276 Use of triaxial accelerometers and a machine learning algorithm for behavioral identification to assess the effectiveness of a joint supplement in old domestic cats (<i>Felis catus</i>)

Abstract Degenerative joint disease (DJD) has a 90% prevalence in older cats and adversely affects quality of life1. Long-term administration of non-steroidal anti-inflammatory drugs is usually required but can have adverse effects2. Nutraceuticals have been shown to relieve joint pain in cats without such side effects; however, due to differences in dose rates and formulations there is no consensus as to their efficacy. Most studies have utilized subjective insensitive measures such as veterinary or owner-assessed changes in activity. Accelerometers can provide objective data on cat activity and can identify specific behaviors3.Triaxial accelerometers (ActiGraph, Pensacola, FL) were fitted to the collars of 16 cats (aged 11 to 16 yr; Massey University, Palmerston North, NZ) for 20 wk to assess their activity and behavior. The cats were fed the same diet (Chef: Kraft Heinz Wattie’s Ltd., Hastings, NZ) for 4 wk to collect baseline acceleration data, before being divided into two groups balanced for physical activity. After 4 wk, a blood sample were taken from all of the cats to establish their baseline levels of a range of inflammatory markers (e.g., IL-6, TNFα, INF- γ, SDF-1). The test group (n = 8) then received a joint supplement added to the baseline diet while the control group (n = 8) were fed the diet unsupplemented for 16 wk. The commercial joint supplement contained glucosamine, chondroitin sulphate, and green-lipped mussel (Tandem, Wellington, NZ), at daily doses shown previously to be effective in cats4,5,6. The aim of the work was to assess the efficacy of the joint supplement in maintaining or promoting activity and specific behaviors in the cats. At the end of the period of supplementation a second blood sample was taken and assessed for the same range of inflammatory markers. Overall physical activity was not different between groups at baseline but was greater over the study in the test group compared with the control (P &amp;lt; 0.03). Grooming, as identified by the algorithm, showed a tendency (P = 0.051) to be greater in the test group than the control. These results suggest that the supplement had a positive effect on both overall activity and more specifically on grooming behavior in old cats. However, no differences in the range of inflammatory markers were seen before or after supplementation. While the behavioral measures could indicate the supplement reduced the effects of DJD, further research is required to confirm this.1Cunningham et al (2022) Journal of Feline Medicine and Surgery 24, 962-974. 2Lemetayer and Taylor (2014) Journal of Feline Medicine and Surgery 16, 547-562; 3Smit et al., (2023) Sensors 23, 7165; 4Corbee (2022) Veterinary Medicine and Science 8, 1025–1035; 5Corbee et al. (2013) Journal of Animal Physiology and Animal Nutrition 97, 846–853; 6Lascelles et al. (2010) Journal of Veterinary Internal Medicine, 24, 487–495.