Accounting for seasonal patterns in syndromic surveillance data for outbreak detection
2006

Impact of Seasonal Patterns on Outbreak Detection in Syndromic Surveillance

Sample size: 1000 publication Evidence: moderate

Author Information

Author(s): Burr Tom, Graves Todd, Klamann Richard, Michalak Sarah, Picard Richard, Hengartner Nicolas

Primary Institution: Los Alamos National Laboratory

Hypothesis

The study aims to evaluate the impact of inconsistent seasonal effects on performance assessments in detecting anomalous counts in syndromic surveillance data.

Conclusion

The assumption that 'one season fits all' is violated, leading to optimistic performance claims for some forecasting methods.

Supporting Evidence

  • The study found that moving average methods performed well on real data.
  • Detection probabilities were generally lower for real data compared to data simulated from the nonhierarchical model.
  • The hierarchical model provided more realistic estimates of outbreak detection probabilities.

Takeaway

The study shows that different years can have different patterns in disease outbreaks, and using a one-size-fits-all approach can lead to mistakes in detecting these outbreaks.

Methodology

Synthetic outbreaks were injected into real and simulated data to evaluate detection probabilities using various forecasting methods.

Potential Biases

The nonhierarchical model may lead to optimistic performance assessments compared to real data.

Limitations

The study is based on one data set and one type of simulated outbreak, which may not generalize to other contexts.

Participant Demographics

Data collected from the Emergency Center of the University Hospital, Albuquerque NM.

Statistical Information

Confidence Interval

± 2

Digital Object Identifier (DOI)

10.1186/1472-6947-6-40

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication