Skip to main content
. 2023 Oct 17;23:732. doi: 10.1186/s12884-023-06018-1

Table 1.

Sensitivity analysis of the model to predict preterm birth: Comparison of the regression coefficients, standard errors (SE), and p-values for complete case analysis (CCA) and multiple imputed data (MI).

Predictor Variable Complete Case Analysis Multiple Imputation
β SE P-value β SE. P-value
Age (< 20) 1.656 0.669 0.013 1.62 0.648 0.012
Parity (multipara) 1.067 0.431 0.013 1.31 0.424 < 0.001*
Recent pregnancy complications(Yes) 3.469 0.31 < 0.001 3.61 0.304 < 0.001*
Time of ANC initiation( Late) 0.763 0.385 0.047 0.78 0.383 0.032*
Comorbid(Yes) 0.265 0.924 0.774 0.28 0.924 0.762
Pregnancy status(unplanned) 1.309 0.297 < 0.001 1.18 0.288 < 0.001*
Weight (underweight) 0.312 0.439 0.477 0.32 0.441 0.461
MUAC(< 24 cm) 0.623 0.375 0.097 0.67 0.371 0.069*
Hemoglobin(< 11 g/dl) 4.171 0.482 < 0.001 4.19 0.482 < 0.001*

*Variables retained in the reduced model using the likelihood ratio test are: age, timing ANC initiation, pregnancy status, hemoglobin, parity, and recent pregnancy complications. Recent pregnancy complications include antepartum hemorrhage, PROM (premature rupture of membrane), PIH (pregnancy-induced hypertension), gestational diabetic mellitus, and threatened abortion. ANC Antenatal care