May 11, 2009 (CIDRAP News) – A rapid assessment team from the World Health Organization (WHO) today predicted that the novel H1N1 (swine flu) virus may spread similarly to 20th century pandemics, but said it's too early to gauge the future impact on human health.
In making their modeling projections, the group used surveillance data from Mexico, focusing on La Gloria, Veracruz, where the outbreak is believed to have started in mid February, sickening over half of the village's population. The research team published its findings in an early online edition of Science.
"Our early analysis would suggest that this is going to be an outbreak comparable to that of 20th century pandemics regarding the extent of its spread—it's very difficult to quantify the human health impact at this stage, however," said senior author Neil Ferguson of Imperial College London in a statement released by Science.
Though the authors cautioned that the data on the outbreak are incomplete and many uncertainties remain, they wrote that providing an early picture of the outbreak can help governments and health organizations guide policy decisions such as closing schools and deploying other community mitigation measures.
As of May 5, Mexico had reported 11,357 suspected and 822 laboratory-confirmed cases, which likely underestimates the true impact of the epidemic, given the surveillance focus on severe cases, the group wrote. Calculating back from the number of infections in travelers, they estimated the number of people infected in Mexico by late April varied between 18,000 and 32,000.
At a media briefing today, Anne Schuchat, MD, interim deputy director for the Centers for Disease Control and Prevention's (CDC's) science and public health program, said the number of cases in the United States also is likely the "tip of the iceberg," because testing has focused on confirming the more serious cases.
With 9 confirmed and 92 suspected deaths in Mexico reported by Apr 30, the team estimated, using the interval-censored case count model, that the case-fatality rate ranged from 0.3% to 0.6%, a figure that the WHO used when it discussed raising the pandemic alert phase from 4 to 5. The team wrote that the severity seems less than the 1918 influenza pandemic and comparable to the milder one that occurred in 1957.
In making their transmissibility projection, the group used diversity in genetic sequences from confirmed cases to guide their analysis rather than focusing on the La Gloria outbreak.
Using a Bayesian coalescent model, assuming exponential viral population growth, the team estimated the outbreak began on or around Jan 12, 2009 (with a 95% credible interval of Nov 3, 2008 to Mar 2, 2009). The finding is consistent with the epidemiologic investigation of the outbreak's start and magnitude, they wrote.
Using different methods, they estimated the reproductive number (RO), a key measure of transmissibility, to range from 1.4 to 1.6. The Science statement said those estimates are similar to or lower than those for the pandemics of 1918, 1957, and 1968. However, a genetic analysis produced an RO estimate of 1.2, consistent with sustained human-to-human transmission in Mexico.
The group reported that their transmissibility estimates are at the low end of ranges used in earlier computer simulations to study pandemic mitigation strategies, meaning the resulting conclusions could be relevant to the current epidemic.
"However, the key trade-off remains the balancing of the economic and societal cost of interventions such as school closure, against the number of lives saved through such measures," they suggested. Health officials might also have to weigh large-scale antiviral prophylaxis, where stockpiles are available, against the possible development of antiviral resistance.
They cautioned that current estimates of disease severity aren't robust enough to evaluate the trade-offs, but they said estimates will become more certain in coming weeks as more data on severe cases in the United States and other countries become available.
To fine-tune the pubic health response, authorities will need detailed investigations of transmission in households and schools, regular and reliable access to electronic medical records, and better information about incubation and infectious periods, the team wrote.
Researchers will be combing through the data for more clinical severity clues and to detect potential regional variations. "As the epidemic spreads further, it is likely that severity will vary from country to country depending on healthcare resources and the public health measures adopted to mitigate impact," they concluded.
High-quality epidemiologic data, combined with analysis and modeling, are useful tools for exploring other issues such as existence of any cross-immunity, whether the new strain will displace circulating influenza A subtypes, and if reduced seasonal transmission in North America and Europe is playing a role in the current moderate transmissibility.
Fraser C, Connelly CA, Cauchemez S, et al. Pandemic potential of a novel strain of influenza A (H1N1): early findings. Science 2009 May 11; early online edition [Abstract]
