obstetrics and gynecology residents, 11 (3.6%) were maternal-fetal medicine fellows, 7 (2.3%) were advanced practice providers or certified nurse midwives, 4 (1.3%) were family medicine specialists, and 1 (0.3%) identified as “other” (Table 1). Of the providers, 55.8% reported practicing in academic centers, with the remainder practicing in private (19.1%), community (12.9%), multispecialty (10.2%), or “other” (2.0%) practices (Table 1). Of the providers, 45.2% practiced in the western part of the United States, 26.6% in the south, 14.1% in the northeast, and 14.1% in the midwest (Table 1). Of the providers, 38.2% had been in practice 10 years or less, whereas the remaining 61.8% had been in practice for 11 or more years (Table 1). Of the providers, 70.1% reported caring for more than 25 GDM patients per year, and all provided care to at least one GDM patient per year (Table 1).

Table 1.

Respondent Demographics

	N	%
Total	304
Years in practice
0–10 years	116	38.2
11–20 years	188	61.8
Race
Asian	57	18.8
Black or African American	14	4.6
Pacific Islander	6	2.0
White	203	67.0
Mixed or multiracial	10	3.3
Other	13	4.3
Missing	1
Gender
Female	229	75.3
Male	72	23.7
Other	3	1.00
Provider type
Generalist	165	54.5
Maternal-Fetal Medicine attending	62	20.5
Maternal-Fetal Medicine fellow	11	3.6
Resident	53	17.5
Family medicine provider	4	1.3
Advanced practice providers/certified nurse midwives	7	2.3
Other	1	0.33
Missing	1
Practice type
Academic	169	55.8
Community	39	12.9
Multispecialty	31	10.2
Private	58	19.1
Other	6	2.00
Missing	1
Region
Midwest	41	14.1
Northeast	41	14.1
South	77	26.6
West	131	45.2
Missing	14
Patients treated with GDM per year
0–10 patients	18	5.9
11–25 patients	73	24.0
26–50 patients	96	31.6
51–100 patients	66	21.7
101+ patients	51	16.8

GDM, gestational diabetes mellitus.

Current GDM screening and diagnostic methods compared with the pre-HAPO period

The rates of universal screening for GDM were high in both the current study and the 2003 survey results (99.7% versus 95.5%, respectively, p < 0.001). Compared to the 2003 survey, a similar number of providers used the two-step method for diagnosing GDM (95.0% versus 95.2% in 2003, p = 0.18) (Table 2). Among physicians using the two-step method, Carpenter and Coustan criteria (glucose cutoff of 95/180/155/140 mg/dL) were more frequently used in 2021 (86.6% versus 37.9% in 2003, p < 0.001) than the National Diabetes Data Group criteria (105/190/165/145 mg/dl) (13.1% versus 59.0% in 2003, p < 0.001) (Table 2).

Table 2.

Provider Screening, Diagnosis, and Treatment Practices of 2021 Survey Versus 2003 Survey

