Assessment of the Lame Horse

Summary

# Assessment of the Lame Horse — Editorial Summary Accurate documentation of clinical outcomes forms the foundation of quality improvement in equine practice, yet little is known about the reliability of data collection when responsibilities fall to junior staff members. Researchers at a surgical centre implemented a prospective quality database managed by residents, then audited its accuracy over six months—initially without disclosure, followed by targeted training and continued auditing with the residents' knowledge—whilst simultaneously surveying outcome-recording practices across 108 European medical centres. The findings were sobering: approximately 80% of postoperative complications went unrecorded in both periods regardless of training intervention, though when complications were documented, grading accuracy reached 97%; comorbidities were missed in 20% of cases initially and 14% subsequently, improvements that fell short of statistical significance. With residents and junior staff responsible for outcome recording in 80% of European centres surveyed, the authors argue that dedicated personnel rather than rotating junior doctors should manage clinical databases, as even focused teaching failed to substantially improve detection rates. For equine practitioners relying on outcome data to benchmark performance and drive clinical improvements, these findings underscore the need for independent assessment protocols—whether through dedicated staff, standardised video review, or structured owner follow-up questionnaires—to capture the true complication landscape and avoid false reassurance from incomplete datasets.

Read the full abstract on PubMed

Key Findings

• •This paper addresses surgical quality assessment methodology in human medicine, not equine lameness despite its title attribution
• •Residents failed to report 80% of postoperative complications in both audited periods, with minimal improvement after training
• •When complications were recorded, grading accuracy was 97%, indicating recognition ability but not systematic capture