Since the dawn of our collective memory, plague and contagious disease has been a frightening, unseen, implacable enemy. It is not surprising that efforts to understand contagious illness extend back nearly as far as the written word. Hippocrates, the father of the western medical tradition, observed patterns of illness associated with temporal, geographic and environmental conditions. The Persian physician Avicenna, in the Eleventh Century, further speculated on the nature of contagion and disease spread, proposed the use of syndrome definitions for diagnosing diseases, and introduced the practice of quarantining. As the bubonic plague ravaged Europe and the Middle East in the 14th century, Islamic scholars advanced theories of tiny, unseen entities spreading the illness through contact, and by the 16th century, the germ theory of disease was taking root in Western Europe. John Graunt’s analysis of mortality rolls in the aftermath of the Great Plague of 1666 pioneered the use of statistical analysis for understanding the behavior of contagious disease. For most modern epidemiologists, the turning point was Dr. John Snow’s analysis of the Broad Street Cholera outbreak of 1854, which combined statistical and spatial analysis, allowing for the localization of the source of infection and management of the outbreak. The lesson became clear—the sooner an outbreak was identified and localized the greater the probability to contain and lessen the effects of the outbreak.
At the beginning of the 21st century, the shadow of a pandemic still casts fear across our modern generation. Several factors have conspired to increase the potential destructive power of epidemics. Both human and animal populations are densely packed in artificial environments. Pathogens have developed resistance to established treatments, and the Anthrax attacks of 2001 have illustrated that deliberate propagation of an infectious disease is a real threat in the modern age. The geographic barriers that have provided protection in the past have been bypassed. The geographical isolation between plants and animals has been gradually broken by the intentional or natural transport of organisms caused by human travel, tourism or trade. Foods and materials we use may come from anywhere around the globe. The acceleration of transportation has increased the speed and reach of a disease. Steam powered transport made it possible for the 1918 Influenza Pandemic to cross oceans within weeks. As evidenced by the SARS outbreak of 2003, an infection may now travel from continent to continent in a matter of hours. A Pathogen does not respect geographic or national boundaries. It is a global threat that all nations of the world need to face.
Although technology has contributed to the speed and distance which disease can spread, it also provides the key for meeting current and future epidemic threats. One of the most exciting innovations in public health is the development of automated biosurveillance systems that can churn through vast amounts of health-related data to support early identification, situation awareness, and response management for epidemiologists and public health officials. The aim of this book is to capture the story of these modern-day pioneers who are walking in John Snow’s footsteps, and to portray the current state of the art in biosurveillance, where some of the most promising aspects of modern information technology can be applied to this age-old challenge of combating the spread of disease and illness.
Despite the excitement that has greeted the initial generation of biosurveillance systems, several difficult challenges remain. As this book illustrates, biosurveillance operates in a complex, multidimensional problem space, and can be viewed from many different perspectives. Perhaps no single factor will affect the design of a biosurveillance system more than the data it uses. There is a tension between the most reliable sources of data (such as confirmed laboratory results) and the syndromic approaches which offer an earlier opportunity for detection (Emergency Department (ED) chief complaints, or over the counter sales). Animal data, vital statistics, and environmental data may identify the type of endemic relationships that Hippocrates first documented. Collecting and aggregating data from multiple sites raises issues of data granularity, quality, latency, and underlying meaning. Experience in developing detection algorithms has shown this to be a formidable undertaking, given the effort needed to obtain or create data sets for calibration, as well as tuning algorithms to reduce the false alarm rate.
Experience from fielding the first generation of biosurveillance systems, has shown a host of operational issues that need to be addressed for a successful system. One of the most frequently noted shortcomings was that the role of humans (epidemiologists, public health officials, and the system developers themselves), was not well defined as these systems primarily focused on automation. How were alerts to be validated and communicated? How could the system better support multiple organizations at the local, state/province, national, and even global level? What elements from the “Information Storm” are most important to convey accurate situation awareness at these different levels? How can such a system support different communities of interest; such as respiratory or foodborne illnesses, and different types of users? How easily can the system incorporate new data sources, and identify new types of threats? Practical considerations, particularly cost, are also important. Some existing systems are proving too costly to operate and maintain, let alone expand to incorporate more data sources. Quantifying the added value provided by a biosurveillance system will be necessary for increasing public investments to support them and making widespread deployment of these systems a reality.
Innovators today are pushing the envelope as to what we would consider a biosurveillance system. Some are taking advantage of new technologies such as web-crawling, remote sensing, and SMS texting to provide flexible, lower cost input channels for surveillance. Others are adopting a “system of systems” approach, where a given system will need to reason based on its own raw data as well as conclusions from other automated systems. The scope of these systems is also expanding from early detection to support the entire outbreak and management cycle. For Dr. Snow, this was easy, as he only needed to remove the handle of the water pump that was dispensing the contaminated water. For all of us in the global health community, this is a challenge we dare not ignore.
I have tried to organize this book practically, following natural sequence from theory to application. The initial chapters of the book build a foundation of knowledge for the reader. Later chapters contain more applied case studies written by experts from the field relating their practical experience within their professional domain. I would like to thank all of the authors in this volume for not only sharing their insights, but also their candid appraisals of lessons learned and unresolved issues that will help chart the future of biosurveillance. I would also like to thank the many dedicated experts and pioneers in the public health community who, although they may not be listed as authors in this collection, made indispensable contributions to our understanding of biosurveillance. I am also grateful to the many organizations and companies which have supported my involvement with biosurveillance over the years, including the US Centers for Disease Control and Prevention (CDC), Innovative Support to Emergencies, Diseases and Disasters (InSTEDD), the Rockefeller Foundation and Google.org.+Taha Kass-Hout
[Taha A. Kass-Hout, MD, MS] (with contributions from Walton J. Page, Jr.)https://books.google.com/books?id=WVvo3npf_EoC&lpg=PP1&dq=biosurveillance&pg=PP1#v=onepage&q&f=false