The contact structure of Great Britain's salmon and trout aquaculture industry.

We analyse the network structure of the British salmonid aquaculture industry from the perspective of infectious disease control. We combine for the first time live fish transport (or movement) data covering England and Wales with data covering Scotland and include network layers representing potential transmission by rivers, sea water and local transmission via human or animal vectors in the immediate vicinity of each farm or fishery site. We find that 7.2% of all live fish transports cross the England-Scotland border and network analysis shows that 87% of English and Welsh nodes and 72% of Scottish nodes are reachable from cross-border connections via live fish transports alone. Consequently, from a disease-control perspective, the contact structures of England and Wales and of Scotland should not be considered in isolation. We also show that large epidemics require the live fish movement network and so control strategies targeting movements can be very effective. While there is relatively low risk of widespread epidemics on the live fish transport network alone, the potential risk is substantially amplified by the combined interaction of multiple network layers.

The evolution and phylodynamics of serotype A and SAT2 foot-and-mouth disease viruses in endemic regions of Africa.

Foot-and-mouth disease (FMD) is a major livestock disease with direct clinical impacts as well as indirect trade implications. Control through vaccination and stamping-out has successfully reduced or eradicated the disease from Europe and large parts of South America. However, sub-Saharan Africa remains endemically affected with 5/7 serotypes currently known to be circulating across the continent. This has significant implications both locally for livestock production and poverty reduction but also globally as it represents a major reservoir of viruses, which could spark new epidemics in disease free countries or vaccination zones. This paper describes the phylodynamics of serotypes A and SAT2 in Africa including recent isolates from Cameroon in Central Africa. We estimated the most recent common ancestor for serotype A was an East African virus from the 1930s (median 1937; HPD 1922-1950) compared to SAT2 which has a much older common ancestor from the early 1700s (median 1709; HPD 1502-1814). Detailed analysis of the different clades shows clearly that different clades are evolving and diffusing across the landscape at different rates with both serotypes having a particularly recent clade that is evolving and spreading more rapidly than other clades within their serotype. However, the lack of detailed sequence data available for Africa seriously limits our understanding of FMD epidemiology across the continent. A comprehensive view of the evolutionary history and dynamics of FMD viruses is essential to understand many basic epidemiological aspects of FMD in Africa such as the scale of persistence and the role of wildlife and thus the opportunities and scale at which vaccination and other controls could be applied. Finally we ask endemic countries to join the OIE/FAO supported regional networks and take advantage of new cheap technologies being rolled out to collect isolates and submit them to the World Reference Laboratory.

Assessing the performance of a Fasciola gigantica serum antibody ELISA to estimate prevalence in cattle in Cameroon.

BACKGROUND: Cattle rearing in Cameroon is both economically and culturally important, however parasitic diseases detrimentally impact cattle productivity. In sub-Saharan Africa bovine fasciolosis is generally attributed to F. gigantica, although understanding of Fasciola species present and local epidemiology in individual countries is patchy. Partly limited by the lack of representative surveys and understanding of diagnostic test perfromance in local cattle populations. The aims of this paper were to determine the Fasciola species infecting cattle, develop a species specific serum antibody ELISA, assess the performance of the ELISA and use it to assess the prevalence of F. gigantica exposure in two important cattle-rearing areas of Cameroon. RESULTS: A random sample of Fasciola parasites were collected and were all identified as F. gigantica (100%, CI:94.0-100%, n = 60) using RAPD-PCR analysis. A F. gigantica antibody ELISA was developed and initially a diagnostic cut-off was determined using a sample of known positive and negative cattle. The initial cut-off was used as starting point to estimate an optimal cut-off to estimate the best combination of sensitivity and specificity. This was achieved through sampling a naturally infected population with known infection status (cattle slaughtered at Bamenda abattoir, North West Region (n = 1112) and Ngaoundere abattoir, Vina Division, Adamawa Region (n = 776) in Cameroon). These cattle were tested and results analysed using a Bayesian non-gold standard method. The optimal cut-off was 23.5, which gave a sensitivity of 65.3% and a specificity of 65.2%. The prevalence of exposure to F. gigantica was higher in cattle in Ngaoundere (56.4% CI: 50.2-60.0%) than Bamenda (0.6% CI: 0.0-1.4%). CONCLUSION: Fasciola gigantica was identified as the predominant Fasciola species in Cameroon. Although the sensitivity and specificity F. gigantica antibody ELISA requires improvement, the test has shown to be a potentially useful tool in epidemiological studies. Highlighting the need for better understanding of the impact of F. gigantica infections on cattle production in Cameroon to improve cattle production in the pastoral systems of Central-West Africa. This paper also highlights that non-gold standard latent class methods are useful for assessing diagnostic test performance in naturally-infected animal populations in resource limited settings.

