Skip to main content

Table 2 Number of predictors retained in each model developed and tested in Born in Bradford from total possible (n (%)). Percentages are rounded to the nearest whole number

From: Do nuclear magnetic resonance (NMR)-based metabolomics improve the prediction of pregnancy-related disorders? Findings from a UK birth cohort with independent validation

Outcome Model (retained predictors/total number of variables possible [%])
Gestational diabetes mellitus Risk factor (5/5 [100%])
Metabolite (140/156 [90%])
Combined (152/161 [94%])
Hypertensive disorder of pregnancy Risk factor (4/5 [80%])
Metabolite (50/156 [32%])
Combined (38/161 [24%])
Small for gestational age Risk factor (4/5 [80%])
Metabolite (86/156 [55%])
Combined (101/161 [63%])
Large for gestational age Risk factor (4/5 [80%])
Metabolite (65/156 [42%])
Combined (56/161 [35%])
Preterm birth Risk factor (4/5 [80%])
Metabolite (19/156 [12%])
Combined (18/161 [11%])