Regional quality control of coagulation assay procedures (normotest, simplastin-a and thrombotest).

The report is based on the first 8 months' operation of an extended regional quality control program involving 20 hospital laboratories in the Stockholm area. The test material was commercially available lyophilized human reference plasma containing uniformly reduced levels of factors II, VII and X, and optimal concentrations of fibrinogen and factor V. Each participant received 2 batches of the test material with different activity at each 4-week period. Participants were requested to analyze the specimens, independently and in duplicates, along with ordinary patient samples three times weekly during each period. The levels of the factors in the batches were changed in each period. The total number of reported results was 3113. The statistical analysis of values included the calculation of means and variances within and between different thromboplastins (Normotest, Simplastin-A, Thrombotest), different techniques (venous and capillary blood, automatic and manual end-point determination), and laboratories classified according to the category and number of personnel and of performed coagulation tests. Systematic differences between thromboplastins, techniques and laboratories, even using the same type of thromboplastin, contributed a significantly greater amount than within-laboratory random variation to the total variation of determinations. The obtained results indicate the urgency of the quality control of coagulation activity tests.