Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Feb 20.
Published in final edited form as: J Matern Fetal Neonatal Med. 2025 Feb 20;38(1):2467996. doi: 10.1080/14767058.2025.2467996

Maternal and neonatal outcomes with different screening strategies for gestational diabetes mellitus: a retrospective cohort study

Shirley J Shao a, Lucy J Fu b, Llyke Ching c, Katelin P Kramer d, Nasim C Sobhani e
PMCID: PMC12256285  NIHMSID: NIHMS2058510  PMID: 39978928

Abstract

Objective

To evaluate the effect of an expanded gestational diabetes mellitus (GDM) screening strategy on perinatal outcomes.

Methods

This retrospective cohort study included gravidas screened for GDM at a single academic center. The “before” cohort (estimated due dates (EDD) March 2018 – April 2019) was screened using the standard 2-step method. The “after” cohort (EDD November 2019 – July 2023) was screened using an expanded strategy that included the potential for GDM diagnosis based on 1–2 weeks of home glucose monitoring following isolated fasting hyperglycemia on the 3-hour glucose tolerance test (GTT). The primary outcomes were primary cesarean delivery (PCD) and neonatal intensive care unit (NICU) admission. Binomial regression and Kruskal Wallis tests were used to compare perinatal outcomes between the two cohorts in the general population and in a subgroup of those with isolated fasting hyperglycemia.

Results

Outcomes for the “before” cohort (n=1,733) were compared with those in the “after” cohort (n=6,280). In the general population, A2GDM incidence increased after expansion of the screening protocol (4.8% vs. 6.4%, RR 1.34, 95% CI 1.07–1.69), but PCD and NICU admission rates were unchanged. In the subgroup of patients with isolated fasting hyperglycemia (n=233), there was a significant increase in the incidence of any GDM (12.5 vs. 46.0%, RR 3.68, 95% CI 1.95 – 6.93) and A2GDM (8.3% vs. 37.3%, RR 4.47, 95% CI 2.03 – 9.87). There was no difference in PCD between cohorts, but NICU admission increased significantly in the “after” cohort (8.3% vs 22.4%, RR 2.68, 95% CI 1.18–6.08) in this subgroup.

Conclusion

An expanded approach to GDM diagnosis using home blood glucose monitoring following isolated fasting hyperglycemia on 3-hour GTT was associated with increased A2GDM incidence but no improvements in primary maternal or neonatal outcomes. In the absence of clinical benefit, we do not recommend widespread implementation of this expanded strategy.

Keywords: fasting hyperglycemia, gestational diabetes screening, primary cesarean delivery, self-monitored blood glucose testing, 3-hour oral glucose tolerance test

MAIN TEXT

Introduction

Gestational diabetes mellitus (GDM) is the most common medical morbidity of pregnancy, with a global prevalence of 4.4–16.9% [1–3]. GDM is associated with adverse maternal and neonatal outcomes, including cesarean delivery (CD), hypertensive disorders of pregnancy (HDP), preterm birth (PTB), birth trauma, neonatal intensive care unit (NICU) admission, and long-term metabolic complications [4,5]. The optimal screening strategy has been deliberated for decades, as treatment of GDM can reduce serious perinatal morbidity and improve quality of life, but overdiagnosis of GDM can lead to excessive monitoring, unnecessary interventions, and psychosocial distress [1,6–10].

Currently, the most common screening strategy in the United States is a 2-step process using glucose tolerance tests (GTT) as recommended by the American Diabetes Association (ADA), the American College of Obstetricians and Gynecologists (ACOG), and the National Institutes of Health (NIH) [11–13]. Patients found to have two or more elevated values during the 3-hour GTT are diagnosed with GDM (Figure 1).

Figure 1:

Figure 1:

Open in a new tab

Gestational diabetes mellitus screening protocol

GTT, glucose tolerance test; SMBG, self-monitoring of blood glucose; GDM, gestational diabetes mellitus.

This figure depicts the standard screening protocol as recommended by ACOG. Patients at intermediate risk with isolated fasting hyperglycemia are targeted in this expanded screening strategy.

Although patients with 1 abnormal 3-hour GTT value are not typically diagnosed with GDM, they may still experience an increased risk of adverse outcomes compared to individuals with normal glucose tolerance [14–19]. ACOG suggests that further research is needed to assess risk for patients with a single abnormal value on the 3-hour GTT [12]. Self-monitoring of blood glucose (SMBG), though not widely used for GDM diagnosis, has been proposed as a GTT complement that can identify additional patients susceptible to the adverse outcomes of gestational hyperglycemia [20,21].

Prior to March 2019, pregnant individuals at our institution were diagnosed with GDM if 2 or more values on the 3-hour GTT were abnormal, while those who had only 1 abnormal value continued with routine prenatal care. Starting in March 2019, this screening strategy was expanded such that individuals with an isolated abnormal fasting value on the 3-hour GTT were asked to perform SMBG for one week, with GDM diagnosed if more than half of those values were elevated. In this study, we aimed to compare GDM incidence and adverse perinatal outcomes in the general pregnant population before and after implementation of this expanded screening strategy. Additionally, we explored GDM incidence and adverse perinatal outcomes in the subgroup of patients with isolated fasting hyperglycemia who were most affected by screening expansion. We hypothesized that expanded screening with appropriate treatment for those diagnosed with GDM would lead to lower rates of primary cesarean delivery (PCD, the primary maternal outcome) and lower rates of NICU admission (the primary neonatal outcome) in both the general population and in the subgroup with isolated fasting hyperglycemia.

Materials and Methods

This was a retrospective cohort study at a single tertiary care academic institution with a high-risk obstetrics service and a level IV Intensive Care Nursery. An expanded GDM screening strategy was implemented in March 2019 (Figure 1). Cohort assignment was determined by estimated delivery date (EDD) (Appendix 1):

  1. “Before” cohort: Individuals with EDD between March 1, 2018 to April 30, 2019, screened for GDM using a standard two-step screening protocol.

  2. “After” cohort: Individuals with EDD between November 1, 2019 to July 30, 2023, screened using an expanded SMBG-supplemented screening protocol.

