Experts are frustrated that investigations into a cluster of bird flu infections in Indonesia are moving too slowly, fearing a failure to pin down the source quickly may mean they miss a dangerous mutation of the virus.
Seven members of a family in Kubu Simbelang village in north Sumatra were infected with the H5N1 virus and six of them died between May 4 and 12. But experts and local health authorities have come no closer to finding the culprit.
"This case shows surveillance work should intensify. When there are human infections, you have to find the source. It's too slow," said microbiologist Guan Yi, who has studied the H5N1 virus since it made its first known jump to humans in Hong Kong in 1997. "Early detection in animals is essential."
"If it was a pandemic strain, we'd be finished," he said.
Experts have been warning for over a year that the H5N1 may trigger a pandemic that could kill millions once it gains the ability to transmit efficiently among humans.
The Indonesian infections, like any cluster, are deeply worrying as they raise the possibility of the virus mutating into the much dreaded pandemic beast.
But experts say that has probably not happened because the disease has not spread beyond the seven people. The virus has killed 123 people around the world since late 2003.
"It's good news that there is no further transmission," said Sari Setiogi, WHO's spokeswoman in Indonesia.
Field epidemiologists trying to establish the source of the infections have left nothing to chance and have tested pigs, chickens, ducks and geese for traces of the virus.
Bird flu antibodies were detected in pigs raised by the family, but nasal swabs taken from the swine, which all appear healthy, showed nothing. All the other animals tested negative.
"This could mean that the pigs were previously infected with H5N1 and developed antibodies. But the infection did not persist, so you don't detect the virus in the respiratory tract," said Leo Poon, a microbiologist at the University of Hong Kong.
But establishing sources of infection are by no means easy.
While genetic sequencing may show that two virus samples taken from a human and an animal are very similar, that could just mean that either one could have infected the other or that a third agent might have infected both the human and the animal.
And it is often impossible to determine if human-to-human transmission has occurred since family members are exposed to the same animal and environmental sources as well as to one another.
"I don't think it is human-to-human, because they shared the same environment, poultry, pigs, so they may have been exposed to the same source repeatedly," Guan said of the Sumatran cluster.
Concerned about the country's sporadic cases of human infections, a number of WHO experts were already in Indonesia before tragedy befell the family in Sumatra. But their investigations have yielded nothing so far.
"We can't rule either way (human-to-human or animal-to-human) at this stage, we're still doing investigations," Setiogi said.
Investigations are also fraught with logistical, and heartbreaking, difficulties.
"The family is in mourning. You have lost your children and there are strangers asking questions, if you remember this or that, it's not easy to deal with," Setiogi said.