RESUMEN
A collaborative study to assign values to two enzyme reference materials (ERMs) was performed by 18 laboratories whose spectrophotometers were checked by us, just before the study. We measured the wavelength accuracy and repeatability, the accuracy and linearity of the absorbance curves, the cuvette pathlength, equilibration time, equilibrium temperature, and a few other variables. Five spectrophotometers exhibited a marked wavelength-dependent nonlinearity. Most instruments were rather slow in bringing the sample to the correct temperature and the final temperature was often too high. In the collaborative study, each participant performed the same manual, well-described methods on four occasions in triplicate, using reagents prepared locally. The relation between the photometric checks and the analytical results is discussed, as well as the treatment of outliers and the effects on the variances. Suggestions are made about various facets of collaborative studies. The values assigned to the two ERMs carry a 95% uncertainty interval of +/- 1-4% of the mean.