To separate cohorts, we designated a six-month washout period that excluded patients with EDD between May 1, 2019 to October 31, 2019 and patients screened for GDM between March 1, 2019 to August 30, 2019. Individuals who completed the first step of the 2-step screen before 29 weeks and delivered at our institution were included. Individuals with pregestational diabetes, multifetal pregnancies, or delivery < 24 0/7 weeks were excluded. The length of the “before” cohort was constricted to the thirteen months where the standard 2-step screening protocol was employed, as our institution used 1-step screening before this time. As the expanded screen had been in place for 2.5 years when the study was designed, we chose to maximize the sample size by including all patients who met criteria that 2.5 year span, leading to unequal sizes of the two cohorts.

Individuals diagnosed with GDM were referred to our institution’s diabetes and pregnancy program led by maternal-fetal medicine (MFM) specialists. GDM management was unchanged between cohorts and consisted of medical nutrition therapy, SMBG at least 4 times per day, weekly review of blood glucose logs by MFM, and pharmaceutical treatment as needed. For both cohorts, patients at high risk for pre-eclampsia were recommended to start low-dose aspirin between 12–16 weeks as part of routine prenatal care. Labor management did not change between cohorts. NICU admission policies changed on April 2021 to allow neonates born between 35 0/7 to 35 6/7 weeks to trial admission to the regular nursery; prior to that, any neonate born < 36 0/7 weeks was automatically admitted to the NICU. In October 2022, high intensity mattress phototherapy was made available in the regular nursery for initial treatment of hyperbilirubinemia. Prior to this, any infant requiring phototherapy was admitted to the NICU. At the end of 2022, our institution also transitioned from the Northern California Neonatal Consortium (NCNC) guidelines to the new 2022 American Academy of Pediatrics (AAP) thresholds for phototherapy [22–24]. These two guidelines are very similar, and both recommend increased phototherapy thresholds compared to 2004 AAP thresholds [22–24]. NICU admission policies did not change because of the COVID-19 pandemic. Despite the temporal overlap with the start of the COVID-19 pandemic, our institution’s policies regarding antepartum care for GDM and management of labor/delivery remained unchanged. At our institution, GDM management was performed virtually both before and after the COVID-19 pandemic, and there were no explicit protocol changes to the frequency of follow-up visits or the availability of counseling and review of blood sugar logs.

The primary maternal outcome was PCD, and the primary neonatal outcome was NICU admission. Secondary maternal outcomes included any GDM, A2GDM, any HDP, preeclampsia, gestational hypertension, and obstetric anal sphincter injuries (OASIS). Secondary neonatal outcomes included PTB, shoulder dystocia, large-for-gestational age (LGA, defined as birthweight > 90th percentile), small-for-gestational age (SGA, defined as birthweight <10th percentile), neonatal demise, severe neonatal hypoglycemia, and severe hyperbilirubinemia. Severe neonatal hypoglycemia was defined as hypoglycemia requiring management with dextrose infusions in infants born ≥ 34 0/7 weeks. Severe hyperbilirubinemia was defined as receipt of phototherapy during a NICU admission or high-intensity mattress phototherapy in the well-baby nursery. These outcomes were selected to maximize coverage of the core outcome set for GDM recommended by a Delphi consensus [25]. All analyses of secondary outcomes were considered exploratory and hypothesis-generating. This study was approved by our local Institutional Review Board (IRB# 11–07364, renewed July 20, 2023).

Demographic, obstetric, and neonatal details were obtained from a research database of all patients who delivered at this tertiary academic institution. This database is populated after each birth by the health care professionals involved in the delivery and obstetric and neonatal hospital course and regularly validated. GTT values and test dates were extracted from the electronic medical record (EMR) by an informatics specialist. Additional missing variables were collected via manual chart review. Covariates of interest included maternal age, early body mass index (BMI), self-reported race and ethnicity, chronic hypertension, and history of prior CD. Early BMI was defined as a BMI taken within 6 months before conception or within the first trimester as calculated based on EDD. Despite its inherent limitations, self-reported race and ethnicity was included as a covariate and an imperfect proxy for sociodemographic factors for health disparities, as previous studies have reported ethnic differences in perinatal outcomes [26].

Baseline demographics were compared between the before and after cohorts using chi-square and Fisher’s exact tests. To evaluate the effect size for these baseline demographics, Cohen’s d was used to describe standardized mean differences for continuous variables (such as age and early BMI) and Cohen’s h was calculated to describe the effect size for proportions (such as percent nulliparous or with a history of chronic hypertension). The risk ratios for binary maternal and neonatal outcomes were computed using binomial regression. Continuous neonatal outcomes (gestational age at delivery, birthweight) were compared between cohorts using the Kruskal-Wallis test. An a priori directed acyclic graph (DAG) was used to understand the relationship between the exposure (method of GDM screening), the primary outcomes, and typical covariates (including maternal age, BMI, and self-reported race/ethnicity) (Appendix 2). Adjustment for these covariates was not performed as none were identified by the DAG as confounders between the exposure and the outcome. In fact, time of conception was the only variable affecting exposure. Preterm birth, severe hyperbilirubinemia, severe hypoglycemia, and respiratory distress syndrome were considered mediators of NICU admission, and accordingly were not adjusted for when assessing the effect of the exposure (screening strategy) on this outcome (NICU admission).

We chose to conduct primary analyses in the general pregnant population in order to evaluate the intervention at a population level. We also performed a secondary analysis in the population of individuals who had isolated fasting hyperglycemia on the 3-hour GTT, referred to as the “intermediate” risk group. In post-hoc analyses, Lowess plots and interrupted time-series analyses were performed as part of exploratory investigations to identify potential trends present in the specified primary outcomes that may have overwhelmed any effect of the screening expansion. Post hoc power analyses using the Fleiss-Tytun-Ury continuity correction were performed to determine whether our sample sizes were sufficient to detect clinically significant differences in the outcomes reported in our manuscript.[27] In the general population, the final sample size had an 80% power to detect a 15% reduction (RR 0.85) in the incidence of primary CD and 17% reduction in incidence of NICU admission (RR 0.83). In the “intermediate” risk group, the final sample size an 80% power to detect a 67% reduction (RR 0.33) in the incidence of primary CD and a 280% increase in the incidence of NICU admission (lacking 80% power to detect a decrease in NICU admission).

