Significant features of the epidemiology of equine influenza in Queensland, Australia, 2007

An outbreak of equine influenza (EI) caused by influenza A H3N8 subtype virus occurred in the Australian states of Queensland and New South Wales in August 2007. Infection in the Australian horse population was associated with the introduction of infection by horses from overseas. The first case of EI in Queensland was detected on 25 August 2007 at an equestrian sporting event. Infection subsequently spread locally and to other clusters through horse movements prior to the implementation of an official standstill. There were five main clusters of infected properties during this outbreak and several outliers, which were investigated to find the potential mechanism of disease spread. To contain the outbreak, Queensland was divided into infection status zones, with different movement controls applied to each zone. Vaccination was implemented strategically in infected areas and within horse subpopulations. Control and eventual eradication of EI from Queensland was achieved through a combination of quarantine, biosecurity measures, movement control, rapid diagnostic testing and vaccination.     

Clinical signs of equine influenza in a closed population of horses at a 3-day event in southern Queensland, Australia

This report describes the clinical signs of equine influenza (EI) during an epidemic in a closed, predominantly immunologically na?ve population of horses. It included 254 study horses, few of which exhibited all three signs of pyrexia, nasal discharge and cough simultaneously. We conclude that although the majority of affected horses exhibit temperature patterns resembling those most often described in the published literature, clinicians should be aware that other profiles are quite common.     

Epidemic curve and hazard function for occurrence of clinical equine influenza in a closed population of horses at a 3-day event in southern Queensland, Australia, 2007

The risk of individuals becoming infected during an epidemic of infectious disease can vary as the disease progresses. Monitoring this risk may provide information about the dynamics of transmission. This study describes the epidemic curve for an epidemic of equine influenza (EI) in a closed population of horses predominantly immunologically na?ve to EI at a 3-day event at Morgan Park in southern Queensland, Australia. The hazard function suggested that a subset of horses were at reduced risk of becoming infected. This highlights the importance, when modelling infectious disease in populations, of considering possible differences in the risk of infection among subgroups in the population.     

Pathological changes in horses dying with equine influenza in Australia, 2007

Analysis of pathology results from the 2007 equine influenza (EI) outbreak in Australia indicate that young foals in na?ve horse populations are prone to developing broncho-interstitial pneumonia, and that this can be a rare manifestation of EI virus infection in mature horses. All horses may develop secondary bacterial bronchopneumonia, with mature horses more likely to die. EI outbreaks among heavily pregnant mares can result in increased neonatal losses because of premature placental separation and dystocia causing fetal hypoxia.     

Significant features of the epidemiology of equine influenza in New South Wales, Australia, 2007

Equine influenza (EI) was first diagnosed in the Australian horse population on 24 August 2007 at Centennial Park Equestrian Centre (CPEC) in Sydney, New South Wales (NSW), Australia. By then, the virus had already spread to many properties in NSW and southern Queensland. The outbreak in NSW affected approximately 6000 premises populated by approximately 47,000 horses. Analyses undertaken by the epidemiology section, a distinct unit within the planning section of the State Disease Control Headquarters, included the attack risk on affected properties, the level of under-reporting of affected properties and a risk assessment of the movement of horses out of the Special Restricted Area. We describe the epidemiological features and the lessons learned from the outbreak in NSW.     

Diagnosis of equine influenza virus infections in quarantine stations in Australia, 2007

In August 2007, several horses showed pyrexia and respiratory signs while in post-arrival quarantine in Australia. Subsequent investigations diagnosed equine influenza by serology and PCR in two quarantine stations. A common origin in a shipment of horses from Japan was indicated.     

Innovative zoning to support equine influenza eradication from New South Wales, Australia

Following detection of equine influenza (EI) in New South Wales, a complete standstill was imposed the next morning on the movement of all horses and donkeys within the state. Premises' biosecurity guidelines became progressively more stringent with time in an effort to stop local spread. The standstill was highly effective as a primary response to stop EI becoming widespread across Australia, but did not prevent spread to properties contiguous to infected premises in areas of high horse density and small property size, nor transmission by fomites nor possible local airborne transmission. Within 2 weeks of the start of the outbreak, a zoning system of Purple (Special Restricted), Red (Restricted), Amber (Control), and Green (Protected) Zones was implemented, and progressively modified as disease distribution changed. The colour coding system proved to be an easy way to communicate zone changes, the approximate level of disease risk and the stringency of movement restrictions to the general public and should be adopted more generally in AUSVETPLAN.     

