The Bhattacharya method and the ‘average of normals’ method for internal quality control were compared. Both are based on unselected patient test results. The Bhattacharya method is mostly used in clinical chemistry for calculation of reference intervals but could be modified for quality control procedures. Using the data of a coagulation test and simulating systematic errors, it was concluded that the Bhattacharya method is more sensitive to expected shifts and is more flexible than the ‘average of normals’ method.
HoffmanRGWaidME. The ‘average of normals’ method of quality control. Am J Clin Pathol1965; 43: 134–41.
2.
CembrowskiGSChandlerEPWestgardJO. Assessment of ‘average of normals’ quality control procedures and guidelines for implementation. Am J Clin Pathol1984; 81: 492–9.
3.
BhattacharyaC. A simple method of resolution of a distribution into Gaussian components. Biometrics1967; 23: 115–35.
4.
NausAJM. De berekening van referentiewaarden in de klinische chemie uit analyseresultaten van een patientenpopulatie. (In Dutch, summary in English) Thesis, Maastricht, The Netherlands1982.
5.
OosterhuisWPModdermanThAPronkC. Reference values. Bhattacharya or the method proposed by the IFCC?Ann Clin Biochem1990; 27: 359–65.
6.
BlandJMAltmanDG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet1986; i: 307–10.
7.
WestgardJOHuntMR. Use and interpretation of common statistical tests in method-comparison studies. Clin Chem1973; 19: 49–57.
8.
HotellingH. The selection of variates for the use in prediction with some comments on the general problem of nuisance parameters. Ann Mathemat Statist1970; 11: 271–83.
