An Indication of Reliability of the Two-Level Approach of the AWIN Welfare Assessment Protocol for Horses.
Authors: Czycholl Irena, Büttner Kathrin, Klingbeil Philipp, Krieter Joachim
Journal: Animals : an open access journal from MDPI
Summary
# Editorial Summary: AWIN Two-Level Welfare Assessment Protocol for Horses The AWIN protocol streamlines equine welfare assessment through a two-tiered approach—initial visual inspection from distance, followed by detailed close-up evaluation of all horses—but its reliability hinged on whether the abbreviated first level captured meaningful data. Researchers compared 112 paired assessments conducted by a single observer on consecutive days, using correlation coefficients, intraclass correlations, and agreement limits to evaluate consistency between levels across multiple welfare indicators. The protocol proved generally robust, with most indicators showing sufficient reliability to justify the two-level structure; however, four key measures—the Horse Grimace Scale, Avoidance Distance Test, Voluntary Human Approach Test (ICC: 0.38) and manure consistency assessment (ICC: 0.38)—demonstrated weak between-level agreement that warrants methodological refinement. For practitioners conducting welfare audits, this work validates the efficiency gains of the two-level approach whilst flagging that behavioural indicators and faecal scoring require either closer attention during preliminary visual inspection or recalibration of first-level assessment criteria. The findings suggest scope for adapting these problem areas without abandoning the framework's practical feasibility for farm-based welfare monitoring.
Read the full abstract on PubMed
Practical Takeaways
- •The two-level AWIN protocol can reduce assessment time by using initial visual inspection for most indicators without losing reliability in welfare data
- •Behavioral avoidance tests and grimace scale assessments require close-up inspection and cannot be reliably completed from distance
- •Manure consistency assessment needs protocol refinement if only level 1 (distance) inspection is to be used
Key Findings
- •Most AWIN protocol indicators showed sufficient reliability between level 1 (visual distance inspection) and level 2 (close-up inspection) assessments
- •Horse Grimace Scale, Avoidance Distance Test, and Voluntary Human Approach Test demonstrated poor reliability between levels (e.g., Voluntary Human Approach Test: RS 0.38, ICC 0.38)
- •Manure consistency indicator showed weak reliability (RS 0.31, ICC 0.38) between assessment levels
- •Two-level assessment structure is reliable overall and could enhance feasibility of welfare assessment protocols without substantial information loss