We have adapted for glucose determination a new approach to kinetic analyses [Anal. Chem. 50, 1611 (1978)]; it is 50-fold less dependent upon some experimental variables than is a more conventional rate method. Modification of a commercially available hexokinase/glucose-6-phosphate dehydrogenase reagent system for glucose provides that the rate of production of NADH be first-order in total glucose concentration within about 30 s after sample and reagent are mixed. In the kinetic method, absorbance vs. time data recorded after 30 s and a multiple-linear-regression program are used to compute the absorbance change that would occur if the reaction were monitored to completion. Results demonstrate a linear relationship between glucose concentration and computed absorbance change. Application of the method to 51 human sera without rigorous control of either temperature or reagent composition yielded a regression equation of y = 1.01x -0.3 when kinetic results (y) were compared with equilibrium results (x) for the same samples analyzed in a hospital laboratory.