First international workshop on the standardisation of insulin autoantibodies

1. Although certain features distinguished ELISA and radiobinding systems, there was considerable inter-laboratory variation in both. As long as such variation persists, assays should be considered individually rather than grouped according to type. 2. The assays showed less concordance when measuring insulin autoantibodies than insulin antibodies from insulin treated diabetic patients. 3. Non-specific binding was a very important source of variation in some assays (ELISA as well as radiobinding assays). 4. Specificity for human insulin was clearly demonstrated in some IAA sera. 5. The observed variation in IAA assays could account for the differences in published reports on the frequency and associations of IAA.