merging infections the World Health Organization (WHO) describes them as part of the "unfinished agenda of communicable diseases" for public health. The latest example of their importance was the explosion of severe acute respiratory syndrome (SARS) on the global scene. SARS, which we now know originated in Guangdong province of China in late 2002, was found early this year to have spread to Hong Kong, Hanoi, Singapore, Taiwan and Toronto, and later to 32 countries around the world, including the U.S. Inevitably, some key questions arise from the outbreak of any new infectious disease that kills human beings. how serious is the outbreak? Can it become an epidemic? How far will it spread and how rapidly? Can it be contained, and if so, what interventions have the best chance of working?
It is the nature of epidemics to be unpredictable. We have learned a great deal from epidemiologic study of past epidemics. In particular, the 1918 influenza pandemic, which killed 20-40 million people worldwide, and the measles outbreaks of the last century, which wiped out a high percentage of the population of Hawaii, taught us that infectious diseases that are spread by the respiratory route can be devastating and that SARS had to be taken very seriously.
The critical step in controlling an epidemic is to stop transmission. What do we need to know in order effectively to block transmission of a respiratory infection? The key is provided by a fundamental epidemiologic parameter called the 'basic reproductive number,' or Ro, which measures the potential for the spread of a disease. Formally, Ro is defined as the expected number of secondary infectious cases generated by an average infectious case in an entirely susceptible population. It not only tells epidemiologists the potential of the outbreak to spread in the absence of interventions but also allows them to predict the ability of control measures to reduce transmission. If Ro is greater than 1, each patient will on average infect more than one individual, and thus an epidemic will be propagated. If interventions can reduce the reproductive number to less than 1, the epidemic will be reduced and hopefully dissipate.
The School is privileged to have two brilliant young infectious disease epidemiologists, Marc Lipsitch and Megan Murray, who do very sophisticated modeling of communicable infections. Marc's areas of interest include the epidemiology of pneumococcal infections, and the factors influencing the spread of antibiotic resistance in a range of infectious diseases. These factors range from human behavior, such as prescribing antibiotics, to the biology of infectious agents, so Marc's research extends from analysis and modeling of epidemiologic data to laboratory work to determine whether drug resistance genes and mutations put the bacteria carrying them at a disadvantage.
During the course of her research, Megan Murray developed a complex mathematical model for predicting the transmission of tuberculosis, a major respiratory infectious disease that still causes eight million new cases and more than two million deaths each year. Together with the Mitchell L. Dong and Robin LaFoley Dong Professor of Epidemiology and Biostatistics Jamie Robins, some very bright students and fellows, and an alumnus in a SARS-affected country who was able to provide information on the spread of the infection in real time, Marc and Megan decided that the epidemic was sufficiently menacing that they should dedicate themselves to trying to solve the key epidemiologic question: estimating the Ro. They recognized that some of the larger and more established scientific groups (with greater access to the data in Hong Kong, for example) would be working on the same question, but the challenge of taking on an urgent problem and testing the predictive value of models that constituted their own research proved too strong to pass up. In the three weeks that followed, as I can as dean attest, they worked flat out, getting almost no sleep, to analyze the fragmentary and constantly changing data emerging from the epidemic. They came up with an Ro of 2-4. In itself that result told them a lot. It indicated that SARS was not the most infectious disease ever seen, since measles has an R0 of about 15 and tuberculosis an R0 of about 10. But it was in the same league as smallpox, and the Ro told them that in the absence of appropriate interventions, SARS would infect millions of people within six months. On the other hand, because the R0 was sufficiently close to 1, by taking the proper steps, it should be possible to reduce the reproductive number below that limit, and the epidemic would be controlled. In fact, when modeling possible interventions, they determined that the most effective means of blocking transmission would be to isolate infected individuals and quarantine people likely to be exposed. Indeed, just those strategies, along with preventing transmission in hospital facilities, would be responsible for allowing every affected country to contain the epidemic, at least for now.