Molecular epidemiology of Mycobacterium bovis in Cameroon.

We describe the largest molecular epidemiological study of Bovine Tuberculosis (bTB) in a sub-Saharan African country with higher spatial resolution providing new insights into bTB. Four hundred and ninety-nine samples were collected for culture from 201 and 179 cattle with and without bTB-like lesions respectively out of 2,346 cattle slaughtered at Bamenda, Ngaoundere, Garoua and Maroua abattoirs between 2012-2013. Two hundred and fifty-five M. bovis were isolated, identified and genotyped using deletion analysis, Hain® Genotype MTBC, spoligotyping and MIRU-VNTR. African 1 was the dominant M. bovis clonal complex, with 97 unique genotypes including 19 novel spoligotypes representing the highest M. bovis genetic diversity observed in Africa to date. SB0944 and SB0953 dominated (63%) the observed spoligotypes. A third of animals with multiple lesions had multiple strain infections. Higher diversity but little evidence of recent transmission of M. bovis was more common in Adamawa compared to the North-West Region. The Adamawa was characterised by a high frequency of singletons possibly due to constant additions from an active livestock movement network compared to the North-West Region where a local expansion was more evident. The latter combined with population-based inferences suggest an unstable and stable bTB-endemic status in the North-West and Adamawa Regions respectively.

Sputum bacterial load predicts multidrug-resistant tuberculosis in retreatment patients: a case-control study.

BACKGROUND: Rapid and effective diagnosis of multidrug-resistant tuberculosis (MDR-TB) is an essential component of global tuberculosis (TB) control, but most MDR-TB cases are still not diagnosed. OBJECTIVE: To assess whether patient sputum bacterial load can be used to identify patients at increased risk of MDR-TB. METHODS: We used a case-control study and multivariable logistic regression models to investigate associations between MDR-TB and sputum bacterial load, as measured by semi-quantitative microscopy and automated time to detection (TTD) of liquid culture. We assessed data from retreatment TB patients with MDR-TB (cases) and from those without MDR-TB (controls) at a reference laboratory in Cameroon. RESULTS: MDR-TB was associated with a smear microscopy grade of 3+ (OR 21.9, 95%CI 6.2-76.8) or 2+ (OR 10.8, 95%CI 2.9-40.7), compared to a result of 1+, scanty or smear-negative among 80 MDR-TB cases and 521 controls. MDR-TB was associated with automated TTD of â©¿160 h (OR 2.2, 95%CI 1.1-4.7) compared to >160 h among a subpopulation of 47 cases and 350 controls. CONCLUSIONS: A higher sputum bacterial load is associated with MDR-TB in retreatment patients in Cameroon.

Abattoir-based estimates of mycobacterial infections in Cameroon.

Mycobacteria cause major diseases including human tuberculosis, bovine tuberculosis and Johne's disease. In livestock, the dominant species is M. bovis causing bovine tuberculosis (bTB), a disease of global zoonotic importance. In this study, we estimated the prevalence of Mycobacteria in slaughter cattle in Cameroon. A total of 2,346 cattle were examined in a cross-sectional study at four abattoirs in Cameroon. Up to three lesions per animal were collected for further study and a retropharyngeal lymph node was collected from a random sample of non-lesioned animals. Samples were cultured on Lowenstein Jensen media and the BACTEC MGIT 960 system, and identified using the Hain® Genotype kits. A total of 207/2,346 cattle were identified with bTB-like lesions, representing 4.0% (45/1,129), 11.3% (106/935), 23.8% (38/160) and 14.8% (18/122) of the cattle in the Bamenda, Ngaoundere, Garoua and Maroua abattoirs respectively. The minimum estimated prevalence of M. bovis was 2.8% (1.9-3.9), 7.7% (6.1-9.6), 21.3% (15.2-28.4) and 13.1% (7.7-20.4) in the four abattoirs respectively. One M. tuberculosis and three M. bovis strains were recovered from non-lesioned animals. The high prevalence of M. bovis is of public health concern and limits the potential control options in this setting without a viable vaccine as an alternative.

Falls and injuries to Polo players: risk perception, mitigation and risk factors.

Polo, one of the world's oldest sports, is played in over 80 countries. It is unique in combining the skills of a person with the agility and performance of an animal in a contact sport. There is only one report of the frequency and type of injuries in this population. Here we report risk perception, mitigation and risk factors for injuries and falls in UK polo players. Data were collected retrospectively from a random sample of 112 UK polo players by telephone questionnaire. Injuries (commonly to a shoulder or wrist) requiring a hospital visit were sustained by 17.3% (95% confidence interval [CI] 9.1-25.5) of players. Falls (odds ratio [OR] 6.6, 95% CI 1.4-31.9) and higher self-assessed fitness levels increased the risk (OR 1.7, CI 1.2-2.4). Use of wrist supports (OR 0.2, CI 0.03-0.9) and gym exercise (OR 0.1, CI 0.02-0.9) reduced it. Falls were reported by 58% (CI 47.3-68.8) of players. Women were less at risk than men (OR 0.3, CI 0.1-0.9). Aiming for a better handicap increased the risk (OR 8.4, CI 1.2-57.0). Pre-season rider and horse training were also risk factors. Helmets are compulsory, but players reported that safety certification was not their most important criterion for helmet selection; 49.4% (CI 38.5-60.3) chose appearance. Attendance of a doctor at polo games was not considered important by 65.4% (CI 55.1-75.8) of players; attendance of paramedics and ambulances was volunteered as being of greater consequence. The findings of this study suggest that the protective effect of wrist supports needs testing, helmet manufacturers should incorporate both style and safety into their designs, and paramedics and ambulances should attend polo games.

Molecular epidemiology of rotavirus in cats in the United Kingdom.

Rotaviruses are leading causes of gastroenteritis in the young of many species. Molecular epidemiological studies in children suggest that interspecies transmission contributes to rotavirus strain diversity in people. However, population-based studies of rotaviruses in animals are few. We investigated the prevalence, risk factors for infection, and genetic diversity of rotavirus A in a cross-sectional survey of cats housed within 25 rescue catteries across the United Kingdom. Morning litter tray fecal samples were collected during the winter and summer in 2012 from all pens containing kittens and a random sample of those housing adult cats. Group A rotavirus RNA was detected by real-time reverse transcription-PCR, and positive samples were G and P genotyped using nested VP4 and VP7 PCR assays. A total of 1,727 fecal samples were collected from 1,105 pens. Overall, the prevalence of rotavirus was 3.0% (95% confidence interval [CI], 1.2 to 4.9%). Thirteen out of 25 (52%; 95% CI, 31.3 to 72.2%) centers housed at least one rotavirus-positive cat. The prevalence of rotavirus was associated with season (odds ratio, 14.8 [95% CI, 1.1 to 200.4]; P = 0.04) but not age or diarrhea. It was higher during the summer (4.7%; 95% CI, 1.2 to 8.3%) than in winter (0.8%; 95% CI, 0.2 to 1.5%). Asymptomatic epidemics of infection were detected in two centers. G genotypes were characterized for 19 (33.3%) of the 57 rotavirus-positive samples and P genotypes for 36 (59.7%). Two rotavirus genotypes were identified, G3P[9] and G6P[9]. This is the first population-based study of rotavirus in cats and the first report of feline G6P[9], which questions the previous belief that G6P[9] in people is of bovine origin.

Polo pony injuries: player-owner reported risk, perception, mitigation and risk factors.

