The detection of rheumatoid factors (RF) is used as one point of criteria for the diagnosis of rheumatoid arthritis (RA), by American Rheumatism Association (ARA). A conventional method used in routine laboratory for RF detection is latex agglutination test which mainly measure IgM RF and has been found to be positive in only 60-70% of RA patients, as reported from many countries including Thailand. A recent plubication reported that ELISA for RF detection had higher sensitivity and specificity than the latex agglutination test. In addition, this technique was able to detect RF classes which may be useful for the differential diagnosis and the explanation of pathogenesis of RA. In this study 2 methods of ELISA Microplate and Dot Blot ELISA were established for the determination of RF classes in RA and other diseases and the results were compared. Microplate ELISA and latex agglutination for RF detection were also compared for their sensitivities and specificities. The studied sera were obtained from 77 patients with RA, 200 normal individuals, 14 SLE patients 9 patients with juvenile rheumatoid arthritis 24 patients with other rheumatic diseases 56 patients with viral infection and 30 cancer patients. It was found that some sera of subjects other than RA had detectable amounts of IgM RF and IgA RF by ELISA, the levels of these RF classes were significantly lower than those in RA sera. The levels of IgM RF was found to correlate with IgA RF (r=0.62, p<0.001) but such a correlation was not found either between IgG RF and IgA RF or IgM RF and IgG RF. In this study the sensitivity of IgM RF, IgG RF, IgA RF, IgD RF and IgE RF determined by ELISA were 46.75%, 59.74%, 46.75%, 27.67% and 9.1% with 97.60% 90.99%, 92.49%, 97.60% and 10% specificity respectively. Dot Blot ELISA for RF detection developed in this study showed good correlation with the the Microplate ELISA for all three classes of RF (IgM, IgG, IgA) determination (r>0.85, p<0.001) and also with the high agreement index (Kappa>0.9, p<0.001). Furthermore, it could be performed more rapidly than Microplate ELISA. The dot blotted membrane could be stored for 6 weeks at 4 C before used in the test without causing any deterioration of the rabbit IgG which was adsorbed on this membrane, acting as the antigen in this method. In this study, latex agglutination test for RF detection was found to have 83.1% sensitivity and 92.5% specificity which were higher than those reported previously. Both IgM RF and IgA RF detected by ELISA were found to correlate with rfdetected by latex agglutination test (=0.0.6, 0.7, p<0.001). The results indicated that the latex agglutination should still be used as the routine laboratory test for diagnosis of RA according to its high senditivity and specificity provided that the reagents used had the same quality as the ones used in this atudy. The determination of RF by ELISA did not show any advantage over the latex agglutination for the diagnosis of RA. However results of Dot Blot ELISA in this study showed good correlation with the ELISA performed in microplate but it could be performed more rapidly and easily. This suggested the uselfulness of Dot Blot method of ELISA for other systems of antigen or antibody detection inwhich ELISA would be useful.