Feeding behaviour and activity as early indicators of disease in pre-weaned dairy calves.

Across the industry, there is large variation in health status of dairy calves and as a result, disease incidence and antibiotic use is high. This has significant implications for animal welfare, productivity and profitability of dairy and dairy-beef production systems. Technology-based early detection systems could alleviate these issues; however, methods of early detection of disease in dairy calves have not been widely explored. This study aimed to determine whether changes in activity and feeding behaviour can be used as early warning indicators of respiratory disease in calves. In total, 100 pre-weaned male Holstein calves (age: ~8-42 days) were used. Calves were group-housed and provided with starter diet, straw bedding and ad libitum water. Calves were fed milk replacer ad libitum through an automatic calf feeder, and each calf was fitted with a leg-mounted activity monitor. Daily activity and feeding behaviour variables were calculated for each calf. Each calf was assessed daily using a modified version of the Wisconsin Scoring System to assess respiratory disease status. Calves were classed as 'Diseased', 'Intermediate' or 'Healthy' based on their cumulative health score. The peak day of the most extreme illness event was identified for each calf. Data from Diseased and Healthy calves were paired for analysis based on age and BW. Data were compared for the day of peak illness, and for the 3 days previous and post. Compared to healthy calves, diseased calves lay for longer and tended to have longer lying bouts (daily lying: 17.6 ± 0.3 vs 16.7 ± 0.2 h, P < 0.01; bout length: 74.8 ± 10.6 vs 56.0 ± 3.7 min, P = 0.09 for diseased and healthy calves, respectively). Diseased calves fed for a shorter time and had fewer feeder visits (with intake) each day compared to healthy calves (feeding time (min): 19.3 ± 1.4 vs 22.8 ± 1.5; P < 0.05; visits: 2.1 ± 0.2 vs 3.2 ± 0.4; P < 0.05). Importantly, differences between diseased and healthy calves were evident in both activity and feeding behaviour on the days prior to the peak day of disease. Lying bout length was greater in diseased calves for the 2 days prior to the peak day (P < 0.05), lying time was longer on day -1 (P < 0.05) and feeder visits with milk intake were less frequent on day -3 (P < 0.05). Thus, measurement of feeding and activity using precision technology within early detection systems could facilitate early intervention and optimized treatment.