Statistical disclosure control: Applications in healthcare

Miller, James (2005) Statistical disclosure control: Applications in healthcare. MSc(R) thesis, University of Glasgow.

Full text available as:
[thumbnail of 10753993.pdf] PDF
Download (4MB)

Abstract

Statistical disclosure control is a progressive subject which offers techniques with which tables of data intended for public release can be protected from the threat of disclosure. In this sense disclosure will usually mean information on an individual subject being revealed by the release of a table. The techniques used centre around detecting potential disclosure in a table and then removing this disclosure by somehow adjusting the original table. This thesis has been produced in conjunction with Information and Services Division (Scotland) (ISD) and therefore will concentrate on the applications of statistical disclosure control in the field of healthcare with particular reference to the problems encountered by ISD. The thesis predominately aims to give an overview of current statistical disclosure control techniques. It will investigate how these techniques would work in the ISD scenario and will ultimately aim to provide ISD with advice on how they should proceed in any future update of their statistical disclosure control policy. Chapter 1 introduces statistical disclosure and investigates some of the legal and social issues associated with the field. It also provides information on the techniques which are used by other organisations worldwide. Further there is an introduction to both the ISD scenario and a leading computing package in the area, Tau-Argus. Chapter 2 gives an overview of the techniques currently used in statistical disclosure control. This overview includes technical justification for the techniques along with the advantages and disadvantages associated with using each technique. Chapter 3 provides a decision rule approach to the selection of disclosure control techniques described in Chapter 2 and much of Chapter 3 revolves around a description of the implications derived from the choices made. Chapter 4 presents the results from an application of statistical disclosure control techniques to a real ISD data set concerned with diabetes in children in Scotland. The results include a quantification of the information lost in the table when the disclosure control technique is applied. The investigation concentrated on two and three- dimensional tables and the analysis was carried out using the Tau-Argus computing package. Chapter 5 concludes by providing a summary of the main findings of the thesis and providing recommendations based on these findings. There is also a discussion of potential further study which may be useful to ISD as they attempt to update their statistical disclosure control policy.

Item Type: Thesis (MSc(R))
Qualification Level: Masters
Additional Information: Adviser: E Marian Scott
Keywords: Information technology, Health care management
Date of Award: 2005
Depositing User: Enlighten Team
Unique ID: glathesis:2005-74083
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 23 Sep 2019 15:33
Last Modified: 23 Sep 2019 15:33
URI: https://theses.gla.ac.uk/id/eprint/74083

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year