Comparing allostatic load scoring methods and timing: their links to preterm birth outcomes

Area of Science:

• Reproductive Health
• Biomarkers
• Pregnancy Complications

Background:

• Allostatic load, a measure of cumulative physiological stress, is a promising indicator for identifying preterm birth risk.
• Current methods for calculating allostatic load scores lack standardization, hindering consistent application.
• The impact of different scoring algorithms and assessment timing on preterm birth prediction remains unclear.

Purpose of the Study:

• To compare the predictive performance of various allostatic load scoring algorithms for preterm birth.
• To evaluate how the timing of biomarker assessment influences the association between allostatic load and preterm birth.

Main Methods:

• Prospective longitudinal cohort study of 1576 healthy pregnant women in Pakistan (2018-2020).
• Biomarkers assessed in early pregnancy (10-19 weeks) and mid-pregnancy (22-29 weeks).
• Seven allostatic load scoring algorithms were applied and compared using logistic regression, AIC, and AUC.

Main Results:

• 13.1% of participants experienced preterm birth.
• Six of seven scoring algorithms significantly predicted preterm birth when assessed in early pregnancy.
• Prediction accuracy generally decreased when biomarkers were assessed later in pregnancy; clinical limits scoring was strongest early on.

Conclusions:

• The association between allostatic load and preterm birth is dependent on the scoring algorithm and assessment timing.
• Early pregnancy assessment using the clinical limits scoring method provides a robust prediction of preterm birth risk.
• Standardizing cut-offs for certain algorithms could yield measures less sensitive to timing, complementing clinical thresholds for screening.