Development of a prognostic model for predicting spontaneous singleton preterm birth

Jelle M. Schaaf, Anita C J Ravelli, Ben Willem J Mol, Ameen Abu-Hanna

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Objective: To develop and validate a prognostic model for prediction of spontaneous preterm birth. Study design: Prospective cohort study using data of the nationwide perinatal registry in The Netherlands. We studied 1,524,058 singleton pregnancies between 1999 and 2007. We developed a multiple logistic regression model to estimate the risk of spontaneous preterm birth based on maternal and pregnancy characteristics. We used bootstrapping techniques to internally validate our model. Discrimination (AUC), accuracy (Brier score) and calibration (calibration graphs and Hosmer-Lemeshow C-statistic) were used to assess the model's predictive performance. Our primary outcome measure was spontaneous preterm birth at <37 completed weeks. Results: Spontaneous preterm birth occurred in 57,796 (3.8%) pregnancies. The final model included 13 variables for predicting preterm birth. The predicted probabilities ranged from 0.01 to 0.71 (IQR 0.02-0.04). The model had an area under the receiver operator characteristic curve (AUC) of 0.63 (95% CI 0.63-0.63), the Brier score was 0.04 (95% CI 0.04-0.04) and the Hosmer Lemeshow C-statistic was significant (p < 0.0001). The calibration graph showed overprediction at higher values of predicted probability. The positive predictive value was 26% (95% CI 20-33%) for the 0.4 probability cut-off point. Conclusions: The model's discrimination was fair and it had modest calibration. Previous preterm birth, drug abuse and vaginal bleeding in the first half of pregnancy were the most important predictors for spontaneous preterm birth. Although not applicable in clinical practice yet, this model is a next step towards early prediction of spontaneous preterm birth that enables caregivers to start preventive therapy in women at higher risk.

LanguageEnglish
Pages150-155
Number of pages6
JournalEuropean Journal of Obstetrics and Gynecology and Reproductive Biology
Volume164
Issue number2
DOIs
Publication statusPublished - 1 Jan 2012

Keywords

  • Prediction model
  • Preterm birth
  • Prognostic model
  • Risk assessment
  • Screening
  • Spontaneous

ASJC Scopus subject areas

  • Reproductive Medicine
  • Obstetrics and Gynaecology

Cite this

Schaaf, Jelle M. ; Ravelli, Anita C J ; Mol, Ben Willem J ; Abu-Hanna, Ameen. / Development of a prognostic model for predicting spontaneous singleton preterm birth. In: European Journal of Obstetrics and Gynecology and Reproductive Biology. 2012 ; Vol. 164, No. 2. pp. 150-155.
@article{6aff4deb47f94f20b1b97b3b97fa63d1,
title = "Development of a prognostic model for predicting spontaneous singleton preterm birth",
abstract = "Objective: To develop and validate a prognostic model for prediction of spontaneous preterm birth. Study design: Prospective cohort study using data of the nationwide perinatal registry in The Netherlands. We studied 1,524,058 singleton pregnancies between 1999 and 2007. We developed a multiple logistic regression model to estimate the risk of spontaneous preterm birth based on maternal and pregnancy characteristics. We used bootstrapping techniques to internally validate our model. Discrimination (AUC), accuracy (Brier score) and calibration (calibration graphs and Hosmer-Lemeshow C-statistic) were used to assess the model's predictive performance. Our primary outcome measure was spontaneous preterm birth at <37 completed weeks. Results: Spontaneous preterm birth occurred in 57,796 (3.8{\%}) pregnancies. The final model included 13 variables for predicting preterm birth. The predicted probabilities ranged from 0.01 to 0.71 (IQR 0.02-0.04). The model had an area under the receiver operator characteristic curve (AUC) of 0.63 (95{\%} CI 0.63-0.63), the Brier score was 0.04 (95{\%} CI 0.04-0.04) and the Hosmer Lemeshow C-statistic was significant (p < 0.0001). The calibration graph showed overprediction at higher values of predicted probability. The positive predictive value was 26{\%} (95{\%} CI 20-33{\%}) for the 0.4 probability cut-off point. Conclusions: The model's discrimination was fair and it had modest calibration. Previous preterm birth, drug abuse and vaginal bleeding in the first half of pregnancy were the most important predictors for spontaneous preterm birth. Although not applicable in clinical practice yet, this model is a next step towards early prediction of spontaneous preterm birth that enables caregivers to start preventive therapy in women at higher risk.",
keywords = "Prediction model, Preterm birth, Prognostic model, Risk assessment, Screening, Spontaneous",
author = "Schaaf, {Jelle M.} and Ravelli, {Anita C J} and Mol, {Ben Willem J} and Ameen Abu-Hanna",
year = "2012",
month = "1",
day = "1",
doi = "10.1016/j.ejogrb.2012.07.007",
language = "English",
volume = "164",
pages = "150--155",
journal = "European Journal of Obstetrics Gynecology and Reproductive Biology",
issn = "0301-2115",
publisher = "Elsevier Ireland Ltd",
number = "2",

}

Development of a prognostic model for predicting spontaneous singleton preterm birth. / Schaaf, Jelle M.; Ravelli, Anita C J; Mol, Ben Willem J; Abu-Hanna, Ameen.

