THE CALCULATION OF THE MINIMUM NUMBER OF DAILY NORMAL RESULTS IN ORDER THEIR MEAN VALUE TO BE USED AS A QUALITY CONTROL METHOD

P. Karkalousos, S. Psarakis

Purpose

The “Average Of Normals” (AON) method has been proposed since 60’s as an alternative quality control method. According to this method the average of the daily normal results must be varied within certain limits. The violation of these limits results to analytical systematic error. The purpose of this study is to find the minimum number (N) of normal results for which their mean value can be used as a substitute of the common control samples. Former studies proved that this number depends on the ratio Sp/Sa where Sp is the variance between health people and Sa is the analytical variance.

 

Method

A specific software was constructed based on a large number of simulations according to different analytes, minimum numbers N, human variance (Sp), analytical variance (Sa), control chart limits (L) and the truncation limits of the normal results (T).

 

Results

Many different cases of Sp/Sa were studied. The relations between Sp/Sa and minimum N for the detection of analytical systematic error of the size of two standard deviations with probability 0,9 are: 1 to 5, 2 to 15, 3 to 32, 4 to 47, 5 to 80, 6 to 170, 7 to 210, 8 > 300. For instance: Τ3 (N=5), Τ4 (N=90), TSH (N=210), Chol (N=210), GOT (N=320).

 

Conclusions

The number N depends on L, T and Sp/Sa. Because L and T are constants and the variance Sa is extremely small on the modern analyzers, the number N depends mostly on the Sp value or better on the reference values of the analyte.

Tελευταία ενημέρωσηΔευτέρα Σεπτεμβρίου 07, 2009