A collaborative study was carried out to determine the reproducibility of the inhibition form of the radioallergosorbent test (RAST). Seven laboratories were provided with three coded extracts of cocksfoot (Dactylis glomerata) pollen. All laboratories used a common method of RAST inhibition and were supplied with all reagents required to carry out the assays. In addition, four laboratories used local methods of RAST inhibition. Potency estimates were in reasonable agreement between laboratories when the study method was used but insufficient data were available to assess reproducibility when local methods were used.
Two extracts prepared from the same source material were found to have parallel log dose-response lines but the line for the unrelated extract was steeper in all but one of the assays. This finding was an indication that the third extract differed qualitatively from the other extracts.
It was concluded that the use of a study method of RAST inhibition resulted in good agreement between potency estimates obtained by different laboratories.