Evaluating the walking ability of Pekin ducks using a treadmill performance test

Christopher J Byrd, Purdue University


Reduced walking ability is one of the most important welfare concerns facing the commercial duck industry. This is due to the likelihood that impaired walking is associated with pain, and may inhibit a duck’s ability to reach needed resources, such as food and water. Additionally, ducks with severe walking impairments are culled from the flock, which poses an economic challenge for the producer. Therefore, strategies for evaluating reduced walking ability are important for monitoring the prevalence of affected ducks within a flock. Currently, gait scoring is the most popular method for assessing walking ability on-farm. Gait scoring requires an observer to rank walking ability by assigning a categorical description of walking behavior to a duck as it walks on a flat surface (severity of reduced walking ability increases with category number). These systems are criticized for being subjective and their validity is often questioned. Further, observed differences in gait, alone, do not indicate whether a duck’s welfare is impacted. To address these concerns, a treadmill performance test was developed and used to assess whether quantitative differences in performance (the amount of time a duck could remain walking on a treadmill) existed among ducks with different gait scores (experiment 1, Chapter 2). Ducks with a gait score of 0 (GS 0; smooth gait, n=55), 0.5 (GS 0.5; labored walk without easily identifiable impediment, n=56) and 1 (GS 1; obvious impediment, n=59) were selected. Ducks were individually placed on the treadmill and allowed to walk until meeting 1 of 3 elimination criteria. In addition to measuring the amount of time ducks spent walking on the treadmill at a speed of 3.1 m/s, video was collected and later analyzed for incidences of sitting, stumbling and leaning behaviors. Results from the experiment showed that differences existed among all gait scores (P < 0.001) in regard to time spent walking. Specifically, GS0 ducks walked the longest (least square mean = 475 s) followed by GS0.5 (least square mean = 392 s) and GS1 ducks (least square mean = 287 s), respectively. No differences were found among gait scores in relation to counts of sitting, stumbling and leaning behaviors (all P > 0.05). The use of time spent walking on the treadmill as an indicator of walking ability supports the notion that a level of validity does exist for observed differences in gait. Given the treadmill’s ability to quantify differences in walking ability, experiment 2 (Chapter 3) utilized the treadmill performance test to investigate the association between reduced walking ability and Pekin duck welfare by evaluating the effect of meloxicam, a non-steroidal anti-inflammatory drug, on walking performance. One hundred twenty ducks were used for this study (60 GS0, 60 GS1) and were assigned to 1 of 3 treatments: 1) subcutaneous injection of 0.5mg/kg meloxicam, 2) saline injection identical in volume to treatment one, and 3) handling only. Ducks were individually tested on the treadmill using the same procedure outlined in experiment 1. No effect of meloxicam on walking ability was found (P > 0.05). However, gait score was associated with the amount of time spent walking on the treadmill (P < 0.001). Specifically, GS0 ducks walked for longer periods of time (least square mean = 452 s) when compared to GS1 ducks (least square mean = 314 s). While the treadmill test was able to quantify differences in time spent walking based on gait score, no effect of meloxicam on time spent walking or interaction between gait score and meloxicam were found. Given the inconclusive results found in experiment 2, further research is needed to assess the relationship between gait score and Pekin duck welfare. A number of research areas that need to be addressed specifically for Pekin ducks are presented to guide future studies on this topic.




Makagon, Purdue University.

Subject Area

Animal sciences

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server