Mastitis is a problem in the sheep industry, and its incidence varies widely with how it is recorded, the breed of sheep and the farm. Virtually all the published information about the genetics of mastitis refers to dairy breeds of cattle and sheep, and there is little information for meat sheep breeds. Many dairy breeding programmes worldwide use the somatic cell count (scc) in milk as an indicator of resistance to clinical and subclinical mastitis, but it is difficult to measure in meat sheep breeds. Molecular genetic technologies may therefore be a more practical way to assess susceptibility to mastitis. This paper reviews the genetics of mastitis and considers the opportunities for breeding for resistance to mastitis, with particular reference to sheep. In addition, to investigate the potential economic effects of mastitis in a purebred sheep population, a computer model of flock dynamics was developed. By making a modest set of assumptions about the key farm parameters that influence lowland sheep production, the model showed that breeding for resistance (or other control methods), if it could reduce the risk of contracting mastitis by 10 per cent, would be worth pounds 8.40 per ewe, equivalent annually to pounds 2.7 million for the purebred sector of the Texel breed alone.

Culling dairy cows with clinical mastitis reduces the incidence of the disease in dairy herds, but the costs of such action, in terms of reduced milk production and increased replacement costs, are generally thought to outweigh the benefits. To test this hypothesis a stochastic dynamic programming model was developed to establish the economically optimum time of replacement for dairy cows subject to variable levels of clinical mastitis infection, using average United Kingdom production and price parameters. The optimal stage at which to replace a dairy cow was found to be sensitive to changes in mastitis incidence and in critical price parameters within the bounds of commercial experience. This result indicates that an objective culling policy based on clinical mastitis records in addition to milk production potential may be economically viable.

Generic frameworks for the economic analysis of farm animal disease are now well established. The paper, therefore, uses bovine viral diarrhoea (BVD) as an example to explore how these frameworks need to be adapted to fit the characteristics of a particular disease and the specific objectives of the analysis. In the case of BVD, given the relative strength of tests available to correctly identify virus-positive animals, thus enabling them to be culled, the emphasis has been on cost-benefit analysis of regional and national certification/eradication schemes. Such analyses in turn raise interesting questions about farmer uptake and maintenance of certification schemes and the equity and cost-effective implementation of these schemes. The complex epidemiology of BVD virus infections and the long-term, widespread and often occult nature of BVD effects make economic analysis of the disease and its control particularly challenging. However, this has resulted in a wider whole-farm perspective that captures the influence of multiple decisions, not just those directly associated with disease prevention and control. There is a need to include management of reproduction, risk and enterprise mix in the research on farmer decision-making, as all these factors impinge on, and are affected by, the spread of BVD.

No abstract

Finland's cold loose housing systems (CLH) for dairy cows have been compared to the more traditional warm loose housing systems (WLH) in regard to the incidences of udder and ovarial disorders, ketosis, parturient paresis and metritis. For both systems approximately 10,000 calvings on 210 farms during the years 1996 and 1997 were modelled using survival analysis and two-level logistic regression in an observational cohort study. The results of the different multivariable models, the modelling process, model diagnostics and sample size and confidence limit analysis were combined in an attempt to estimate an unbiased risk ratio. The estimated risk ratios show that there are some differences in the incidence of production diseases between the 2 farming systems. It could be seen from the diagnostics that they were influenced considerably by the modelling process and some 'influential farms'. Even though some disease incidences differed statistically significantly, the observed differences were small so that we concluded that they should not influence the farmers decision making about whether or not to build a CLH in Finland. ASSOCIATIONS BETWEEN DAIRY COW SOMATIC CELL COUNT AND FOUR TYPES OF BEDDING IN FREE STALLS Kirkegaard P, Agger JF, Bjerg B.Department of Animal Science and Animal Health. The Royal Veterinary and Agricultural University, Grønnegårdsvej 8, DK-1870 Frederiksberg C, Denmark.The purpose was to test the hypothesis of no difference in associations between four types of bedding material in free stalls and the cow somatic cell count (SCC) in a study population of primiparous cows in 52 Danish dairy herds. The types of bedding were 1) sand (20-25 cm deep), 2) straw (20 cm deep), 3) channel mattresses (channel bags filled with rubber granulate and a permeable top shield), and 4) two-layer mattresses with foam rubber and a rubber top cover. These bedding types were chosen because they all provided the necessary cow comfort. The herds were randomly selected within each of the four types of bedding systems and represented all regions of Jutland in Denmark. Information on the stable interior and hygiene, cow hygiene, feeding, health management, milking and ventilation were collected at farm visits. Data on milk production and somatic cell count were provided from The Danish Cattle Data Base.