Simple Four-Point Score Predicts Preterm Birth Risk at Hospital Admission

Preterm birth remains the leading risk factor for neonatal morbidity and mortality worldwide, affecting approximately 11% of all live births and contributing to long-term developmental impairment. Yet many existing prediction models are complex and rely on specialized testing that is not always practical at the bedside. A new retrospective cohort study published in BMC Pregnancy and Childbirth reports the development and validation of a simplified cumulative complication burden score, or sCCBS, designed to estimate preterm birth risk using only four variables routinely collected when a patient is admitted to hospital.

The researchers drew on electronic medical records from 66,357 deliveries at a tertiary hospital in Zhejiang Province, China, spanning 2014 to 2024. After excluding cesarean sections and records with incomplete data, 42,868 vaginal deliveries formed the initial validation cohort. The score itself was built using Least Absolute Shrinkage and Selection Operator regression, a statistical technique that selects the most informative predictors, and was tested through temporal validation: data from 2014 to 2020 served as the training set with 31,372 deliveries, while data from 2021 to 2024 formed an independent validation set of 11,496 deliveries.

The resulting tool assigns integer weights to four admission variables. Premature rupture of membranes contributes seven points, gestational diabetes mellitus two points, advanced maternal age of 35 years or older two points, and non-local origin one point. Because every component is a routinely recorded clinical variable, the score can be calculated at admission without specialized testing, which the authors identify as the key advantage over more cumbersome existing models.

Performance was assessed on three fronts: discrimination, calibration, and clinical usefulness. Discrimination, measured by the area under the receiver operating characteristic curve, came in at 0.714 (95% CI: 0.699–0.728) in the training set and 0.735 in the temporal validation set. At the optimal Youden cutoff of a score of three or more, the model showed a sensitivity of 62.4% and a specificity of 74.8%. Calibration in the validation set was moderate, with a Brier score of 0.0392, a calibration slope of 0.938, a calibration intercept of 0.005, and a Hosmer–Lemeshow p value of 0.000.

The score also sorted patients into three risk tiers with clearly separated preterm birth rates. Patients scoring zero to one had a 1.8% preterm rate and were classified as low risk. Those scoring two to four fell into the moderate-risk group with a 3.5% preterm rate, corresponding to a relative risk of 2.05. The high-risk group, with scores of five to twelve, experienced an 11.9% preterm rate, a relative risk of 6.76 compared with the reference group.

Decision curve analysis, which evaluates the net benefit of acting on a model’s predictions across a range of threshold probabilities, showed positive net benefit from 1% to approximately 11%, with a net benefit of 0.016 at a 5% threshold and 0.003 at 10%. This suggests the score could support admission-level triage decisions within that probability range, though the analysis is specific to this cohort.

A sensitivity analysis underscored which predictor carries the most weight. When premature rupture of membranes was excluded, the AUC dropped substantially from 0.714 (95% CI: 0.699–0.728) to 0.558 (95% CI: 0.528–0.551). This confirms PROM as the dominant predictor in the score, while indicating that the remaining three variables still provide moderate risk stratification on their own.

The authors are careful about scope. The study validated the score in vaginal deliveries as a methodologically rigorous first step, and they note that external validation in cesarean delivery cohorts, diverse geographic settings, and independent centers is required before broader applicability can be claimed. Within those limits, the sCCBS stands as a simple, admission-based tool built from four routinely collected variables, offering a transparent and easily computed first estimate of preterm birth risk at the moment it matters most in clinical workflow.

Subject of Research: Development and temporal validation of a simplified admission-based score for predicting preterm birth risk

Article Title: Development and validation of a simplified cumulative complication burden score (sccbs) for preterm birth prediction at hospital admission

Article References: Lu, J., Xu, H., & Xu, Q. (2026). Development and validation of a simplified cumulative complication burden score (sccbs) for preterm birth prediction at hospital admission. BMC Pregnancy and Childbirth. https://doi.org/10.1186/s12884-026-10159-4

Image Credits: AI Generated

DOI: 10.1186/s12884-026-10159-4

Keywords: preterm birth, risk prediction score, hospital admission, retrospective cohort, LASSO regression, calibration, decision curve analysis, PROM, gestational diabetes, advanced maternal age, temporal validation, obstetrics