Mathematical modellers at Harvard´s School of Public Health may have identified the catch-22 of pandemic flu.

Future influenza pandemics might in theory be containable through aggressive early control measures, including wide use of vaccine and anti-viral drugs, they reported Wednesday in a letter to the journal Nature.

The hole in the argument, the authors admit, is that the world is not laying the groundwork needed to have those expensive tools ready for use in the early days of a pandemic.

"In theory it´s possible," senior author Marc Lipsitch said of the containment hypothesis, which is based on modelling that suggests the influenza strain responsible for the worst pandemic in known history was not highly contagious.

While attractive, the theory would not gain universal acceptance from those who specialize in studying the way infectious diseases spread.

"This means that the disease can be stopped or slowed down by appropriate interventions, but those interventions require significant planning and investment and a high level of compliance among the population to those control measures," said Babak Pourbohloul, a mathematical modeller at the British Columbia Centre for Disease Control.

Still, Lipsitch said the results support the argument that it´s worth a try.

In particular, he highlighted the notion that developed countries should donate a stockpile of anti-viral drugs that could be used in a bid to extinguish an emerging pandemic.

"We as rich countries . . . need to stop seeing it as altruism and see it as enlightened self-interest," Lipsitch, an epidemiologist and a microbiologist, said in an interview.

Governments ought to apply the lessons that SARS taught the world, he added.

"It showed that sparks lead to fires and that if you stop them early, you do better."

The stockpile idea is one that influenza experts and the World Health Organization have been mulling over as they have ramped up pandemic preparedness efforts in recent months.

Their work has taken on a sense of increased urgency, fuelled by the widespread and ongoing outbreak of highly pathogenic H5N1 avian influenza in parts of Southeast Asia.

They fear the longer the virus - which has infected at least 44 people this year, killing 32 of them - is in close proximity to people, the greater the chance it will acquire the ability to spread easily to and among humans, sparking a pandemic.

"We´re living on borrowed time," Dr. Klaus Stohr, head of the WHO´s global flu program said Wednesday following a 2½-day pandemic preparedness meeting in Geneva.

"But that gives us a chance to do things we should perhaps have done before."

According to the Harvard modelling, the strain responsible for the Spanish flu of 1918 was far more deadly than strains before or since, but was not significantly more infectious.

Looking at data from 45 U.S. cities, Lipsitch and colleagues Christina Mills and James Robins calculated that on average each person who caught that flu infected between two and four other people.

(By comparison, measles and chickenpox are highly infectious. In unvaccinated populations, a case of the former would be expected to spark 15 to 17 new cases, and the latter 11.)

That would suggest pandemic flu is about as infectious as SARS, which was contained through measures such as quarantine, isolation and voluntary limits on travel.

But flu is a different beast. For one thing, a person can transmit disease before they even know they´re sick. For another, the incubation period is significantly shorter than that of SARS, giving public health officials much less time to track down and isolate people who might be infected.

"The difficulty is all of us want to believe we can control it," said Dr. Allison McGeer, an infectious diseases expert at Toronto´s Mount Sinai Hospital.

"But there´s nothing new we know about pandemics than people knew about pandemics in ´68 or ´57" - the last two pandemics - "and the precautions didn´t work then."

McGeer suggested early interventions would probably at best buy time, limiting spread at the onset as the parts of the world that have vaccine production wait for a vaccine to become available.

Bolder steps - such as grounding global air travel - might achieve greater time gains but would be politically harrowing decisions to make.

"We will do too little too late, because we won´t have the courage to do what might possibly work at the beginning," McGeer said.

"And just like we don´t have the courage to buy oseltamivir (a costly anti-viral), we won´t have the courage to shut down travel when it might delay the onset, because it will be too early."

Governments aren´t just lacking courage, they´re lacking the data they would need to make those tough decisions, Pourbohloul said.

He said mathematical models can help governments assess whether measures will slow or contain the spread of pandemic flu, but too little work is being done in this area.

"Every country has to develop this capacity for itself. So that they can come up with the best and most rapid solution at a time of crisis," Pourbohloul said.

"And I think in Canada, this has to be addressed."