Marc and Megan's group feverishly wrote up their data and conclusions, hoping that they would be published rapidly to help public health officials in affected countries plan their programs for containing the epidemic. Having myself been asked to write an editorial for Science on the Lessons from SARS for inclusion in the issue containing the genome sequences of the corona-virus that causes the disease, I had the opportunity to ask a senior editor whether she would consider publishing an epidemiologic modeling paper on the subject in a subsequent issue. The answer was yes, but with the caveat that the completed manuscript had to be received within four days.
Our dauntless junior faculty put their noses to the grindstone to submit the manuscript on schedule. They, in turn, received the reviewer's comments in record time and were told that if they included additional data from Singapore and submitted a revised version within days, there was a good chance it could be published. In a wonderful example of scientific collaboration, colleagues from across the globe, whom Marc and Megan had never met, agreed to contribute their Singapore data to the paper. But that meant that all the calculations had to be redone. After yet another stretch of sleepless nights, they submitted their final version. The paper was not only accepted but, because of the urgency of the situation, it appeared just two days later in Science Express, the Web version of Science that is used to publish important articles before they can appear in printed form. Their data were published together with those of the great English epidemiologist Roy Anderson and his group. Because the results, obtained completely independently, were so similar, the world learned the key epidemiologic principles about the SARS epidemic in a breathtakingly short time. Our wonderful young scientists took on a global challenge and with intellectual brilliance, extraordinary dedication, and great spirit made a great contribution to the world. Now, I hope they will find a little time to get some sleep and visit their families.
Their thinking also had a huge impact on Harvard University. In an epidemiologic sense, a university, with 18,000 students living in very close proximity, can be seen as an epidemic waiting to happen. From the very earliest notice from WHO that SARS was a global problem, Harvard President Larry Summers and Provost Steven Hyman were concerned about the potential threat of SARS to the University and how to prevent its introduction into our community. I was privileged to be able to work with them and with the Director of University Health Services, Dr. David Rosenthal, on a SARS team that considered what advice on student and faculty travel and what precautions on contact with visitors from SARS-affected countries would be the most prudent and least disruptive to protect the University. With information that the School of Public Health contributed to the decision making, the team created and revised its policies on almost a daily basis to keep up with the changing epidemiologic picture. We were most fortunate to have had no cases. Despite the fact that a number of individuals inevitably were inconvenienced, the University community was extraordinarily cooperative in working together to protect the health of everyone.
I am enormously proud of our young epidemiologists for taking on one of the most urgent and challenging public health problems facing the world. I should mention that they had no financial support whatsoever to do so, but felt it to be an obligation to use their knowledge to inform public policy in an urgent circumstance. It was also gratifying that they were able to elicit the cooperation and collaboration of international colleagues who had some of the basic data required to analyze and model the epidemic. In fact, one of the great lessons of the SARS epidemic has been the value of scientific networks in addressing the problem of emergent infectious diseases. In the 1980s, it took two years to identify HIV as the agent that caused AIDS and two or more years to characterize its genome sequence. In the case of SARS, WHO created a global network of 13 laboratories in ten countries, which identified the coronavirus as the cause of SARS in two weeks and sequenced the genome in another two--a truly extraordinary achievement.
I believe we can learn two other lessons from SARS. The first is that one cannot make public policy or reasoned allocation of scarce health resources without adequate and accurate information. While the economic impacts of outbreaks are serious for any country, epidemics are not contained by suppressing information--the price of trying to hide outbreaks is to allow them to become epidemics with far greater costs. In the future, I hope any country that suspects an outbreak of an emerging infectious disease will seek help from WHO and the Centers for Disease Control and Prevention (CDC) and enlist the best scientists in the world as early as possible to contain it and to provide early warning to other countries.
I think the final lesson that SARS has taught us is that infectious diseases do not respect national boundaries. The security of the United States increasingly depends on expertise from around the world to increase our ability to identify potential health threats and address those threats locally. Investment in improving health infrastructures around the world, training epidemiologists and strengthening laboratories, and supporting WHO more generously would be great first steps to creating a global health network. And, I believe that such an investment by this country would, as I concluded in my Science editorial, "protect our country and every other against global epidemics, save millions of lives, and help to change the U.S. image from self-interest to human interest."
page is maintained by Development Communications in the Office of Resource
To contact us with suggestions, comments, and questions, please e-mail: email@example.com
Copyright, 2005, President and Fellows of Harvard College