A collaborative study of the measurement of glycosylated haemoglobin by several methods in seven laboratories in the United Kingdom.

A collaborative study compared methods for measuring glycosylated haemoglobin in seven laboratories in the United Kingdom. No satisfactory standard for general use was found. Satisfactory internal quality control systems were in use for each assay which allowed the maintenance of a normal range in each participating laboratory. No satisfactory quality control system suitable for general use could be identified. Costs and convenience of the assays are reported. The technical problems with each type of assay and precautions for their use were identified, such as the need for standardization in incubation times, the practicability of automation of colorimetric assays, and the precision of pH needed for buffers in column separation methods. The relevance of the technical problems to interpretation of measurements is also considered. It is concluded that laboratories measuring glycosylated haemoglobin should maintain a normal range, use 'in-house' quality controls to monitor assay performance and keep clinical colleagues informed of the findings and of any changes in methodology that might affect the interpretation of results.