[System of internal laboratory control over the quality of clinical biochemical research]

An analysis and testing of the system of on-laboratory control on the quality of biochemical investigations with the use of control cards, has been performed. The system provides for the working out of control cards, the carrying out of control investigations, the evaluation of the results, and the correction of the control cards. Data for the built-up of the cards has been obtained with the practice of as many as 30 replications with one and the same sample, under the same conditions, the values of the average quadratic deviation (S) being in compliance with Tonks criterion. The action of control is substantiated through doubling the study of 10 per cent of the samples of each batch for each analysis and evaluation of the differences with regard to S by a two-mark system. An average mark from 0.00 to 2.00 per batch of samples is formed for each analysis. Filled in control cards for the total calcium and inorganic phosphorus are presented as an example. The changes in the S values are followed up after the introduction of control cards for 9 indices. With 7 indices reproducibility has proved higher. A total of 90 series of analyses have been controlled and evaluated--54 series (60 per cent) have proved 'good' (from 1.5 to 2.00), 22 series (24.44 per cent)--'satisfactory' (from 1.00 to 1.49), and 14 series (15.56 per cent)--'bad' (from 0.00 to 1.00 or at differences greater than 2.5 S) which implied that the respective analyses of the series of samples had to be replicated. A total of 24 control cards for various indices have been built up.