Communicating risk and uncertainty in relation to development and implementation of disease control policies

Scientific evidence is one of the key factors to be considered in the development of disease control policies. It is generated using investigations into cause-effect relationships, which usually produce results that are associated with a varying degree of uncertainty. Experience has shown that taking account of these uncertainties can become a formidable challenge for policy makers when devising the strategies and when communicating them to stakeholders. The situation has been further complicated by a reduction in public trust in scientific evidence. It is now recognised that this challenge cannot be managed by simply providing more information, but it is also necessary to consider the influence that variation in risk perception amongst stakeholders has on their response to and commitment towards the policies.

Estimation of the relative risk of developing clinical scrapie: the role of prion protein (PrP) genotype and selection bias

Prion protein (PrP) genotype data from statutory confirmed cases and from three non-case datasets have been used to calculate the odds ratio (or) for the development of clinical scrapie for an individual sheep of a given PrP genotype, compared with one possessing the "wild-type" ARQ/ARQ genotype. Logistic regression has been used to estimate the ors, and a multiple-test procedure has been used to evaluate the statistical significance of each comparison. The results are similar to those observed in other studies: the VRQ/VRQ genotype has or point estimates greater than 20; the ARQ/VRQ and ARH/VRQ genotypes have or point estimates between 5 and 20; AHQ/VRQ between 0.03 and 0.1; ARR/VRQ 0.4 and 0.5; all the other PrP genotypes, excluding ARR/ARR, ARR/ARH and AHQ/ARH for which no clinical cases have been recorded have or point estimates of less than 0.3. The estimates derived from each dataset are comparable, but not identical. This can be explained by plausible biases inherent in the sampling of the non-case populations.

Risk factors for respiratory diseases in New Zealand pig herds

AIMS

A survey of lung lesions and risk factors for respiratory diseases was conducted in order to estimate the prevalence of respiratory diseases in the New Zealand pig population and to identify influential management practices.

METHODS

Eighty-nine New Zealand pig farms with a minimum herd size of 50 sows participated in the survey, and risk factor data were collected using a mailed questionnaire. Abattoir data were recorded once in winter 1995 and once in summer 1996. A total of 6887 lungs was inspected.

RESULTS

The prevalence of enzootic pneumonia, pleuropneumonia and pleurisy in winter was 63.4%, 2.7% and 19.1%, respectively. Enzootic pneumonia was significantly less frequent in summer. Pleuropneumonia/pleurisy was found to be more prevalent in the South Island. The univariate risk factor analysis was consistent with earlier published evidence on the importance of environmental factors related to housing and management of the farm. The multivariate models for enzootic pneumonia and pleuropneumonia or pleurisy had a reasonably good predictive power of 81-91% for farms with high disease prevalence.

CONCLUSION

The results are useful to model the disease process on high-risk farms, which account for a considerable proportion of the New Zealand pig population.

Bayesian spatio-temporal modelling of national milk-recording data of seasonal-calving New Zealand dairy herds

A spatio-temporal analysis was undertaken with the aim of identifying the dynamics of herd mean individual cow SCCs (MICSCC) in seasonally calving New Zealand dairy herds. Two datasets were extracted from the Livestock Improvement Corporation's extensive national dairy recording database: (1) milk-recording data aggregated at the herd-level and (2) sales questionnaire data containing information on the size, location and infrastructure of each farm. A Bayesian spatio-temporal modelling approach was applied to the analysis. The data were aggregated by 10 km(2) grid cells and linear regression models were developed with spatially structured and unstructured random effects, a linear temporal trend random effect and spatial-temporal interactions for log-transformed median MISCC (ln(median MISCC)). Significant associations were found between ln(median MISCC) and milk yield, milk fat, milk protein, farm area and number of cups in the dairy. This led us to suggest that SCCs should be adjusted for volume and constituents prior to determining a threshold MISCC for identification of subclinical mastitis (SCM) problem herds. Part, or all, of the temporal trend in MISCC in the spatio-temporal model was accounted for by inclusion of yield and milk constituents as independent variables. This supports the hypothesis of a dilution effect with potential consequences for misdiagnosis of SCM, particularly in late lactation. Unmeasured covariates were similarly likely to be spatially structured and unstructured.

