Prediction models for the risk of gestational diabetes: a systematic review

Abstract

Background: Numerous prediction models for gestational diabetes mellitus (GDM) have been developed, but their methodological quality is unknown. The objective is to systematically review all studies describing first-trimester prediction models for GDM and to assess their methodological quality. Methods: MEDLINE and EMBASE were searched until December 2014. Key words for GDM, first trimester of pregnancy, and prediction modeling studies were combined. Prediction models for GDM performed up to 14 weeks of gestation that only include routinely measured predictors were eligible. Data was extracted by the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies (CHARMS). Data on risk predictors and performance measures were also extracted. Each study was scored for risk of bias. Results: Our search yielded 7761 articles, of which 17 were eligible for review (14 development studies and 3 external validation studies). The definition and prevalence of GDM varied widely across studies. Maternal age and body mass index were the most common predictors. Discrimination was acceptable for all studies. Calibration was reported for four studies. Risk of bias for participant selection, predictor assessment, and outcome assessment was low in general. Moderate to high risk of bias was seen for the number of events, attrition, and analysis. Conclusions: Most studies showed moderate to low methodological quality, and few prediction models for GDM have been externally validated. External validation is recommended to enhance generalizability and assess their true value in clinical practice.

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

“…In addition, prognostic prediction models may also constitute a basis for personalized medicine-based medicine guiding planning of antenatal care and targeting preventive strategies. 11 A substantial number of prediction models for the risk of GDM have been developed, 12 but to our knowledge none of these is routinely used in clinical practice. Validation of prediction models in independent populations is a crucial step before implementation in clinical practice.…”

External validation and clinical usefulness of first‐trimester prediction models for small‐ and large‐for‐gestational‐age infants: a prospective cohort study

“…high BMI or GDM) or ‘vascular’ (i.e., hypertensive disorder) origin. The predictors in the included models for SGA and LGA overlap considerably with those of models predicting hypertensive pregnancy disorder and GDM, respectively . Although most SGA and LGA infants are born to mothers without a hypertensive pregnancy disorder or GDM, respectively, the conditions share common pathophysiological aspects …”

“…Performance measures at different risk thresholds for recalibrated model from Frick et al, 31 predicting the risk of large-for-gestationalage considerably with those of models predicting hypertensive pregnancy disorder and GDM, respectively. 67,68 Although most SGA and LGA infants are born to mothers without a hypertensive pregnancy disorder or GDM, respectively, the conditions share common pathophysiological aspects. 24,65 Regarding primary prevention strategies, recent metaanalyses have demonstrated that aspirin modestly reduces the risk of delivering an SGA infant in women at high risk, with most benefit derived by starting treatment before 16 weeks of gestation and using a dose of ≥100 mg (risk ratio 0.56-0.76).…”

Personalizing obstetric care

“…high body mass index, GDM) origin, respectively. Although the included prediction models predict overall SGA and LGA, the predictors considerably overlap with those of models predicting hypertensive pregnancy disorders and GDM, respectively 51,52 . Recent meta-analyses demonstrated that low-dose aspirin modestly reduces the risk for delivering a SGA infant in women at high risk (risk ratio 0.56-0.76) 25,26 .…”

“…Published studies showed that the predictive performance of a model, particularly predicting PE, might improve by taking into account biomarkers or ultrasound imaging (i.e. uterine artery Doppler) 51,52 . However, these promising predictors are not always routinely performed or readily available in general antenatal care, and it is unknown whether models incorporating non-routine predictors are more (cost-)effective compared to non-invasive models 51,85,86 .…”