All analyses were performed using Stata 17.0 (StataCorp LLC, College Station, TX). The results of this study have been reported in accordance with STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines.

Results

A total of 8,013 individuals met inclusion criteria, with n=1,733 in the before cohort and n=6,280 in the after cohort (Appendix 3). In the general population, there were statistically significant differences in maternal age (p<0.01), early BMI (p<0.01), self-reported race/ethnicity (p<0.01), and gestational age at initial glucose loading test screening (p<0.01) with individuals in the “after” cohort likelier to be older, heavier, screened half a week later, and less likely to be non-Hispanic White (Table 1). For all of these baseline demographic variables, the effect size was less than 0.1, suggesting that the statistically significant differences between the two groups were clinically negligible. There were no differences in history of chronic hypertension or prior CD.

Table 1.

Baseline characteristics of pregnant individuals before and after implementation of an expanded screening protocol for gestational diabetes (GDM)

Cohort 1 (Standard Screening) Cohort 2
(Expanded Screening)		 p value
(test)		 Effect size (test)
General Population n=1,733 n=6,280
Maternal age (years) 34.8 (32.5,37.6) 35.4 (32.8,38.1) <0.01
(H-test)		 0.09 (Cohen’s d)
Early BMI (kg/m2)a 22.9 (20.9,25.9) 23.3 (21.2,26.5) <0.01
(H-test)		 0.08 (Cohen’s d)
Self-reported race/ethnicity
 Asian
 Hispanic
 Non-Hispanic Black
 Non-Hispanic White
 None of the above
481 (27.8%)
114 (6.6%)
51 (2.9%)
836 (48.2%)
251 (14.5%)
1615 (25.7%)
408 (6.5%)
249 (4.0%)
2463 (39.2%)
1,545 (25.6%)		 <0.01
(Chi)
Chronic hypertension 127 (7.3%) 484 (7.7%) 0.60
(Chi)		 0.01 (Cohen’s h)
Nulliparous 1,023 (59.0%) 3,532 (56.2%) 0.04
(Chi)		 −0.06 (Cohen’s h)
History of prior cesarean 178 (10.3%) 691 (11.0%) 0.39
(Chi)		 0.02 (Cohen’s h)
Gestational age (weeks) at glucose loading test screening 25.9 (24.9, 27.0) 26.4 (25.1, 27.4) < 0.01
(H-test)		 0.02 (Cohen’s d)
Subpopulation of intermediate risk individuals n=72 n=161
Maternal age (years) 36.5 (33.9,38.8) 36.2 (33.2,38.9) 0.84
(H-test)		 0.02 (Cohen’s d)
Early BMI (kg/m2)a 24.9 (21.8,29.9) 26.3 (22.8,29.6) 0.29
(H-test)		 −0.07 (Cohen’s d)
Self-reported race/ethnicity
 Asian
 Hispanic
 Non-Hispanic Black
 Non-Hispanic White
 None of the above
22 (30.6%)
3 (4.2%)
1 (1.4%)
34 (47.2%)
12 (16.7%)
60 (37.3%)
9 (5.6%)
3 (1.9%)
52 (32.3%)
37 (23.0%)		 0.30
(Fisher’s exact)
Chronic hypertension 7 (9.7%) 18 (11.2%) 0.74
(Chi-square)		 0.05 (Cohen’s h)
Nulliparous 42 (58.3%) 84 (52.2%) 0.38
(Chi-square)		 −0.12 (Cohen’s h)
History of prior cesarean 8 (11.1%) 23 (14.3%) 0.51
(Chi-square)		 0.06 (Cohen’s h)
Gestational age (weeks) at glucose loading test 26.0 (24.4, 27.0) 25.6 (19.6, 26.9) 0.32
(H-test)		 0.27 (Cohen’s d)
Open in a new tab

Data presented as number (percentage) or median (interquartile range).

Maternal age, early BMI, and gestational age and glucose loading test screen were compared using the Kruskal-Wallis (H-test).

Self-reported race/ethnicity, proportion of patients with chronic hypertension, nulliparity, and history of prior cesarean section were compared using chi-square and Fisher’s exact tests.

BMI, body mass index.

a

BMI defined as pre-pregnancy or first-trimester BMI.

In the general population, there was no difference in the risk of PCD (20.4% vs 18.4%, RR 0.9, 95% CI, 0.81 – 1.01) or NICU admission (16.7% vs 18.6%, RR 1.11, 95% CI 0.99 – 1.25) between the two cohorts.

For the secondary outcomes in the general population, the incidence of A2GDM significantly increased after implementation of expanded screening (4.8% vs. 6.4%, RR 1.34, 95% CI 1.07–1.69). There was a trend towards increased HDP (18.5% vs 20.6%, RR 1.11, 95% CI 1.00–1.24), with a statistically significant increase in preeclampsia risk (8.7% vs 10.4%, RR 1.20, 95% CI 1.01–1.42). There were statistically significant but clinically insignificant decreases in gestational age at delivery (median 39.4 weeks vs 39.3 weeks, p<0.01) and in birthweight (3315 grams vs 3295 grams, p=0.04) after implementation of the expanded protocol. Risk of PTB and incidence of severe hyperbilirubinemia significantly increased (6.9% vs 8.4%, RR 1.22, 95% CI 1.01–1.47 and 5.2% vs 7.7%, RR 1.49, 95% CI 1.20–1.85, respectively). There were no statistically significant differences in risk of other maternal or neonatal outcomes (Table 2).

Table 2.