	2021		2003		p
	N	%	N	%	p
Total	304
Universal screening					<0.001
Yes	302	99.7	421	95.5
No	1	0.33	20	4.5
Missing	1
Early (first trimester) screening rates and methods			Not asked		—
Two-step method (1-hour 50-g GLT followed by 3-hour 100-g OGTT if positive)	201	66.8
One-step method (2-hour 75-g OGTT)	4	1.3
Hemoglobin A1c	86	28.6
Other	7	2.3
Does not perform early screening	3	1
Missing	3
Diagnosis method					0.18
Two-step method (1-hour 50-g GLT followed by 3-hour 100-g OGTT if positive)	286	95.0	420	95.2
Other	15	5.0	21	4.8
Missing	3
Cutoff for positive 1-hour GLT			Not asked		—
130 mg/dL	32	11.3
135 mg/dL	79	27.9
140 mg/dL	170	60.1
Other	2	0.71
Missing	21
Criteria used for 3-hour OGTT					<0.001
Carpenter and Coustan (95/180/155/140 mg/dL)	245	86.6	167	37.9
National Diabetes Data Group (105/190/165/145 mg/dL)	37	13.1	260	59.0
Other	1	0.35	14	3.2
Missing	21
First-line medication					<0.001
Insulin	184	64.1	363	82.3
Glyburide	24	8.4	58	13.2
Metformin	76	26.5	—	—
Other	3	1.1	20	4.5
Missing	17
Antenatal testing—diet controlled			Question format differed
None	125	41.9
Nonstress tests	72	24.2
Amniotic fluid index	49	16.4
Growth ultrasound	150	50.3
Missing	6
Antenatal testing—medication controlled			Question format differed
None	2	0.67
Nonstress tests	281	94.3
Amniotic fluid index	225	75.5
Growth ultrasound	249	83.6
Missing	6
Postpartum glucose evaluation					<0.001
2-hour 75-g OGTT	261	87.9	167	37.9
Fasting glucose level	10	3.4	90	20.4
Hemoglobin A1c	12	4.0	—	—
Other	11	3.7	72	16.3
Does not perform postpartum screening	3	1.0	112	25.4
Missing	7

GLT, glucose loading test; OGTT, oral glucose tolerance test.

Among providers surveyed in 2021, 99% of respondents reported performing early GDM screening (for high-risk patients performed in the first trimester), with the majority (66.8%) using the two-step method (1-hour, 50-g GLT followed by 3-hour, 100-g OGTT if positive) followed by Hemoglobin A1c (28.6%) (Table 2). This was not queried in the 2003 study.

Provider choice of first-line medication and postpartum diabetes screening

Providers in 2021 differed in their first-line medication selection, with 64.1% using insulin as first-line compared with 82.3% in 2003 (p < 0.001) (Table 2). Of all providers surveyed in 2021, providers in practice for 10 or fewer years used insulin as first-line more often (78.7%) compared with providers in practice for 11 or more years (55.3%) (p < 0.001). Glyburide was reportedly used less frequently in 2021 as a first-line agent (8.4% versus 13.2% in 2003), with 26.5% of providers using metformin as first-line therapy (not assessed in the 2003 study) (Table 2). Providers in 2021 more frequently tested for type 2 diabetes postpartum (99.0% versus 74.6%) and more frequently used the 2-hour OGTT (87.9% versus 37.9%) (p < 0.001) to make the diagnosis postpartum (Table 2).

Provider perceptions about the effect of oral intake before the GDM screen

We surveyed providers about their perception regarding the effect of overnight fasting on the result of the 1-hour, 50-g GLT compared with eating a regular meal immediately before the test. Of the providers who responded to the following questions, 110 (39.3%) perceived that fasting would make the serum glucose level lower, 96 (34.3%) perceived it would make the serum glucose level higher, and 74 (26.4%) perceived it would make no difference in the serum glucose level (Fig. 1). While 237 (83.8%) reported that they had never advised a patient to delay the GDM screen based on oral intake, 40 (14.2%) would recommend a delay after eating a chicken wrap (a proxy for a meal with high protein), 81 (28.7%) would recommend delay after eating a doughnut (a proxy for a meal with high sugar content), and 13 (4.6%) would delay if a patient reported fasting (Fig. 2). Providers in practice for longer (>10 years) were more likely to recommend delaying the GDM screen based on prior oral intake (35 (20.0%) versus 11 (10.2%), p = 0.03) and were less likely to recommend a normal diet before the test (141 (80.6%) versus 94 (87.0%), p = 0.01). Perceptions and patient counseling practices did not significantly differ by geographic location.

FIG. 1.

Provider perceptions of prolonged fasting compared with eating a regular meal immediately before the glucose loading test (GLT).

FIG. 2.