REASONS FOR PERFORMING STUDY: Polo, one of the world's oldest sports, is unique in merging human skill and balance with animal agility and performance in a contact sport. These modern-day 'centaurs' offer medical, dental and veterinary scientists an unrivalled, if quirky, opportunity to collaborate. Collection of epidemiological data on injuries to UK polo riders and ponies is the first step. OBJECTIVES: To measure the reported risk and risk factors for injuries to UK polo ponies, their perception and mitigation by player-owners. STUDY DESIGN: A retrospective cohort design and telephone interviews were used. METHODS: Data on equine injuries, preseason training and risk perception were collected from a random sample of player-owners using a structured questionnaire. Injuries were defined as requiring veterinary treatment. Frequencies were represented as percentages and 95% confidence intervals (CIs). Risk factors for injuries were identified by univariable and multivariable analyses. RESULTS: The cumulative incidence of player-owner-reported injury was 10.6% (95% CI 8.4-12.7). Tendon injuries were most common (4.3%; 95% CI 2.9-5.7), followed by wounds and splints. The only risk factor was stabling all season (odds ratio 4.79; 95% CI 1.46-15.73). Tendon injuries were perceived as the major risk and hard ground the most important risk factor. Risk mitigation practices were bandaging before exercise (45.7%; 95% CI 34.8-56.5), checking tendons (84.0%; 95% CI 76.0-91.9), cold hosing (40.7%; 95% CI 30.0-51.4), bandaging (38.3%; 95% CI 27.7-48.9) and using clays and coolants after exercise (24.7%; 15.3-34.1). Cuts and wounds were considered most frequent by only 2.5% (95% CI 0.0-3.6) of players but were the second most common injury, accounting for 21.6% of veterinary treatments. Splints accounted for 12.5% of injuries. CONCLUSIONS: The risk of injury to polo ponies is similar to that in the general horse population; musculoskeletal injuries, particularly tendon injuries, are most common, followed by wounds and splints. The association between stabling and injury, risk factors for wounds and splints and efficacy of cooling tendons post exercise warrant further study.

Hock burn: an indicator of broiler flock health.

This study aimed to identify risk factors for the prevalence of hock burn, a common disease of broiler chickens that influences the welfare and profitability of affected flocks, using hierarchical logistic regression analysis of routine flock management data. The study identified an association between hock burn and other diseases detected at slaughter and found that the management of flocks around the slaughter period was of particular importance for the control of hock burn, providing a rational basis for intervention.

Epidemics and control strategies for diseases of farmed salmonids: a parameter study.

The susceptibility of the English and Welsh fish farming and fisheries industry to emergent diseases is assessed using a stochastic simulation model. The model dynamics operate on a network comprising directed transport and river contacts, as well as undirected local and fomite transmissions. The directed connections cause outward transmission risk to be geographically more confined than inward risk. We consider reactive, proactive, and hybrid methods of control which correspond to a mixture of policy and the ease of disease detection. An explicit investigation of the impact of laboratory capacity is made. General quantified guidelines are derived to mitigate future epidemics.

Risk factors for Thoroughbred racehorse fatality in jump starts in Victoria, Australia (1989-2004).

REASONS FOR PERFORMING STUDY: The risk of fatality is greater in jump than in flat racing in Victoria, Australia. This is the first study to identify risk factors specific to jump starts in Victoria. OBJECTIVE: To identify risk factors for fatality of Thoroughbred racehorses in jump starts on all racecourses in Victoria, Australia between 1989 and 2004. METHODS: Fatalities comprised all horses that died during or immediately after a jump (hurdle or steeplechase) race or official jump trial and all horses that were subjected to euthanasia within 24 h of an event in which an injury was sustained. The retrospective study involved 191 case starts and 2324 control starts. Univariable and multivariable backward stepwise logistic regression was used to identify risk factors for fatality at any one start. A multiple level model was used with racecourse included as a random effect. RESULTS: In the final multivariable model, the duration of the racing career of the horse, the number of flat, hurdle and steeple starts accumulated in the 60 days prior to the case or control start, the number of flat and jump starts accumulated over the racing career, if the horse had had a start between 1 and 14 days prior to the case or control start, the type of jump race (hurdle or steeple), the calendar year of the start and the location of the racecourse were associated with fatality. CONCLUSIONS: The findings highlight the need to investigate further the differences between hurdle and steeplechase events and the adverse effect of prolonged prior flat racing careers on the risk of fatality in jump starts. POTENTIAL RELEVANCE: This is the first study to examine risk factors for fatality in jump starts in Victoria. The results should shape the development of interventions to reduce the risk in jump starts in the future.

