Computer-based detection of central nervous system syndromes for public health surveillance.

The early detection of infectious disease outbreaks is key to their management and initiation of mitigation strategies. This is true whether the disease is naturally occurring or due to intentional release as an act of terrorism. In recent times, this has become evident with the anthrax bioterrorism attacks of October 2001, the occurrence of emerging infections such as West Nile Virus and Severe Acute Respiratory Syndrome of the concern for a new pandemic of influenza based on H5N1 avian influenza. Public health surveillance efforts at the University of Utah have been place for several years and came to the forefront during the 2002 Winter Olympic Games. At that time, an electronic medical record-based system was developed and deployed to perform daily surveillance of patients visiting the clinics and emergency department of the University of Utah Health Care System. This effort was then followed by a detailed validation of the computer rules used in the surveillance system, with special emphasis on the early detection of central nervous system (CNS) syndromes such as meningitis and encephalitis. These syndromes are of importance to both emerging infections such as West Nile Virus and for NIH/CDC Category B threat agents such as Eastern and Western Equine Encephalitis. True CNS syndromes caused by infectious agents represent a small proportion of patients seen at the emergency department of a large tertiary hospital. "Reason for visit" chief complaint data were poor predictors for the early detection of CNS syndromes. Orders and early results from the laboratory testing of cerebro-spinal fluid were useful for the early detection of meningitis and encephalitis. Overall, computer-based surveillance methods have a role to play in the early detection of infectious diseases. In particular, this project has contributed to public health surveillance by moving the field beyond complaint data and has shown the validity of suing computer-based rules for the detection of meningitis and encephalitis.