Provider advice regarding taking or delaying the 1-hour 50-g glucose loading test (GLT) according to timing and content of last meal.

Antenatal testing

Providers were asked to select all antenatal fetal monitoring tools that they utilize for either diet-controlled GDM or GDM requiring pharmacologic management. For diet-controlled GDM, 41.9% reported not performing any method of antenatal testing, whereas 24.2% reported obtaining nonstress tests, 16.4% reported obtaining amniotic fluid index measurements, and 50.3% reported obtaining growth ultrasounds (Table 2). For GDM requiring pharmacologic management, 0.67% reported not performing any antenatal testing, whereas 94.3% reported obtaining nonstress tests, 75.5% reported obtaining amniotic fluid index measurements, and 83.6% reported obtaining growth ultrasounds (Table 2). The 2003 study by Gabbe et al. did not differentiate between diet-controlled and medication-controlled GDM in their antepartum fetal monitoring survey results, precluding a direct comparison between studies.

Discussion

Our study provides insight into current U.S. GDM screening and diagnosis practice patterns in addition to adding a unique comparison between practice patterns before and after the publication of the HAPO trial and ACOG’s most recent Practice Bulletin on GDM. Despite the results of the HAPO trial leading to the IADPSG’s recommendation to use the one-step method, 95% of the U.S. providers we surveyed continue to use the ACOG-recommended two-step method to diagnose GDM, as the conflicting recommendations from these expert groups continue to be discussed as recently as in the 2024 American Diabetes Association (ADA) Standards of Care in Diabetes Practice Committee.¹⁷ In addition, compared with the study conducted by Gabbe et al. in 2003, providers are prescribing insulin less frequently as a first-line agent and using oral metformin more frequently for the treatment of GDM.¹⁵ This is despite data published since 2003 suggesting insulin’s benefits and both ACOG and the ADA’s recommendation of using it as a first-line agent.¹⁸ In contrast, providers in this study reported prescribing glyburide less frequently, following ACOG’s recommendations.⁹ Finally, our study shows that providers vary significantly in their perceptions and counseling in terms of the effect of fasting versus oral intake before the GDM screen.

One- versus two-step method

There are conflicting recommendations about the optimal way to screen for and diagnose GDM.⁹ The HAPO study utilized the one-step method and showed a continuous association between maternal glucose levels and adverse perinatal outcomes.¹⁰ Consequently, the IADPSG recommended that the one-step method be used to diagnose GDM, citing a preset odds ratio for adverse outcomes to guide their recommendations surrounding blood glucose thresholds for diagnosing GDM.¹⁹ However, both ACOG and the Eunice Kennedy Shriver National Institute of Child Health and Human Development Consensus Development Conference on Diagnosing Gestational Diabetes recommend continued use of the two-step method for diagnosing GDM, citing that use of the one-step method would likely double the diagnosis of GDM without proven clinical benefit, given its lower glucose thresholds for diagnosis and requiring only one elevated value to meet criteria rather than two.¹²

More recently, the new 2024 ADA standards of care document underscores these conflicting recommendations and recognizes that establishing a uniform approach to diagnosing GDM could benefit many stakeholders. This document elaborates on the finding that compared with the two-step method, the one-step method leads to an increased diagnosis of GDM and increased medicalization of pregnancy without demonstrating improved patient and neonatal outcomes.¹⁷ However, arguments in favor of the one-step method include that the patients who met criteria for glucose intolerance using the one step (but would not have met criteria had they been screened using the more stringent two-step method) may benefit beyond pregnancy by identifying patients who may be at increased risk of developing prediabetes and type 2 diabetes and could therefore make lifestyle modifications postpartum to reduce this risk.^20,21 A recent randomized trial conducted in the United States comparing the two methods found that using the one-step method doubled the number of participants who were diagnosed with GDM from 8.5% using the two-step approach to 16.5% using the one step, but clinical outcomes were not significantly different between the two groups.²² The same trial also reported lower adherence rates to the one-step screening (66%) compared with the two-step (92%). Patient preferences are also an important consideration to achieve better treatment outcomes. In one large study, most patients preferred the two-step method (46%) compared with the one-step method (26%), although patients with more adverse metabolic profiles or a previous history of GDM typically preferred the one-step method.²³ Other relevant factors that vary between tests may include access to transportation, coordinating fasting times, and convenience of laboratory hours or locations.²⁴

Overall, the current study provides reassurance that GDM screening is being performed by >99% of providers and that most providers comply with ACOG’s recommendation to utilize the two-step screening method. These findings are consistent with the survey performed by Bimson et al. of 40% (937) of the Society for Maternal-Fetal Medicine members in 2014–2015, which reported similar rates of use of the two-step method (90.1%).¹⁴ After the publication of these survey results, ACOG released a new GDM Practice Bulletin in 2018, reinforcing the recommendation of a two-step method, which may explain the even higher utilization of the two-step method in the current study.

Early screening

Screening for GDM is recommended by ACOG at the initiation of prenatal care for certain individuals at particularly high risk of having undiagnosed type 2 diabetes. The ADA’s 2024 recommendations outline a fasting glucose of >110 or a Hemoglobin A1c of 5.9% may identify patients at higher risk of adverse outcomes, whereas ACOG does not list a testing method of choice.^17,25 Respondents in our study most commonly utilize the two-step method (66.8%) or Hemoglobin A1c (28.6%), and this was not reported in prior provider surveys. Given the ongoing obesity epidemic and increasing age of obstetric patients, an increasing number of patients will likely qualify for early GDM screening, which may prompt the call for more standardized screening recommendations.

Insulin as first-line medication

Both ACOG and the ADA recommend insulin as the first-line medication for GDM not controlled with diet and lifestyle modifications.¹⁸ The survey study conducted in 2003 by Gabbe and colleagues showed that over 80% of providers utilized insulin as a first-line medication therapy.¹⁵ Since that survey, the use of oral hypoglycemic therapies to treat GDM has increased, including the use of glyburide and metformin.⁹ ACOG’s 2013 Practice Bulletin stated that “insulin and oral agents are equivalent in efficacy, and either can be an appropriate first-line therapy,” which was reflected by the findings of the 2017 study performed by Bimson et al. reporting that 57% of providers used glyburide as a first-line agent, 36% used insulin, and 5% used metformin.^13,14 While oral hypoglycemic medications are an active area of research, metformin and glyburide are not yet approved by the United States Food and Drug Administration for treatment of GDM, as they lack long-term neonatal safety data and existing trials have not been appropriately designed and powered to evaluate equivalence or noninferiority when comparing these agents to insulin.⁹ In addition, metformin has been shown to cross the placenta with levels as high as those in maternal blood concentrations—long-term consequences of which are currently unknown.²⁶ Metformin use in pregnancy has been associated with multiple benefits, including a patient preference for metformin over insulin, lower rates of patients requiring insulin, and a reduction in gestational weight gain and fetal size.^27,28 However, it has also been associated with higher body mass index z-scores among exposed offspring, higher subcutaneous fat levels in children with metformin exposure, and there are mixed data surrounding fetal anomalies, including pulmonary valve atresia and the embryonic development of reproductive organs.^29,30

ACOG’s most recent 2018 GDM Practice Bulletin more clearly recommends insulin as a first-line medication for treatment of GDM requiring medication. Despite this, our study suggests that provider use of insulin in the U.S. as a first-line agent has decreased since 2003, with only 60% of providers reporting doing so. As mentioned previously, these trends have been shown to fluctuate over time and between provider cohorts, given Bimson et al. reported that 36% of providers used insulin as first-line and these practices varied by region.¹⁴ These fluctuations are further supported by our findings that, of the providers surveyed in 2021, those in practice for 10 or fewer years used insulin as first-line more often (78.7%) compared with providers in practice for 11 or more years (55.3%).

