Optimizing moving average control procedures for small-volume laboratories: can it be done?

INTRODUCTION: Moving average (MA) means calculating the average value from a set of patient results and further using that value for analytical quality control purposes. The aim of this study was to examine whether the selection, optimization and validation of MA procedures can be performed using the already described bias detection simulation method and whether it is possible to select appropriate MA procedures for a laboratory with a small daily testing volume. MATERIALS AND METHODS: The study was done on four analytes: creatinine, potassium, sodium and albumin. All patient results of these tests processed during six months were taken from the laboratory information system. Using the MA Generator software, different MA procedures were analysed. Different inclusion criteria, calculation formulas, batch sizes and weighting factors were tested. Selection of optimal MA procedures was based on their ability to detect simulated biases of different sizes. After optimization, the validation of MA procedures was done. The results were presented by bias detection curves and MA validation charts. RESULTS: Simple MA procedures for albumin and sodium without truncation limits were selected as optimal. Exponentially weighted MA procedures were found optimal for creatinine and potassium, with the upper truncation limits of 150 μmol/L and 6 mmol/L, respectively. CONCLUSIONS: It has been experimentally confirmed that it is possible to perform the selection, optimization and validation of MA procedures using the bias detection simulation method. Also, it is possible to define MA procedures optimal for a laboratory with a small daily testing volume.