Increases in the completeness of disease records in dairy databases following changes in the criteria determining whether a record counts as correct

Evaluation of the usefulness at national level of the dairy cattle health and production recording systems in Great Britain

The aim of this study was to formally evaluate, qualitatively, the ability of existing recording systems to generate accurate and reliable estimates of the frequency of selected health conditions in the dairy herd of Great Britain. Fifty-nine recording systems were identified, of which 36 had their key characteristics defined through a web-based questionnaire. Nineteen of them were further assessed following the SERVAL, a SuRveillance EVALuation framework against a set of 12 attributes: benefit, bias, communication, coverage, data collection, data management, data analysis, data completeness, flexibility, multiple utility, representativeness and stability/sustainability. The evaluated systems showed considerable differences in their coverage, implementation and objectives. There were overlaps in recorded conditions, with Johne's disease, bovine viral diarrhoea, mastitis and lameness being recorded by most of the systems. Selection bias, data ownership and lack of integration of data from different systems appeared to be a key limitation on the future use of existing systems for nationwide monitoring. The results showed that even though the individual systems can provide reliable estimates of dairy health for individual farmers, none of the systems alone could provide accurate and reliable estimates for any of the conditions of interest at national level.