Background
The Fetal Medicine Foundation developed a multiple logistic regression algorithm for risk prediction of delivering a small for gestational age neonate.

Aim
To validate this algorithm in an Australian population.

Methods
At the combined first trimester screen participants’ medical histories, demographic data, mean arterial pressure, uterine artery pulsatility index and pregnancy‐associated plasma protein‐A were assessed. After delivery, risk of delivering a small for gestational age neonate at <37 or ≥37 weeks gestation was retrospectively calculated using the Fetal Medicine Foundation algorithm. Results Three thousand and eight women underwent prediction of risk for delivering a small for gestational age neonate. The algorithm detected 15.0% (95% CI: 3.2–37.9) of small for gestational age neonates delivered <37 weeks gestation at a fixed 10% false positive rate (or 35.0% (95% CI: 15.4–59.2) at a fixed 20% false positive rate). It detected 23.4% (95% CI: 16.1–30.7) of small for gestational age neonates delivered ≥37 weeks gestation at a fixed 10% false positive rate (or 39.1% (95% CI: 30.7–47.5) at a fixed 20% false positive rate). The algorithm performed significantly better than individual parameters (P < 0.05). The area under the receiver operating characteristic curve was 0.68 (95% CI: 0.56–0.80) and 0.70 (95% CI: 0.65–0.74) for small for gestational age neonates at <37 and ≥37 weeks gestation, respectively. Conclusions The Fetal Medicine Foundation algorithm for first trimester prediction of small for gestational age neonates does not perform as well in an Australian population as in the original United Kingdom cohort. However, it performs significantly better than any individual test parameter in both preterm and term neonates. Incorporation of further variables may help improve screening efficacy.