AIM:To determine the relationship between clinical practice and publication of an Australian consensus statement for management of extremely preterm infants in 2006. METHODS:A population-based study using linked data from New South Wales, Australia for births between 22 + 0 and 26 + 6 weeks of gestation between 2000 and 2011. RESULTS:There were 4746 births of whom 2870 were liveborn and 1876 were stillborn. Of the live births, 2041 (71%) were resuscitated, 1914 (67%) were admitted into a neonatal intensive care unit (NICU) and 1310 (46%) survived to hospital discharge. Thirty-nine (2%) stillbirths were resuscitated but none survived. No 22-week infant survived to hospital discharge. Fewer 23-week gestation infants were resuscitated between 2004 (52%) and 2005 (20%) but resuscitation rates increased by 2008 (44%). There was no difference at other gestations. Adjusted odds ratio (OR) for resuscitation was increased by birthweight (OR: 1.01), tertiary hospital birth (OR: 3.4) and Caesarean delivery (OR: 11.3) and decreased by rural residence (OR: 0.4) and male gender (OR: 0.7). CONCLUSION:Expert recommendations may be shaped by clinical practice rather than the converse, especially for 23-week gestation infants. Recommendations should be revised regularly to include clinical practice changes.