Dossier AI Transmission: Per-Contact Probability of Infection by Highly Pathogenic Avian Influenza

Estimates of the per-contact probability of transmission between farms of Highly Pathogenic Avian Influenza virus of H7N7 subtype during the 2003 epidemic in the Netherlands are important for the design of better control and biosecurity strategies.

We used standardized data collected during the epidemic and a model to extract data for untraced contacts based on the daily number of infectious farms within a given distance of a susceptible farm.
With these data, the ‘maximum likelihood estimation’ approach was used to estimate the transmission probabilities by the individual contact types, both traced and untraced.

The outcomes were validated against literature data on virus genetic sequences for outbreak farms. The findings highlight the need to

1) Understand the routes underlying the infections without traced contacts and
2) To review whether the contact-tracing protocol is exhaustive in relation to all the farm’s day-to-day activities and practices.


Dossier AI Transmission: Supplementary information wind-mediated transmission HPAI

A comparison between the transmission risk pattern predicted by the model and the pattern observed during the 2003 epidemic reveals that the wind-borne route alone is insufficient to explain the observations although it could contribute substantially to the spread over short distance ranges, for example, explaining 24% of the transmission over distances up to 25 km.

In this generic overview, you will find the date used in the publication “Modelling the Wind-Borne Spread of Highly Pathogenic Avian Influenza Virus between Farms”, published February 2012 (see also For the outbreak of avian influenza A(H7N7) in the Netherlands in 2003, much data are available. The overview gives a description of the data used in the analyses of the mentioned publication:

Epidemiological data

There were 5360 poultry farms in the Netherlands in 2003, for all of which geographical information x is available. For 1531 farms the flocks were culled, for all of these the date of culling Tcull is known. For 227 of the 241 infected farms the date of infection tinf has been estimated, based on mortality data. The remaining 14 farms are hobby farms, defined as farms with less than 300 animals, for which no mortality data are available.

The geographic and temporal data together have previously been used to estimate the critical farm density, i.e. above what density of farms outbreaks are can occur.

Genetic data
The HA, NA and PB2 genes of viral samples from 231 farms have previously been sequenced. Sequence data RNA can be found in the GISAID database under accession numbers EPI ISL 68268-68352, EPI ISL 82373-82472 and EPI ISL 83984-84031. These data have previously been used to give general characteristics of the outbreak, to reconstruct the transmission tree and to assess the public health threat due to mutations of the virus in the animal host.

Meteorological data

Available meteorological data include wind speed wv and direction wdir (with a ten degree precision) and the fraction of time r without precipitation for every hour of every day of the outbreak, measured at five weather stations close to the infected farms. These data are available from the Royal Dutch Meteorological Institute at


Dossier AI Transmission: Modelling the Wind-Borne Spread of Highly Pathogenic Avian Influenza Virus between Farms

The mechanisms of HPAI virus spread between farms are poorly understood; it has been hypothesized that the indirect between-farm contacts play a role [9,14–17].

The frequency and the transmission effectiveness of these contacts determine their virus transmission rates. Here we perform a quantitative assessment of the contribution of indirect contacts to the spread of the virus between farms during the 2003 HPAI epidemic in the Netherlands.

During this epidemic, potentially infectious contacts to both infected and escaping farms were traced. In this paper, the collected data is used to quantify the per-contact probability of virus transmission between farms.

A quantitative understanding of the spread of contaminated farm dust between locations is a prerequisite for obtaining much-needed insight into one of the possible mechanisms of disease spread between farms.

A model was developed to calculate the quantity of contaminated farm-dust particles deposited at various locations downwind of a source farm and apply the model to assess the possible contribution of the wind-borne route to the transmission of Highly Pathogenic Avian Influenza virus (HPAI) during the 2003 epidemic in the Netherlands.

The model is obtained from a Gaussian Plume Model by incorporating the dust deposition process, pathogen decay, and a model for the infection process on exposed farms. Using poultry- and avian influenza-specific parameter values we calculate the distance-dependent probability of between-farm transmission by this route.

A comparison between the transmission risk pattern predicted by the model and the pattern observed during the 2003 epidemic reveals that the wind-borne route alone is insufficient to explain the observations although it could contribute substantially to the spread over short distance ranges, for example, explaining 24% of the transmission over distances up to 25 km.


