Preterm birth continues to be a major cause of infant morbidity and mortality worldwide, but advances have recently been made in its prediction and prevention. A short cervix (<25 mm) in the second trimester on transvaginal ultrasound scan and fetal fibronectin are important predictive tests. For over ten years, the Preterm Labour Clinic at the Royal Women's Hospital, Melbourne, Australia has provided care for women at high risk of preterm birth, including those with a previous preterm birth, previous cervical surgery, uterine malformation or incidental finding of short cervix at routine ultrasound. The purpose of this study was to review this clinic's outcomes for the first decade. Methods This was a retrospective cohort study of all referrals to the Preterm Labour Clinic during the period 2004–2013 inclusive. Seven hundred and fifty‐six cases met the study inclusion criteria of appropriate risk factor, singleton pregnancy, surveillance undertaken and outcome data available. Results The preterm birth rate (<37 weeks) was 21.4%. The rate of preterm birth by year decreased significantly when adjusted for risk (P = 0.002). A short cervix was diagnosed in 32% of the sample, and positively correlated with lower gestational age at delivery. Fetal fibronectin and serum alkaline phosphatase were independent predictors of preterm birth <34 weeks and <37 weeks. Conclusion The adjusted preterm birth rate at the Royal Women's Hospital's Preterm Labour Clinic has decreased significantly over the decade studied. Positive fetal fibronectin at 26 weeks and elevated serum alkaline phosphatase are independent predictors of preterm birth.