Laboratory-based surveillance is fundamental to effective rabies prevention and control. The direct fluorescent antibody (AB) test (FAT) is the gold standard for rabies diagnosis. Recently, additional tests besides the FAT have been developed, such as the direct rapid immunohistochemical test (DRIT). In this study, our objective was to further refine technical aspects of the DRIT using a combination of two monoclonal ABs (MABs), 502 and 802, conduct additional testing among rabies reference laboratories using a diversity of animal species and rabies virus (RV) variants and compare the potential utility of the DRIT for end users via proficiency testing (PT) against the FAT. Considering the ideal molar ratios of biotin to AB in formulation of the DRIT conjugate, 3.9 was found to be superior to 7.4, for detection of RV antigens in the brain of a naturally infected raccoon. Optimization of the DRIT conjugate may also be dependent upon the apparent choice of specific viral antigens for testing, as a gray fox RV variant reacted less strongly than a raccoon RV variant in determining the working dilution of the MAB cocktail. Using the same MABs and protocol, the DRIT was compared to the FAT using more than 800 samples of mammalian brains, representative of more than 25 taxa, including in excess of 250 animal rabies cases from Europe and North America. Sensitivity was determined at 98% (96–100%, 95% CI) and specificity was calculated at 95% (92–96%, 95% CI). In a comparison among end users, PT of laboratory personnel resulted in values of 77–100% sensitivity and 86-100% specificity. Based upon these and previously reported results, the DRIT appears to be a suitable alternative to the FAT for use in lyssavirus diagnosis.