This may, in part, be due to provider comfort and ease with managing metformin compared with the skill set needed to manage and adjust insulin, new and emerging data regarding efficacy and long-term safety data, or as some studies have shown, patient preference for oral hypoglycemic medications.²⁸ Our survey data highlight the importance of ongoing research efforts to identify barriers in utilizing/recommending insulin as first-line for patients and/or providers to help increase provider use of insulin as a first-line agent. As new safety data are published, it will also be important to revisit previously published patient preferences with regard to treatment strategies to better provide patient-centered care resulting in improved clinical outcomes.

Antenatal testing

ACOG recommends initiating fetal testing for medication-requiring GDM (A2GDM) beginning at 32 weeks, whereas testing before 40 weeks of gestation may not be required in patients with well-controlled GDM not requiring medication management (A1GDM).²⁵ The current study indicated that most respondents utilized some form of antepartum fetal testing in both A1GDM (58.1%) and A2GDM (99.3%) and that growth ultrasounds were much more commonly used in A1GDM (50.3%) and A2GDM (83.6%) than prior studies. The study by Gabbe et al. published in 2003 reported that 80.3% of respondents utilized antepartum fetal monitoring in patients with GDM, although medication versus nonmedication controlled GDM was not specified.¹⁵ However, in contrast to our study, nonstress tests were most widely applied (74.1%) rather than assessment of fetal growth by ultrasound (7.3%).¹⁵ In 2017, Bimson et al. reported that 38.8% of respondents considered A1GDM an indication for antenatal testing (less than our study) and 98.6% for A2GDM, although the authors did not survey the type of testing performed.¹⁴ While ACOG does not specify the type of test and frequency of testing, our study suggests that there has been an increase in use of growth ultrasound as a component of antepartum surveillance and also in the amount of surveillance recommended to patients with A1GDM.

Provider counseling and recommendations regarding oral intake before the 1-hour 50-g GLT

Our study contributes important information on current provider counseling and recommendation practice patterns with regard to preparation for the 1-hour 50-g GLT. The ADA recommends that the test be performed without regard to time of day or time of last meal.³¹ Recently, we published results from a patient survey, which was conducted as a part of a prospective randomized trial at our institution with regard to counseling before the GDM screen in which 19% of participants reported that a health care professional gave recommendations about oral intake before the test in a previous pregnancy.³² This study also queried patient perceptions regarding the effect of oral intake on the GDM screen and found that 19.6% believed fasting would increase their glucose level, 49.0% believed it would decrease the glucose level, and 31.4% believed it would not affect it.³² Participants in the randomized trial who fasted had a significantly higher screen positive rate than those who ate within 2 hours of the 50-g GLT (32.0% versus 13.3%) with higher mean glucose values, suggesting that there may be real-world consequences to the heterogeneity of provider counseling with regard to preparation for the test. While the rate of diagnosis of GDM was also higher in the patients who fasted, this was not statistically significant given the study was not powered to detect this difference; thus, further studies are required to investigate this effect.

Our current study showed a similar wide range of provider perceptions and recommendations, with 72% of respondents believing that fasting before the test will alter the results either by increasing or decreasing the glucose value and that between 14% and 32% of providers would recommend their patient delay the test if they reported eating certain foods immediately before the test. The current study highlights the variability in counseling practices with regard to preparation for the 1-hour 50-g GLT despite the recommendation that the test be taken without regard to last meal, suggesting that either providers are unaware of the recommended counseling or believe that timing and content of oral intake before the 1-hour GLT will affect the results. Neither ACOG nor the ADA address these discrepancies in their published clinical guidelines. Our study findings highlight the need for the development of more widely accepted and standardized patient counseling resources in preparation for the GDM screen.

