Abstract
Seven- and 14-day net protein ratio (NPR) data were obtained from 7 laboratories for 6 protein sources: ANRC casein, lean beef, Iactalbumin, textured vegetable protein, and peanut flour were fed as 10% protein (N × 6.25) in the test diet. Wheat flour, casein, and textured vegetable protein were fed as 6% protein (N × 6.25) in the test diet. Weighed dry ingredients for each diet were sent to each collaborator, who mixed the dry ingredients, then added specified amounts of corn oil and water and mixed each complete diet thoroughly. Rats were adapted for 0, 2, or 4 days, and then were fed the test diets for 28 days for protein efficiency ratio (PER) diets. The animal weight gain and feed consumption data obtained after 7 or 14 days of feeding were used to calculate NPR values. Analyses of data were done before [net protein ratio (NPR)] and after (R-NPR [relative-NPR]) adjustment of the data from each laboratory by its results for the reference protein casein. From the analysis of variance for NPR, significant (P <0.05) interactions were observed among laboratories, protein sources, and adaptation times of the animals (0,2, or 4 days). Inter- and intralaboratory variability were decreased by use of 14-day values compared with 7- day values. Adjustment of the NPR data to R-NPR did not lower the intralaboratory variability but did lower the interlaboratory variability of the data. Increasing adaptation time did not consistently decrease interlaboratory or intralaboratory variability or decrease coefficients of variation (CV) of R-NPR values. The 14-day NPR inter- and intralaboratory variations for the 10 % protein diet over all factors (5 protein diets, 3 adaptation periods, and 7 laboratories), as measured by CV values, were 13.2 and 7.7%, respectively. The corresponding R-NPR values were 9.2 and 8.0%, respectively.