<p>Aim To evaluate the performance of the simplified Indian Diabetes Risk Score (IDRS) and the Australian Type 2 Diabetes Risk Assessment (AUSDRISK) instruments in predicting diabetes in Indian-Australians.</p> <p>Background Screening for diabetes in the general community is common and numerous scoring systems are being used to predict the risk of diabetes.</p> <p>Data sources For this cross-sectional study, data were obtained from people attending the Australia India Friendship Fair.</p> <p>Review methods Data relating to risk factors for diabetes were obtained using a questionnaire and a random blood glucose level. The IDRS and AUSDRISK scores were calculated. Student’s t-test, Pearson chi-square, and receiver-operating characteristic curves were used to compare the performance of the predictive scores.</p> <p>Discussion Data were analysed for 136 participants: 28 per cent of individuals considered to be low-risk and 35 per cent considered to be moderate-risk according to AUSDRISK were classified as moderate-risk and high-risk respectively by IDRS.</p> <p>Conclusion The two models were not congruent in predicting diabetes risk among Asian-Indians.</p> <p>Implications for practice/research The results of this study have significant implications for education relating to diabetes screening.</p>