A meta-analysis approach to evaluate the effects of early group housing on calf performance, health, and behavior during the preweaning period.

This study aimed to evaluate the effects of early group housing on the performance, health, and behavior of dairy calves during the preweaning period using systematic and meta-analysis approaches. Peer-reviewed articles written in English that compared dairy calves individually and group-housed with performance, health, or behavior outcomes were collected from Web of Science, PubMed, and CABDirect databases. The resulting articles (n = 850) underwent a 4-step appraisal process following the Preferred Reporting Items for Systematic Review and Meta-Analyses protocols, resulting in a final sample of 51 articles containing 85 studies. The weighted or standardized mean differences between individually housed and group-housed (pair or group with >2 calves) calves were analyzed for each variable using the DerSimonian and Laird methods. Heterogeneity between calf housing systems was evaluated by the χ ² test and I ² statistics. A meta-regression analysis was conducted to identify categorical covariate effects for variables with high heterogeneity. Most of the studies included in this review evaluated female calves (45.8%) weaned at 8 wk old (52.6%). Housing systems were mainly paired housing (55.3%), followed by groups of 3 to 6 calves (30.65) and groups of 7 to 15 calves (4.7%). We did not find studies with comparable outcomes for more robust health parameters (as occurrence of diseases), only blood parameters, that were not influenced by the housing system. However, weight gain and feed intake parameters were higher in group-housed calves. Through the meta-regression, we found that ADG was positively affected by the group housing in studies with calves housed in small pen areas (<1.5 m ² per calf). Group-housed calves presented more active behaviors (feeding and playing) and less stress-related behaviors (self-grooming and interacting with the pen) than individually housed calves. The behavioral tests most used were novel object, human approach, novel environment, and social tests. Individually housed calves presented fewer vocalizations on these tests and spent more time interacting with humans during the human approach test than group-housed calves. Our findings provide consistent evidence that group housing improves the welfare of dairy calves; however, the effects on health parameters are still scarce and unclear.
(The Authors. Published by Elsevier Inc. on behalf of the American Dairy Science Association®. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).)