An albumin selective urine strip based on bis (3β²,3β²β²-diiodo-4β²,4β²β²-dihydroxy-5β²,5β²β²-dinitrophenyl)-3,4,5,6-tetrabromo sulfonphthalein dye (DIDNTB) dye was examined in populations with clinical proteinuria. The relationship of albumin to the sum concentration of all protein in urine was found to vary widely even though the albumin concentration generally increased with the total protein concentration. The albumin reagent strips correlated well with immuno-nephrometric assays for albumin on specimens from hypertensives, diabetics, and renal disease which tended to have albumin contents of β₯ 50.0%. High proteinuria concentrations of > 250 mg/l, with low albumin contents of β€ 30%, occurred more frequently in cases of cancer, infection, and myeloma. The albumin strip read higher than the immuno assay in samples with high proteinuria and low albuminuria. The albumin strip was also less affected by albumin fragmentation than by the immunological assay. Overall, the albumin strip gave a lower risk of false negatives than a protein strip based on tetrabromophenol blue (TBPB) dye and was more sensitive to disease condition. The protein strip was not sensitive to low levels of albumin and the agreement between TBPB dye strip and the quantitative analysis was not as affected by the albumin content.