The purpose of the study was to examine the validity of alpha1-microglobulin (alpha1-MG) in comparison with popularly used beta2-microglobulin (beta2-MG). A database was revisited to select ca. 7,500 spot urine samples (of adequate urine density) from non-pregnant, non-lactating and never-smoking adult women. The validity of the MGs was examined in terms of stability of the MG-uria prevalence in urine samples of various creatinine (CR or cr) concentration or specific gravity (SG or sg). Comparisons were made for MGs as observed (e.g., alpha1-MGob), as corrected for CR (e.g., alpha1-MGcr) and as corrected for SG of 1.016 (e.g., alpha1-MGsg). A cut-off value of 5.7 mg/g cr (or mg/l) for alpha1-MG was deduced from a cut-off value of 400 microg/g cr (or mcirog/l) for beta2-MG, because the correlation between alpha1-MGcr and beta2-MGcr was statistically significant. The prevalence of a 1-MGsg-uria was essentially unchanged (i.e., from a low of 13.6% to a high of 17.0%, or 1.2 times) except for in very dense or very thin urine samples, in contrast, beta2-MGcr-uria showed a substantial increase (from 0.0% to 2.8% with an infinite rate) as a reverse function of a decrease in CR in urine. The prevalence of uncorrected markers, i.e., alpha1-MGob-uria and beta2-MGob-uria, showed even greater CR- or SG-dependent changes. Thus, it appeared prudent to consider a alpha-MGsg rather than beta2-MGcr as a marker of tubular dysfunction among a general population with various urine density.