First Bird Flu Death in 2003, Admits China

China today admitted that its first human case of bird flu occurred two years earlier than previously reported, a disclosure that rewrites the history of the disease and raises questions about Beijing's willingness to share data about the epidemic.

A 24-year-old soldier identified only by his surname, Shi, died of the H5N1 virus at a military hospital in Beijing in November 2003, the health ministry said on its website. The online statement confirmed claims made in June by eight Chinese researchers in the New England Journal of Medicine.

The government blamed the long delay in reporting the case on the initial confusion about the causes of the man's illness. After treatment for pneumonia failed, Shi was tested for severe acute respiratory syndrome (Sars), which shares many avian flu symptoms. He died four days after being admitted to hospital.

It was only in February 2004 - amid an outbreak of avian flu - that researchers took his samples out of the freezer and checked them for the H5N1 virus. The ministry said it took more than a year for the results to be confirmed. In other countries diagnosis takes just days or weeks.

The World Health Organisation said Shi's death was the first confirmed fatality from a new strain of the disease that subsequently killed at least 135 people, most of them in Vietnam and Thailand. China had claimed its first human case occurred in November 2005. Since then it has reported 19 cases, 12 of them fatal.

Julie Hall, the coordinator of WHO China's epidemic and alert response team, said Shi's case showed how the virus had mutated from an earlier strain of H5N1 found in Hong Kong in 1997 and February 2003. "This helps us to understand how the disease has evolved," she said.

After the New England Journal of Medicine report, the Chinese health ministry invited the WHO to help retest the samples. Dr Hall said she was pleased at the opportunity to cooperate, but questioned the delay between diagnosis and disclosure. "This has highlighted a gap in China's reporting system," she said.

Unlike hospitals, research institutes are not legally obliged to report suspected cases of infectious diseases, which means that other H5N1 positive samples may not yet have come to light. China's agriculture ministry has also yet to make public its records about the spread of avian flu.

Shi is thought to have contracted the disease in Beijing in October or November 2003 but there is no record of any avian flu outbreak in poultry stocks.

Dr Hall called on the ministry to release more information so that the WHO could chart the disease's history. She said: "The clearer the picture we have globally, the easier it is to see how the virus has been changing, what it might do next, and how to respond."