Recent guidelines call for lower blood pressure targets in patients with proteinuria of 1 g/d or greater. Dipstick test for proteinuria is widely available, but no information on its sensitivity or specificity is available when implementing current guidelines. To determine performance characteristics of dipstick proteinuria, we studied 332 patients attending a Veterans Administration renal clinic who underwent simultaneous measurement of urine protein and creatinine, as well as automated urinalysis with a dipstick graded from 0 to 4+ with a machine reader. There was increasing severity of proteinuria with each increment in dipstick result. When the spot urine protein-creatinine ratio was less than 1, a lower specific gravity was associated with greater severity of proteinuria. Areas under the receiver operating characteristic curve for 1 and 3 g protein/g creatinine were 0.945 and 0.905, respectively. This attests to the excellent performance of the urine dipstick. The likelihood of having a protein-creatinine ratio of 1 or greater was 7% when urine dipstick protein value was 1 + or 2+, 62% when dipstick protein value was 3+, and 92% when dipstick protein value was 4+. Thus, dipstick proteinuria assessed by an automated reader remains a useful test to predict the severity of proteinuria and can be used to guide therapy. A cutoff dipstick proteinuria value of 3+ has the best combination of sensitivity and specificity (96% and 87%, respectively) in predicting a protein-creatinine ratio of 1 or greater and a cutoff value of 4+ in predicting a protein-creatinine ratio of 3 or greater (sensitivity, 94%; specificity, 83%). These results suggest that patients with 3+ or greater proteinuria should not only trigger appropriate evaluation of proteinuria, but also have mean arterial pressures targeted to less than 92 mm Hg. (C) 2002 by the National Kidney Foundation, Inc.