Summary

Participants were drawn from a list of medical practitioners in Queensland, NSW and Victoria who provided more than 1500 general practice Medicare items of service during the previous year. The sample was stratified within States by population of postcode, into metropolitan areas and three country strata: "small country towns" (population less than 5000); "medium country towns" (5000-15000); and "large country towns" (more than 15000). The total country sample is referred to as "country areas". Planned sample size was 180 country GPs (20 in each stratum in each State) and 60 metropolitan GPs (20 in each of the three State capital cities). The final data set was weighted to be representative of the distribution of the source population. Data collection: The variables studied included: GP characteristics; practice isolation factors; patient age, sex and status to the practice; patient reasons for encounter (up to three per encounter); problems managed (up to four); drugs prescribed and other treatments provided (up to four per problem); tests and investigations ordered and referrals made at these encounters; and planned follow-up. Data were centrally coded. Participation rate: The final sample of 231 GPs (177 country and 54 metropolitan) recorded information during 435 recording weeks (336 country and 99 metropolitan). These practitioners represented 50.7% of those contacted and available, the response rate being better in country (57.5%) than in metropolitan (36.5%) areas. A practice profile questionnaire which included questions regarding the doctor and the practice was completed by 97.4% of participants, while a questionnaire on distance from support services was completed by 93.8% of country participants. The final weighted data set included 51741 encounters with country GPs and 11 351 with metropolitan GPs.