Influence of weather conditions on fly abundance and its implications for transmission of rabbit haemorrhagic disease virus in the North Island of New Zealand

Blowflies (Diptera: Calliphoridae) and flesh flies (Diptera: Sarcophagidae) are potential vectors of rabbit haemorrhagic disease virus (RHDV) in New Zealand. The associations between habitat and weather factors on the abundance of these flies were investigated. Between October 1999 and June 2001, flies were trapped on open pasture and in dense vegetation patches on farmland in the Himatangi area of the North Island. Five calliphorid species were trapped commonly at scrub edges and the most abundant sarcophagid, Oxysarcodexia varia Walker, was trapped mainly on open pasture. An abundance peak of O. varia was probably associated with the occurrence of a rabbit haemorrhagic disease (RHD) outbreak in the study area. Overall abundance of flies varied according to habitat and species, and species numbers differed between seasons and years. The all-day minimum temperature 3 weeks before trapping was a significant variable in all models of fly abundance, whereas average rainfall did not affect fly abundance. The all-day temperature range was significant only for O. varia. The influence of other climatic factors varied between fly species. Climate dependent variations in fly abundance may contribute to the risk of transmission of RHD, which occurred intermittently on the site during the study period.

Linear and temporal stride characteristics of 3-day event horses at a CCI*** 3-day event horse inspection

AIMS

To quantify the linear and temporal characteristics of the trot of 3-day event horses during the preliminary horse inspection of an advanced grade CCI*** 3-day event.

METHODS

A cross sectional study was made of the kinematics of 24 3-day-event horses during the first horse inspection at a CCI*** level S-day event. Video footage was digitised and linear and temporal measurements were made.

RESULTS

The horses trotted for an average of 10.44+/-1.55 strides. Spatial measurements were made on an average of 5.66+/-0.92 consecutive strides when the horses were within the calibration zone. The horses increased and then obtained a constant velocity within the calibration zone. Trotting on the asphalt track did not alter the relationships between stride length, stride duration and velocity when compared with previously published values. Horse-specific differences in stance and retraction percentages were identified.

CONCLUSION

Horse specific differences were identified that may contribute to each horse's unique gait or kinematic fingerprint. It is proposed that the initiation of, and completion of, stance by the hind limb first may represent engagement of the hind quarters and be a response to dressage training.

Factors associated with the wastage and achievements in competition of event horses registered in the United Kingdom

The aims of this study were to estimate the wastage of horses registered for eventing in Britain, to investigate the reasons for this wastage and to evaluate factors affecting the horses' achievement of grade I status (at least 61 points) while registered. An analysis of the database of the British Eventing register found that 33.7 per cent of horses registered for the first time in 1999 were not re-registered for eventing in subsequent years. By using multivariable logistic regression analysis, it was shown that horses that were kept at an event yard were more likely to be re-registered than those kept on other premises (odds ratio [or] 2.0, 95 per cent confidence interval [ci] 1.2 to 3.2), and those that took part in showjumping while registered were also more likely to be re-registered (or 1.5, 95 per cent ci 1.1 to 2.2). Horses that took part in unaffiliated eventing while registered were less likely to be re-registered the following year (or 0.7, 95 per cent ci 0.5 to 0.9), as were those that were not insured (or 0.7, 95 per cent ci 0.5 to 1.0) and those from outside the British Isles (or 0.6, 95 per cent ci 0.3 to 1.0). Veterinary problems were the most commonly cited explanation (35.1 per cent) why horses that remained in their original ownership were not re-registered with British Eventing the following year. Horses from Australia were more likely to achieve grade I status than horses from the British Isles (or 9.7, 95 per cent ci 7.1 to 13.2), as were horses from New Zealand (or 6.4, 95 per cent ci 5.0 to 8.2), the usa (or 5.2, 95 per cent ci 3.8 to 7.2) and France (or 2.8, 95 per cent ci 2.1 to 3.7), but horses from the Netherlands (or 0.5, 95 per cent ci 0.3 to 0.9) and Belgium (or 0.3, 95 per cent ci 0.1 to 0.9) were less likely to achieve grade I status. Mares were less likely to achieve grade I status than geldings (or 0.4, 95 per cent ci 0.4 to 0.5).