Outcomes among pregnant individuals before and after implementation of an expanded screening protocol for gestational diabetes (GDM)

Cohort 1 (Standard Screening)
(n=1,733)		 Cohort 2
(Expanded Screening)
(n=6,280)		 p value Relative Risk (95% CI)
Maternal outcomes
Gestational diabetes mellitusa 191 (11.0%) 748 (11.9%) 0.31 1.08 (0.93 – 1.26)
Treatment with insulina 83 (4.8%) 404 (6.4%) 0.01 1.34 (1.07 – 1.69)
Hypertensive disease of pregnancya 320 (18.5%) 1,292 (20.6%) 0.055 1.11 (1 – 1.24)
Gestational hypertension 169 (9.8%) 638 (10.2%) 0.62 1.04 (0.89 – 1.22)
Pre-eclampsia 151 (8.7%) 655 (10.4%) 0.04 1.2 (1.01 – 1.42)
Mode of deliverya
 Spontaneous VD 1126 (65.0%) 4177 (66.5%) 0.24 1.02 (0.98 – 1.06)
 Operative VD 138 (8.0%) 460 (7.3%) 0.37 0.92 (0.77 – 1.1)
 Cesarean 469 (27.1%) 1643 (26.2%) 0.45 0.97 (0.89 – 1.06)
Primary CD 353 (20.4%) 1157 (18.4%) 0.07 0.9 (0.81 – 1.01)
Obstetric anal sphincter injuries 47 (2.7%) 212 (3.4%) 0.17 1.24 (0.91 – 1.7)
Neonatal outcomes
NICU admission 290 (16.7%) 1166 (18.6%) 0.08 1.11 (0.99 – 1.25)
Gestational age at delivery (weeks)a 39.4 (38.6,40.4) 39.3 (38.4,40.1) <0.01
Preterm birtha,b 120 (6.9%) 529 (8.4%) 0.04 1.22 (1.01 – 1.47)
Shoulder dystocia 17 (1.0%) 68 (1.1%) 0.71 1.1 (0.65 – 1.87)
Birthweight (grams)a 3315 (3000,3620) 3295 (2965,3615) 0.04
Large for gestational age a,c 137 (7.9%) 509 (8.1%) 0.79 1.03 (0.86 – 1.23)
Neonatal death a 1 (0.1%) 2 (0.0%) 0.63 0.55 (0.05 – 6.08)
Severe neonatal hypoglycemia a,d 25 (1.4%) 132 (2.1%) 0.08 1.46 (0.95 – 2.23)
Severe hyperbilirubinemiae 90 (5.2%) 485 (7.7%) <0.01 1.49 (1.2 – 1.85)
Respiratory distress syndrome 21 (1.2%) 89 (1.4%) 0.52 1.17 (0.73 – 1.88)
Open in a new tab

Data presented as number (percentage) or median (interquartile range).

VD, vaginal delivery; CD, cesarean delivery; NICU, neonatal intensive care unit.

Categorical variables were compared using binomial regression.

Continuous variables (gestational age at delivery, birthweight) were compared using Kruskal-Wallis tests (H-test)

a

GDM core outcome set

b

Preterm birth defined as birth <37 0/7 weeks

c

Large for gestational age defined as birthweight ≥ 90th percentile

d

Severe neonatal hypoglycemia defined as hypoglycemia requiring management with IV dextrose

e

Severe hyperbilirubinemia defined as hyperbilirubinemia requiring phototherapy and/or exchange transfusion

As for the results of our pre-specified subgroup analysis: of the 8,013 individuals included in the primary analysis, a total of 233 were part of the “intermediate” subgroup with isolated fasting hyperglycemia, with n=72 in the “before” cohort (who received routine prenatal care following their 3-hour GTT results) and n=161 individuals in the “after” cohort (who were instructed to perform SMBG for 1–2 weeks following their 3-hour GTT results, with the potential for GDM diagnosis based on those values). There was no significant difference in baseline characteristics between the two groups in this subpopulation, and the effect size for all characteristics was < 0.15, with the exception of gestational age at glucose screening (Cohen’s d 0.27). (Table 1). There was a significant increase in the risk of any GDM and A2GDM in the “after” cohort (12.5 vs. 46.0%, RR 3.68, 95% CI 1.95 – 6.93, and 8.3% vs. 37.3%, RR 4.47, 95% CI 2.03–9.87, respectively).

The prevalence of GDM in the intermediate “before” cohort was non-zero. There were still patients in the “before” cohort with isolated fasting hyperglycemia on initial evaluation who technically would not be diagnosed with GDM based on standard two-step screening, who ended up with GDM diagnoses. On post-hoc review of the data, these diagnoses were made based on repeat GTT in the setting of excessive fetal growth identified on obstetric ultrasound and/or by elective self-monitoring of blood glucose at home by patients with prior history of GDM and/or anxiety regarding inaccurate GTT. The remainder of the 63 patients with isolated fasting hyperglycemia in the “before” cohort were not diagnosed with and not managed as patients with GDM.

There was not a significant change in risk of PCD (23.6% vs 20.5%, RR 0.87, 95% CI 0.52–1.45), but there was a significant increase in risk of NICU admission in the “after” cohort (8.3% vs 22.4%, RR 2.68, 95% CI 1.18–6.08). The most common indication for NICU admission in the “before” cohort was respiratory distress, whereas the most common indication for NICU admission in the “after” cohort was treatment of hyperbilirubinemia. Overall, there was a significant increase in risk of severe hyperbilirubinemia in the “after” cohort of this “intermediate” group (1.4% to 10.6%, RR 7.6, 95% CI 1.03–56.04). There were no significant differences in any other secondary neonatal or maternal outcomes (Table 3).

Table 3.