Insights into the economic consequences of the 2007 equine influenza outbreak in Australia

We estimate and describe the financial costs of the equine influenza (EI) outbreak in Australia, including the costs of emergency response measures and lost income/assets to businesses, associations and private horse owners. Costs to associations, governments and industry are discussed. We identify a lack of reliable data about the financial costs of the EI outbreak to the non-racing sectors of the horse industry.     

Role of the diagnostic laboratories during the 2007 equine influenza outbreak in Australia

During the 2007 equine influenza (EI) outbreak in Australia, diagnostic laboratories and the use of appropriate tests played a pivotal role in the response to the crisis. This role began with the detection of EI virus in New South Wales (NSW) on the evening of 24 August 2007 and culminated in providing the final 'proof of freedom' from EI in March 2008. The tests that were used during the EI response were able to provide results quickly, and with high sensitivity and specificity. This section of the supplement describes the roles and functions of the Australian laboratories; tests used and their performance characteristics; improvements made and methods used to validate assays; strategies that were followed during the response; the characterisation of viruses detected; and the resources that were available to laboratories.     

Horse owners' biosecurity practices following the first equine influenza outbreak in Australia

A cross-sectional study was conducted involving 759 Australian horse owners to determine their biosecurity practices and perceptions one year after the 2007 equine influenza outbreak and to investigate the factors influencing these perceptions and practices. A web link to an online questionnaire was sent to 1224 horse owners as a follow-up to a previous study to obtain information about biosecurity perceptions and practices, impacts of the 2007 EI outbreak, demographic information and information about horse industry involvement. Ordinal logistic regression analyses were conducted to determine factors associated with poor biosecurity practices. Biosecurity compliance (low, medium, high), as determined by horse owners' responses to a 16-item question on the frequency of various biosecurity measures, was used as the outcome variable in ordinal logistic regression analyses. Variables with a univariable p-value ≤0.2 were eligible for inclusion in multivariable models built using a manual stepwise approach. Variables with a p-value <0.05 in multivariable models were retained in the final model. Two potential confounders - age and gender of participants - were included in the final model irrespective of their p-values. Thirty percent of the respondents had low biosecurity compliance and were performing biosecurity practices 'not very often' or 'never'. Younger people, people with two or more children, those who were not involved with horses commercially and those who had no long-term business impacts resulting from the 2007 EI outbreak were more likely to have lower biosecurity compliance. People who were not fearful of a future outbreak of equine influenza in Australia and those who thought their current hygiene and access control practices were not very effective in protecting their horses also had poor biosecurity practices. In this observational study we identified factors associated with a group of horse owners with low levels of biosecurity compliance. As this cross-sectional study only assesses associations, the identified factors should be further investigated in order to be considered in the design of extension activities to increase horse owners' biosecurity compliance.     

Vaccination program in the response to the 2007 equine influenza outbreak in Australia

At the time of the initial notification of the occurrence of equine influenza (EI) in Australia in August 2007, vaccination was restricted to horses for which it was an import requirement and only with the approval of the state or territory Chief Veterinary Officer. This paper describes the complexities involved in the selection of a vaccine and its distribution. A combination of ring, predictive and blanket vaccination was implemented during the response. The specific vaccination programs, including its use in buffer zones and for movement of horses, the performance of the vaccine, any adverse reactions and the effect on exposure of vaccinated horses to circulating virus, are also described. Vaccination is considered to have made a valuable contribution to the containment and subsequent eradication of EI from Australia and to risk management in the resumption of horse activities in affected areas from December 2007. Movement restrictions and other biosecurity measures were critically important in controlling the spread of EI and contributing to its eventual eradication, and vaccination was an aid to these measures.     

