External validation and clinical utility of prognostic prediction models for gestational diabetes mellitus: A prospective cohort study

Abstract

Introduction: We performed an independent validation study of all published first trimester prediction models, containing non-invasive predictors, for the risk of gestational diabetes mellitus. Furthermore, the clinical potential of the best performing models was evaluated. Material and methods: Systemically selected prediction models from the literature were validated in a Dutch prospective cohort using data from Expect Study I and PRIDE Study. The predictive performance of the models was evaluated by discrimination and calibration. Clinical utility was assessed using decision curve analysis. Screening performance measures were calculated at different risk thresholds for the best model and compared with current selective screening strategies. Results: The validation cohort included 5260 women. Gestational diabetes mellitus was diagnosed in 127 women (2.4%). The discriminative performance of the 12 included models ranged from 68% to 75%. Nearly all models overestimated the risk. After recalibration, agreement between the observed outcomes and predicted probabilities improved for most models.

Implementation of a first-trimester prognostic model to improve screening for gestational diabetes mellitus

“…The performance of the implemented model in our study was higher than previous development, external validation and update studies with a c-statistic of 0.85 (95%CI 0.81–0.90) compared to c-statistics ranging from 0.70 (95%CI 0.68–0.73) to 0.80 (95%CI 0.76–0.84) [ 10 – 12 , 37 ]. In clinical practice, this resulted in less pregnant women requiring testing for GDM while more GDM cases were timely identified, which may reduce perinatal morbidity and subsequent healthcare expenditure [ 7 , 9 ].…”

Adherence rates to a prediction tool identifying women with an increased gestational diabetes risk: An implementation study

“…The consortium achieved consensus regarding a suitable risk‐threshold. A predicted risk of 3.5% or more was used as the cut‐off value to identify women with an increased GDM risk (sensitivity 80%, specificity 51%) 13 …”

“…14,15 Women's GDM risk was assessed during the first prenatal visits by means of an online prediction tool. This tool embedded the externally validated model of van Leeuwen et al, 13,16 a prediction model based on maternal characteristics (age, body mass index, ethnicity, family history, and obstetric history). The consortium achieved consensus regarding a suitable risk-threshold.…”

“…A predicted risk of 3.5% or more was used as the cut-off value to identify women with an increased GDM risk (sensitivity 80%, specificity 51%). 13 For women with an increased GDM risk estimation, additional risk-based care included the recommendation of a 75-g 2-h OGTT within the gestational window of 24-28 weeks, in a shared decisional approach. The results of the OGTT were evaluated in accordance with the IADPSG criteria.…”

“…Recently, healthcare professionals in the southeastern part of the Netherlands implemented an externally validated prediction model to assess women's risk of developing GDM during the first trimester of pregnancy. 13 The detection rate was set at 80%, leading to a false-positive rate of 49%. After risk assessment by means of the model-based prediction tool, women with an increased risk were offered an OGTT in a shared decision setting.…”