Postpartum diabetes screening—fasting versus the 2-hour method

While just over half of the respondents in 2003 reported recommending patients undergo a 2-hour OGTT at the postpartum visit to screen for T2DM, nearly 90% in the current study report recommending this test, compared with either using a fasting glucose test or not performing glucose testing. Studies have shown that patients with a history of GDM have an increased risk of developing T2DM later in life as high as 70%, prompting a strong recommendation from ACOG to recommend postpartum glucose testing to identify diabetes in patients after pregnancies complicated by GDM.^1,9 The 75 g, 2-hour OGTT is the preferred method given the poor sensitivity of fasting glucose levels alone in diagnosing diabetes postpartum.³³ Our current data on postpartum diabetes screening practices add to the existing literature on this topic.

Strengths and limitations

Strengths of the study include the number of participants who completed the survey, the inclusion of providers across all U.S. regions, and a diverse set of provider practice settings and career stages. We were also able to make direct comparisons to pre-HAPO data given that our questions were written to match those administered by Gabbe et al. in 2003. We assessed provider perceptions about the effect of oral intake before GDM screening, as well as all aspects of GDM care ranging from pre- and postdelivery screening to treatment. Our study, however, is not without limitations. First, the 2003 survey and our current study surveyed different cohorts. Our response rate of 30% represents a possible bias of results, including a potential selection bias of respondents who completed the survey. Second, participants were mostly contacted via the PRCRN and departmental listserves, rather than a single national survey distributed randomly. That said, our findings may be generalizable as we were able to obtain a geographically diverse pool of respondents. The majority of providers reported practicing in academic settings, which may not necessarily be representative of all provider practice settings nationally, although we did survey a broad range of providers ranging from specialists to residents. In addition, the current study did not investigate provider rationale behind their decisions regarding screening, treatment, and counseling, and it will therefore be important to explore further why this may be the case and subsequently what resources or education providers or patients need to shift this practice. Finally, while we matched some questions with those of a 2003 survey, it is possible that differences between the two reports are not necessarily due to the results of the HAPO trial, but rather due to differences in providers, differences in provider settings, or other unidentified differences between the two cohorts.

Conclusions

Since the publication of the HAPO trial in 2008, there has been an international trend toward utilizing the one-step method to screen for GDM. Despite this, our study demonstrates that the majority of U.S. survey respondents continue to follow ACOG’s recommended two-step method. Our study also shows that fewer providers use insulin as a first-line agent, despite ACOG’s recommendations to do so. Finally, there remains wide variability in provider counseling with regard to patient preparation for the 1-hour 50-g GLT and provider perceptions regarding the effect of oral intake before the 1-hour GLT.

Footnotes

Acknowledgment

The authors would like to acknowledge Dr. Jay Schulkin for his significant contributions to this article’s design and preparation, who unfortunately passed away before the study’s publication.

Authors’ Contributions

S.E.M.: Conceptualization and writing. M.M.S.: Conceptualization and writing—review and editing. J.A.M.: Data curation, formal analysis, and writing—review and editing. S.A.L.: Data curation, formal analysis, and writing—review and editing. D.J.L.: Conceptualization and writing—review and editing. T.H.: Methodology and writing—review and editing. Y.J.B.: Conceptualization, methodology, supervision, and writing—review and editing.

Author Disclosure Statement

T.H. is a clinical advisor for and is an investor in Malama Health, a gestational diabetes management company, which was not involved in the design or implementation of the study. The remaining authors report no conflict of interest.

Funding Information

This research did not receive any specific grant funding.

