The research team, led by investigators at the National Research Institute for Family Planning in Beijing and collaborating institutions, categorized preconception fasting plasma glucose (FPG) into four groups: hypoglycemia (below 3.9 mmol/L without a diabetes history), normal (3.9 to below 5.6 mmol/L without a diabetes history), impaired fasting glucose (5.6 to below 7.0 mmol/L without a diabetes history), and diabetes mellitus (7.0 mmol/L or above, or with a diabetes history). Body mass index (BMI) was similarly divided into underweight (below 18.5 kg/m²), normal (18.5 to below 24.0 kg/m²), overweight (24.0 to below 28.0 kg/m²), and obesity (28.0 kg/m² or above). These cutoffs align with standards commonly applied in Chinese and Asian populations.

Of the more than eight million participants, 40.69 percent had at least one abnormality in either preconception FPG or BMI — a strikingly high proportion that underscores how common metabolic deviations are among women of reproductive age even before pregnancy begins. The researchers then tracked seven adverse pregnancy outcomes: medically indicated termination, spontaneous abortion, preterm birth, large for gestational age (LGA), small for gestational age (SGA), birth defects, and perinatal death.

Using robust Poisson regression models adjusted for a range of maternal characteristics, the team found that women who entered pregnancy with both obesity and diabetes mellitus faced the highest relative risks across most outcomes. Compared with women who had normal FPG and normal BMI, those with combined obesity and diabetes had an adjusted risk ratio of 1.49 for medically indicated termination, 1.38 for spontaneous abortion, 1.39 for preterm birth, 1.87 for large for gestational age, and 2.31 for perinatal death. In absolute terms, the excess risks translated to 2.38 percentage points higher risk of preterm birth and 9.08 percentage points higher risk of large for gestational age relative to the reference group.

Interestingly, the pattern reversed for two outcomes. The highest risks of small for gestational age (risk ratio 1.41) and birth defects (risk ratio 2.19) were observed not among women with obesity and diabetes, but among underweight women with diabetes mellitus. This asymmetry hints that different biological mechanisms may underlie fetal growth restriction and congenital anomalies compared with the complications more typically associated with maternal hyperglycemia and excess adiposity.

Beyond simple relative risks, the researchers formally tested for interaction between the two exposures — both on the additive scale (using metrics such as the relative excess risk due to interaction, or RERI) and on the multiplicative scale. They found that positive additive interactions between hyperglycemia and elevated BMI were outcome-specific: they appeared for medically indicated termination, spontaneous abortion, preterm birth, and large for gestational age, but were not observed for small for gestational age, birth defects, or perinatal death. This means that for some complications, the combined effect of having both high blood sugar and high body weight exceeds what would be expected from simply adding the individual risks together.

The team also employed restricted cubic splines to map the dose–response relationship between each exposure and each outcome. These curves revealed a bidirectional amplification: the dose–response relationship of elevated FPG with adverse pregnancy outcomes was steeper among women with higher BMI, while the relationship of elevated BMI with adverse outcomes was similarly more pronounced among women with higher FPG. In contrast, the risk of small for gestational age was predominantly associated with preconception underweight, with much less variation across FPG levels — suggesting that for fetal growth restriction, maternal nutritional status may matter more than glycemic control.

The study’s scale is notable. The National Free Preconception Checkup Project is a government-funded program offering free health examinations to couples planning marriage or conception across all 31 provinces of China. Because the database captures women before they become pregnant, it provides a rare opportunity to examine metabolic status at a truly preconceptional time point, rather than relying on measurements taken during early pregnancy or after complications have already emerged. The researchers conducted multiple sensitivity analyses — excluding women with a history of adverse pregnancy outcomes, restricting to nulliparous women, using region-specific and year-specific meta-analyses, employing imputed datasets, and varying the time interval from preconception examination to conception — and the core findings remained consistent.

The authors caution that the study is observational and retrospective, meaning it can identify associations but cannot prove that modifying preconception FPG or BMI would improve pregnancy outcomes. Residual confounding — for instance, from unmeasured dietary patterns, physical activity, or underlying conditions such as polycystic ovary syndrome — cannot be excluded. The findings are also specific to a Chinese population, and BMI cutoffs differ across ethnic groups, so the absolute risks may not generalize directly to other populations. Nonetheless, the consistency of the joint effects across multiple sensitivity analyses strengthens the plausibility of a genuine interaction.

From a public health standpoint, the results argue for a dual-assessment approach in preconception care: rather than treating elevated blood glucose and elevated body weight as separate risk factors to be flagged independently, clinicians may improve risk stratification by considering both measurements together. The authors suggest that prospective interventional studies are now needed to determine whether actively modifying these factors before conception — through weight management, glycemic optimization, or both — can reduce the burden of adverse pregnancy outcomes among singleton pregnancies. Given that more than 40 percent of women in this enormous cohort had at least one metabolic abnormality before conception, even modest improvements in preconception metabolic health could translate into substantial population-level benefits.

Subject of Research: Joint association of preconception fasting plasma glucose and body mass index with adverse pregnancy outcomes: a retrospective cohort study in China

Article Title: Joint association of preconception fasting plasma glucose and body mass index with adverse pregnancy outcomes: a retrospective cohort study in China

Article References: Wu, H., Yang, Y., Lei, J., Lyu, X., Zhao, C., Li, J., Liu, M., Hu, X., He, Y., Wang, Y., Peng, Z., Zhang, Y., Zhang, H., Wang, Q., Shen, H., Zhang, Y., Yan, D., & Ma, X. (2026). Joint association of preconception fasting plasma glucose and body mass index with adverse pregnancy outcomes: a retrospective cohort study in China. BMC Public Health. https://doi.org/10.1186/s12889-026-29692-0

Image Credits: AI Generated

DOI: 10.1186/s12889-026-29692-0

Keywords: preconception health, fasting plasma glucose, body mass index, pregnancy outcomes, diabetes, obesity, retrospective cohort study, China, Joint, association, preconception, fasting