Researchers have long warned of the threat of an avian flu pandemic. The message has been taken on board by some politicians, notably in Canada, Britain and Australia. But the past few weeks have seen an unprecedented flurry of top-level US diplomatic activity about flu, prompted in large part by recognition of the fact that the public would not tolerate an inadequate response to another major catastrophe after Hurricane Katrina.

The frontline would be in communities and hospitals, so this is where governments need to focus their disaster planning.

Last month, George Bush told the United Nations that the United States would be at the centre of an international coalition to fight the threat of a pandemic. Last week, he convened officials from more than 80 countries to take the plan forward, and then sat down with the heads of the vaccine industry. The Senate has approved $3.9 billion to fight avian flu, with $3 billion of it going on a US stockpile of antivirals. And Michael Leavitt, the health secretary, has gone on a ten-day trip to meet officials in affected countries in southeast Asia.

But behind this activity, there are disconcerting gaps between the official discourse and reality. US officials have correctly identified promptly stamping out outbreaks as a top priority, emphasizing the need for affected countries to share data and samples. This is an important issue, but what can the coalition do if China and Vietnam, for example, refuse to share data? And even more urgently required are funds to build surveillance capacity in vulnerable countries, and to eradicate the disease in livestock.

The United Nations' Food and Agriculture Organization estimated in February that a minimum of $100 million is needed to begin tackling the problem effectively. But so far, a few countries (including the United States) have pledged a total of just $16.5 million. The outbreak teams of the World Health Organization (WHO) are dwarfed by the challenge facing them.

Many scientists in affected countries are reluctant to cooperate with what often seems to them a one-way street. Hospitals in these countries often have barely enough antivirals to treat existing cases, and lack diagnostics. A better atmosphere for sharing will only come if rich nations offer these countries true cooperation and substantial aid in research and health infrastructure to deal with outbreaks.

US officials have in the past weeks made strident warnings about the lack of preparedness for a pandemic, while simultaneously giving overly reassuring messages that the job is now well in hand. Much of the $3.9 billion in the Senate bill would go towards buying drugs, for example. But US officials have neglected to mention that the United States currently has only enough drugs ordered to cover 1% of its population. Ten countries already have drugs ordered to cover 25–40% of their populations, but it will be several years before the United States can match this, as it is at the back of the queue at the door of Roche, the sole supplier of Tamiflu.

Roche's monopoly on the drug, and its inability to ramp up production swiftly to meet demand, are themselves cause for concern. UN secretary-general Kofi Annan last week hinted that countries might use compulsory licences to produce the drug off-patent, arguing that intellectual-property rights should not be allowed to get in the way of access by the poor to medication. This option has been rejected by the WHO, which recently obtained a donation of 3 million courses from Roche.

With few drugs on the horizon, US officials are stressing that the key weapon in the event of a pandemic will be a vaccine, and that the biggest bottleneck is in industry's capacity to produce it in sufficient quantities. But the vaccine in question is a prototype. It requires such huge doses that, even if the entire world vaccine production capacity of 900 million doses of seasonal flu vaccine antigen were switched to making it, just 75 million doses could be manufactured. In contrast, antigen-sparing strategies using adjuvants could allow from 1 billion to 7 billion doses to be produced, and this is why all efforts should be directed to this goal.

Although research into new vaccines and drugs may help us fight pandemics better several years down the line, the frontline of a pandemic would be in communities and hospitals, so this is where governments need to focus their disaster planning. For example, taps and door knobs in washrooms are a significant route of flu transmission. Converting them to be pushed or opened with an elbow, as in surgical areas of hospitals, could cut transmission during a pandemic.

Vaccines and antiviral drugs deserve top-level attention. But so too do much simpler means of protecting citizens.