Outcomes among subpopulation of intermediate risk individuals screened for gestational diabetes (GDM) before and after implementation of an expanded screening protocol

Cohort 1 (Standard Screening)
(n=72)		 Cohort 2
(Expanded Screening)
(n=161)		 p value Relative Risk (95% CI)
Maternal outcomes
Gestational diabetes mellitusa 9 (12.5%) 74 (46.0%) <0.001 3.68 (1.95 – 6.93)
Treatment with insulina 6 (8.3%) 60 (37.3%) <0.001 4.47 (2.03 – 9.87)
Hypertensive disease of pregnancya 19 (26.4%) 40 (24.8%) 0.80 0.94 (0.59 – 1.51)
Gestational hypertension 8 (11.1%) 15 (9.3%) 0.67 0.84 (0.37 – 1.89)
Pre-eclampsia 11 (15.3%) 25 (15.5%) 0.96 1.02 (0.53 – 1.95)
Mode of deliverya
 Spontaneous VD 44 (61.1%) 107 (66.5%) 1.35 1.09 (0.44 – 0.88)
 Operative VD 6 (8.3%) 6 (3.7%) 0.44 0.45 (0.15 – 0.15)
 Cesarean 22 (30.6%) 48 (29.8%) 0.91 0.98 (0.64 – 1.49)
Primary CD 17 (23.6%) 33 (20.5%) 0.59 0.87 (0.52 – 1.45)
Obstetric anal sphincter injuries 0 (0%) 1 (0.6%) -- --
Neonatal outcomes
NICU admission 6 (8.3%) 36 (22.4%) 0.02 2.68 (1.18 – 6.08)
Gestational age at delivery (weeks)a 39.4 (38.4,40.1) 39.3 (38.4,39.7) 0.20
Preterm birtha,b 4 (5.6%) 12 (7.5%) 0.6 1.34 (0.45 – 4.02)
Shoulder dystocia 0 (0%) 3 (1.9%) --- --
Birthweight (grams)a 3350 (3055, 3683) 3390 (3065,3685) 0.17
Large for gestational age a,c 6 (8.3%) 20 (12.4%) 0.37 1.49 (0.63 – 3.55)
Neonatal death a 0 (0%) 0 (0%) --- --
Severe neonatal hypoglycemia a,d 0 (0%) 3 (1.9%) --- --
Severe hyperbilirubinemiae 1 (1.4%) 17 (10.6%) 0.05 7.6 (1.03 – 56.04)
Respiratory distress syndrome 0 (0%) 4 (2.5%) --- --
Open in a new tab

Data presented as number (percentage) or median (interquartile range). The relative risk of some secondary outcomes could not be computed using binomial regression due to complete collinearity with cohort assignment.

VD, vaginal delivery; CD, cesarean delivery; NICU, neonatal intensive care unit.

Categorical variables were compared using binomial regression.

Continuous variables (gestational age at delivery, birthweight) were compared using Kruskal-Wallis tests (H-test)

a

GDM core outcome set

b

Preterm birth defined as birth <37 0/7 weeks

c

Large for gestational age defined as birthweight ≥ 90th percentile

d

Severe neonatal hypoglycemia defined as hypoglycemia requiring management with IV dextrose

e

Severe hyperbilirubinemia defined as hyperbilirubinemia requiring phototherapy and/or exchange transfusion

In our post-hoc Lowess plots and interrupted time-series analyses (Appendices 4–7), we did not see any overtly overwhelming trends. However, there were trends towards decreasing primary cesarean delivery and NICU admission rates that preceded screening expansion.

Discussion:

Principal findings

Our study examined the effect of changing GDM screening from the standard two-step approach to an expanded approach that includes home SMBG for those with isolated fasting hyperglycemia on the 3-hour GTT. This expanded screening strategy was associated with a significantly increased risk of A2GDM but no change in risk of PCD or NICU admission in the general population. In a subgroup analysis restricted to patients with isolated fasting hyperglycemia on the 3-hour GTT, the expanded screening strategy was associated with significant increases in the overall relative risk of GDM, A2GDM, NICU admission, and severe neonatal hyperbilirubinemia.

Comparison with existing literature

Finding an increased risk of A2GDM in the general population and intermediate subgroup was unsurprising, as the expanded strategy targets patients with isolated fasting hyperglycemia, and fasting hyperglycemia is a known indicator of the need for pharmacological intervention for GDM [28–30]. This expanded screening strategy was an institutional change that was implemented based on the assumption that it would capture more individuals at risk of adverse perinatal outcomes can be diagnosed with and treated for GDM. Multiple studies have suggested that fasting hyperglycemia is a predictor for LGA neonates and that treating fasting hyperglycemia results in decreased risk of LGA neonates and better detection of newborns with neonatal hypoglycemia [14,26,29,31–35]. If treating fasting hyperglycemia routinely reduces the risk of LGA, we would have expected a reduction in risk of LGA in the “after” cohort compared to the “before” cohort, since those in the “after” cohort were more likely to be identified as having GDM and more likely to have their fasting hyperglycemia treated. However, we found no difference in risk of LGA between the two cohorts. This may be explained by prior findings that there are residual risk associations between GDM and adverse perinatal outcomes, even with adequate treatment of hyperglycemia [36].

While we were surprised to find that the relative risk of adverse perinatal outcomes did not decrease with an expanded GDM screening approach, this is consistent with prior investigations that have shown that general increases in GDM incidence do not necessarily correlate with improved outcomes and may be associated with increased healthcare costs and inefficiencies [7–10,37–40].

In the intermediate subgroup, we found a significant increase in risk of NICU admission in the “after” cohort, driven largely by an increase in hyperbilirubinemia requiring phototherapy. Hyperbilirubinemia is a known risk of poorly controlled diabetes in pregnancy, and if treatment of hyperglycemia lowers the likelihood of poorly controlled diabetes, then capturing and treating the intermediate group should have treated more people and decreased the likelihood of hyperbilirubinemia [41]. In addition, the aforementioned narrow increases in phototherapy thresholds as per new 2022 AAP guidelines would be expected to decrease phototherapy administration, rather than increase [23]. A systematic meta-analysis of 7 randomized controlled trials comparing 1-step and 2-step screening recently found that in addition to the increased incidence of GDM diagnoses with 1-step screening, 1-step screening was also associated with increased rates of NICU admission and neonatal hypoglycemia, demonstrating that that increased GDM diagnoses have previously been correlated to increased rates of adverse events rather than improved outcomes [40].

