In a collaborative study between the University laboratory and central laboratory, accuracy and precision of milk analyses for fat and protein content were studied. Each week during a 2-year period, milk from individual cows of four breeds (SRB, SLB, SJB and SRB×SLB crosses) was analysed by reference methods in use at the University laboratory (situated near the herd) and by fully automated indirect methods (Milko-Scan 300 and Milko-Tester) at a central laboratory. Systematic inter-laboratory errors were determined for both fat and protein analyses. The central laboratory gave lower (0.12 percentage units) fat values (p<0.001) and higher (0.07 percentage units) protein values (p<0.001). The overall accuracy (standard deviation of difference between laboratories) of fat and protein determinations was 0.21 percentage units (C.V.=4.4%) and 0.14 percentage units (C.V.=4.1%), respectively. The overall inter-laboratory variation in analysis differences for both fat and protein was explained by biological factors (breed, cow, stage of lactation), level variations (calibration) and a small proportion of extremely large deviating results (outliers). In addition, protein determinations were affected by somatic cell count level due to low lactose content.
No takes yet. Share an insight, caveat, or question.
Sjaunja et al. (1984) studied this question.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: