Introduction Proper identification of healthy subjects is essential in case–control studies. However, standardized definitions of healthy controls are lacking in equine orthopaedic research.
Objectives The aim of this study was to define the non-invasive methods used for selecting healthy control horses in osteoarthritis (OA), desmitis and tendinitis research.
Methods Systematic review. Case–control studies with a healthy control group and longitudinal studies where horses had to be healthy at the start were included. Studies where joints were visualized by arthroscopy or post-mortem examination were excluded.
Results: From 2,472 OA papers and 2,746 desmitis/tendinitis papers, 127 and 84 papers met the inclusion criteria respectively. For OA, 11 methods were identified for defining healthy subjects with a median of three methods used per paper. Dynamic examination, radiographic evaluation and clinical examination were the most frequent. Eight different methods were identified in the desmitis/tendinitis papers with a median of three methods per paper; ultrasonography, clinical- and dynamic examination were the most frequent.
Conclusions Overall, the OA and desmitis/tendinitis studies used similar methods for defining subjects as healthy, but the way the examinations were performed and interpreted was inconsistent. In several studies, healthy controls were not examined for lameness. The most common methods have limitations for detecting horses with early OA, which may have implications for interpretation of results. Standardized use of more sensitive and objective methods could be beneficial.