Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Drug sales data analysis for outbreak detection of influenza-like illness in China

Zhang, Yufei (2018) Drug sales data analysis for outbreak detection of influenza-like illness in China. Master Essay, University of Pittsburgh.

[img] Microsoft Word
Submitted Version

Download (1MB)


Data analysis is increasingly being used in syndromic surveillance system. This paper studies the drug sales analysis because of its public health relevance to outbreak ILI disease detection. Influenza-like illness (ILI) currently has a particularly high public health significance in China because the 2009 swine flu and the 2013 H7N9 outbreaks caused hundreds of deaths. Drug sales analysis can use real-time drug sales data to provide early detection of potential flu outbreaks. One example of syndromic surveillance system is the use of OTC medication sales in estimating annual influenza epidemics. Moreover, there is no established syndromic surveillance system that leverages drug sales data in China.
This essay aims to investigate the correlation between influenza surveillance data and drug sales data to help public health department make a timely outbreak detection in China. Specifically, this study collected and compared OTC drug sales data from a drug retailing pharmacy chain in Danyang, China with ILI trends surveillance system data from local CDC department. The collected ILI drugs categories includes medications for the treatment or control of cough, cold, and other respiratory ailments both in children and adults. The incidence rate of ILI or influenza was provided by the local CDC. To compare weekly lagged sales of selected drugs with ILI% data from 2015 to 2017, we computed Pearson's correlation coefficients. The correlation coefficient between the ILI % and drug sales data from 2016 Oct to 2017 Mar was 0.848 (p<0.05) which showed that the drug sales data can be used as a source to predict outbreaks. Moreover, we observed that peaks for drug sales appear 1-2 weeks before the local ILI% peak.
In conclusion, we observed a strong correlation between sales of influenza drugs and the number of influenza-like illness visits.
Public health significance/relevance. These results suggest that the analysis of OTC medication sales can potentially predict disease outbreaks, which will allow for advanced preparation by the local health authorities. With increasing concern over influenza-like disease outbreaks, public health is being required to have early outbreak detection systems in China.


Social Networking:
Share |


Item Type: Other Thesis, Dissertation, or Long Paper (Master Essay)
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Zhang, Yufeiyuz118@pitt.eduyuz118
ContributionContributors NameEmailPitt UsernameORCID
Committee MemberBarron, Geraldgbarron@pitt.eduGBARRON0000-0002-2771-3671
Committee MemberHernandez, Inmaculadainh3@pitt.eduUNSPECIFIEDUNSPECIFIED
Date: 3 April 2018
Date Type: Submission
Number of Pages: 39
Institution: University of Pittsburgh
Schools and Programs: School of Public Health > Health Policy & Management
Degree: MPH - Master of Public Health
Thesis Type: Master Essay
Refereed: Yes
Date Deposited: 19 Aug 2019 21:38
Last Modified: 19 Aug 2019 21:43


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item