Multivariate analysis of risk factors for feline hyperthyroidism in New Zealand

AIM

To search for putative risk factors for feline hyperthyroidism in New Zealand, using a case-control study.

METHODS

A questionnaire-based case-control study involving the owners of 375 cats in New Zealand (125 hyperthyroid cats, 125 randomly selected control cats, and 125 age- and sex-matched control cats) was conducted to examine associations between potential risk factors and occurrence of feline hyperthyroidism. Data were collected between December 1996 and February 1998, relative to cat and owner demography and medical history, cats' indoor and outdoor environments, and cats' diets. A range of statistical techniques was employed to analyse the data, including descriptive analyses, univariate logistic regression for each variable and multivariate stepwise forward logistic regression.

RESULTS

Multivariate analysis revealed that affected cats were more likely to be female (odds ratio (OR)=3.3; 95% confidence interval (CI)=1.2-9.0) and older than unaffected random control cats. Purebred cats were at a much lower risk of being diagnosed as hyperthyroid than were domestic short- and long-haired cats (OR=0.01; 95% CI=0.001-0.20). If more than one cat was present in a household, hyperthyroidism was less likely to be identified (OR=0.15; 95% CI=0.05-0.44) compared with single-cat households. Hyperthyroid cats were 6.6 times more likely (95% CI=1.8-23.9) to be reported to sleep predominantly on the floor than control cats. Cats whose bedding was regularly treated with anti-flea products appeared to be at a considerably higher risk for hyperthyroidism (OR=57.6; 95% CI=3.8-->200); and, to a lesser extent, so were cats living in households where fly sprays were reported to be used regularly (OR=3.3; 95% CI=1.2-9.3). The interaction between drinking water from puddles and regular use of organic garden fertilisers, such as compost or animal manure, was associated with a 5.3-fold (95% CI=1.1-25.6) increase in the risk of cats being diagnosed with the disease. Hyperthyroid cats were twice as likely (95% CI=0.3-12.9) to have eaten at least half of their daily food requirements as canned commercial cat food compared with unaffected cats. Cats exposed to a variety of flavours of canned cat food were more likely to be diagnosed with hyperthyroidism than were those fed only one flavour (OR=3.8; 95% CI=1.5-9.6). The presence of dental disorders was associated with a 5.5-fold increase in the risk of being diagnosed as hyperthyroid and this association was independent of the cat's age (95% CI=1.7-17.5).

CONCLUSIONS

The results of this study support and extend those in several earlier reports and show that cats in New Zealand are, in many respects, similar to cats in Europe and North America in terms of their susceptibility to hyperthyroidism. The finding that female cats are predisposed to hyperthyroidism is at variance with most previously published work. It remains unclear which, if any, of the identified disease associations are causal, so further studies of this increasingly prevalent feline endocrinopathy are warranted.

Clinical evaluation of postpartum vaginal mucus reflects uterine bacterial infection and the immune response in cattle