Supplementary Material

Supplementary Appendix

References

Casagrande

, Linder

, Cowie

. Prevalence of gestational diabetes and subsequent Type 2 diabetes among U.S. women. Diabetes Res Clin Pract, 2018; 141:200–208; doi: 10.1016/J.DIABRES.2018.05.010

Zhou

, Du

, Sun

, et al. Prevalence and Trends in Gestational Diabetes Mellitus Among Women in the United States, 2006–2017: A Population-Based Study. Front Endocrinol (Lausanne), 2022; 13:1092; doi: 10.3389/FENDO.2022.868094/BIBTEX

Ehrenberg

, Durnwald

, Catalano

, et al. The influence of obesity and diabetes on the risk of cesarean delivery. Am J Obstet Gynecol, 2004; 191(3):969–974; doi: 10.1016/j.ajog.2004.06.057

Black

, Sacks

, Xiang

, et al. The relative contribution of prepregnancy overweight and obesity, gestational weight gain, and IADPSG-defined gestational diabetes mellitus to fetal overgrowth. Diabetes Care, 2013; 36(1):56–62; doi: 10.2337/DC12-0741

Voormolen

, De Wit

, Van Rijn

, et al. Neonatal hypoglycemia following diet-controlled and insulin-treated gestational diabetes mellitus. Diabetes Care, 2018; 41(7):1385–1390; doi: 10.2337/DC18-0048

Yogev

, Xenakis

EMJ

, Langer

. The association between preeclampsia and the severity of gestational diabetes: The impact of glycemic control. Am J Obstet Gynecol, 2004; 191(5):1655–1660; doi: 10.1016/j.ajog.2004.03.074

, Song

, Zou

, et al. Association between neonatal hyperbilirubinemia and hypoglycemia in Chinese women with diabetes in pregnancy and influence factors. Sci Rep, 2022; 12(1):16975–16978; doi: 10.1038/s41598-022-21114-6

Pillay

, Donovan

, Guitard

, et al. Screening for gestational diabetes: Updated evidence report and systematic review for the U.S. preventive services task force. JAMA, 2021; 326(6):539–562; doi: 10.1001/JAMA.2021.10404

Caughey

, Turrentine

. ACOG practice bulletin no. 190: Gestational diabetes mellitus. Obstetrics and Gynecology, 2018; 131(2):E49–E64; doi: 10.1097/AOG.0000000000002501

10.

Metzger

, Lowe

, Dyer

, et al.; HAPO Study Cooperative Research Group. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med, 2008; 358(19):1991–2002; doi: 10.1056/NEJMOA0707943/SUPPL_FILE/NEJM_HAPO_1991SA1.PDF

11.

Agarwal

, Dhatt

, Shah

. Gestational diabetes mellitus. Diabetes Care, 2010; 33(9):2018–2020; doi: 10.2337/dc10-0572

12.

Bodmer-Roy

, Morin

, Cousineau

, et al. Pregnancy outcomes in women with and without gestational diabetes mellitus according to the international association of the diabetes and pregnancy study groups criteria. Obstet Gynecol, 2012; 120(4):746–752; doi: 10.1097/AOG.0B013E31826994EC

13.

Practice Bulletin No. 137. Obstetrics & Gynecology, 2013; 122(2):406–416; doi: 10.1097/01.AOG.0000433006.09219.f1

14.

Bimson

, Rosenn

, Morris

, et al. Current trends in the diagnosis and management of gestational diabetes mellitus in the United States*. J Matern Fetal Neonatal Med, 2017; 30(21):2607–2612; doi: 10.1080/14767058.2016.1257603

15.

Gabbe

, Gregory

, Power

, et al. Management of diabetes mellitus by obstetrician-gynecologists. Obstet Gynecol, 2004; 103(6):1229–1234; doi: 10.1097/01.AOG.0000128045.50439.89

16.

Pregnancy Related Care Research Network. Available from: https://mchb.hrsa.gov/research/project_info.asp?ID=348#:∼:text=The PRCRN program assesses the, to mothers and their children

17.

Diagnosis and Classification of Diabetes: Standards of Care in Diabetes—2024. Diabetes Care, 2024; 47(Supplement 1):S20–S42; doi: 10.2337/dc24-S002

18.

