A more reliable estimate of the prevalence of Prader-Willi syndrome in Australian babies is now available, following the first population study to assess genetically proven cases.

The birth prevalence of 4/100 000 live births (95% confidence interval 2.7 to 5.7) or ~1/25 000 live births /year is based on 30 notified cases between January 1998 and December 2000 in children aged under 1 year. It is a minimum estimate because of the possibility of missing data for the target age and failure to recognise cases clinically.

The 30 cases (71%) were among 42 cases in children aged under 15 years, all confirmed by DNA methylation, fluorescent in situ hybridisation, or DNA polymorphisms. A paternal deletion on chromosome 15q(11–13) was the cause in 21 (70%) of the 30 cases and disomy in one parent in three (10%) but was unknown in six.

The results were based on notifications to the Australian Paediatric Surveillance Unit. Paediatricians and child health specialists were asked to report newly diagnosed cases, whether on clinical or genetic criteria, in the past month in children aged under 15. They were sent a copy of clinical diagnostic criteria and asked for demographic data, growth and genetic test results, perinatal and neonatal histories, and clinical details. Data on children aged under 1 year were used. The Australian Bureau of Statistics provided data on live births.

Prader-Willi syndrome is not always recognised, so is underestimated among babies and overestimated among obese retarded teenagers, and all previous estimates of prevalence have been retrospective.