The overall goal of this research is to improve disaster management though early detection of outbreaks of disease. We will contribute to this goal by conducting basic research that identifies over-the-counter (OTC) healthcare products whose sales correlate and precede clinical events such as hospital admissions. The results of the research will find immediate application in the National Retail Data Monitor, an existing system that monitors sales of OTC healthcare products that was developed by our laboratory. The proposed research will answer questions about which over-the counter (OTC) healthcare products to monitor for which types of diseases, and how large and sudden must an outbreak be to be detectable over background. This proposal has the following specific aims: 1. To discover new OTC healthcare product categories for use in public health surveillance by searching the very large space of OTC product categories for those that best correlate with a range of diseases and to refine the OTC product categories in current use 2. To measure the time when naturally occurring disease outbreaks are detected from sales of the best OTC healthcare product categories (as determined by product screening studies) relative to the time when they are detected from other types of surveillance data. 3. Determine the smallest increase in sales of the best OTC healthcare product categories (as determined by product screening studies) above baseline that we can detect by studying the ability of different outbreak detection algorithms to detect simulated outbreaks that we add to the sales data. 4. Determine whether what is known about self-treatment behavior is consistent with results of correlation studies by interviewing manufacturers of market-leading products in each category about their marketing research (Manufacturer Interviews) The research team has worked in this and closely related areas for the past four years and as a result have assembled a unique set of resources for this study. These resources include sets of retail data, sets of reference disease data, and analytic methods.

Agency
National Institute of Health (NIH)
Institute
National Library of Medicine (NLM)
Type
Exploratory/Developmental Grants (R21)
Project #
5R21LM008278-02
Application #
6888283
Study Section
Biomedical Library and Informatics Review Committee (BLR)
Program Officer
Florance, Valerie
Project Start
2004-07-01
Project End
2007-06-30
Budget Start
2005-07-01
Budget End
2006-06-30
Support Year
2
Fiscal Year
2005
Total Cost
$167,063
Indirect Cost
Name
University of Pittsburgh
Department
Internal Medicine/Medicine
Type
Schools of Medicine
DUNS #
004514360
City
Pittsburgh
State
PA
Country
United States
Zip Code
15213
Cami, Aurel; Wallstrom, Garrick L; Fowlkes, Ashley L et al. (2009) Mining aggregates of over-the-counter products for syndromic surveillance. Pattern Recognit Lett 30:255-266
Que, Jialan; Tsui, Fu-Chiang (2008) A multi-level spatial clustering algorithm for detection of disease outbreaks. AMIA Annu Symp Proc :611-5
Tsai, Ming-Chi; Tsui, Fu-Chiang; Wagner, Michael M (2007) An evaluation of biosurveillance grid--dynamic algorithm distribution across multiple computer nodes. AMIA Annu Symp Proc :746-50
Wallstrom, Garrick L; Hogan, William R (2007) Unsupervised clustering of over-the-counter healthcare products into product categories. J Biomed Inform 40:642-8
Hogan, William R; Wallstrom, Garrick L; Wagner, Michael M (2005) An evaluation of three policies for updating product categories in the National Retail Data Monitor. AMIA Annu Symp Proc :325-9
Wallstrom, Garrick L; Wagner, M; Hogan, W (2005) High-fidelity injection detectability experiments: a tool for evaluating syndromic surveillance systems. MMWR Morb Mortal Wkly Rep 54 Suppl:85-91
Li, Ran; Wallstrom, Garrick L; Hogan, William R (2005) A multivariate procedure for identifying correlations between diagnoses and over-the-counter products from historical datasets. AMIA Annu Symp Proc :450-4