The effects of feeding pelleted timothy hay to ameliorate the duration, frequency, and latency to start cross-sucking behavior in paired dairy calves

Cross-sucking is an abnormal oral behavior that is a potential barrier to the adoption of social housing of calves. The objective of this randomized control trial was to assess whether providing pelleted timothy hay to pair-housed dairy calves could affect the duration, frequency, and latency to start cross-sucking behavior surrounding the daily milk feedings for the first 21 d after pairing. We enrolled 24 calves (n = 12 pairs; 6 pairs negative control [CON], 6 pairs timothy hay pellets [Hay]) at pairing at 5 ± 3 d of age. Calves were fed 7.4 L/d of milk replacer by bottle in 2 equal feedings and access to water by bucket from birth and had access to calf starter grain by bucket and by bottle ad libitum after pairing. Orts were recorded daily by scale for hay and calf starter intake. Calves were video recorded for 21 consecutive days after pairing. Two trained observers (ĸ = 0.80) watched cross-sucking events twice daily for the hour before and 2 h after each milk feeding for cross-sucking behavior using BORIS software. A cross-sucking bout was defined as one calf suckling on the navel, udder area, or between the back legs of the receiving calf and ended when the cross-sucker broke contact with the defined areas of the receiving calf for ≥10 s. The cross-sucking bouts were summarized into the daily frequency for each pair, and the duration of the bouts was summed into the daily duration by pair. The latency to first cross-suck was the first day on which a cross-sucking bout was observed within a pair. Data were non-normally distributed (Shapiro-Wilk criteria <0.95) and were common log-transformed with a 1.0 correction factor for model fit. We used mixed linear regression models to evaluate the association of Hay with the cross-sucking duration, cross-sucking frequency, day, and the Hay × day interaction, repeating by day, and using pair as the subject. Latency to begin cross-sucking was assessed using Pearson's chi-squared test. The back-transformed geometric means minus the correction factor and 95% CI of the back-transformed data are reported. We observed that Hay and the Hay × day interaction were not associated with the duration or frequency of daily cross-sucking bouts surrounding milk feeding. The geometric mean daily duration of cross-sucking surrounding milk feeding for CON was 54.7 (95% CI: 49.3 to 60.5 s) and for Hay was 59.4 (95% CI: 53.3 to 65.5 s). The geometric mean daily frequency of cross-sucking bouts surrounding milk feeding for CON was 7.8 (95% CI: 7.5 to 8.1 bouts) and for Hay was 8.1 (95% CI: 7.7 to 8.4 bouts). The latency to cross-suck was not associated with the provision of the Hay pellets and occurred by a maximum of 2 d after pairing (95% CI: 0.81 to 2.52 d). However, the average pelleted hay DMI throughout the study was negligible (Hay 89.41 ± 99.89 g/d; mean ± SD), reaching 220.51 ± 108.81 g/d on d 21. Our results suggest that offering timothy hay to pair-housed calves was not associated with cross-sucking behavior, though the behavior occurred for a short duration surrounding milk feeding, and was variable by pair.