On exploratory analyses of secondary outcomes, we also found an increase in the risk of preeclampsia and PTB in the general population. It is unlikely that these trends were driven by increased identification and treatment of GDM – if anything, treatment of hyperglycemia would be expected to lower these risks among patients with diabetes in pregnancy, rather than increase them. Statistically, patients in the “after” cohort were older, heavier, and less likely to be non-Hispanic White patients – all factors that are known to be associated with a higher risk of preeclampsia and PTB [42–44]. Another likely explanation includes other secular trends, namely the COVID-19 pandemic. COVID-19 infection in pregnancy is associated with a higher risk of neonatal hyperbilirubinemia and an increase in neonatal readmissions for phototherapy, possibly secondary to vascular malperfusion, compensatory polycythemia, or reduced lactation support [45,46]. Previous studies have also shown an association between COVID-19 in pregnancy and increased risks of preeclampsia and PTB [47–50].

Clinical Implications

Although this study did not directly examine healthcare utilization, we can assume that healthcare utilization was higher in the “after” cohort, since more patients were diagnosed with GDM. For providers, this likely contributed to increased workload with review of additional glucose logs. For patients, this likely contributed to higher healthcare burden than would be expected with standard GDM criteria, including SMBG and dietary logging, nonstress tests and obstetric ultrasounds for fetal surveillance, and pharmacologic treatments for blood glucose control. Despite this presumed increase in monitoring and healthcare services, we found no evidence of maternal or neonatal benefits and thus would not recommend continued use of this approach in the clinic.

Research Implications

A detailed investigation of the costs of the assumed increases in healthcare utilization would be helpful in refining our understanding of the cost-effectiveness of this screening strategy. Qualitative studies would be required to understand the experience of patients whose GDM screening was supplemented with SMBG. Our study revealed statistically significant increases in secondary neonatal outcomes such as severe hyperbilirubinemia and preterm birth, and additional studies would be required to investigate hypotheses related to these secondary outcomes. Finally, further investigation is required to identify the optimal GDM screening strategy that balances the risks of overdiagnosis with the beneficial reductions in perinatal morbidity availed by appropriate GDM treatment- such future studies may include assessing risk or treatment benefits for patients with a single abnormal value on the 3-hour GTT, or a different threshold for treating patients with isolated fasting hyperglycemia as patients with GDM.

Strengths and Limitations

Our study is limited by its retrospective nature and reliance on data collected within the electronic medical record. It was also conducted at a single tertiary care academic institution, which also inherently limits generalizability. Our GDM screening strategy was expanded in March 2019, while the first stay-at-home order was issued in California in March 2020. Although there were no official changes to the care offered to patients with GDM with the onset of the COVID-19 pandemic, this retrospective analysis had limited ability to account for the unmeasured effects of COVID-19, including the impact of social distancing, economic trends, changes in daily activities, trends in gestational weight gain, access to and utilization of healthcare, and the consequences of COVID-19 infections. When reviewing post-hoc Lowess plots visualizing general trends in our two primary outcomes, we did not see apparent trends correlating with the onset of COVID-19. Our post-hoc interrupted time-series analyses also suggested there were trends towards decreased PCD and NICU admission rates that preceded screening expansion.

Although our study included more than 8,000 patients in the overall population, the intermediate risk subgroup was smaller in size, subjecting our subgroup analyses to limitations related to underpowering, including the possibility of a type II error. This is reflected in our post-hoc power analyses in the intermediate risk population, which revealed an 80% power to detect a 67% reduction (RR 0.33) in the incidence of primary CD and a 280% increase in the incidence of NICU admission. This small sample size also meant that the relative risk of some secondary outcomes could not be computed using binomial regression due to complete collinearity with cohort assignment, further limiting investigation. Notably, although it has been reported that even a single abnormal GTT value can portend long-term glucose intolerance and metabolic disruption, our study looked only at immediate perinatal outcomes, and insight into long-term metabolic dysfunction in individuals with isolated elevated fasting glucose values treated for GDM is remains poorly characterized [19]. Our study also excludes people who were screened at other institutions or who did not receive screening, and likely disproportionately excludes people with limited prenatal care from our study.

Despite these limitations, our study has multiple strengths. The large size of the general population allowed for adequate power to detect clinically significant differences in the primary outcomes, and as described in methods, post-hoc sample size calculations revealed an 80% power to detect a 15% reduction (RR 0.85) in the incidence of primary CD and 17% reduction in incidence of NICU admission (RR 0.83). Our primary and secondary outcomes were drawn from an expert recommended core outcome set for GDM [25]. We contribute to the literature results about the clinical utility of basing treatment decisions on this contested, isolated measure of abnormal glucose tolerance.

Conclusions

In conclusion, a useful screening strategy should reduce morbidity in the population in which it is applied. Although previous literature had suggested that fasting hyperglycemia is a predictor of perinatal adverse outcomes, a GDM strategy that expanded the standard two-step screening process to capture more individuals with fasting hyperglycemia did not lead to maternal or neonatal health benefits. Given the expected increase in healthcare utilization and burden with this screening strategy, we would not recommend widespread implementation of SMBG as a screening supplement to selectively target patients with isolated fasting hyperglycemia. This strategy was discontinued in our clinic after reviewing the results of our study.

Supplementary Material

Supp 1

Acknowledgements

We thank Miriam Kuppermann, PhD MPH; Thomas Hoffmann, PhD; and Chuck McCulloch, PhD, MS, MA for lending their expertise in epidemiology and biostatistics. We thank Dr. Ilse Larsen for providing her expertise on neonatal management.

Funding

No external funding sources. This project was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through UCSF-CTSI Grant Number UL1 TR001872 and TR 001871. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.

Footnotes

Declaration of Interest Statement:

The authors report no conflicts of interest.

Data availability statement

The data that support the findings of this study are available from the corresponding author, SJS, upon reasonable request.

