Displaying results 1 - 3 of 3
-
The Predictive Accuracy of Non-Regression Data Analysis Methods
Content Type: Abstract
Analysis of time series data requires accurate calculation of a predicted value. Non-regression methods such as the Early Aberration Reporting System CuSum are computationally simple, but most do not adjust for day of week… read more… days were analyzed at 75 (56 VA and 19 DoD) Texas and 26 (19 VA and 17 DoD) Virginia facilities. The mean observed … <.01 for severe injury/death to 21.0 for respiratory. For all facilities and syndromes, the mean observed daily count … -
Performance Characteristics of Control Chart Detection Methods
Content Type: Abstract
To recognize outbreaks so that early interventions can be applied, BioSense uses a modification of the EARS C2 method, stratifying days used to calculate the expected value by weekend vs weekday, and including a rate-based method… read more… syndrome/days with the standard deviation (SD) ≥0.5 by all methods were included. For calculating expected values, … vs rate methods, the latter accounting for total visits (i.e., both visits assigned and not assigned to a syndrome); … -
Evaluation of Spatial Estimation Methods for Cluster Detection
Content Type: Abstract
CDC’s BioSense system provides near-real time situational awareness for public health monitoring through analysis of electronic health data. Determination of anomalous spatial and temporal disease clusters is a crucial part of… read more… data, we then applied our program to determine 1-day (i.e., spatial) clusters significant at a p-value threshold …