Bacteria contaminate the uterus of most dairy cattle after parturition and endometritis causes infertility. An endometritis score can be ascribed based on the vaginal mucus character and odour but it is not clear if the clinical score reflects the number of uterine bacteria or the inflammatory response. The present study tested the hypothesis that clinical evaluation of endometritis reflects the number of bacteria present in the uterus, and the acute phase protein response. Swabs (n = 328) were collected from the uterine lumen of dairy cattle, 21 and 28 days postpartum, vaginal mucus was scored for character and odour, and blood samples collected for acute phase protein measurement. Bacteria were identified following aerobic and anaerobic culture, and the bacterial growth density was scored semi-quantitatively. When bacteria were categorised by their expected pathogenic potential in the uterus, purulent or fetid odour vaginal mucus was associated with the growth density of pathogenic bacteria but not opportunist contaminants. When bacteria were analysed independently, Arcanobacterium pyogenes, Proteus and Fusobacterium necrophorum growth densities were associated with mucopurulent or purulent vaginal mucus. The bacterial growth densities for A. pyogenes, Escherichia coli, non-hemolytic Streptococci, and Mannheimia haemolytica were associated with a fetid mucus odour. Peripheral plasma concentrations of alpha(1)-acid glycoprotein were higher if there was a fetid compared with a normal vaginal mucus odour (1.50 +/- 0.09 mg/mL versus 1.05 +/- 0.02 mg/mL, P < 0.001), but did not differ significantly between vaginal mucus character scores. The evaluation of the character and odour of vaginal mucus reflects the number of bacteria in the uterus, and the acute phase protein response.

Serum insulin concentrations in horses with equine Cushing's syndrome: response to a cortisol inhibitor and prognostic value

REASONS FOR PERFORMING STUDY

Serum insulin concentration and its use as a prognostic indicator in horses with equine Cushing's syndrome (ECS) have been poorly documented.

OBJECTIVES

To examine daily insulin variations in horses with ECS and the effect of treatment using trilostane, a competitive inhibitor of 3beta-hydroxysteroid dehydrogenase. Further, we aimed to examine the relationship between baseline serum insulin concentration and survival in horses with ECS.

METHODS

Basal serum insulin concentrations were measured in 20 confirmed ECS cases by taking blood at regular 4 h intervals for 24 h (1200, 1600, 2000, 2400, 0400 and 0800 h) before treatment (Day 0) and 10 days, and 30 days and 1-2 years after the onset of trilostane therapy. The temporal pattern of insulin was analysed using a linear mixed model approach, and the prognostic value of measurements on Day 0 assessed using receiver-operating characteristic analysis.

RESULTS

Horses with ECS showed a diurnal pattern of serum insulin concentration, highest value at 1200 h, and this pattern was not altered by treatment with trilostane. Furthermore, despite a mild increase of serum insulin concentrations after 10 days of trilostane therapy, insulin concentration was unaffected in the long-term. Low serum insulin concentrations at the beginning of the trial were significantly associated with improved survival to 1-2 years. The 1200 h sampling before treatment had the highest prognostic value for prediction of survival with a sensitivity and specificity of at least 90% for serum insulin at < 62 and > 188 microu/ml to predict survival and nonsurvival, respectively.

CONCLUSIONS AND POTENTIAL RELEVANCE

Insulin is a useful prognostic indicator for ECS, but potentially large variations can occur throughout a 24 h period, indicating a single sample may not be representative. Serum insulin concentration did not increase over 1-2 years in horses receiving trilostane therapy.

Herd-level seroprevalence and risk-mapping of bovine hypodermosis in Belgian cattle herds

Our objective was to determine the seroprevalence of Hypoderma spp. and to develop a spatial model describing the risk surface of warble-fly infection in Belgian cattle herds (adjusting for herd size, herd type, local temperature, rainfall, relative air humidity and land-cover). This survey was carried out in 390 selected herds of all types (dairy, mixed and beef) from December 1997 to March 1998, which were included in a national infectious bovine rhinotracheitis and paratuberculosis (Johne's-disease) survey. All animals >24 months old were blood sampled and an ELISA was used on pooled serum samples (10 animals per pool). The herd seroprevalence was 48.7% (95% confidence interval: 43.6-53.8); positive herds were mainly in the south of the country and along the North Sea coast. The logistic multiple-regression model of herd-level seropositivity indicated that mixed-type and beef-cattle herds have more than four-fold and two-fold increases in the odds of being Hypoderma-positive, respectively, compared with dairy herds.

