A selected number of PCR protocols were evaluated to determine if they could serve as a universal protocol for detecting and identifying all arboviruses. In this study, four parameters that affect the efficacy of RT-PCR (RNA extraction method, choice of reverse transcriptase, choice of DNA polymerase and thermocycling program) were evaluated in combination. The most optimal combination of those parameters employed use of silica gel membrane spin column, RAV-2 reverse transcriptase, Tth DNA polymerase, and a simple modification of a published thermocycling program. By this modified protocol, viral RNA could be amplified satisfactorily with more than 50 pairs of primers designed for diagnosis of arboviruses representing five families. The sensitivity and specificity obtained by this universal protocol were comparable to those obtained by the original protocol for each primer pair tested; and for some primers, improved sensitivity was observed. It was also found that a simple modification of a suggested protocol of a commercial RT-PCR kit could produce nearly identical results and serve as another universal protocol. With the use of a universal diagnostic reverse transcriptase-polymerase chain reaction (RT-PCR) protocol, simultaneous screening of clinical or biological specimens against a large number of RNA viruses belonging to many families can be performed more efficiently for etiologic determination in the situations complicated by the difficulty of differential diagnosis. Furthermore, such a universal protocol facilitates reducing the cost of PCR-based diagnostic operation and standardizing the qualities of PCR-based diagnosis within an institution or among collaborating institutions. A logical strategy is to conduct diagnosis in two stages by using broadly group-reactive primers in the first stage to narrow the range of possible etiologic agents and using virus-specific primers in the second stage for identification. Before such a strategy is employed, however, more group-reactive primers for a large number of arboviruses, for which no such primers currently exist, must be made available. Furthermore, the best pair or pairs of primers need to be selected for each virus for the second stage of the strategy.