ABSTRACT
The variance of between-laboratory results for rheumatoid factor tests was reduced through the use of a reference serum preparation. Two separate proficiency testing surveys have shown reduction of 58 to 81% in variance with samples of differing levels and composition. Reduction of variance was seen in data from laboratories in which the same methods and reagents were used, but greater reduction in variance was seen in results from laboratories using different methods and reagents. This study demonstrates that comparability of results of rheumatoid factor tests can be significantly improved through the use of a standard reference serum preparation and that confidence in such results can therefore be increased. A serum rheumatoid factor level of approximately 50 IU/ml is equivalent to a titer of 160 in the Singer-Plotz latex test, which has traditionally been accepted as evidence of rheumatoid arthritis.