Vaccination coverage and epidemiological parameters of the owned-dog population in Thungsong District, Thailand

Canine rabies vaccination is delivered in Thungsong District, Thailand, as an annual campaign between March 1 and 31, and also at other times through private veterinary clinics, para-veterinarians and health-care staff residing in the villages. The current questionnaire-interview survey was conducted between June 23 and July 18, 2002 to determine: rabies-vaccination coverage amongst the owned-dog population; basic dog-population information; and community awareness about rabies. The modified expand programme on immunization cluster-survey method was used to collect information about dog demography and management characteristics. Household knowledge about rabies and sources of rabies information were assessed. Vaccinated dogs were identified from vaccine certificates or owner reports confirmed by vaccinators. Seventy percent (95%CI 62-78) of 364 eligible owned dogs were vaccinated within the 6 months prior to data collection. Of these 255 vaccinated dogs, 44, 21, 13, 17 and 5% were vaccinated through the annual vaccination campaign, veterinary clinics, para-veterinarians, other vaccinators and owners, respectively. Fifty-four percent of households owned dogs. The sex ratio in dogs was 2 males per female; the dog: human ratio was 1: 4.6 with an average of 0.9 dogs per household (1.7 dogs per dog-owning household). Most dogs roamed freely and these were less likely to be vaccinated compared to dogs being kept on premises or on a leash. Almost all households were aware of rabies and the need for dog rabies vaccination as a control method. Seventy-six percent believed that rabies only occurred in summer. There was little awareness about cat rabies amongst households. Vaccination coverage in the total dog population clearly has not yet reached the 80% target level set by Thailand's official rabies-control programme. Improved effectiveness of the owned-dog rabies-vaccination campaigns in each community is needed-perhaps by more community education about dog management or by better management of ownerless dogs.

Behavioural characteristics of dairy cows with lameness in Taranaki, New Zealand

AIM

To investigate possible relationships between the incidence of lameness in dairy cows, and walking order from the paddock into the milking shed, milking order, and dominance structure in the herd.

METHODS

Ten dairy herds from the Taranaki region in New Zealand, with an average herd size of 183 cows, were observed during early, mid and late lactation. Each observation period consisted of five observations within 1 week, totalling 15 observations per farm. Three observers recorded the daily movement of cows between paddocks and the milking shed at each of three different locations. Additionally, the dominance structure of 21 cows per herd, randomly selected at the beginning of the study, was determined at the beginning of each of the three observation periods, on the basis of antagonistic behaviour.

RESULTS

Within herds, walking and milking order showed a Kappa value of 0.31, reflecting fair agreement beyond chance. Based on the multivariate model, an increased risk of lameness was found for cows that walked or were milked in the last quarter of the herd (OR=1.8 and 1.5, respectively) compared with animals that walked/were milked in the first three quarters of the herd. The dominance structure of the randomly selected animals remained stable during the course of the lactation. High dominance animals tended to walk at the front of the herd and get milked first, whereas low dominance animals tended to walk at the back of the herd and get milked last. Although lameness was not significantly related to the dominance group, this could have been due to the low power of the study.

CONCLUSIONS

The results of the study suggested that there is a relationship between the position of the animal in the herd and the risk of lameness in dairy cows, but that the strength of this relationship varied between herds.

CLINICAL RELEVANCE

Walking/milking order was a risk factor for conditions causing lameness in dairy cows, and for the effective management of herd lameness problems it is important for farmers to be aware of this relationship.

Dietary trials with a commercial chicken hydrolysate diet in 63 pruritic dogs

The owners of 63 pruritic dogs were instructed to feed them a chicken hydrolysate diet exclusively for six weeks as part of diagnostic investigations into non-seasonal pruritus. Ectoparasitism and microbial infections were eliminated during the dietary trial. The dogs' skin lesions, gastrointestinal signs and frequency of defecation were assessed and scores for pruritus were assigned before they started the diet and before and after they resumed their original diet. An adverse food reaction was diagnosed if the pruritus resolved while they were on the diet, but recurred when they resumed their original food regimen. Seventeen of the 63 dogs were withdrawn from the trial, including four which found the diet unpalatable; however, its palatability was reported to be good or excellent in 48 of the dogs. An adverse food reaction alone was diagnosed in nine (19.6 per cent) of the 46 dogs and another nine had an adverse food reaction and atopy. There were gastrointestinal signs in six of the nine dogs with an adverse food reaction, which resolved on the trial diet but recurred when they resumed their original diet.