Isolation and characterisation of an H3N8 equine influenza virus in Australia, 2007

Before 2007, equine influenza had never been diagnosed in Australia. On 22 August 2007, infection was confirmed in horses at Eastern Creek Animal Quarantine Station near Sydney. The virus subsequently isolated (A/equine/Sydney/2888-8/2007) was confirmed by sequence analysis of the haemagglutinin (HA) gene as an H3 virus of the variant American Florida lineage that is now referred to as Clade 1. The HA sequence of the virus was identical to that of a virus isolated from a contemporaneous outbreak in Japan and showed high homology to viruses circulating in North America.     

Assessment of the proportion of under-reporting during the 2007 equine influenza outbreak in New South Wales, Australia

During the 2007 equine influenza (EI) outbreak in Australia, there was no objective information about the possible under-reporting of cases by horse owners either so that they would avoid movement restrictions or because of their inability to detect infection. This investigation aimed to estimate the proportion of under-reporting during the outbreak based on the results of surveillance undertaken in conjunction with vaccination. The results provided improved estimates of morbidity during the outbreak and indicated the level of under-reporting likely to occur in future outbreaks of other infectious diseases in horses in Australia.     

Attack risk on infected properties during the 2007 equine influenza outbreak in New South Wales, Australia

The aim of this preliminary study was to estimate the proportions of seropositive horses on infected premises (IPs) in order to assess the attack risk of the disease. Logistic regression analyses were conducted to evaluate the differences in attack risks between enterprise sizes and predefined spatial clusters/regions. The average attack risk experienced during the outbreak was 96.88% (median 100%), but it differed according to the size of the enterprise and other geographic and demographic conditions. The highest attack risks were observed in the Dubbo cluster/region and the lowest in the Narrabri-Northern cluster. Properties with fewer horses were generally more likely to have higher attack risks than larger enterprises, though this was not true for all cluster/regions.     

Industry liaison officers in the infected states of Australia during the 2007 equine influenza outbreak

We describe the activities of persons who were formally employed as industry liaison officers (ILOs) in the infected states of Australia during the 2007 equine influenza outbreak. The knowledge and communication skills of ILOs allowed them to function very effectively, despite most of them having received no prior training. It is arguable, however, that if ILOs had received prior training, they would have performed more effectively sooner. We suggest that more individuals need ILO training in 'peacetime' and that 'just-in-time' training packages require preparation and trial prior to outbreaks to be effective.     

Proof of freedom from equine influenza infection in Australia in 2007-08

The challenges faced in demonstrating Australia's freedom from equine influenza following the outbreak in 2007-08 and the responding strategy developed by the Proof of Freedom Working Group are presented.     

Immune responses to commercial equine vaccines against equine herpesvirus-1, equine influenza virus, eastern equine encephalomyelitis, and tetanus

Horses are commonly vaccinated to protect against pathogens which are responsible for diseases which are endemic within the general horse population, such as equine influenza virus (EIV) and equine herpesvirus-1 (EHV-1), and against a variety of diseases which are less common but which lead to greater morbidity and mortality, such as eastern equine encephalomyelitis virus (EEE) and tetanus. This study consisted of two trials which investigated the antigenicity of commercially available vaccines licensed in the USA to protect against EIV, EHV-1 respiratory disease, EHV-1 abortion, EEE and tetanus in horses. Trial I was conducted to compare serological responses to vaccines produced by three manufacturers against EIV, EHV-1 (respiratory disease), EEE, and tetanus given as multivalent preparations or as multiple vaccine courses. Trial II compared vaccines from two manufacturers licensed to protect against EHV-1 abortion, and measured EHV-1-specific interferon-gamma (IFN-gamma) mRNA production in addition to serological evidence of antigenicity. In Trial I significant differences were found between the antigenicity of different commercial vaccines that should be considered in product selection. It was difficult to identify vaccines that generate significant immune responses to respiratory viruses. The most dramatic differences in vaccine performance occurred in the case of the tetanus antigen. In Trial II both vaccines generated significant antibody responses and showed evidence of EHV-1-specific IFN-gamma mRNA responses. Overall there were wide variations in vaccine response, and the vaccines with the best responses were not produced by a single manufacturer. Differences in vaccine performance may have resulted from differences in antigen load and adjuvant formulation.