Developing welfare assessments for livestock in a commercial environment from a commercial perspective
Developing robust, scientifically validated measures that can assess the welfare of animals for a specific production system requires considerable effort. A common scientific approach is to assess expert opinion on appropriate measures and then test and validate the most likely measures, which are usually a combination of animal, resource and management-based parameters.
In the case of some recently developed welfare assessment schemes a site visit is required, and this provides an overview of the welfare status of animals at a specific point in time. On-animal sensors are increasing in their capability to measure a range of biological and behavioural parameters and are becoming more accessible and affordable. This provides the option to have real-time, continuous data collected and analysed for both welfare and production purposes.
This increase of interest in continuous data collection is occurring in a research environment as well as commercially from a production perspective, particularly in the dairy industry. However, there is not yet a clear path on how best to use sensors in an integrated way to provide valid and actionable data on the welfare state of livestock in a commercial production environment. The outcome for welfare status in a commercial setting is likely to be somewhat simplified e.g. alerts on changes in key indicators or dropping below a specified threshold on a welfare scale. However, there will need to be multiple integrated measures underlying this simple outcome. It is likely some of these measures will cover the most prevalent welfare and health issues for a species within a production environment and reflect various physical and psychological health changes. For example, in ruminant animals’ rumination is a measure that reflects overall health and, when linked to various behavioural changes, can provide information on issues including poor nutrition, disease occurrence and distress. Behaviours exhibited the differ from ‘normal’ patterns are also likely to be a useful measure and may allow negative affect and specific issues, such as stereotypies, to be identified.
Other more specialised measures may include simple alerts providing feedback on environmental issues such as predation events. There is no easy solution to providing a continuous measure of welfare for livestock that is feasible in a commercial environment. The most likely model is one where aggregated data underlies and influences an outcome.
This model will need to be scientifically robust with an outcome that is easily understood, that is useful and informative, has value for the commercial entity producing the product, the producer using the product and those interested in monitoring welfare outcomes. No doubt a difficult task, but well worth the effort!