The Library
The analysis of clustered data in public health and healthcare research
Tools
Ng, Siu Woon Edmond (2021) The analysis of clustered data in public health and healthcare research. PhD thesis, University of Warwick.
|
PDF
WRAP_Theses_Ng_2021_Redacted.pdf - Submitted Version - Requires a PDF viewer. Download (9Mb) | Preview |
Official URL: http://webcat.warwick.ac.uk/record=b3747943
Abstract
Clustered data arise when data are sampled from clusters of individuals randomized to different interventions in a cluster randomised trial (CRT) or from groups of individuals in a population such as primary sampling units in a multistage clustered survey. Examples include: trials to compare different approaches for managing the wellbeing of elderly people clustered by general practices in the community; surveys on family planning or access to clean water where respondents are geographically clustered; and cost effectiveness studies conducted alongside CRTs.1-3
Although individual randomised trials and simple random sampling are statistically more efficient, CRTs and complex survey designs are pragmatic, and sometimes necessary alternatives in the presence of logistical, financial, or ethical constraints, or a risk of intervention contamination across different arms in a clinical trial. However, the dependence of data from the same cluster violates the critical independence assumption on which most standard statistical methods rely. Such dependence must be accounted for appropriately to enable valid inference.
In this thesis of published work, I draw upon seven publications to highlight the challenges in analysing clustered data, demonstrate the application of a variety of appropriate statistical methods, and show my contribution to developing statistical methodologies for clustered data. They are (i and ii) the analysis of a large CRT, (iii) estimation of intraclass correlation coefficients, [i-iii use data from the same CRT in geriatric research in the UK], (iv) the cost effectiveness analysis (CEA) of a CRT of a non-clinical intervention to reduce caesarean rate, and (v to vii) three pieces of methodological work to improve and extend current methods and models for analysing clustered data.
Item Type: | Thesis (PhD) | ||||
---|---|---|---|---|---|
Subjects: | Q Science > QA Mathematics R Medicine > R Medicine (General) R Medicine > RA Public aspects of medicine |
||||
Library of Congress Subject Headings (LCSH): | Cost effectiveness, Multilevel models (Statistics), Group-randomized trials, Clinical trials, Family Planning -- Statistical methods, Older people -- Health and hygiene -- Research, Coefficient of concordance, Monte Carlo method, Medical economics -- Statistical methods | ||||
Official Date: | October 2021 | ||||
Dates: |
|
||||
Institution: | University of Warwick | ||||
Theses Department: | Warwick Medical School | ||||
Thesis Type: | PhD | ||||
Publication Status: | Unpublished | ||||
Supervisor(s)/Advisor: | Stallard, Nigel ; Parsons, Nick | ||||
Format of File: | |||||
Extent: | various pagings : illustrations | ||||
Language: | eng |
Request changes or add full text files to a record
Repository staff actions (login required)
View Item |
Downloads
Downloads per month over past year