Dossier AI Transmission: Evidence that link outbreaks in Eurasia, China, South Korea, USA and Canada

In a new study published in the Journal Virology on March 31, 2015, scientists from the U.S. Geological Survey and U.S. Fish and Wildlife Service harnessed a new type of DNA technology to investigate avian influenza viruses in Alaska. Using a “next generation” sequencing approach, which identifies gene sequences of interest more rapidly and more completely than by traditional techniques, scientists identified low pathogenic avian influenza viruses in Alaska that are nearly identical to viruses found in China and South Korea. This publication provides even more evidence of this intercontinental avian influenza exchange program.
The viruses were found in an area of western Alaska that is known to be a hot spot for both American and Eurasian forms of avian influenza. “Our past research in western Alaska has shown that 70 percent of avian influenza viruses isolated in this area were found to contain genetic material from Eurasia, providing evidence for high levels of intercontinental viral exchange,” said Andy Ramey, a scientist with the USGS Alaska Science Center and lead author of the study. “This is because Asian and North American migratory flyways overlap in western Alaska.”

In this study, led by the USGS, low pathogenic H9N2 viruses were found in an Emperor Goose and a Northern Pintail. Both of the H9N2 viruses were nearly identical genetically to viruses found in wild bird samples from Lake Dongting, China and Cheon-su Bay, South Korea.
“These H9N2 viruses are low pathogenic and not known to infect humans, but similar viruses have been implicated in disease outbreaks in domestic poultry in Asia,” said Ramey. “There is no commercial poultry production in western Alaska and highly similar H9N2 virus strains have not been reported in poultry in East Asia or North America, so it is unlikely that agricultural imports influenced this result.”
The finding provides evidence for intercontinental movement of intact avian influenza viruses by migratory birds. The USGS recently released a publication about the detection of a novel highly pathogenic H5N8 virus in the U.S. that is highly similar to the Eurasian H5N8 viruses. This suggests that the novel re-assortment may be adapted to certain waterfowl species, enabling it to survive long migrations. That virus, and associated strains, has now spread from early detections in wild and domestic birds in Pacific states to poultry outbreaks in Minnesota, Missouri and Arkansas.

“The frequency of inter-hemispheric dispersal events of avian influenza viruses by migratory birds may be higher than previously recognized,” said Ramey. While some of the samples for the project came from bird fecal samples collected from beaches at Izembek National Wildlife Refuge, most of the samples came from sport hunters.

“For the past several years, we’ve worked closely with sport hunters in the fall to obtain swab samples from birds and that has really informed our understanding of wildlife disease in this area,” said Bruce Casler, formerly a biologist with the USFWS Izembek National Wildlife Refuge and a co-author of the study.


Dossier AI Transmission: David Stallknecht doubts the role migratory birds spreading AI

The question what causes the current outbreaks in Minnesota is heavily discussed, especially the role of migratory birds. USDA APHIS tends to the opinion that migratory birds play an important role. In that case, H5N2 will likely remain a threat to U.S. poultry for three to five years, Olson said, citing information from wildlife experts. That is how long it will take wild birds to develop immunity to the disease. Since the beginning of the year, the flu, which can kill nearly an entire poultry flock within 48 hours, has also been found in birds from Oregon to Arkansas. The discoveries have prompted major overseas buyers such as Mexico and Canada to limit imports of U.S. poultry and companies such as Tyson Foods Inc to strengthen measures to keep the disease off farms.

The number of infections is climbing as migratory ducks, which are believed to be spreading the disease, return to Minnesota to breed after spending the winter farther south, said Beth Thompson, assistant director of the Minnesota Board of Animal Health. The larger number of ducks likely increases the risk for wild birds to transmit the virus.

Farm workers are probably infecting turkeys by tracking the virus into barns after stepping in contaminated duck feces, said John Glisson, vice president of research for the U.S. Poultry and Egg Association. Chicken flocks are also vulnerable. “Minnesota is a real hotbed for returning waterfowl,” Glisson said. The USDA has said it believes migratory ducks are spreading the flu and sent a team to Minnesota to determine how it is moving into poultry flocks. So far, efforts to stop the spread by controlling human and vehicle traffic on farms have not worked.

The number of infections may continue to rise through mid-May, when spring migration ends, said Steve Olson, executive director of the Minnesota Turkey Growers Association. New cases may accelerate again in the autumn when recently hatched ducks, which have never been exposed to the virus, begin migrating south, he added.

In the absence of reliable epidemiological data about the spreading under wildfowl population, the source of infection will continue to spark speculations about the role of migratory birds.

Despite all evidence, there are scientists who have reason to doubt the thesis that migratory birds are to blame, like Professor David Stallknecht from University of Georgia’s College of Veterinary Medicine. David is openly questioning the USDA’s conclusion. David Stallknecht called the notion that avian flu originated in wild birds “pure speculation” in a story posted by the University of Minnesota’s Center for Infectious Disease Research and Policy (CIDRAP).

