Risk Scoring Systems for Preterm Birth and Their Performance: A Systematic Review

Area of Science:

• Obstetrics and Gynecology
• Perinatal Medicine
• Clinical Risk Prediction

Background:

• Preterm birth (PTB) prediction and risk stratification are significant clinical challenges.
• Numerous risk factors for PTB have been identified, leading to the development of various risk scoring systems (RSSs).

Purpose of the Study:

• To systematically review existing RSSs for PTB.
• To identify the variables utilized in these RSSs.
• To evaluate the predictive performance of these RSSs.

Main Methods:

• A systematic search of two major databases was conducted.
• Independent screening and data extraction were performed by two authors.
• Studies evaluating RSSs for PTB prediction were included, with reference list searches also performed.

Main Results:

• Fifty-six studies were included in the review.
• Commonly used variables in RSSs included maternal age, weight, smoking history, previous PTB history, and cervical length.
• Performance metrics varied widely, with sensitivity ranging from 4.2% to 92.0% and AUC from 0.59 to 0.95.

Conclusions:

• Current RSSs for PTB demonstrate poor predictive ability in most clinical scenarios.
• Despite advancements in understanding PTB-related variables, a single universal RSS is not yet feasible for clinical practice.
• Future efforts should focus on developing novel RSSs, identifying new predictive variables, and creating larger reference datasets.