Indonesia's health minister announced Thursday that her country will no longer report H5N1 avian flu deaths as they occur — a policy shift experts said would put the country in violation of the International Health Regulations, a key global health treaty.

And an international law expert said Indonesia, which has refused for more than a year to share sample H5N1 viruses with the global scientific community, is playing a risky game if it intends to make good on the threat to only report H5N1 deaths at six-month intervals.

"Indonesia is rolling the dice here with this, both with the virus sharing and then again if these press reports are true with this as well," said David Fidler, director of the Center on American and Global Security at Indiana University.

"If H5N1 mutates — I don't care where it is — it's going to be known as the Jakarta strain. Because people are going to associate it with the recalcitrance of Indonesia to participate fully in the global surveillance needs with regards to H5N1."

Fidler was reacting to comments made by Indonesian Health Minister Siti Fadilah Supari in interviews published Thursday.

The outspoken health minister seemed to be suggesting the ongoing announcements of Indonesian deaths were undermining the image of the country and overshadowing "the positive steps and achievements made by the government in fighting bird flu."

"How does it help us to announce these deaths?" Supari asked in an interview with The Associated Press.

Since early 2007, Supari has dramatically reduced the amount of co-operation her country has given to the World Health Organization's Global Influenza Surveillance Network, a group of influenza laboratories around the world. The group tracks changes in human flu viruses, advising on what strains should be included in annual flu shots.

It also keeps a close eye out for novel flu strains that might pose a pandemic threat. For the last five years, the highly virulent H5N1 virus has been at the centre of its radar screen.

Indonesia refused to share virus samples

But its efforts to assess what is happening with H5N1 have been hampered since early last year by Indonesia's refusal to share sample viruses. Supari has insisted the country won't release H5N1 viruses until it is assured a fair share of any vaccines that are derived from Indonesian viruses.

Throughout the standoff, the country has continued to report new H5N1 cases and deaths. And the suggestion it may now only report the latter from time to time rang alarm bells.

A spokesperson for the WHO said it had not received official notification of a change in policy but is seeking clarification of the minister's remarks.

"We want to confirm whether or not what's been reported in the press is in fact a change in Indonesian policy," John Rainford said from Geneva.

Fidler suggested Indonesia would be in violation of the International Health Regulations if Supari carries through on her threat.

The IHR, as the treaty is commonly known, requires prompt reporting of cases of diseases that have been designated as global health threats. When the IHR were revised in 2005, SARS, polio, smallpox and human cases of avian influenza infection were specified as notifiable diseases. The revised treaty was ratified by the WHO's 193 member states, including Indonesia.

The IHR stipulate that countries must report new H5N1 cases within 24 hours of confirmation and must report deaths in a timely manner.

Twice-a -year reporting of deaths would not meet the test of timeliness, Fidler said.

Rainford concurred that if Indonesia plans to go this route, it would not be fulfilling its commitments under the International Health Regulations.

"It would be, I think, the position of any signatory to the IHR, 193 states ... including Indonesia that this is a notifiable illness and that there are duties and obligations under that to report that detail," he said from Geneva.

"But again, this is speculation if we don't have confirmation of that. But if in fact it was the case, yes, that would be a cause of serious concern."

Dr. Michael Osterholm, director of the Center for Infectious Diseases Research and Policy at the University of Minnesota, also expressed dismay.

"This is a very slippery slope that flies in the face of good public health," he said from Minneapolis. "And lessons around SARS should have been very instructive to countries in Asia about what happens when you do not have full disclosure of public health events in your country."

Since late 2003, there have been 383 confirmed human cases of H5N1 infection in 15 countries and 241 of those people — or 63 per cent — have died.

Indonesia has recorded about a third of all of the cases but has a substantially higher death rate — 81 per cent. The country has recorded 133 confirmed human cases and 108 deaths since July 2005.