Evaluation of a commercial activity monitor for determining step counts in horses

Abstract Monitoring activity level has become popular in people and animals. In the horse, these monitors may be helpful for detection of disease, monitor limb loading after surgery, and assist in rehabilitation. The objective of this study was to evaluate the accuracy of a commercial activity monitor (AM) (FitBit ZipTM) in horses. In Part 1, four stalled horses had five AMs placed (head, chest, withers, and left fore- and hind-limbs) with stride length setting of 60.96cm and were monitored for 24hrs with AM and video. In Part 2, four stalled horses had two AMs placed (left fore- and hind-limbs) with stride length setting of 30.48cm and were monitored for 24hrs with AM and video. In Part 3, three horses with two AMs (left fore- and hind-limbs) had step data collected at the walk and trot in hand by AM and video. AM and video data were compared using Pearson’s correlation coefficients and by calculating AM to video step count ratios. In Part 1, there was considerable movement of AMs at the chest and withers but head and limb mounted AMs were more secure. In Part 2, video and AM activity counts were moderate to high for fore- and hind-limbs with R2 values of 0.82 and 0.74, respectively. From Part 3, AMs mounted on either fore- or hind-limbs approximately doubled the step counts compared to video. While this AM did not accurately count steps, it may be able to provide general information on activity level in stalled horses.