It has been reported that mixed meals are used in clinics in developing and developed countries in screening and diagnosis of diabetes. Thus, we aimed to determine the differences in 2-h plasma glucose values after non-diabetic subjects ingested 75 g pure glucose and its equivalent content in frequently consumed carbohydrate foods in Caribbean subjects. Twenty-seven apparently healthy non-diabetic subjects (nine males, 18 females) consumed 75 g pure glucose and its carbohydrate equivalent in three ethnic test foods (bread, rice and roti) at 7 days apart. Plasma glucose and insulin levels were determined in blood samples collected before and after 60, 90, 120 and 150 min of ingestion of these foods. In comparison with each of the test foods, the postprandial 1-h and 2-h plasma glucose values and the 60, 90, 120 and 150 min incremental glucose concentrations after oral glucose load were significantly higher than the corresponding values for each of the test foods (all P<0.01). In spite of these higher postprandial glucose concentrations, the postprandial insulin responses following the oral glucose load and the test foods did not significantly differ at any time point (all P>0.05). However, the test food, roti, tended to stimulate higher absolute and incremental insulin secretions than pure glucose or any other test food (all P>0.05). Generally, the correlation between 2-h plasma glucose value after the ingestion of the pure glucose and each of the test foods was significant (all correlation coefficients were greater than 0.70, P<0.01). In conclusion, different ethnic mixed meals could serve as an alternative to glucose in routine screening and diagnosis of diabetes if its available carbohydrate content is known and quantified.