Skip to main content
. 2024 Aug 14;3(1):e000784. doi: 10.1136/bmjmed-2023-000784

Table 4. Predictive performance of developed birth weight model with average intercept in each internal-external cross validation cycle. UK (Allen et al, 2017),32 Australia (Rumbold et al, 2006),31 and Norway (STORK Groruddalen research programme, 2010)33 cohorts, and pooled estimate.

Pooled estimate Allen et al, 2017 Rumbold et al, 2006 STORK Groruddalen, 2010
No of pregnancies for model development 236 183 235 351 236 405
No of pregnancies for external validation 1045 1877 823
Calibration slope
 Point estimate 1.00 0.90 1.07 1.04
 Confidence interval 0.78 to 1.23 0.82 to 0.97 1.02 to 1.12 0.96 to 1.12
 Prediction interval −0.25 to 2.26
 τ2 (95% CI) 0.01 (0.00 to 0.14)
Calibration-in-the-large (g)
 Point estimate 9.72 −22.32 −33.42 86.41
 Confidence interval −154.3 to 173.8 −48.36 to 3.7 −53.4 to −13.5 57.3 to 115.5
 Prediction interval −943.23 to 962.67
 τ2 (95% CI) 4200 (801 to 76000)
Observed to expected birth weight ratio
 Point estimate 1.00 1.00 0.99 1.03
 Confidence interval 0.94 to 1.07 0.95 to 1.04 0.94 to 1.05 0.97 to 1.09
 Prediction interval 0.81 to 1.20
 τ2 (95% CI) 0.00 (0.00 to 0.01)
R2* (%)
 Median 45.7 32.6 47.4 45.7
 Range 32.2-47.8 32.2-32.8 47.1-47.8 45.0-46.2
 interquartile range 32.7-47.4 32.5-32.7 47.4-47.5 45.5-45.8
*

Reported as median, range, and interquartile range for imputations because R2 cannot be summarised for all imputations with Rubin’s rules.

CIconfidence interval