SARS Model Shows Virus Has Potential To Spread Widely In The Absence Of Effective Public Health Interventions

May 23, 2003

BOSTON (Harvard School of Public Health) -- Using a computer model to quantify and predict the spread of severe acute respiratory syndrome (SARS), researchers at Harvard School of Public Health have determined that the virus has a real potential to spread widely but can be effectively controlled-- even without a vaccine or drug treatment-- through public health measures including quarantine, isolation of sick patients and stringent protection of hospital personnel.

The results appear in an expedited release by Science magazine at www.sciencemag.org or http://www.sciencemag.org/cgi/rapidpdf/1086616v1.pdf

that is free to the public. The paper is entitled “Transmission Dynamics and Control of Severe Acute Respiratory Syndrome.”

Using detailed epidemiologic data from Singapore and the epidemic curves from other settings, the researchers, led by Marc Lipsitch and Megan Murray, both assistant professors of epidemiology at HSPH, estimated that a single infectious case of SARS will infect about three secondary cases in a population that has not yet instituted control measures.

SARS has spread widely since November, 2002 and probable cases have been reported in 31 countries, with extensive ongoing transmission in Taiwan and China and major outbreaks that are now under control in Singapore, Vietnam and Canada. The World Health Organization today also lifted its travel advisory to Hong Kong and Guangdong Province in China, where local transmission has been controlled, leaving transmission only in hospital settings.

The cause of SARS appears to be a new kind of coronavirus.

“Our study shows in a quantitative way that SARS could have spread very widely and still could,” said Murray. “It has the potential to do so, and we need to take stringent control measures, quickly and effectively.”

“It’s very close to unprecedented to have stopped an outbreak of a respiratory disease without any biological interventions such as a vaccine or drugs,” said Lipsitch. “While we have a lot of uncertainty that SARS can be completely controlled, this model gives us reasonable hope.”

While highly infectious, SARS is less infectious than other, common respiratory diseases such as measles where a single case might generate 15 to 17 other cases; a case of chicken pox would result in 6 to 8 secondary cases. Smallpox, which was eradicated in 1977, was just slightly less infectious than chicken pox. A disease would have to be kept to an infection rate of less than 1 to be eventually eliminated.

Still, said Lipsitch, SARS, with a rate of 3 infections, would be expected to create a very large number of cases in the absence of any control methods. “We still need to know in the big outbreaks in China and Taiwan a person-by-person accounting of the transmission circumstances,” he said. “And in order to make control more effective, we need to find out more precisely when people become infectious, how that infectiousness changes over time and when they stop being infectious.”

Other uncertain aspects of SARS crucial to knowing if the virus might return include whether the virus proves to be seasonal; whether it exists in an animal reservoir from which it could be reintroduced; and whether people who recover from SARS retain any immunity. With other coronaviruses, including some kinds of the common cold, people acquire short term, temporary immunity.

The researchers analyzed the first 205 probable cases of SARS reported in Singapore to obtain relevant epidemiologic parameters. They saw a rapid decline in secondary cases in the second week of the outbreak there, coinciding with the application of control measures, including isolation of SARS cases and quarantine of their asymptomatic contacts. Ironically, because control measures were rapidly applied in Singapore, the researchers found too few data there to provide a reliable estimate of the infectiousness of SARS prior to the institution of control measures. They therefore estimated the rate of secondary infections from the rate of exponential growth in the number of cases in several other settings.

The researchers conclude that an achievable combination of control measures can contain SARS as demonstrated in Canada, Singapore, Vietnam and, most recently, Hong Kong and Guangdong. But, they warn that “On the other hand, in the absence of such effective measures, SARS has the potential to spread very widely.”

Other contributors to the article included James Robins, Professor of Epidemiology at HSPH, and researchers from the Ministry of Health, Singapore, McMaster Hospital in Hamilton, Ont., the City of Hamilton Public Health and Community Service Department, the University of Utah and Massachusetts General Hospital.