Risk factors for Thoroughbred racehorse fatality in flat starts in Victoria, Australia (1989-2004).

REASONS FOR PERFORMING STUDY: Research into risk factors specific for fatality in flat racing should be focused at a regional level as the risk factors may differ among countries and even regions within countries. OBJECTIVE: To identify risk factors for fatality of Thoroughbred racehorses in flat starts on all racecourses in Victoria, Australia between 1989 and 2004. METHODS: Fatalities comprised all horses that died during or immediately after a flat race or official flat trial, and all horses that were subjected to euthanasia within 24 h of an event in which an injury was sustained. The retrospective study involved 283 case starts and 3307 control starts. Univariable and multivariable backward stepwise logistic regression was used to identify risk factors for fatality at any one start. RESULTS: In the final multivariable model, horse gender, prior racing history, race length, racing year, racecourse location and track rating were associated with fatality. CONCLUSIONS: This study identified specific risk factors for fatality. The findings are consistent with the hypothesis that horses accumulating high speed exercise are predisposed to catastrophic injury. The study has also highlighted the need to investigate further the adverse effects of different track ratings on the incidence of injury and subsequent fatality. RELEVANCE: The results will facilitate the development of effective strategies to improve overall safety of horses and jockeys in flat racing in Victoria, Australia.

Prevalence and associated risk factors of necrotic enteritis on broiler farms in the United Kingdom; a cross-sectional survey.

In order to determine the prevalence and risk factors for necrotic enteritis in broilers, a cross-sectional survey was conducted among 857 farms, rearing broilers for nine UK poultry companies. The main data collection tool was a postal questionnaire directed at farm managers. Additional information on disease occurrence on the farm was collected from veterinary postmortem reports. The response rate to the questionnaire was 75%, ranging from 54% to 90% within companies. During 2001, 32.8% of the respondents indicated that they had observed a case of necrotic enteritis (95% confidence interval, 29.1 to 36.8) in at least one flock. The disease was most often reported during the months October to February. The point prevalence (necrotic enteritis occurrence in the most recently reared flock) reported by farm managers was 12.3% (95% confidence interval, 9.8 to 15.2). Multilevel logistic regression was performed with the poultry company as the random effect, using the occurrence of necrotic enteritis in the farm's most recently reared flock as the dependent variable. Strong associations were found between the outcome variable and the occurrence of wet litter (odds ratio, 2.39; 95% confidence interval, 1.27 to 4.52; P = 0.007) and coccidiosis (odds ratio, 4.68; 95% confidence interval, 1.74 to 12.55; P = 0.002). In addition, the use of ammonia as a disinfectant for coccidial oocysts appeared to be an independent risk factor (odds ratio, 3.44; 95% confidence interval, 1.53 to 7.71; P = 0.003). Finally, the positive association between the use of plasterboard walls in poultry houses and the occurrence of necrotic enteritis might point to an important role of cleaning and disinfection in the epidemiology of this disease (odds ratio, 3.72; 1.38 to 10.00; P = 0.009).

Descriptive analysis of retirement of Thoroughbred racehorses due to tendon injuries at the Hong Kong Jockey Club (1992-2004).