References

  • 1.Behboudi-Gandevani S, Amiri M, Bidhendi Yarandi R, Ramezani Tehrani F. The impact of diagnostic criteria for gestational diabetes on its prevalence: a systematic review and meta-analysis. Diabetol Metab Syndr 2019;11:11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Guariguata L, Linnenkamp U, Beagley J, Whiting DR, Cho NH. Global estimates of the prevalence of hyperglycaemia in pregnancy. Diabetes Res Clin Pract 2014;103(2):176–185. [DOI] [PubMed] [Google Scholar]
  • 3.Modzelewski R, Stefanowicz-Rutkowska MM, Matuszewski W, Bandurska-Stankiewicz EM. Gestational Diabetes Mellitus-Recent Literature Review. J Clin Med 2022;11(19). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Billionnet C, Mitanchez D, Weill A, et al. Gestational diabetes and adverse perinatal outcomes from 716,152 births in France in 2012. Diabetologia 2017;60(4):636–644. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Ye W, Luo C, Huang J, Li C, Liu Z, Liu F. Gestational diabetes mellitus and adverse pregnancy outcomes: systematic review and meta-analysis. BMJ 2022;377:e067946. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Crowther CA, Hiller JE, Moss JR, et al. Effect of treatment of gestational diabetes mellitus on pregnancy outcomes. N Engl J Med 2005;352(24):2477–2486. [DOI] [PubMed] [Google Scholar]
  • 7.Ghaffari N, Gonzalez JM, Rosenstein MG. Does the 1-step method of gestational diabetes mellitus screening improve pregnancy outcomes? Am J Obstet Gynecol MFM 2020;2(4):100199. [DOI] [PubMed] [Google Scholar]
  • 8.Hillier TA, Pedula KL, Ogasawara KK, et al. A Pragmatic, Randomized Clinical Trial of Gestational Diabetes Screening. N Engl J Med 2021;384(10):895–904. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Davis EM, Abebe KZ, Simhan HN, et al. Perinatal Outcomes of Two Screening Strategies for Gestational Diabetes Mellitus: A Randomized Controlled Trial. Obstet Gynecol 2021;138(1):6–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Fitria N, van Asselt ADI, Postma MJ. Cost-effectiveness of controlling gestational diabetes mellitus: a systematic review. Eur J Health Econ 2019;20(3):407–417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.National Institutes of Health Consensus Development Conference Statement: Diagnosing Gestational Diabetes Mellitus, March 4–6, 2013. Obstetrics & Gynecology. 2013;122(2 PART 1):358–369. [DOI] [PubMed] [Google Scholar]
  • 12.ACOG Practice Bulletin No. 190: Gestational Diabetes Mellitus. Obstet Gynecol 2018;131(2):e49–e64. [DOI] [PubMed] [Google Scholar]
  • 13.American Diabetes A. 2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2021. Diabetes Care. 2021;44(Suppl 1):S15–S33. [DOI] [PubMed] [Google Scholar]
  • 14.Roeckner JT, Sanchez-Ramos L, Jijon-Knupp R, Kaunitz AM. Single abnormal value on 3-hour oral glucose tolerance test during pregnancy is associated with adverse maternal and neonatal outcomes: a systematic review and metaanalysis. Am J Obstet Gynecol 2016;215(3):287–297. [DOI] [PubMed] [Google Scholar]
  • 15.Retnakaran R, Qi Y, Sermer M, Connelly PW, Zinman B, Hanley AJ. Isolated hyperglycemia at 1 hour on oral glucose tolerance test in pregnancy resembles gestational diabetes mellitus in predicting postpartum metabolic dysfunction. Diabetes Care. 2008;31(7):1275–1281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Koning SH, van Zanden JJ, Hoogenberg K, et al. New diagnostic criteria for gestational diabetes mellitus and their impact on the number of diagnoses and pregnancy outcomes. Diabetologia 2018;61(4):800–809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Salman L, Pardo A, Krispin E, Oron G, Toledano Y, Hadar E. Perinatal outcome in gestational diabetes according to different diagnostic criteria. J Perinat Med 2019;47(5):553–557. [DOI] [PubMed] [Google Scholar]
  • 18.Shah BR, Sharifi F. Perinatal outcomes for untreated women with gestational diabetes by IADPSG criteria: a population-based study. BJOG 2020;127(1):116–122. [DOI] [PubMed] [Google Scholar]
  • 19.Kautzky-Willer A, Bancher-Todesca D, Weitgasser R, et al. The impact of risk factors and more stringent diagnostic criteria of gestational diabetes on outcomes in central European women. J Clin Endocrinol Metab 2008;93(5):1689–1695. [DOI] [PubMed] [Google Scholar]
  • 20.Ardilouze A, Bouchard P, Hivert MF, et al. Self-Monitoring of Blood Glucose: A Complementary Method Beyond the Oral Glucose Tolerance Test to Identify Hyperglycemia During Pregnancy. Can J Diabetes. 2019;43(8):627–635. [DOI] [PubMed] [Google Scholar]
  • 21.Allard C, Sahyouni E, Menard J, et al. Gestational diabetes mellitus identification based on self-monitoring of blood glucose. Can J Diabetes. 2015;39(2):162–168. [DOI] [PubMed] [Google Scholar]
  • 22.Cahill C, Jegatheesan P, Song D, et al. Implementing Higher Phototherapy Thresholds for Jaundice in Healthy Infants 35 Plus Weeks. Hosp Pediatr 2023;13(9):857–864. [DOI] [PubMed] [Google Scholar]
  • 23.Kemper AR, Newman TB, Slaughter JL, et al. Clinical Practice Guideline Revision: Management of Hyperbilirubinemia in the Newborn Infant 35 or More Weeks of Gestation. Pediatrics. 2022;150(3). [DOI] [PubMed] [Google Scholar]
  • 24.Landman G, Hoffman K, Sun Y, et al. Northern California Neonatal Consortium. NCNC hyperbilirubinemia treatment guideline. https://bchsfoutreach.ucsf.edu/sites/bchsfoutreach.ucsf.edu/files/Neo%20Hyperbilirubinemia%20Guideline%20FINAL%20APRROVED%204.2018.pdf. Updated 11/2017. Accessed 6/8/2024.
  • 25.Egan AM, Bogdanet D, Griffin TP, et al. A core outcome set for studies of gestational diabetes mellitus prevention and treatment. Diabetologia 2020;63(6):1120–1127. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Farrar D, Fairley L, Santorelli G, et al. Association between hyperglycaemia and adverse perinatal outcomes in south Asian and white British women: analysis of data from the Born in Bradford cohort. Lancet Diabetes Endocrinol 2015;3(10):795–804. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Kohn MA, Senyak J. Sample Size Calculators. UCSF CTSI. https://www.sample-size.net/ Published 2025. Accessed 27 January 2025. [Google Scholar]
