Development of software for reference values validation by the Monte Carlo technique

Veli Kairisto, University of Turku, Department of Clinical Chemistry, Turku, Finland

Allan Poola, ProExpert Ltd., Tallinn, Estonia

Production of reference values is cumbersome and only few laboratories actually

produce reference values by themselves. Most clinical laboratories depend on

reference limits produced in other laboratories. However, it is customary that

the validity of the "borrowed" reference intervals is checked by analyzing a small

number of samples from healthy individuals. There is an apparent need for a powerful

statistical method to extract as much information as possible from these few samples.

With increased capacity of computers computing intensive statistical methods have

become a practical choice for treatment and estimation of data. These methods can

be applied also for the validation of reference values.

Holmes et al. (Clin Chem 1994;40:2216-2222) presented a method based on Monte Carlo

sampling technique for the verification of reference intervals. We developed

software which applies the method suggested by them. The lowest number of control

samples accepted is 7, however, we recommed that at least 20 control samples were

examined before attempting the calculations. Our software functions under

Microsoft Windows. Data for the program can be imported by standard Windows

procedures. The graphical output includes both the reference distribution and

the distribution of the control samples in a double histogram display. The Monte

Carlo sampling based distributions of the test statistics are also displayed in

the graph. As numerical output the program produces probability values for the

null hypothesis that the control samples are nothing more than a small sample

randomly drawn from the reference distribution. These p-values are calculated

for all of the selected test statistics. Probably the greatest practical problem

in the application of the method is the poor availability of original sets of

reference values. This problem could be solved by the international generation