The level of secretory immunoglobulin A (SIgA) in saliva is associated with the risk of upper respiratory infection and has been identified as a condition indicator for athletes and an indicator of mucosal immune function in the elderly. In addition, it has been suggested that non-specific polyreactive SIgA is present in human saliva and colostrum. The purpose of this study was to establish a new detection method of polyreactive salivary SIgA. Using saliva samples from eight healthy individuals, indirect ELISA was used to detect anti-LPS antibodies (LPS ELISA), and competitive ELISA was used to detect LPS antibodies after competition with DNA (LPS-DNA ELISA). The difference in SIgA concentration between LPS ELISA and LPS-DNA ELISA is thought to be the concentration of the polyreactive SIgA, which responds to both DNA and LPS. First, we determined the concentration for coating LPS and DNA as 10 μg/ml, and the saliva sample dilution was determined as 1/2. The concentration of anti-LPS antibody contained in the saliva sample, which is strongly reactive to LPS, was defined as the standard. Compared to the SIgA concentration determined by LPS ELISA, SIgA concentration of LPS-DNA ELISA did not decrease in two samples, but was reduced in 6 samples. The decrease in SIgA concentration is thought to be the concentration of polyreactive SIgA that reacts with both DNA and LPS. Notably, we found sample-to-sample differences in the concentration of polyreactive SIgA.