The WHO MCS collected data on various socio-demographic and clinical-obstetric characteristics, severity markers, and complications that were analysed as candidate predictors for CS. First, we assessed the reference population and the WHO MCS country data using the Robson classification. Secondly, we carried out univariate analyses of the reference population to explore the relationship of several candidate predictors and the occurrence of CS.
Third, a logistic regression random effects model was used to determine the relationship between candidate predictors
and CS. As part of the regression analysis, the reference population was divided in two subpopulations: ‘A1’ (used
for model building), and ‘A2’ (used for testing the internal validity of the prediction model). To maximise representativeness and at the same time ensure the ability to test the internal validity of the model, 90% of the reference
population was randomly allocated to population ‘A1’ and the remainder (10%) to population ‘A2’.
The different versions of C-Model estimate the probability of CS based on the presence or absence of significant predictors
The WHO MCS collected data on various socio-demographic and clinical-obstetric characteristics, severity markers, and complications that were analysed as candidate predictors for CS. First, we assessed the reference population and the WHO MCS country data using the Robson classification. Secondly, we carried out univariate analyses of the reference population to explore the relationship of several candidate predictors and the occurrence of CS.Third, a logistic regression random effects model was used to determine the relationship between candidate predictorsand CS. As part of the regression analysis, the reference population was divided in two subpopulations: ‘A1’ (usedfor model building), and ‘A2’ (used for testing the internal validity of the prediction model). To maximise representativeness and at the same time ensure the ability to test the internal validity of the model, 90% of the referencepopulation was randomly allocated to population ‘A1’ and the remainder (10%) to population ‘A2’.The different versions of C-Model estimate the probability of CS based on the presence or absence of significant predictors
การแปล กรุณารอสักครู่..