Simulated hazards of loosing infection-free status in a Dutch BHV1 model

A compulsory eradication programme for bovine herpesvirus 1 (BHV1) was implemented in the Netherlands in 1998. At the start of the programme, about 25% of the dairy herds were certified BHV1-free. Simulation models have played an important role in the decision-making process associated with BHV1 eradication. Our objective in this study was to improve understanding of model behaviour (as part of internal validation) regarding loss by herds of the BHV1-free certificate. Using a Cox proportional hazards model, the association between farm characteristics and the risk of certificate loss during simulation was quantified. The overall fraction of herds experiencing certificate loss amongst initially certified during simulation was 3.0% in 6.5 years. Factors that increased risk for earlier certificate loss in the final multivariable Cox model were higher 'yearly number of cattle purchased', 'farm density within a 1 km radius' and 'cattle density within a 1 km radius'. Qualitative behaviour of risk factors we found agreed with observations in field studies.

Evaluation of the basal and post-adrenocorticotrophic hormone serum concentrations of 17-hydroxyprogesterone for the diagnosis of hyperadrenocorticism in dogs

Serum concentrations of 17-hydroxyprogesterone and cortisol were measured before and after the administration of exogenous adrenocorticotrophic hormone (ACTH) to three groups of dogs: 27 healthy dogs (group 1), 19 dogs with non-adrenal illness (group 2) and 46 dogs with hyperadrenocorticism (group 3). The median (range) post-ACTH concentrations of 17-hydroxyprogesterone were 5.0 (22.2 to 16.8), 6.9 (2.0 to 36.2) and 14.4 (1.7 to 71) nmol/litre in groups 1, 2 and 3, respectively. There were no significant differences in the basal or post-ACTH concentrations of cortisol or 17-hydroxyprogesterone between groups 1 and 2. The post-ACTH concentrations of 17-hydroxyprogesterone in group 3 were significantly (P<0.001) greater than those in groups 1 and 2 combined. The area under the receiver operating curve (ROC) for the post-ACTH concentration of cortisol (0.94) was significantly greater than that for the post-ACTH concentration of 17-hydroxyprogesterone (0.76). Using a two-graph ROC analysis, a cut-off of 8.5 nmol/litre was found to maximise both the sensitivity and specificity of the post-ACTH concentration of 17-hydroxyprogesterone for the diagnosis of hyperadrenocorticism at 71 per cent. With a cut-off of 4.5 nmol/litre the sensitivity increased to 90 per cent but the specificity decreased to 40 per cent; with a cut-off of 16.7 nmol/litre the specificity increased to 90 per cent but the sensitivity decreased to 47 per cent.

Neurological signs and results of magnetic resonance imaging in 40 cavalier King Charles spaniels with Chiari type 1-like malformations

In human beings a Chiari type 1 malformation is a developmental condition characterised by cerebellar herniation and syringohydromyelia. Abnormalities compatible with such a malformation were identified by magnetic resonance imaging in 39 cavalier King Charles spaniels with neurological signs and in one neurologically normal cavalier King Charles spaniel that was examined postmortem. The dogs with these abnormalities had a wide variety of neurological signs, but there was no apparent correlation between the neurological signs and the severity of cerebellar herniation, syringohydromyelia or hydrocephalus.

