This study analyses the incidence of Neonatal Abstinence Syndrome (NAS) in a large geographically defined population in Australia.Database linkage analysis of all births between 2000 and 2011 in New South Wales (NSW), Australia. The diagnosis of NAS was derived from hospital coding P96.1, 'Neonatal withdrawal symptoms from maternal use of drugs of addiction'. Temporal trends were studied by comparing epoch 1 (2000-05) with epoch 2 (2006-11). The relationship with changes in maternal factors was further analysed.The NAS was coded in 3842 of 1 022 263 live born infants (0.38%). NAS incidence peaked at 5.07 per 1000 live births in 2002, decreasing to 3.18 in 2011 and was negatively correlated with maternal age (r = -0.7). The rate of NAS in epoch 2 (3.4 per 1000 births, 95% CI 3.28, 3.58) was significantly lower than in epoch 1 (4.1 per 1000 births, 95% CI 3.96, 4.33). Epoch 2 mothers were significantly older (mean 29.8 years vs. 28.3 years), less likely to be multiparous (OR 0.7, 95% CI 0.6, 0.9) or smoke (OR 0.4, 95% CI 0.4, 0.5). They were more likely to engage in antenatal care earlier (mean first visit: 14.1 vs. 18.9 weeks). Most infants (~80%) were born at term (>37 weeks gestation).The incidence of NAS as a discharge diagnosis has decreased in our population since 2002. Mothers are also older and engaging earlier in prenatal care. Whether these changes alter NAS presentation and diagnosis or whether pregnant women are using drugs that do not cause typical NAS (e.g. amphetamines) is uncertain and requires further study.