Science Editor

The World Health Organization announced late Monday it has followed the advice of its emergency committee and raised the Swine flu pandemic alert to Phase 4, meaning that there is sustained human-to-human transmission of a novel influenza virus.

Though the level has not been raised to 5 because community outbreaks are limited to Mexico for the time being, with cases now confirmed on three continents, there are clearly other things to fear from the H1N1 swine flu virus besides fear itself.

But WHO Assistant Director-General Keiji Fukuda tried to put those fears into perspective over the weekend and on Monday, stressing that there is no pandemic to date and it is still possible that a pandemic will not develop.

"A pandemic is not considered inevitable at this time," Fukuda told reporters at a press briefing on Monday. But "given the current situation, the current focus of effort should be on mitigation efforts, rather than trying to contain this virus. At this time, containment is not a feasible option," he said.

It is for this reason that the WHO did not recommend closing borders or restricting travel, though Fukuda did say that it would be "prudent" for sick individuals to delay travel.

The emergency committee also recommended against switching from the production of seasonal influenza to swine influenza vaccine at this point, but did recommend putting efforts into developing a swine flu vaccine "as quickly as possible."

Fukuda sounded a positive note late Monday, saying that vaccine production "has changed considerably over the past few years," based partly on the experience with avian influenza viruses. He said that that the development of a vaccine is expected to take at least four to six months, with several additional months for production.

The WHO defines a pandemic as occurring "when a new influenza virus appears against which the human population has no immunity, resulting in epidemics worldwide with enormous numbers of deaths and illness."

If a pandemic does develop - as now appears likely - there still may be some good news. The apparently relatively mild pathogenic potential of the H1N1 virus and its susceptibility to currently available antivirals means that such a pandemic would not necessarily overwhelm public health systems and/or lead to a massive number of deaths.

The worst-case scenario for a flu pandemic is a pandemic with a highly pathogenic virus in the mold of the 1918/1919 "Spanish flu," for which death toll estimates range from 20 million to 100 million. That death toll clearly exceeded that of World War I, which directly preceded it. The leading suspect over the past few years to cause such a pandemic has been the H5N1 avian influenza virus, which currently does not spread easily from human to human, but whose mortality rate of more than 50 percent has put its monitoring high on the priority of public health officials.

But the 20th century actually saw three pandemics; in 1918, 1957, and 1968. The pandemics of 1957 and 1968 were caused by milder influenza strains and were largely forgotten as soon as they were over, precisely because they did not lead to the scale of death seen in the 1918 pandemic - whose death rate was lower than that of H5N1, but which spread very widely due to a combination of factors. Even the 1918 strain would presumably lead to a lower death rate still today than it did in 1918, before the advent of antivirals, antibiotics to combat the pneumonia, which ultimately kills many flu victims, or mechanical ventilators.

If the new swine flu strain does start spreading uncontrollably, the consequences are impossible to predict at this point. On Sunday, World Health Organization Director-General Margaret Chan succinctly summarized the bottom line: "A new disease is, by definition, poorly understood," she told reporters at a press conference. "We do not yet have a complete picture of the epidemiology or the risks."

But to date at least, the swine flu virus that's responsible for the current outbreak appears to be relatively benign - outside of Mexico, most cases have been mild. In the U.S., there have been 40 confirmed cases, but only one brief hospitalization and no deaths - though the WHO's Fukuda cautioned at a press conference that "we are in an evolving situation so we cannot be too focused on numbers. These numbers can change quite rapidly."

Still, even in Mexico, where the current outbreak began in mid-March and where cases appear to be more serious, "the confirmed number of cases is much, much, much lower than what's being bandied about in some public forums," a WHO spokesperson noted - with some irritation - on Monday.

As of late Monday, Mexico had seven confirmed deaths. Estimates of 150 deaths and more than a thousand infections have not been clinically confirmed, though with "suspect clinical cases" reported in 19 of 32 Mexican states, the number of confirmed cases will clearly increase as suspect cases are tested.

Why the virus seems to be deadly in Mexico, but nowhere else to date, is still unclear. One possibility is that Mexico has greater numbers of undetected cases, and/or that only the most serious cases are coming to the attention of doctors. During a 2006 outbreak of H5N1 avian influenza in Turkey, the death rate was comparatively low; at the time, public health officials suggested that this lower mortality rate might be due to improved surveillance and detection compared to earlier outbreaks in Southeast Asia.

But a more worrisome possibility is that Mexico is simply the tip of the iceberg and death rates will rise elsewhere as the virus spreads. "There is no standard picture of how [the disease] develops," Fukuda said. And the 1918 pandemic provides a grim reminder that in virology, too, past results are no guarantee of future performance that a gradual development toward worse outcomes is also possible. That pandemic occurred in several waves, with the strain becoming more virulent over time.