Social-network analysis of Mycobacterium bovis transmission among captive brushtail possums (Trichosurus vulpecula)

Wild brushtail possums (Trichosurus vulpecula) are the main source of Mycobacterium bovis infection for New Zealand livestock. The disease is spread principally by infectious aerosol; therefore, social interactions determine disease transmission. In captive possums, den-sharing behaviour provided the greatest risk of tuberculosis transmission between animals. Den sharing between individual possums was used as the basic measurement for quantifying close proximity between animals over extended periods. Social-network analysis (SNA) was used to model patterns of social behaviour and to predict tuberculosis transmission. There was great diversity between groups in their social behaviour-but there were consistent trends in the SNA measures (closeness and flow-betweenness). With time, the social distance between possums in the same group increased, the social network became more homogeneous and the possums less differentiated from each other. Alteration of the physical environment of the pens (such as changing the number of dens or relocating the group to a new pen) had an inconsistent effect on social structure when comparing different groups. During the infection-transmission study, the possums that became infected had greater closeness and flow-betweenness scores than those that remained free of infection. Although standard statistical descriptive measures (such as the number of partners and the frequency of den-sharing events) were greater for the infected than the infection-free possums, the SNA-specific measures were more precise and could be compared across time and between groups.

A global livestock production and health atlas (GLiPHA) for interactive presentation, integration and analysis of livestock data

An interactive electronic atlas has been developed with the purpose of providing a scaleable overview of spatial and temporal variation in animal production and health-related information for decision and policy makers in national and international institutions. The information contained in the atlas is currently managed and presented using the Key Indicators Mapping System (KIMS), and will also be integrated using the Key Indicators Database System (KIDS). Both systems were developed by the World Agricultural Information Centre of the FAO (FAO-WAICENT), the former as a stand-alone application and the latter for access via the Internet. Components of the atlas include vector maps, livestock disease and production databases, rules for country-level disease risk classification and 'disease cards' containing basic background information on diseases included in the atlas. The disease data is currently based primarily on Office International des Epizooties (OIE) disease reports, and the livestock production data on the FAO-WAICENT database. The atlas is highly interactive and allows visual presentation of information using maps, tables and charts. It also contains links to relevant resource information on the Internet. Diseases covered in the animal health layer include most OIE List A diseases and a subset of OIE List B diseases. Extensive analyses have been conducted to develop a set of qualitative and semi-quantitative criteria that allow improved disease status classifications based on 5-years cumulative OIE disease reports, and official disease control declarations. Classification rules were determined depending on the epidemiological features of each disease and considering spatial heterogeneity of disease presence in local regions.

An analysis of the performance characteristics of serological tests for the diagnosis of Neospora caninum infection in cattle

AIMS

To analyse the performance characteristics (sensitivity/specificity) of two enzyme-linked immunosorbent assays (ELISA) against the indirect fluorescent antibody test (IFAT).

METHODS

A total of 1199 sera were tested in two ELISAs and the IFAT and results analysed utilising software that performed a receiver-operating characteristic (ROC) analysis.

RESULTS

Sensitivity and specificity for the two ELISAs were calculated for a range of different cut-offs. Minimal misclassification was achieved at cut-offs that, in the case of the Central Animal Health Laboratory (CAHL)-ELISA were in line with previously published cut-off values. In the case of the IDEXX-ELISA lower cut-off values than suggested by the manufacturer were calculated.

CONCLUSIONS

ROC-analysis resulted in optimised, as compared to the IFAT, cut-off thresholds for the CAHL and IDEXX-ELISA which can be further adjusted depending on the purpose of the investigation.

Natural transmission ofMycobacterium bovisinfection in captive brushtail possums (Trichosurus vulpecula)

AIMS

To examine natural transmission of bovine tuberculosis (Mycobacterium bovis infection) in captive brushtail possums (Trichosurus vulpecula) and to determine if this mode of transmission could be employed to challenge possums in vaccination studies.

METHODS

Three experiments were conducted. In Experiment 1, 11 pairs of possums were housed together in cages, one of the pair having been experimentally infected with M. bovis. Of the in-contact possums 5/11 had been vaccinated with bacille Calmette-Guerin (BCG). In Experiment 2, three susceptible possums were placed in a colony of 19 possums that had been experimentally infected with M. bovis. In Experiment 3, the four most socially active possums in each of two colonies (24 possums in one colony and 23 in the other) were experimentally infected with M. bovis, and 10 of the remaining possums in each colony were vaccinated with BCG.

