The Westmead HIV-1 antibody testing strategy showed that, regardless of ELISA screening kit manufacturer, sera which were repeatedly positive by two ELISA screening assays (one indirect and the other competitive format) had a 97-98% chance of being confirmed positive by Western blot for HIV-1 antibody or a less than 3% chance of either being identified as a seroconverter (1%) or a late stage AIDS patient (1.2%). Sera which were discordant by two ELISA screening assays had a less than 4% chance of either being confirmed positive by Western blot (2.5%) or identified as a seroconverter (1.3%). The incidence of non-specific indeterminate Western blot profiles were shown to be inversely proportional to the specificity of the ELISA screening kits used. The use of a recombinant envelope ELISA was able to confirm the viral specificity of HIV-1 envelope bands (gp160, 120 or 41) on Western blot. Guidelines suggested by the Australian National HIV Reference Laboratory, Fairfield Hospital, Melbourne, which categorized indeterminate or typical Western blot profiles into four reaction groups were found to be useful for the interpretation of Western blot patterns. A Western blot profile which is reactive for HIV-1 viral glycoproteins (gp160, 120 and 41) alone or in combination with not more than two other viral proteins (Indeterminate Group 4) and which is confirmed viral envelope specific by a recombinant envelope ELISA can be used as a predictor of seroconversion.