The identification and management of H5N1 avian influenza in Denmark in 2006 has been facilitated by close collaboration between the animal and human health authorities and institutions, at national and local level. Public expressions of concern have been well informed and calm, and media reporting has been responsible.

Avian influenza A/H5N1 is currently not a very contagious virus for humans, but there is a small and real risk of infection for people who have close contact with sick birds. Even in an industrialised country with a well-informed population, delays in case ascertainment affect the management of an outbreak. At present, it is impossible to know whether avian influenza H5N1 has become endemic in Danish wild birds, or, if it has not, whether it is likely to be reintroduced later. It is therefore important to maintain timely surveillance, preparedness and communication lines between the relevant stakeholders.

H5N1 in wild birds
The first detection of highly pathogenic avian influenza A H5N1 in Denmark was on 14 March 2006 in a common buzzard (Buteo buteo) found on Svinø near the town of Næstved in the southern part of Sjælland. Subsequently highly pathogenic avian influenza A H5N1 was found in 43 wild birds through passive surveillance in various locations in Denmark, mainly areas bordering the Baltic Sea along the Eastern migratory flyways. Avian influenza has also been detected in a pooled sample of herring gull (Larus argentatus) faeces. Among the various types of bird tested, the rates of positive findings have been highest in waterfowl, particularly swans and ducks. The most recent positive findings in wild birds were made on 2 May 2006. More detailed information is available in Danish and English from the Danish Veterinary and Food Administration [1].

H5N1 in a backyard poultry holding
An outbreak of highly pathogenic avian influenza A H5N1 in a poultry flock was confirmed on 18 May 2006. It occurred in a backyard holding in Funen county, in an area where avian influenza A H5N1 had previously been detected in wild birds. The flock consisted of 102 birds (51 chickens, 41 ducks, 5 geese, 3 guinea fowls, and 2 peacocks) of which 47 died between 3 and 12 May 2006 (Figure). The birds were kept in four different sections located on different parts of the farm, and the sick birds all belonged to the same section which held about 50 birds. The birds in the affected section were all free-ranging and thus exposed to wild birds. When the owner became aware of the outbreak, he implemented biosecurity measures. Changes of clothing and footwear seemed to be sufficient to prevent the spread of the disease to the other sections. All the birds from the three non-affected sections were culled, but were tested H5N1 negative.

The two people living on the farm both had close contact with the birds. They culled sick birds on their own initiative without using any personal protective devices such as masks or gloves. They reported that they got bird blood on their bare hands during this process. In addition, there had been visitors at the farm on more than one occasion, including a person who came on 8 May, and three visitors who purchased eggs for home cooking use on 10 May (Figure). There was a considerable delay between the onset of the outbreak on 3 May 2006 and the notification to the district veterinary officer on the evening of 17 May 2006.

Figure. Daily number of dead poultry in a backyard holding with avian influenza A H5N1 in Funen county, Denmark, 2006. Arrows indicate days with visitors at the farm and reporting to the veterinary authority.

By 18 May, a total of six people had been directly exposed to sick poultry since 3 May. These included the two people living on the farm, one of the three visitors who came to purchase eggs on 10 May, and a veterinarian and two cullers (personal protective devices were used by the veterinarian and the cullers). Following guidance from the World Health Organization and the European Centre for Disease Prevention and Control, these six persons were all recommended to take oseltamivir for post-exposure prophylaxis (which they duly accepted), and were asked to report any illness with fever to the local public health authority [2,3]. The owners were also instructed to keep out of the areas where the birds had been until these areas had been cleaned and disinfected by a contracted cleaning company. No one developed signs or symptoms compatible with human infection with avian influenza virus.

A family of five had received a delivery of fresh table eggs from the farm on 16 May, and ate these eggs fried (that is, with runny, partially cooked yolks) the following day. As a precautionary measure, these individuals were recommended oseltamivir prophylaxis on 20 May. One of the family members developed a sore throat on 18 May, but the family doctor, in consultation with specialists in infectious disease, did not judge the signs to be compatible with avian influenza in humans and it was decided that testing was not justified. No other symptoms have been reported from this family.

The visitor who came on 8 May has since been well, with no symptoms of influenza-like illness, and was not prescribed antiviral drugs, since more than ten days had already passed since exposure. Two of the three visitors who came on 10 May stayed at the farmhouse and were not directly exposed to the birds, and were consequently not offered antiviral drugs. Another delivery of fresh table eggs was made from the farm to a household on 15 May; and although it is theoretically possible that these individuals may have been exposed to avian influenza virus, they were not offered antiviral drugs because they reported that they did not cook, eat or otherwise handle the eggs but kept them in their packaging in the refrigerator.

H5N2 in a game bird holding
On 2 June 2006, low pathogenic avian influenza H5N2 was detected in a game bird holding, also in Funen county. There were no reported clinical signs of disease among the animals. Mallard (Anas platyrhynchos) ducklings from the holding were tested as part of the Danish surveillance programme for avian influenza in poultry. At the initial testing, the virus was found to be A/H5, but the neuraminidase type was unknown. Although a highly pathogenic avian influenza (such as A H5N1) was considered to be unlikely, given the absence of bird deaths, it was considered reasonable to offer the owners and the personnel working at the holding antiviral drugs for post exposure prophylaxis. Nearly 90 people had recently purchased live ducklings from the holding and 50 people had been involved in the culling of the 19 500 birds, and all were eligible for post exposure prophylaxis. However, before prophylaxis could be offered to the duckling purchasers and cullers, it became clear that the viral strain was a low pathogenic H5 type that had not been described as a cause of illness in humans. Therefore, the prophylaxis was not distributed to this group, and treatment of the owners and the personnel working at the holding who had already started a course of prophylaxis was discontinued.