RESULTS

In Experiment 1, transmission of M. bovis infection occurred in only 1/11 pairs. In Experiment 2, none of the three in-contact possums became infected. In Experiment 3, infection was transmitted to 5/20 in-contact possums in one colony and 12/19 in-contact possums in the other. The possums that became infected by natural transmission were significantly more socially interactive than those that remained free of infection (p<0.05).

CONCLUSIONS

When susceptible and infected possums were randomly mixed, the rate of transmission of M. bovis was low, but when highly sociable possums were infected the rate of transmission increased markedly. The risk of transmission was dependent on the close proximity of infected and susceptible possums and the frequency and duration of their social interactions. Natural transmission from experimentally infected to incontact possums in a colony would be a useful way of studying the pathogenesis of tuberculosis in this species, and the social behaviour of the possums studied should be taken into account. The high degree of variation in the rate of natural transmission of M. bovis infection between possums makes this mode of transmission unreliable for assessing vaccine efficacy.

Influence of uterine bacterial contamination after parturition on ovarian dominant follicle selection and follicle growth and function in cattle

First postpartum dominant follicles are preferentially selected in the ovary contralateral to the previously gravid uterine horn. The aim of the present study was to test the hypothesis that uterine bacterial contamination alters the location of ovarian follicle emergence and selection, and inhibits follicle growth and function. Swabs were collected from the uterine body lumen of cattle on days 7, 14, 21 and 28 after parturition. Bacteria were identified by aerobic and anaerobic culture; bacterial growth was scored semiquantitatively and animals were categorized into standard or high bacterial contamination categories on the basis of the number of colonies detected. Follicular growth and function were monitored by daily transrectal ultrasonography, and estimation of plasma FSH, oestradiol and progesterone concentrations. There was no effect of bacterial contamination on plasma FSH concentration profiles or emergence of the ovarian follicle wave. When uterine bacterial growth scores were high on day 7 or day 21 after parturition, fewer first (1/20 versus 15/50; P < 0.05) or second (1/11 versus 13/32; P < 0.05) dominant follicles were selected in the ipsilateral compared with the contralateral ovary, respectively. The diameter of the first dominant follicle was smaller in animals with a high day 7 bacterial score (P < 0.001), dominant follicle growth was slower (P < 0.05) and oestradiol secretion was decreased (P < 0.05). The present study provides evidence for an effect of the uterus on the ovary after parturition, whereby uterine bacteria have a contemporaneous localized effect on ovarian follicle selection and subsequent growth and function, but not on initial emergence.

Influence of uterine bacterial contamination after parturition on ovarian dominant follicle selection and follicle growth and function in cattle

First postpartum dominant follicles are preferentially selected in the ovary contralateral to the previously gravid uterine horn. The aim of the present study was to test the hypothesis that uterine bacterial contamination alters the location of ovarian follicle emergence and selection, and inhibits follicle growth and function. Swabs were collected from the uterine body lumen of cattle on days 7, 14, 21 and 28 after parturition. Bacteria were identified by aerobic and anaerobic culture; bacterial growth was scored semiquantitatively and animals were categorized into standard or high bacterial contamination categories on the basis of the number of colonies detected. Follicular growth and function were monitored by daily transrectal ultrasonography, and estimation of plasma FSH, oestradiol and progesterone concentrations. There was no effect of bacterial contamination on plasma FSH concentration profiles or emergence of the ovarian follicle wave. When uterine bacterial growth scores were high on day 7 or day 21 after parturition, fewer first (1/20 versus 15/50; P < 0.05) or second (1/11 versus 13/32; P < 0.05) dominant follicles were selected in the ipsilateral compared with the contralateral ovary, respectively. The diameter of the first dominant follicle was smaller in animals with a high day 7 bacterial score (P < 0.001), dominant follicle growth was slower (P < 0.05) and oestradiol secretion was decreased (P < 0.05). The present study provides evidence for an effect of the uterus on the ovary after parturition, whereby uterine bacteria have a contemporaneous localized effect on ovarian follicle selection and subsequent growth and function, but not on initial emergence.