Thesis

Development and testing of early detection systems for seasonal and pandemic influenza

Downloadable Content

Download PDF
Creator
Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2019
Thesis identifier
  • T15240
Person Identifier (Local)
  • 201454560
Qualification Level
Qualification Name
Department, School or Faculty
Abstract
  • Medical and statistical studies and research have shown that respiratory viruses are important and dangerous to human life.Daily data were studied for Acute Respiratory Illness (ARI) and influenza-like illness (ILI)consultations for surveillance for influenza from all 14 health boards (HBs) in Scotland.The National Health Service (NHS) provided the data from 2009 to 2014.The weekly case ratio (WCR) method, developed for pandemic detection, may be a useful way of detecting seasonal influenza. The target is to find a simple way to extend the WCR method and compare it to other well established systems. This method is based on two terms, the value of WCR defined as the total influenza rates reported to all GPs in week w divided by the total rates in week w - 1 and NHB defined as the number of HBs which have a WCR > 1. We use daily data for ILI consultations. The starting point is using Scotland data to investigate how effective the WCR method would be for Scotland data.We then extend this situation, through simulation based upon the Scottish data to have more than 14 HBs with the same structure as Scotland.The next step is to develop the WCR algorithm for smaller spatial areas. We created another data structure using 30, 40 and 50 HBs from the original Scottish data, using different population size structures, then simulated more than 3,000,000 cases of ILI,considering the rate as at during the year, then we got the joint distribution for WCR and NHB in the case of no epidemic.We attempted to identify a rejection region for a test using the null joint distribution between WCR and NHB. The modified WCR system exceeds performance of other systems in some circumstances.
Advisor / supervisor
  • Robertson, Chris
  • Gray, Alison
Resource Type
DOI
Date Created
  • 2018
Former identifier
  • 9912718190002996

Relations

Items

Thumbnail Title Date Uploaded Visibility Actions
file details PDF of Thesis T15240 2021-07-02 Public Download