When a death defies explanation

Even before last fall’s anthrax attacks, physicians and public health experts worried about the nation’s ability to identify and respond to outbreaks of infectious disease. In response to a 1992 report from the Institute of Medicine, the Centers for Disease Control and Prevention (CDC) had begun a surveillance strategy to detect new and unrecognized infectious diseases.

The project, called Surveillance for Unexplained Deaths and Critical Illnesses Due to Possibly Infectious Causes, grew out of two observations: new infectious diseases in this country were breaking out long before they were recognized, and the development of new molecular probes allowed infectious agents to be identified and characterized.

In 1995 the CDC, with partners at Yale and Stanford universities and in San Francisco, Minneapolis and Portland, Ore., began a population-based surveillance using diagnostic tests, death records and hospital discharge records to identify cases that would bear the label UNEX (for unexplained deaths and critical illnesses due to possibly infectious etiologies.) Preliminary results were published earlier this year in the journal Emerging Infectious Diseases.

“We found that in a population of 7.7 million, about 40 people are dying or becoming sick from unexplained illnesses each year,” said Andre N. Sofair, M.D., M.P.H., HS ’90, assistant professor of medicine and one of the authors of the study. “While this number might seem small, it is rather significant when it is put into perspective. Each year, in a well-served population, there are many young people who become critically ill or die without a diagnosis.”

The study, which is continuing, examined unexplained critical illnesses and deaths among people between the ages of 1 and 49 in the San Francisco Bay area, the state of Minnesota and New Haven County in Connecticut. It also looked for cases among people between the ages of 1 and 39 in Oregon. A UNEX case was defined as a previously healthy person who died or was hospitalized in an intensive care unit with a life-threatening illness bearing the hallmarks of an infectious disease for which no cause was identified. The study examined tissue or blood samples from 122 patients who died or became ill for unexplained reasons. The researchers divided patients into syndrome categories that represented their illness, such as a disease of lung or liver, and samples were tested accordingly, with each sample undergoing an average of 28 tests. No new infectious agents were discovered, but the cause of illness was determined in 28 percent of the patients tested.

This pilot study yielded numerous lessons, according to the authors. Future surveillance for UNEX, they found, could benefit from simplified case-finding methods, better quality of specimens and a more focused surveillance of specific syndromes. The authors believe their surveillance approaches will strengthen collaborations among clinicians, laboratory technicians and public health professionals and result in better detection of unexplained deaths and critical illnesses and better monitoring of emerging infectious diseases. “These preliminary findings are being used to direct programs to assist in bioterrorism preparedness and outbreak investigation,” said Sofair. “Having more sophisticated diagnostic testing would be helpful in finding a cause of death or illness.”