The Library
Can routinely collected data be used to inform randomised controlled trial outcomes in oncology?
Tools
Mintz, Harriet Paige (2019) Can routinely collected data be used to inform randomised controlled trial outcomes in oncology? PhD thesis, University of Warwick.
|
PDF
WRAP_Theses_Mintz_2019.pdf - Submitted Version - Requires a PDF viewer. Download (11Mb) | Preview |
Official URL: http://webcat.warwick.ac.uk/record=b3492761~S15
Abstract
Introduction:
Randomised controlled trials (RCT) have supplemented standard data collection with routine healthcare data. However, no RCTs in the United Kingdom have been conducted solely using routine data in oncology or secondary care. This thesis was undertaken to assess methods to enable the replacement or supplementation of standard RCT data. I present examples of routine data follow-up in two clinical settings: prostate and bladder cancer.
Methods:
Routine healthcare datasets were validated against reference patient data (for example, trial data and clinical noting), for their ability to identify trial outcomes of interest. Models were developed to algorithmically identify these outcomes from the routine data. Outcomes included: toxicity (serious adverse events), disease progression, treatments and the last known follow-up interaction.
Results:
Models were developed enabling the identification of outcomes of interest from the routine data, for example sepsis admissions and trial non-survival endpoints, for example, progression. This enabled the estimation of uncollected trial case report form (CRF) events, which subsequently have become of interest. I developed a novel routine data-derived endpoint, which correlated with standard trial endpoints, enabling estimation of treatment effects from routine data. I also developed a method to validate the feasibility of using routine data as the basis for oncology trial follow-up.
Discussion:
The nature of the routine data meant that models had to be developed to enable identification of some events of interest indirectly. Although routine data quality was shown to be improving, techniques had to be implemented, for example, through data querying, to ensure integrity, accuracy and relevance. Routine data can provide a robust method of trial data collection but needs to be used in combination with other data sources, such as, standard trial data or clinical noting.
Conclusion:
I propose that routine data are a feasible source of trial outcomes; however, each individual outcome requires validation.
Item Type: | Thesis (PhD) | ||||
---|---|---|---|---|---|
Subjects: | R Medicine > R Medicine (General) R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer) |
||||
Library of Congress Subject Headings (LCSH): | Clinical trials -- Data processing, Clinical trials -- Mathematical models, Oncology -- Data processing, Oncology -- Mathematical models | ||||
Official Date: | September 2019 | ||||
Dates: |
|
||||
Institution: | University of Warwick | ||||
Theses Department: | Warwick Medical School | ||||
Thesis Type: | PhD | ||||
Publication Status: | Unpublished | ||||
Supervisor(s)/Advisor: | James, Nicholas (Nicholas D.) ; Parsons, Helen (Researcher in medicine) | ||||
Sponsors: | Warwick Medical School | ||||
Format of File: | |||||
Extent: | xxxiii, 307 leaves : illustrations (some colour) | ||||
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