Early Diabetes Screening in Pregnancy: Challenges and Opportunities

Early screening for diabetes in pregnancy is gaining attention as a strategy to improve maternal and child health outcomes. In this interview with The American Journal of Managed Care® (AJMC®), Ellen Francis, PhD, assistant professor of biostatistics and epidemiology at Rutgers School of Public Health, explores the potential of earlier screening, the evidence supporting its benefits, and what health systems need to consider when implementing.

This transcript has been lightly edited for clarity and length.

AJMC: Which key factors should be prioritized when considering earlier screening for diabetes in pregnancy, and how do these markers contribute to better long-term health outcomes for the mother and possibly the child?

Ellen Francis, PhD | Image credit: Rutgers School of Public Health

Francis: There's been massive interest in how to treat, screen, and diagnose diabetes in pregnancy for decades. And I think one of the more important things to start off distinguishing is diabetes that was existing prior to pregnancy and then gestational diabetes (GDM), because they're 2 different scenarios.

For diabetes in pregnancy, there's a lot of consensus on using HbA1C [hemoglobin A1C] as a good screener in early pregnancy if women or females are presenting with risk factors, or even fasting glucose. That has been shown historically to be pretty accurate for catching the women who have diabetes but were not diagnosed prior to becoming pregnant, so they just were unaware, essentially. However, HbA1C has really poor sensitivity for impaired fasting glucose or impaired glucose tolerance. So, it's a really efficient screener, and does really well for this higher band of diabetes, so to speak. But it's more complex when you talk about gestational diabetes, which is a lower level of hyperglycemia.

Another point of context: If you don't present in the US with risk factors for diabetes, we screen for gestational diabetes for everyone at 24 to 28 weeks, so that leaves you about 10 to 15 weeks left of pregnancy by the time we're screening. What's coming out now from the TOBOGM trial is that if you screen for early gestational diabetes using a 1-step approach—so not using HbA1C but using an oral glucose tolerance test—and if you applied the Canadian Diabetes Association criteria rather than the WHO [World Health Organization] criteria, that process of early screening was very effective at identifying the risk of poor perinatal outcomes in women who had gestational diabetes diagnosed using higher hyperglycemic criteria.

How we screen for overt diabetes using HbA1C, it's probably sufficient. But if you want to screen early for gestational diabetes, it's likely looking like it's going to need to be that 1-step 75 g oral glucose tolerance test with a higher hyperglycemic threshold than the WHO [criteria]. But there are a lot of downstream implications for that. When you screen early, our standard interventions were able to reduce some of those poor perinatal outcomes that impact offspring like hypoglycemia and macrosomia, and those particular conditions can have poor health prognosis for the child across their life; they're more likely to have diabetes and obesity earlier on.

The impact of early screening and treatment on mom's outcomes long term is less clear, but I think that's a really important area for future research. I think we just happen to know some of the more immediate impacts of diabetes and pregnancy on offspring health, so I think that's why it's easier to conceptualize the early treatment, because you'll reduce the glucose levels that the fetus is exposed to, so that has a more mechanistic implication.

AJMC: What challenges currently exist in implementing earlier diabetes screening for pregnant women, and how can health care systems address these barriers?

Francis: There are 2 important things to think about. One: what will early screening mean?

The diabetes and pregnancy field is pretty aligned [that] if you screen and diagnose someone for GDM early, then that should be considered early GDM. And if you screen someone with risk factors for diabetes and they have diabetes, then [that’s] considered diabetes in pregnancy.

What all of this means in a scenario where we have universal screening is complex, because what do you then do with the standard of care for screening at 24 to 28 weeks? Do you screen again? How do you then categorize someone who has early GDM, but not when they're screened maybe 12 or 15 weeks later? That hasn't been untangled. So, what are people going to do with these findings in terms of translating to care? I'm not sure. It might be based on individual health systems’ or providers’ capacity to screen twice. I really don't know if anyone has decided how they're going to implement that themselves.

One of the bigger questions then becomes, if you're screened early for GDM, you need to have more emphasis on accessing health care early in pregnancy, right? If women aren't presenting to the antenatal clinic until 20 weeks but universal screening for GDM is at 14 weeks, they'll get screened when they present, but maybe that 6-week difference in intervention is pretty critical. Pregnancy is a long time, but it also has very rapid changes from week to week, so it'll be important to identify at what point does the early screening benefit wear off, if at all, or if screening at any time can result in a benefit for mom and baby.

Leslie Kantor, [PhD, MPH, chair of the Department of Urban-Global Public Health at the Rutgers School of Public Health,] has done some work showing that obstetric care is not equitably distributed geographically across New Jersey. It may be that the areas with the highest prevalence of risk factors for hyperglycemia in pregnancy don't necessarily have the infrastructure and the resources to implement early screening, and so that misalignment in recommendations and availability of resources could potentially exacerbate rates of health disparity that we then report within the state. We would have to be pretty cognizant of how this gets rolled out because of that.