SYNDROMIC SURVEILLANCE SYSTEMS

1
SYNDROMIC SURVEILLANCESYSTEMS
Investigators Justin Chimka, PhD Heather
Nachtmann, PhD Manuel Rossetti,
PhD Assistants Patricia Cabrera, James
Burgmeier, Lauren Holloway
2
USAF PARTNERSHIP

• The project is related to Decision Support for
  Logistics Response to Chemical, Biological or
  Radiological Attacks managed by Edward Boyle,
  senior research psychologist and chief of the
  Logistics Readiness branch of the Air Force
  Research Laboratory (AFRL)
• AFRL contact is 1Lt Jean-Claude Beasley /
  Logistics Research Officer

3
INTRODUCTION

• The earliest detectable sign of a covert germ
  warfare attack may be unusual increases in the
  number of people seeking healthcare (Reis and
  Mandl 2003)
• Research objectives
• Compare and evaluate existing detection
  algorithms
• Develop new methods for improved (influenza-like)
  outbreak detection
• Formal evaluation of detection algorithms
  including enhancements with appropriate attention
  paid to distribution assumptions about relevant
  public health data

4
BROAD APROACH

• Identify existing algorithms for analysis
• Collect public health data
• Develop enhanced surveillance methods based on
  generalized linear models and statistical process
  control
• Evaluate and compare detection with respect to
  sensitivity and specificity

5
MOTIVATION

• The threat of terrorism and high profile disease
  outbreaks has drawn attention to public health
  surveillance systems for early detection of
  outbreaks (Buehler et al. 2004)
• In addition we should recognize that the
  potential of syndromic surveillance and better
  public health data systems in general might be
  greater for naturally occurring diseases than for
  bioterrorism (Soto et al. 2004)

6
MOTIVATION (continued)

• A better choice of data series and detection
  algorithms may be able to improve the balance
  between sensitivity and specificity and maintain
  timeliness (Soto et al. 2004)
• Considerable controversy remains about the best
  methods of data analysis and presentation to
  facilitate public health decision making based on
  surveillance data (Bravata et al. 2004)

7
ACTIVITIES

• Review relevant literature about bio terrorism
• Early statistical detection and monitoring of
  outbreaks
• Syndromic and public health situation awareness
• Bio surveillance and alert systems
• Methods
• Understand and enhance developed surveillance
  systems including current time-series methods
• Develop new models for syndromic surveillance
  based on weekly surveillance reports of Centers
  for Disease Control and Prevention

8
ACTIVITIES (continued)

• Results
• Test model detection capabilities of existing,
  enhanced and new surveillance systems
• Develop and/or identify additional assessment
  capabilities as needed
• Conclusions
• Report on important tools for syndromic
  surveillance
• Recommend alarm strategy based on existing,
  enhanced and/or new systems