(Vienna, 02 September 2026) – Whether treatment for gestational diabetes benefits the newborn can only be assessed by taking into account which endpoints a clinical trial measures and how these are defined. An international research group led by MedUni Vienna demonstrates in a recent review article in the American Journal of Obstetrics and Gynaecology (AJOG) that large-scale studies on gestational diabetes sometimes differ significantly in this respect, and provides valuable guidance for doctors and researchers.
The research team led by Christian Göbl (Department of Obstetrics and Gynecology, MedUni Vienna) identified so-called composite endpoints – in which events of varying frequency and clinical significance are assessed together – as being particularly relevant. A statistically significant overall effect therefore does not automatically mean that the clinically most important complications have also been reduced. Conversely, relevant improvements in individual endpoints can be masked by a non-significant overall result. Examples from large studies on gestational diabetes show that the interpretation of a treatment effect can depend significantly on which specific endpoint is considered.
"When interpreting a study, we must first assess precisely what was actually measured and what clinical significance the respective endpoint has. Only then can we draw conclusions about the benefits of a treatment," says Christian Göbl. The paper is aimed at doctors who use the results of clinical trials to inform treatment decisions and the care of their patients, as well as at researchers who plan clinical trials and interpret their results. The aim is to enable a better understanding of the significance of study results within their respective methodological and clinical contexts.
Another key focus of the review concerns the assessment of birth weight. In clinical trials, absolute birth weight thresholds are used on the one hand, whilst on the other hand birth weight percentiles are employed, which take into account factors such as gestational age and the child’s sex. However, there are different reference and growth curves for this purpose, some of which can lead to significantly divergent classifications. The authors therefore recommend that future studies use clearly predefined and clinically relevant neonatal endpoints. In the case of composite endpoints, the individual components should be reported separately. When assessing infant growth, both absolute measures and those adjusted for gestational age and sex should be taken into account, and the results should be verified using alternative growth curves.
"The aim of our work is not to call previous studies into question, but to interpret their results in a more nuanced way and make them more comparable. Only once it is clear which endpoint was measured and what exactly it represents can we assess which conclusions regarding the benefits of a treatment are actually justified," says Christian Göbl.
Publication: American Journal of Obstetrics and Gynecology
Neonatal outcomes in gestational diabetes trials: What clinicians should consider when interpreting treatment effects.
Göbl C, Linder T, Bozkurt L, et al.
DOI: 10.1016/j.ajog.2026.08.044.
https://pubmed.ncbi.nlm.nih.gov/42680081/