National Bioterrorism Syndromic Surveillance: Daily Syndrome Counts by Zip Code

A demonstration program extracts nightly counts of patient encounters corresponding to specific syndromes from electronic ambulatory records. Daily syndrome counts aggregated by zip code are statistically analyzed for unusual clustering using a model-adjusted SaTScan approach. The system, developed by W. Katherine Yih of Harvard Pilgrim Health Care, is designed to detect localized outbreaks and facilitate rapid public health response.

Use Cases

• Detecting unusual disease clusters based on daily syndrome counts by zip code mentioned in the description
• Evaluating statistical aberration thresholds for automated public health alerts as described
• Simulating and testing outbreak detection capabilities for naturally occurring or terrorism-related events
• Analyzing the utility of electronic health record data for syndromic surveillance

Strengths

• Data is updated nightly, providing near real-time surveillance capability
• The system's software is described as mostly open-source and compatible with commonly used systems
• The methodology excludes repeat patient encounters to avoid double-counting
• The system is flexible, allowing changes in participating organizations, syndrome definitions, and alert thresholds

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment
• Column-level documentation is absent; field semantics must be inferred after download
• Patient-level information is not shared centrally, limiting the granularity of analysis available to external researchers

Provenance

Source

Harvard Pilgrim Health Care, authored by W. Katherine Yih.

Collection Method

Data extracted nightly from electronic ambulatory patient records of participating health plans and practice groups.

Freshness

Data are extracted nightly, but the last update date for the dataset itself is unknown.

Geography

Data is aggregated by zip code, suggesting coverage of participating healthcare organizations' service areas.