A tale of two labs: Comparing antimicrobial resistance data in pets across commercial and academic diagnostic laboratories

Abstract

Antimicrobial resistance *AMR) presents significant challenges in veterinary medicine, necessitating accurate surveillance to inform effective mitigation strategies. Most resistance estimates for cats and dogs are based on a single data source, typically university-affiliated diagnostic laboratories *UADLs), which may limit their generalizability. This study is the first to quantitatively compare AMR data from a UADL and a commercial diagnostic laboratory *CDL) by analyzing antimicrobial susceptibility testing *AST) results for Escherichia coli and Staphylococcus pseudintermedius in cats and dogs from New York State between 2019 and 2022. The analysis focused on first-line and higher-tier antimicrobials and revealed a tendency for the UADL data to observe lower susceptibility rates than the CDL. However, the extent of this difference varied by bacteria-antimicrobial combination, geographic region, and time. A secondary objective was to develop and test a novel Shiny application designed to harmonize and prepare data for comparison without exchanging raw data, addressing several data-sharing concerns that could limit collaboration. These findings highlight how variations in data sources can affect resistance estimates and interpretations. By identifying similarities and differences, this study underscores the importance of considering data source characteristics when analyzing and applying AMR surveillance reports. Integrating data from multiple sources may provide a more balanced and representative understanding of resistance patterns, thereby supporting more effective surveillance and decision-making in companion animal medicine. Here, we demonstrate that user-friendly analysis tools can support data integration without requiring raw data to be publicly available or shared between institutions.