REASONS FOR PERFORMING STUDY: This study was part of a programme to optimise the longevity and maximise the health and welfare of the Thoroughbred racehorses in Hong Kong. Injuries to the superficial digital flexor (SDF) tendon are the most common veterinary reason for premature retirement in this population. OBJECTIVE: To describe the frequency and pattern of retirements associated with SDF tendon injuries in Thoroughbred racehorses and to compare the characteristics of these horses with those that retired for other reasons. METHODS: A retrospective analysis of retirement records documented in the Hong Kong Jockey Club clinical database between 1992 and 2004 was conducted. As this is complete census data, no statistical inference to the population is necessary. RESULTS: The mean annual cumulative incidence of retirements due to tendon injury was 3.2% accounting for 14% of all retirements. The risk of tendon injury increased over the 12 year period from 23-4.2%. The racing career, number of starts and earnings of horses retired with tendon injuries were reduced by 25.6, 41.2 and 53.3%, respectively. Thirteen percent of these horses never raced in Hong Kong. A greater proportion of 3- and 4-year-olds and entire males were retired because of tendon injuries. Ninety-seven percent of injuries affected the forelimb, the right more frequently than the left. Only 19.7% of retired horses that had received ultrasound examination for SDF tendon injury retired for this reason. CONCLUSIONS: This study provides population based data on the frequency, career and economic losses associated with tendon injury induced retirement. POTENTIAL RELEVANCE: This descriptive study has provided a useful resource for further case-control studies to investigate risk factors for retirement from racing due to tendon injury. This is the first step toward the development of management tools to reduce the incidence of tendon injury related retirement in Hong Kong.

An abattoir survey on bovine spongiform encephalopathy (BSE) in Turkey.

The aim of this study was to investigate Bovine Spongiform Encephalopathy (BSE) in Turkish cattle in the Marmara region which borders the European Union (EU). For this, cattle brought to abattoirs in Istanbul were analysed. The high risk group were selected and therefore 384 cattle above 2 years old were included in the study. They were primarily examined for the presence of any clinical signs of nervous system and also other clinical disorders. The whole brains were taken and analysed for the presence of vacuolar degeneration and prion protein by PLATELIA BSE test kit. Only 5 cattle were found to be nervous and showed aggressive behaviour. There were no cattle showing incoordination or other neurological disorders. Cysts were observed in 3 brains. Histopathologically, no vacuolar degeneration indicative of BSE was found in any cattle examined. However, in 8 brains, few vacuoles were observed in neurons in sections taken from the brain, cerebellum, medulla oblongata and medulla spinalis. Slight mononuclear cell infiltration in 9 brain, intensed mononuclear cell infiltration in 1 brain, haemorrhages in 5 brains and gliosis in 11 brains were also found. No infective prion was detected by ELISA in samples taken from 384 cattle brain.

Geographical and age-stratified distributions of foot-and-mouth disease virus-seropositive and probang-positive cattle herds in the Adamawa province of Cameroon.

Six of the seven known serotypes of foot-and-mouth disease (FMD) virus occur in Africa. This paper describes the results of a population-based cross-sectional study of the seroprevalence of FMD and the persistence of the virus in cattle herds and associated sheep flocks in the Adamawa province of Cameroon. Antibody titres measured by the virus neutralising test indicated that serotypes O, A and SAT2 viruses had been circulating in the province. The estimates of apparent seroprevalence in cattle herds, based on five juvenile animals (eight to 24 months old) per herd, were 74.8 per cent for serotype SAT2, 30.8 per cent for serotype A and 11.2 per cent for serotype O, indicating recent exposure; the estimates based on animals more than 24 months of age were 91.1 per cent for SAT2, 83.6 per cent for A and 34.2 per cent for serotype O. Epithelial and oropharyngeal samples were collected from cattle and small ruminants, cultured and typed by ELISA; serotypes A and SAT2 were isolated from both types of sample. The herd-level estimate of apparent prevalence of probang-positive herds was 19.5 per cent and the animal-level estimate of apparent prevalence was 3.4 per cent. The geographical distribution of the seropositive herds based on juveniles suggested that recent SAT2 exposure was widespread and particularly high in the more northern and western parts of the province, whereas recent exposure to serotype A was patchy and more concentrated in the south and east. This distribution corresponded very closely with the distribution of herds from which virus was recovered by probang, indicating recent exposure or infection. No serotype O viruses were recovered from cattle, and the distribution of seropositive herds suggested very localised recent exposure. The apparent prevalence of probang-positive animals declined with the age of the animal and the period since the last recorded outbreak in the herd.

