A quantitative risk assessment for bovine spongiform encephalopathy in Japan

A predictive case-cohort model was applied to Japanese data to analyze the interaction between challenge and stability factors for bovine spongiform encephalopathy (BSE) for the period 1985-2020. BSE risk in cattle was estimated as the expected number of detectable cases per year. The model was comprised of a stochastic spreadsheet calculation model with the following inputs: (1) the origin and quantity of live cattle and meat and bone meal imported into Japan, (2) the age distribution of native cattle, and (3) the estimated annual basic reproduction ratio (R(0) ) for BSE. The estimated probability of having zero detectable cases in Japan in 2015 was 0.90 (95% CI 0.83-0.95). The corresponding value for 2020 was 0.99 (95% CI 0.98-0.99). The model predicted that detectable cases may occur in Japan beyond 2015 because of the assumption that continued transmission was permitted to occur (albeit at a very low level) after the 2001 ban on the importation and domestic use of all processed animal proteins for the production of animal feed and for fertilizer. These results reinforce the need for animal health authorities to monitor the efficacy of control measures so that the future course of the BSE epidemic in Japan can be predicted with greater certainty.

A comparison of different models for assessing variations in the sero-prevalence of infectious bovine rhinotracheitis by farm, area and district in Kenya

The relative variability of the sero-prevalence of antibodies to infectious bovine rhinotracheitis (IBR) due to cow, farm, and agroecological area levels were investigated for three contrasting districts in Kenya: Samburu, an arid and pastoral area; Kiambu, a tropical highland area; and Kilifi, a typical tropical coastal area. Cattle were selected by two-stage cluster sampling and visited once between August 1991 and 1992. Data on animal, farm, and area factors were analyzed using Schall's algorithm and MLn (multi-level, n-level), two generalized mixed-model programs suitable for multi-level analysis. Most variation in IBR sero-prevalence was from farm-to-farm. This was reflected by the many farm-level fixed effects (farm size, disease control measures and type of breeding) significant in models both ignoring and accounting for single variance components (clustering) at farm, area, and district levels. Area-to-area and district-to-district variations were noted but the area and district variance components were one-third and one-fifth the size of the farm variance components for both methods. As farm-to-farm variation differed markedly by farm size and district, models in MLn were extended to allow for multiple farm-level variance components by these categories. For each, sero-prevalence of IBR increased with age and was significantly decreased on small-sized zero-grazing farms. These models, particularly the model with different farm variance components by districts, fit the data better and highlighted well that there was considerable farm-to-farm variation--differing by district--and that the available farm-level fixed effects did not predict IBR sero-prevalence well.

Variations in the prevalence of antibody to brucella infection in cattle by farm, area and district in Kenya

Variations in the sero-prevalence of antibody to brucella infection by cow, farm and area factors were investigated for three contrasting districts in Kenya: Samburu, an arid and pastoral area: Kiambu, a tropical highland area; and Kilifi, a typical tropical coastal area. Cattle were selected by a two-stage cluster sampling procedure and visited once between August 1991 and 1992. Schall's algorithm, a statistical model suitable for multi-level analysis was used. Using this model, older age, free grazing and large herd size (> or = 31) were associated with higher seroprevalence. Also, significant farm-to-farm, area-to-area and district-to-district variations were estimated. The patterns of high risk districts and areas seen were consistent with known animal husbandry and movement risk factors, but the larger than expected farm-to-farm variation within high risk areas and districts could not be explained. Thus, a multi-level method provided additional information beyond conventional analyses of sero-prevalence data.

Assessing infections at multiple levels of aggregation

The patterns of sero-prevalence of antibodies to four infectious diseases, representing a broad range of pathogens (bacteria: brucellosis; mycoplasma: contagious bovine pleuropneumonia; viruses: infectious bovine rhinotracheitis; protozoa: trypanosomosis) were investigated at three levels of organization (farm, area and district). Three contrasting districts in Kenya were compared: an arid and pastoral area (Samburu); a tropical highland area (Kiambu), and a tropical coastal area (Kilifi). Cattle in three districts were selected by two-stage cluster sampling between August 1991 and 1992. Schall's algorithm, a generalized linear mixed model suitable for multi-level analysis, was compared to ordinary logistic regression (OLR), which ignores clustering of responses; generalized estimating equations (GEE) or Jacknife, to account for clustering at the farm level; SAS VARCOMP, which provides normal-theory random-effects models. Schall's algorithm provided similar estimates to GEE (regression effects) and Jackknife (standard errors) for farm-level clustered data. Extending Schall's procedure for additional district and area-within-district random effects usually provided additional information. In general, models that included only a farm-level random effect consistently provided larger estimates of farms' variance components than did models with additional district and area random effects. The four type diseases exhibited various amounts of clustering. Brucellosis had moderate farm clustering plus some area and district clustering. Contagious bovine pleuropneumonia had only a small amount of clustering, mostly by area. Infectious bovine rhinotracheitis exhibited a large amount of clustering, primarily at the farm level. Trypanosomiasis antibody prevalence varied by district, area and farm. We believe that patterns of disease clustering identified by multi-level analysis can be used to better target high-risk units for disease control and guide research to understand disease transmission factors.