David Stallknecht: “It is based on circumstantial evidence that is rapidly becoming accepted dogma. The avian flu saga dates back to late last year, when H5N2 struck several farms in British Columbia. That sparked surveillance programs in the U.S., which turned up cases of pintail duck virus in Washington and a related strain in a captive gyrfalcon nearby. Since then, cases of H5N2 have shown up in poultry in several states, including Idaho, Minnesota and Arkansas.”

In order to stop large-scale outbreaks, more efforts have to be undertaken to protect commercial flocks and to predict future outbreaks. Even more reasons to increase the active surveillance efforts, including the crucial active surveillance programs under wild birds, like Stallknecht suggests.


Dossier AI in the USA: Massive blow to the Minnesota turkey industry

The commercial turkey industry in Minnesota has been hit hard by the current HPAI outbreak. Could this been prevented? Is the USA prepared for large-scale outbreaks of Avian Influenza?

Does the United States have the strongest AI surveillance program in the world, as claimed on the website of USDA APHIS? If so, how come that the current outbreak was detected in Canada, weeks or even longer before the first outbreaks within the poultry population were detected just on the other side of the border in the US?

And why doesn’t the AI surveillance program include large-scale active field surveillance within the wild birds population? Why took it so long before the potential impacts associated with the introduction of HPAI viruses into wild bird populations were understood by the industry?

The current epidemiological efforts are insufficient to determine the risks of other farms getting infected, because the mechanism of infection the current flocks is still unknown. Active laboratory surveillance is not the answer either to predict future outbreaks: Massive efforts in active field-based epidemiological research under wild birds and in wetlands are needed to as part of an early warning system; hunting, fishing and other activities that take place in the wetland areas have to be banned to prevent humans to enter the HPAI infection to the farm; and all commercial turkey farms need to be completely isolated as long as the risks of outbreaks is eminent.

Compared to other poultry species, turkeys need only 1/100 of the normal virus load on contaminated materials (soil, organic materials etc.) to become infected, with lethal consequences for the birds, and enormous financial consequences for the farming industry.

The turkey industry is still completely in the dark about the current outbreak situation. The USDA APHIS active laboratory surveillance programs might be one of the best in the world, is clearly not enough under the current circumstances to determine whether a flock has become infected. Unless the efforts to collect more valid data is substantially increased, followed by rigorous epidemiological analysis and a solid risk assessment, the turkey industry in Minnesota will probably see more outbreaks to come in the near future.


Dossier H5N1: Local veterinary officials failed timely notifying large-scale HPAI outbreak

HYDERABAD: It was only after 39,240 birds perished that the state animal husbandry department woke up to the deadly avian influenza virus doing the rounds on the city outskirts. The deaths of layer birds (used for egg production) took place in two poultry farms in Thorrur village of Hayatnagar mandal a full one week before the government declared the bird flu outbreak.

The owners of the two poultry farms, V Bala Krishna Reddy and Srinivas Reddy, had reported the deaths of the birds from April 6-April 13 to the Bhopal-based High Security Animal Diseases Laboratory (HSADL). The lab, in turn, informed the central government following the samples of the dead birds sent to them by the Reddy brothers turning positive for the H5N1 virus.

The state animal husbandry department, which was in the dark of the developments all this while, was finally informed about the outbreak by the central government’s department of animal husbandry, dairy development and fisheries.

The break-up of poultry deaths that fell prey to the avian influenza include 80 (April 6), 160 (April 7), 400 (April 8), 1,600 (April 9), 2,500 (April 10), 17,500 (April 11), 9,000 (April 12) and 7,500 deaths (April 13) respectively, as per records in possession with TOI.

“This is a glaring example of how the government officials had utterly failed in reporting the disease, which is mandatory for all notified diseases including H5N1, as required under The Prevention and Control of Infectious and Contagious Diseases in Animals Act, 2009. The stakeholders must be held accountable now as the damage has already been done,” said city-based wildlife expert Dr C Srinivasulu.

In fact, the Section 4 of the Act states that it is obligatory for the owner or any person in charge of any animal which he has reason to believe to be infective of a scheduled disease, to report the fact to village panchayat or villager officer, who in turn has to report it to the nearest government veterinarian.

However, it is learnt that nothing of this sort happened with top officials of the directorate of animal husbandry, Telangana, admitting that neither their veterinary assistant surgeon Anand Reddy nor two village livestock officers (VLOs) of Hayatnagar mandal in Ranga Reddy had any inkling of the outbreak in their local jurisdiction.

