Key Point

Given timely antiviral treatment, quarantine, and use of a vaccine, a pandemic stemming from avian influenza H5N1 may be preventable.

LONDON AND ATLANTA—Recent dire predictions about avian influenza A H5N1 becoming capable of transmission between humans may make it seem that another worldwide influenza pandemic is inevitable. However, two new studies suggest otherwise—even if a transmissible form of H5N1 emerges.^1,2

In both of these studies, computer simulations of possible transmission patterns have indicated that pandemic influenza could be contained through a combined strategy of quarantine and antiviral prophylaxis. Prevaccination could be helpful, too, regardless of the actual efficacy of an anti-H5N1 vaccine, one of the studies also found.

The studies focused on rural areas of Southeast Asia due to the abundant data available for that region. Southeast Asia has been a center of H5N1 activity in wild and domesticated animals—mainly birds and poultry—and is where some of the first cases of animal-to-human H5N1 transmission have occurred.

“A large percentage of animals and people infected with the virus have died,” related Neil M. Ferguson, DPhil, lead investigator of one of the studies and Professor of Mathematical Biology at the Imperial College of London. “The consequences of an H5N1-based pandemic could be catastrophic,” he added.

It would take major planning and coordination to avert such a disaster, and success would not be guaranteed, cautioned M. Elizabeth Halloran, MD, an investigator in the other study and Professor of Biostatistics at Emory University in Atlanta. “Early intervention could at least slow the pandemic, helping to reduce morbidity until a well-matched vaccine could be produced,” she said.

AN EPIDEMIC FROM A SINGLE INFECTION

Dr. Ferguson’s simulation began with the infection of one resident in a rural Thailand village by an H5N1 variant transmissible between humans. The ability of containment measures to halt the spread of the virus was examined at different basic reproduction numbers, or R₀s, the average amount of secondary cases generated by a primary case.

The simulation included the serial interval, which is the time from the infection of an individual to the infection of his or her contacts. It also incorporated various demographic data such as national and regional population size and age, the numbers of households, schools, and major companies, distances between residences and school or work, and random contacts in the community. Thailand is one of the few places in Southeast Asia where such data are relatively easy to access, and experts believe that an influenza pandemic would originate in that region.

Assuming an R₀ of 1.5 and no containment strategies, the simulation showed that H5N1 would infect 33% of Thailand’s 65 million-person population; the epidemic would peak after about 150 days and would more or less have ended by day 200. At an R₀ of 1.8, the epidemic peaked around day 100 and infected about 50% of the population.

The outbreak was limited to fewer than 200 cases when the simulation included quarantine of affected areas and oseltamivir prophylaxis for everyone within 5 kilometers of an influenza case. However, containment would be unlikely with these strategies if the R₀ exceeded 1.8, the investigators found. “Although our analysis of past pandemics suggests that transmissibility will fall below this threshold, it is unlikely that sufficient data will exist to verify this before a containment policy has to be introduced,” they acknowledged.

Notably, antiviral prophylaxis in the simulation required about two million doses of oseltamivir. The investigators stressed that antivirals alone would be insufficient to contain a real avian influenza outbreak and that combining them with quarantine would be the only hope.

SOME INFECTED INDIVIDUALS WILL FLEE

In the study that Dr. Halloran helped to perform, an avian influenza pandemic was simulated in a computer model of 500,000 individuals distributed across an area of 5,625 square kilometers—the approximate population density of rural Southeast Asia. In addition to the same types of demographic factors and random contacts incorporated into Dr. Ferguson’s computer model, the simulation accounted for the possibility of some infected individuals leaving the area of the outbreak.

In scenarios that included prevaccination, it was assumed that vaccine recipients were immunized long enough before the influenza epidemic for immunity to develop. It was also assumed that the vaccine worked poorly, showing an efficacy for susceptibility of 0.30 and an efficacy for infectiousness of 0.50. “We consider an epidemic contained if there are fewer than 500 cases in the 500,000 person community,” the investigators said.

With no containment measures and an R0 of 1.7, there were 384 influenza cases per 1,000 persons (192,000 cases). Furthermore, 1,254 infected individuals fled the area.

However, the epidemic was well contained through prevaccination of 50% of the hypothetical study population, oseltamivir prophylaxis for 80% of the contacts of index cases, and quarantine of index cases and 70% of their contacts. The latter two measures were implemented 14 days after the appearance of the first influenza case.

At an R₀ of 1.7, this combined strategy reduced the number of influenza cases to 0.03 per 1,000 (15 cases). Only 275 courses of oseltamivir were required and essentially no one fled the area of the outbreak.

Less effective but still acceptable containment was achieved with various combinations of two of the three containment strategies. At an R0 of 1.7, for example, prevaccination and antiviral prophylaxis limited the number of cases to 80, oseltamivir courses to 1,338, and individuals who fled to essentially none.

Ira M. Longini Jr., PhD, Professor of Biostatistics at Emory University’s School of Public Health, was cautiously optimistic about averting a pandemic. “With fairly limited stock—maybe a million courses of oseltamivir—and a good prepared response, we should be able to contain a potential pandemic flu at the source, given that it is not too infectious.”

—Timothy Begany

Reference
1. Ferguson NM, Cummings DAT, Cauchemez S, et al. Strategies for containing an emerging influenza pandemic in Southeast Asia. Nature. August 3, 2005. E-pub ahead of print.
2. Longini IM Jr, Nizam A, Xu S, et al. Containing pandemic influenza at the source. Science. 2005;309:1083-1087.

Return to table of contents