The main problem areas were: 1) getting good estimates of the size and composition of the Indigenous population; 2) getting relatively complete ascertainment of Indigenous people in routinely gathered administrative data collections (such as hospital records and cancer registrations); 3) getting any useful data from national surveys; and 4) getting regular, reliable data for small areas.

1) Good estimates of the population are needed for planning and resource allocation, as well as for the calculation of rates of health events.

2) Administrative data collections are widely available and commonly used for the total population, but they cannot provide useful information about the Indigenous population unless Indigenous people are identified correctly.

3) Because the Indigenous population makes up only about 2% of the total population, nationally representative samples will not provide sufficient numbers of Indigenous respondents for most purposes unless the Indigenous population is over-sampled.

4) Getting good data for small areas requires good quality data at the national level, plus adequate sample sizes or numbers of events and population estimates for each area. In addition, because there is a greater threat that individuals will be identifiable in small area data, additional privacy safeguards may be required.