CSTE logo
This website uses cookies to store information on your computer. Some of these cookies are used for visitor analysis, others are essential to making our site function properly and improve the user experience. By using this site, you consent to the placement of these cookies. Click Accept to consent and dismiss this message or Deny to leave this website. Read our Privacy Statement for more.
CSTE Features
Blog Home All Blogs
Search all posts for:   


View all (176) posts »

New Ways for Analyzing Surveillance Data

Posted By George Turabelidze, Thursday, December 11, 2014

As epidemiologists, we monitor disease incidence and use these data to detect outbreaks when the historical norm is exceeded. But the key issue is to identify what exactly the historical norm is. Often it can be difficult to call an outbreak when the numbers straddle the gray area between the historical norm and statistical variation. So, epidemiologists might need several more days to make a final determination of outbreak, and that affects timeliness and efficiency of the outbreak management.

On November 11, 2014, the Journal of Applied Mathematics published our new methodology that provides a more accurate and timelier solution to reportable conditions data analysis. Using mathematical models and Monte Carlo simulations, the article details how our methodology helps to find aberrations in the surveillance data. During the past year, real-life pilot testing has shown that the system performed well, identifying outbreaks such as E. coli and Shigella infections, pertussis, and Q fever.

Steven Rigdon, PhD; Ehsan Jahanpour, MS; George Turabelidze, MD, PhD
To create this methodology, we combed through databases and analyzed ten years’ worth of reportable conditions data in Missouri. With the help of the St. Louis University Biostatistics Department, we decided to apply trigonometric analysis to the data using software called R, a free program for statistical computing that is gaining widespread use in public health. R produces data analysis output and the Shiny software, an add-on package to R, provides graphic displays allowing for interactivity. With this versatile platform, our collective efforts have produced what we call DESTEM (Disease Electronic Surveillance with Trigonometric Models).
DESTEM offers epidemiologists a new analytic approach that not only produces a report, but allows data analysis from different “angles.” Users can customize their analysis according to specific disease, geographic region, calendar year, statistical parameters, etc. Users can set automatic communicable disease reports, which are comprehensible to non-technical audiences as they simplify results from the complex formulas underlying the tool. Most health department surveillance reports are merely descriptive; frustratingly, they do not qualify how abnormal the “abnormal” data is. By contrast, DESTEM shows red flags for data aberrations that go beyond what can be reasonably expected based on 10 years of historical data. Analysis is applied to aggregate data, and no personal identifiable information is housed on the servers.
Epidemiologists may be interested in using DESTEM as an analytical add-on to their existing surveillance systems. In Missouri, DESTEM is envisioned as an automated analytical complement to the current communicable disease surveillance system. We are working on making system accessible to the local health departments as well. It is anticipated that the full implementation of electronic lab reporting will greatly increase surveillance data coming to the health departments, thereby increasing need for analytical tools even further.

All CSTE members can use their accounts to access the demo of DESTEM software.
Click here to try the DISTEM demo

George Turabelidze, MD, PhD is a State Epidemiologist at the Missouri Department of Health and Senior Services. For more information about DESTEM, please visit the open-access article “Trigonometric Regression for Analysis of Public Health Surveillance Data.” You can also read more about CSTE’s work in surveillance and informatics.
Are you a member with an important message to tell the CSTE community? Tell us about it!
Do you use social media? Stay tuned to CSTE on Facebook and Twitter for daily updates!

This post has not been tagged.

Share |
Permalink | Comments (0)