Interestingly, Dr Y Thirupathaiah, additional director, directorate of animal husbandry, Telangana gave a clean chit to their officials. “Our people cannot be blamed as the burden of sharing information about the outbreak of infective diseases vests with the village officer and the panchayat. We will be writing to the Ranga Reddy collectorate for action against the erring officials,” he said, acknowledging that they learnt about the outbreak from central government sources on April 13.


Dossier wild ducks: Role wild ducks introducing LP Avian Influenza on free range farms in Holland

Lelystad, April 2015: According to a recently published study (in Dutch) by the University of Wageningen, wild ducks are are identified as a high risk factor for the introduction of Low Pathogen Avian Influenza viruses in free-range laying hens.

Through a case-control study investigated presumed risk factors for introduction of low-pathogenic avian influenza (LPAI) virus in poultry laying farms free range. Under a LPAI virus was defined in this study: an avian influenza virus of each subtype (H1 H16 tm), with the exception of the highly pathogenic avian influenza (HPAI) viruses.

In order to determine the potential risk factors for infection with LPAI virus, forty Dutch free range poultry farms where the introduction of Low Pathogen Avian Influenza virus has been confirmed in the past (cases) were compared with 81 free range poultry farms where no introduction has taken place (controls). Questions about the presence of potential risk factors through surveys submitted to the poultry farmers.

The analysis of the various factors shows that the risk of introduction of LPAI virus on free range laying farms 3.3 (95% CI: 1.2-9.7) times higher as mallards has identified by the farmer entering the free range area at least once a week, in comparison to free-range laying farms where wild ducks have been identified by the farmer once a month or less.

It seems logical that the regular presence of wild ducks in the free-range increases the risk exposure of the chickens LPAI virus since wild waterfowl are the natural reservoir of avian influenza viruses.

The study also revealed that the risk factor for free range layer farms located on clay is 5.8 (95% CI: 2.2-15.1) times have higher risk of introduction of LPAI virus then free range layer farms on sandy soil or a soil other than clay. The soil on which the free range farm is situated is probably an indirect risk factor (association and not causation): especially in case the farm is located near the coast or close to rivers.


Dossier H5N2: Questionable role for hunters in introducing AI at US farms

In the U.S. new strains of avian flu that are highly infectious among birds have been found on commercial turkey farms in many states, as The Wall Street Journal has reported. An initial outbreak of H5N8 detected in a Californian turkey flock in January has mixed with North American avian flu strains to generate two new strains, including H5N2, which has been detected along the Pacific bird migration route in Washington, Idaho, and Oregon.

In addition, the U.S. Department of Agriculture has announced that H5N2 had been detected in Minnesota, the first instance of the disease in the Mississippi bird migration route. The virus has spread further to commercial turkey farms in Missouri and Arkansas, including to suppliers of Butterball turkeys.

In an editorial, published online at – March 25, 2015, Dr. Simon M. Shane questions the role of field researchers in transmitting infections from wildlife reservoirs to commercial farms. A March 13th article published by Reuters authored by Tom Polansek highlights questions by wildlife biologists as to the status of migratory waterfowl as reservoirs of H5N2 Avian Influenza which has been isolated from turkey farms in Minnesota, Missouri and Arkansas.

It is an irrefutable fact that identical strains of the pathogen have been isolated from affected turkey farms and from waterfowl. The weight of molecular evidence extending back to 2012 indicates the role of migratory birds in the process of genetic reassortment and dissemination from Asia along the Pacific Flyway.

Dr. Brian McCluskey, Lead Epidemiologist for the USDA-Animal and Plant Health Inspection Service stated “we are pointing right now to ducks as the problem.” Exhaustive assays using advanced analytical techniques including gene sequencing have indicated that the H5N2 virus is carried by waterfowl along the Mississippi Flyway and now more recently the Central Flyway, northward from the Gulf of Mexico.

Dr. Hon Ip a Microbiologist for the National Wildlife Health Center questions the conclusions of the epidemiologists stating “when you are talking about where I would put my money I would say that North to South movement (of infection-Ed.) in the beginning of March totally does not make sense.” The fact that the sequence of infections as confirmed was from North to South, ie. the first case from Minnesota (March 4th) then Missouri (March 9th and 10th) followed by Arkansas (March 11th) all within a short period does not preclude that birds migrating Northwards are transmitting the disease.

Epidemiologic investigations take into account both spatial (location) and temporal (time sequence) factors. It is possible that different groups of ducks, whether by species or age cohort are transmitting virus intra-flock or excreting virus at different rates. Given the small number of outbreaks reported, it is not possible to draw any conclusions such as those expressed by Dr. Ip.

Lou Comicelli, Wildlife Research Manager for the Minnesota Division of Fish and Wildlife stated “it is extraordinary unlikely that avian influenza in a turkey flock in Minnesota has anything to do with wild birds.” This statement is totally unsubstantiated and in fact is contradicted by established scientific investigations. These involved studies on the patterns of seroconversion following infection of range-housed turkeys and sentinels in Minnesota. Strains of avian influenza of low pathogenicity were common to turkeys and waterfowl and antibodies in turkeys were detected concurrently with the appearance of migratory species.

Apparently after the diagnosis of AI in Minnesota turkey flocks was confirmed on March 4th, wildlife officials scouted a 15 mile radius around the farm using an aircraft. Apparently they observed 100 ducks and 18 swans that they concluded were non-migratory “resident birds”. Wildlife biologists in Arkansas also doubt the role of migratory birds because the area where the first case was detected in the state “does not attract many wild birds and waterfowl.”

It is incontrovertible that migratory waterfowl are the reservoirs of H5N2 in the current series of outbreaks involving backyard farms and the few commercial turkey units where the disease has been diagnosed. Obviously defects in biosecurity have allowed the virus to be introduced on to those farms. Possible routes include supplying contaminated, non-chlorinated water from open sources to poultry flocks or failure to implement appropriate personal biosecurity. In this respect workers or contractors hunting or those coming in contact with fecal material voided by waterfowl could have been responsible.

Naturally wildlife biologists are in a state of denial since they are directly and indirectly funded by hunting even though they claim that they are conservationists. Any factor which interferes with their status quo represents a risk to their “field research” and wildlife “management”.

We are now in a new era of avian influenza. No longer is our risk represented by a worker traveling North from Mexico or Guatemala. The paradigm shift requires an understanding of the factors associated with infection and persistence of shedding in migratory waterfowl and the factors linking these reservoir populations with commercial poultry. Cooperation and mutual understanding between wildlife biologist and avian epidemiologists will be necessary to establish appropriate policies relating to commercial poultry production and the recreational exploitation of waterfowl. Denial of scientific fact and an inability to comprehend basic epidemiologic principles coupled with disharmony will detract from productive solutions.


Dossier AI: Global Poultry Outlook Upbeat

Rabobank has published a new report on the global poultry industry, looking at the impact of avian influenza outbreaks across the sector. In the report, Rabobank’s Food & Agribusiness Research team says that pressure from the avian influenza (AI), or bird flu, is further spreading across the globe, with new outbreaks in Asia, Europe and North America.

Rabobank Animal Protein Analyst, Nan-Dirk Mulder, said: “Avian flu is further spreading across the globe and could affect global trade streams, especially as the virus has moved further in Europe to Hungary, and in the US to central states like Minnesota, Missouri and Arkansas. “Joint global approaches, optimal biosecurity, and strong monitoring and compensation systems are necessary to stop the spread of the disease,” he added

Global Outlook

The margin outlook for the global poultry industry remains upbeat, with continuing bullish drivers like high beef prices, lower feed costs and relatively strong demand in most regions.

Global trade is under pressure from the AI outbreaks. The exchange rate volatility and turmoil in regions such as the Middle East and Eastern Europe, are leading to lower trade volumes, prices and shifts in trade streams.

United States
The poultry sector in the US has a favourable outlook but some uncertainties remain. The margins are expected to remain high in 2015. The biggest unknown in 2015 is industry expansion.

The Brazilian poultry market began 2015 bullish despite of the export challenges. The lower oil price will hit the Middle East demand. Still, exports are expected to be strong in 2015, mostly driven by Asia. Margins will be supported by a reduction in feed costs as well.

The EU poultry industry shows some recovery and, although supply is currently tight, the outlook is threatened by ongoing avian flu concerns and still closed export markets.

Entering the first quarter of 2015, China has an ongoing struggle of oversupply due to AI and food safety issues. As a consequence of human AI cases reported, live bird markets were shut down. During the seasonal peak month, poultry retail prices remained flat.

The Russian market remains very bullish on low supply. The outlook remains strong, with expected ongoing tight market conditions due to expensive, limited imported volumes and restrictions on growth.

Other Regions
Mexico has ongoing AI issues and lower pork prices will soften poultry meat consumption growth.
Japan’s poultry industry is still bullish despite ongoing AI outbreaks.
In Thailand, the export position is supported by a strong Thai baht.