In: European Journal of Obstetrics and Gynecology and Reproductive Biology, Vol. 164, No. 2, 01.01.2012, p. 150-155.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Development of a prognostic model for predicting spontaneous singleton preterm birth

AU - Schaaf, Jelle M.

AU - Ravelli, Anita C J

AU - Mol, Ben Willem J

AU - Abu-Hanna, Ameen

PY - 2012/1/1

Y1 - 2012/1/1

N2 - Objective: To develop and validate a prognostic model for prediction of spontaneous preterm birth. Study design: Prospective cohort study using data of the nationwide perinatal registry in The Netherlands. We studied 1,524,058 singleton pregnancies between 1999 and 2007. We developed a multiple logistic regression model to estimate the risk of spontaneous preterm birth based on maternal and pregnancy characteristics. We used bootstrapping techniques to internally validate our model. Discrimination (AUC), accuracy (Brier score) and calibration (calibration graphs and Hosmer-Lemeshow C-statistic) were used to assess the model's predictive performance. Our primary outcome measure was spontaneous preterm birth at <37 completed weeks. Results: Spontaneous preterm birth occurred in 57,796 (3.8%) pregnancies. The final model included 13 variables for predicting preterm birth. The predicted probabilities ranged from 0.01 to 0.71 (IQR 0.02-0.04). The model had an area under the receiver operator characteristic curve (AUC) of 0.63 (95% CI 0.63-0.63), the Brier score was 0.04 (95% CI 0.04-0.04) and the Hosmer Lemeshow C-statistic was significant (p < 0.0001). The calibration graph showed overprediction at higher values of predicted probability. The positive predictive value was 26% (95% CI 20-33%) for the 0.4 probability cut-off point. Conclusions: The model's discrimination was fair and it had modest calibration. Previous preterm birth, drug abuse and vaginal bleeding in the first half of pregnancy were the most important predictors for spontaneous preterm birth. Although not applicable in clinical practice yet, this model is a next step towards early prediction of spontaneous preterm birth that enables caregivers to start preventive therapy in women at higher risk.

AB - Objective: To develop and validate a prognostic model for prediction of spontaneous preterm birth. Study design: Prospective cohort study using data of the nationwide perinatal registry in The Netherlands. We studied 1,524,058 singleton pregnancies between 1999 and 2007. We developed a multiple logistic regression model to estimate the risk of spontaneous preterm birth based on maternal and pregnancy characteristics. We used bootstrapping techniques to internally validate our model. Discrimination (AUC), accuracy (Brier score) and calibration (calibration graphs and Hosmer-Lemeshow C-statistic) were used to assess the model's predictive performance. Our primary outcome measure was spontaneous preterm birth at <37 completed weeks. Results: Spontaneous preterm birth occurred in 57,796 (3.8%) pregnancies. The final model included 13 variables for predicting preterm birth. The predicted probabilities ranged from 0.01 to 0.71 (IQR 0.02-0.04). The model had an area under the receiver operator characteristic curve (AUC) of 0.63 (95% CI 0.63-0.63), the Brier score was 0.04 (95% CI 0.04-0.04) and the Hosmer Lemeshow C-statistic was significant (p < 0.0001). The calibration graph showed overprediction at higher values of predicted probability. The positive predictive value was 26% (95% CI 20-33%) for the 0.4 probability cut-off point. Conclusions: The model's discrimination was fair and it had modest calibration. Previous preterm birth, drug abuse and vaginal bleeding in the first half of pregnancy were the most important predictors for spontaneous preterm birth. Although not applicable in clinical practice yet, this model is a next step towards early prediction of spontaneous preterm birth that enables caregivers to start preventive therapy in women at higher risk.

KW - Prediction model

KW - Preterm birth

KW - Prognostic model

KW - Risk assessment

KW - Screening

KW - Spontaneous

UR - http://www.scopus.com/inward/record.url?scp=84866730001&partnerID=8YFLogxK

U2 - 10.1016/j.ejogrb.2012.07.007

DO - 10.1016/j.ejogrb.2012.07.007

M3 - Article

VL - 164

SP - 150

EP - 155

JO - European Journal of Obstetrics Gynecology and Reproductive Biology

T2 - European Journal of Obstetrics Gynecology and Reproductive Biology

JF - European Journal of Obstetrics Gynecology and Reproductive Biology

SN - 0301-2115

IS - 2

ER -