Currently, neither the American Diabetes Association nor the Kidney Foundation consider the results of a positive dipstick urine test for protein, a semi-quantitative measurement, in the final evaluation of diabetic nephropathy. Instead, they require a quantitative test. The object of this study was to assess whether a positive semi-quantitative test could accurately substitute for a quantitative one to evaluate renal disease. We determined the proportion of urine samples dipstick positive for protein that had an albumin:creatinine ratio of 30 microg/mg or more, the recommended value for the diagnosis of microalbuminuria (incipient nephropathy). Albumin:creatinine ratios were measured in urine samples from 19 diabetic and 51 nondiabetic patients in which the dipstick test for protein was positive. Twelve of 24 (50%) urine samples trace positive for protein by a dipstick method had albumin:creatinine ratios of 30 microg/mg or more, whereas 42 of 46 (91%) urine samples greater than or equal to 1+ for protein exceeded that ratio. The results were similar in the two groups of patients. The positive predictive value for a test result more than or equal to 1+ for protein was 91%. We conclude that in contrast to the recommendations of the American Diabetes Association and the National Kidney Foundation, dipstick positive proteinuria of more than or equal to 1+ can substitute for an albumin:creatinine ratio. An algorithm for this more cost-effective approach to the diagnosis of diabetic nephropathy is suggested.