  • 28.Rodrigo N, Randall D, Al-Hial FA, Pak KLM, Kim AJ, Glastras SJ. Fasting Glucose Level on the Oral Glucose Tolerance Test Is Associated with the Need for Pharmacotherapy in Gestational Diabetes Mellitus. Nutrients. 2023;15(5). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Kotzaeridi G, Blatter J, Eppel D, et al. Characteristics of gestational diabetes subtypes classified by oral glucose tolerance test values. Eur J Clin Invest 2021;51(9):e13628. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Kalok A, Ong MY, Hasrori A, et al. Correlation between Oral Glucose Tolerance Test Abnormalities and Adverse Pregnancy Outcomes in Gestational Diabetes: A Cross-Sectional Study. Int J Environ Res Public Health. 2020;17(19). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Black MH, Sacks DA, Xiang AH, Lawrence JM. Clinical outcomes of pregnancies complicated by mild gestational diabetes mellitus differ by combinations of abnormal oral glucose tolerance test values. Diabetes Care. 2010;33(12):2524–2530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Berglin M, Anderson M, Weintraub M, et al. 685 Impaired fasting glucose in pregnancy: does treatment improve perinatal outcomes? American Journal of Obstetrics & Gynecology. 2021;224(2):S430. [Google Scholar]
  • 33.Ryan EA, Savu A, Yeung RO, Moore LE, Bowker SL, Kaul P. Elevated fasting vs post-load glucose levels and pregnancy outcomes in gestational diabetes: a population-based study. Diabet Med 2020;37(1):114–122. [DOI] [PubMed] [Google Scholar]
  • 34.Uvena-Celebrezze J, Fung C, Thomas AJ, et al. Relationship of neonatal body composition to maternal glucose control in women with gestational diabetes mellitus. J Matern Fetal Neonatal Med 2002;12(6):396–401. [DOI] [PubMed] [Google Scholar]
  • 35.Farrar D, Simmonds M, Bryant M, et al. Hyperglycaemia and risk of adverse perinatal outcomes: systematic review and meta-analysis. BMJ 2016;354:i4694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Li K, Li X, Morse AN, et al. Residual risk associations between initial hyperglycemia and adverse pregnancy outcomes in a large cohort including 6709 women with gestational diabetes. Diabetes Metab 2022;48(1):101320. [DOI] [PubMed] [Google Scholar]
  • 37.Aubry EM, Raio L, Oelhafen S. Effect of the IADPSG screening strategy for gestational diabetes on perinatal outcomes in Switzerland. Diabetes Res Clin Pract 2021;175:108830. [DOI] [PubMed] [Google Scholar]
  • 38.Farrar D, Simmonds M, Griffin S, et al. The identification and treatment of women with hyperglycaemia in pregnancy: an analysis of individual participant data, systematic reviews, meta-analyses and an economic evaluation. Health Technol Assess. 2016;20(86):1–348. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Crowther CA, Samuel D, McCowan LME, et al. Lower versus Higher Glycemic Criteria for Diagnosis of Gestational Diabetes. N Engl J Med 2022;387(7):587–598. [DOI] [PubMed] [Google Scholar]
  • 40.Gomes C, Futterman ID, Sher O, et al. One-step vs 2-step gestational diabetes mellitus screening and pregnancy outcomes: an updated systematic review and meta-analysis. Am J Obstet Gynecol MFM 2024;6(5):101346. [DOI] [PubMed] [Google Scholar]
  • 41.Group HSCR, Metzger BE, Lowe LP, et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008;358(19):1991–2002. [DOI] [PubMed] [Google Scholar]
  • 42.Tyas BD, Lestari P, Aldika Akbar MI. Maternal Perinatal Outcomes Related to Advanced Maternal Age in Preeclampsia Pregnant Women. J Family Reprod Health. 2019;13(4):191–200. [PMC free article] [PubMed] [Google Scholar]
  • 43.Cnattingius S, Villamor E, Johansson S, et al. Maternal obesity and risk of preterm delivery. JAMA 2013;309(22):2362–2370. [DOI] [PubMed] [Google Scholar]
  • 44.Manuck TA. Racial and ethnic differences in preterm birth: A complex, multifactorial problem. Semin Perinatol 2017;41(8):511–518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Monzani A, Remorgida V, Rabbone I. High rates of neonatal polycythemia and hyperbilirubinemia during the first phase of COVID-19 pandemic in Italy: a single-center experience. Ital J Pediatr 2022;48(1):100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Mokha S, Heintz C, Agrawal V, Reynolds AM, Berkelhamer SK. Increased prevalence of severe neonatal hyperbilirubinaemia during social distancing. J Paediatr Child Health. 2022;58(6):991–995. [DOI] [PubMed] [Google Scholar]
  • 47.Papageorghiou AT, Deruelle P, Gunier RB, et al. Preeclampsia and COVID-19: results from the INTERCOVID prospective longitudinal study. Am J Obstet Gynecol 2021;225(3):289 e281–289 e217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Lai J, Romero R, Tarca AL, et al. SARS-CoV-2 and the subsequent development of preeclampsia and preterm birth: evidence of a dose-response relationship supporting causality. Am J Obstet Gynecol 2021;225(6):689–693 e681. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Karasek D, Baer RJ, McLemore MR, et al. The association of COVID-19 infection in pregnancy with preterm birth: A retrospective cohort study in California. Lancet Reg Health Am 2021;2:100027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Main EK, Chang SC, Carpenter AM, et al. Singleton preterm birth rates for racial and ethnic groups during the coronavirus disease 2019 pandemic in California. Am J Obstet Gynecol 2021;224(2):239–241. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp 1
NIHMS2058510-supplement-Supp_1.docx (1.2MB, docx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, SJS, upon reasonable request.

RESOURCES