Risk of fatality and causes of death of Thoroughbred horses associated with racing in Victoria, Australia: 1989-2004.

REASONS FOR PERFORMING STUDY: Determining the risk of fatality of Thoroughbred horses while racing is essential to assess the impact of intervention measures designed to minimise such fatalities. OBJECTIVES: To measure the risk of racehorse fatality in jump and flat starts on racecourses in Victoria, Australia, over a 15 year period and to determine proportional mortality rates for specific causes of death. METHODS: All fatalities of Thoroughbred horses that occurred during or within 24 h of a race were identified from a database. The risk of a start resulting in a racehorse fatality in all races and within flat and jump races, proportional mortality rates, population attributable risk, population attributable fraction and risk ratios were calculated along with 95% confidence intervals. Poisson regression was also performed to estimate risk ratios. RESULTS: There were 514 fatalities over the 15 year period; 316 in flat races and 198 in jump races. The risk of fatality was 0.44 per 1000 flat starts and 8.3 per 1000 jump starts (18.9 x greater). The risk of fatality on city tracks was 1.1 per 1000 starts whereas on country tracks it was 0.57 per 1000 starts. Of the 316 fatalities in flat races, 73.4% were due to limb injury, 2.5% to cranial or vertebral injury and 19.0% were sudden deaths. Of the 198 fatalities in jump races, 68.7% were due to limb injury, 16.2% to cranial or vertebral injury and 3.5% were sudden deaths. The risk of fatality in flat starts increased between 1989 and 2004 but the risk in jump starts remained unchanged over the 15 year period. CONCLUSIONS: The risk of fatality in flat starts was lower in Victoria than North America and the UK but the risk in jump starts was greater. Catastrophic limb injury was the major reason for racehorse fatality in Victoria but there was a larger percentage of sudden deaths than has been reported overseas. The risk of fatality in jump starts remained constant over the study period despite jump racing reviews that recommended changes to hurdle and steeple races to improve safety. POTENTIAL RELEVANCE: This study provides important benchmarks for the racing industry to monitor racetrack fatalities and evaluate intervention strategies.

The risk of a horse-and-rider partnership falling on the cross-country phase of eventing competitions.

REASONS FOR PERFORMING STUDY: Fatalities resulting from horse falls occurring during the cross-country phase of eventing competitions initiated epidemiological investigation of the risk factors associated with horse falls. OBJECTIVES: To identify variables that increased or decreased the risk of a horse fall during the cross-country phase of an eventing competition. METHODS: Data were collected from randomly selected British Eventing competitions held in Great Britain during 2001 and 2002. Data were obtained for 173 cases (jumping efforts resulting in a fall of the horse-and-rider partnership) and 503 matched controls (jumping efforts not resulting in a fall). The risk of falling was modelled using conditional logistic regression. RESULTS: An increased risk of a horse fall was associated with jumping into or out of water; taking off from good-to-soft, soft or heavy ground; fences with a drop landing; nonangled fences with a spread > or =2 m; and angled fences. Other risk factors included riders who knew that they were in the lead within the competition before the cross-country phase; an inappropriate speed of approach to the fence (too fast or too slow); horse-and-rider partnerships that had not incurred refusals at earlier fences; and riders who received cross-country tuition. CONCLUSIONS: This study has identified modifiable course- and fence-level risk factors for horse falls during the cross-country phase of eventing competitions. The risk of horse and rider injury at eventing competitions should be reduced by 3 simple measures; maintaining good to firm take-off surfaces at fences, reducing the base spread of fences to <2 m and reducing the use of fences at which horses are required to jump into or out of water. Risk reduction arising from course and fence modification needs to be confirmed by intervention studies. POTENTIAL RELEVANCE: Knowledge of factors that increase or decrease the risk of a horse fall can be used by UK governing bodies of the sport to reduce the risk of horse falls on the cross-country phase of eventing competitions, and reduce the risk of horse and rider injuries and fatalities. As one in 3 horses that fall injure themselves and one in 100 horse falls results in fatality to the horse, we suggest that immediate consideration is given to these recommendations.