Objectives. This study explored associations between demographic, socioeconomic, behavioural risk, and health factors (study factors) and detection of type 2 diabetes. Methods. A secondary analysis of data extracted from the AusDiab study was undertaken. Participants were classified as known diabetes (KDM), newly detected diabetes (NDDM), impaired fasting glucose (IFG), impaired glucose tolerance (IGT), or normal according to results of a glucose tolerance test. Results. The weighted prevalence of diabetes was 6.9%; 49.6% of participants with diabetes (NDDM + KDM) were classified as NDDM. Although study factors were associated with diabetes prevalence, most were not associated with proportion of diabetic participants classified as NDDM. Among participants with diabetes, NDDM was more likely among those who spoke English at home, were in good general health and did not report past history of cardiovascular disease. Conclusions. Although a range of personal and socioeconomic factors are associated with diabetes prevalence, these factors are not similarly associated with prior detection of diabetes. These findings highlight the importance of systematic approaches to screening for diabetes risk focussed on the whole population, with selective screening based on multi-factorial assessment of diabetes risk using the AUSDRISK Assessment Tool. What is known about the topic? Type 2 diabetes is an increasing public health problem and early detection is associated with improved outcomes. About half of the participants with diabetes in the AusDiab study had been diagnosed previously with diabetes, with the remainder being newly diagnosed in the study. What does this paper add? A range of demographic, socioeconomic, behavioural risk and health factors were associated with increased risk of diabetes. However, these factors did not differentiate between those with a prior diagnosis of diabetes and those with newly diagnosed diabetes. Among participants with diabetes, NDDM was more likely among those who spoke English at home, were in good general health or reported no past history of cardiovascular disease. What are the implications for practitioners? Screening for type 2 diabetes should be broadly focussed on the whole population, with selective blood testing based on multi-factorial assessment of diabetes risk using the AUSDRISK Assessment Tool.