Committee

. 15. Management of diabetes in pregnancy: Standards of medical care in diabetes—2022. Diabetes Care, 2022; 45(Supplement_1):S232–S243; doi: 10.2337/DC22-S015

19.

, Sg G

, et al. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care, 2010; 33(3):676–682; doi: 10.2337/DC09-1848

20.

Lowe

, Scholtens

, Lowe

, et al.; HAPO Follow-up Study Cooperative Research Group. Association of Gestational Diabetes With Maternal Disorders of Glucose Metabolism and Childhood Adiposity. JAMA, 2018; 320(10):1005–1016; doi: 10.1001/jama.2018.11628

21.

Aroda

, Christophi

, Edelstein

, et al.; Diabetes Prevention Program Research Group. The effect of lifestyle intervention and metformin on preventing or delaying diabetes among women with and without gestational diabetes: The diabetes prevention program outcomes study 10-year follow-up. J Clin Endocrinol Metab, 2015; 100(4):1646–1653; doi: 10.1210/jc.2014-3761

22.

Hillier

, Pedula

, Ogasawara

, et al. A pragmatic, randomized clinical trial of gestational diabetes screening. N Engl J Med, 2021; 384(10):895–904; doi: 10.1056/NEJMOA2026028

23.

Raets

, Vandewinkel

, Van Crombrugge

, et al. Preference of women for gestational diabetes screening method according to tolerance of tests and population characteristics. Front Endocrinol (Lausanne), 2021; 12:781384; doi: 10.3389/fendo.2021.781384

24.

Mobin

, Obeid

, El‐Kebbi

, et al. Beyond one size fits all: Probing patient choices in gestational diabetes management, from screening to postpartum. Chronic Dis Transl Med. Published Online October, 2024; 25; doi: 10.1002/cdt3.153

25.

Caughey

, Turrentine

. ACOG PRACTICE BULLETIN: Gestational Diabetes Mellitus. Obstetrics and Gynecology, 2018; 131(2):E49–E64; doi: 10.1097/AOG.0000000000002501

26.

Eyal

, Easterling

, Carr

, et al. Pharmacokinetics of metformin during pregnancy. Drug Metab Dispos, 2010; 38(5):833–840; doi: 10.1124/DMD.109.031245

27.

Dunne

, Newman

, Alvarez-Iglesias

, et al. Early metformin in gestational diabetes. JAMA, 2023; 330(16):1547–1556; doi: 10.1001/jama.2023.19869

28.

Rowan

, Hague

, Gao

, et al.; MiG Trial Investigators. Metformin versus Insulin for the Treatment of Gestational Diabetes. N Engl J Med, 2008; 358(19):2003–2015; doi: 10.1056/NEJMoa0707193

29.

Rowan

, Rush

, Obolonkin

, et al. Metformin in gestational diabetes: The offspring follow-up (MiG TOFU). Diabetes Care, 2011; 34(10):2279–2284; doi: 10.2337/dc11-0660

30.

Newman

, Dunne

. Treatment of diabetes in pregnancy with metformin. Obstet Gynecol, 2024; 144(5):660–669; doi: 10.1097/AOG.0000000000005705

31.

Metzger

, Coustan

. Summary and recommendations of the second international workshop. Conference on Gestational Diabetes Mellitus. Diabetes, 1985; 34(Supplement_2):123–126; doi: 10.2337/DIAB.34.2.S123

32.

Sperling

, Leonard

, Miller

, et al. Fasting compared with fed and oral intake before the 1-hour oral glucose tolerance test. Obstet Gynecol, 2023; 141(1):126–133; doi: 10.1097/AOG.0000000000005013

33.

McClean

, Farrar

, Kelly

, et al. The importance of postpartum glucose tolerance testing after pregnancies complicated by gestational diabetes. Diabet Med, 2010; 27(6):650–654; doi: 10.1111/J.1464-5491.2010.03001.X

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.10 MB