UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Using Primary Care Electronic Health Records to Investigate Preclinical Dementia: Development of a Methodological Toolkit

Mackintosh, Maxine; (2024) Using Primary Care Electronic Health Records to Investigate Preclinical Dementia: Development of a Methodological Toolkit. Doctoral thesis (Ph.D), UCL (University College London). Green open access

[thumbnail of Mackintosh_10196335_Thesis_sig_removed.pdf]
Preview
Text
Mackintosh_10196335_Thesis_sig_removed.pdf

Download (12MB) | Preview

Abstract

Introduction: Dementia is a syndrome characterised by a deterioration in memory, behaviour and ability to carry out normal activities. Despite the substantial toll of dementia on patients and society, it is poorly recognised, diagnosed and managed. Timely diagnosis of dementia can result in better quality of life for patients and cost savings for the health service, however the current symptoms most associated with dementia are indicative of late-stage cognitive decline. Electronic health records (EHR) are a rich source of data which allow us to investigate conditions based on routinely-collected information from clinical care. Aim: We aimed to investigate how EHR data can be utilised to explore the earliest stages of cognitive decline, up to 20 years before diagnosis, by developing a methodological toolkit to characterise both cognitive and non-cognitive features of preclinical dementia. Methods: We analysed data from the Clinical Practice Research Datalink (CPRD); a UK primary care EHR database. Firstly, we phenotyped dementia cases and engineered potential preclinical features of dementia. Secondly, we conducted a descriptive analysis on the trajectories of preclinical features of dementia using incidence rates and temporal multivariable modelling. This analysis was then extended by applying common machine leaning methods for diagnostic purposes along the preclinical period. These were evaluated in terms of diagnostic accuracy, net benefit and economic viability. We then explored the application of graph theory and sequential pattern mining in order to identify influential and unique features and motifs of early dementia, which might be used under clinical decision support scenarios. Our final analysis within our emerging toolkit sought to apply novel statistical methods to evaluate real-world implications incentivising early detection of dementia at a national policy level. We compared a data-driven and hypothesis-driven approach to multiple changepoint detection to assess the impact of dementia policies on the recognition and treatment of dementia. Results: We identified 89,102 people with a diagnosis of dementia in CPRD and engineered 816 potential preclinical features. Our chapters demonstrated that there are substantive differences in the presentations and trajectories of potential preclinical features as hallmark symptoms of cognitive decline emerge. Whilst a definition of a dynamic and broad phenotype of preclinical dementia is still in its early stages, our analyses showed that preclinical dementia symptoms can be identified up to 20 years to diagnosis. There are however practical challenges in screening or incentivising recognition of dementia in its earliest stages. Conclusion: EHR can potentially enable novel insights into the preclinical period of dementia due to the number of dementia patients, range of preclinical features and length of pre-diagnostic period held within them. Our chapters validate the role known features of dementia play (cognitive and cardiovascular conditions) and surface novel potential features of dementia that are physiological, behavioural and administrative (health system utilisation). Our analyses demonstrate the strengths and limitations of applying a range of methodological approaches to heterogeneous conditions in novel datasets, and the breadth, concordance and stability of results across a range of methods.

Type: Thesis (Doctoral)
Qualification: Ph.D
Title: Using Primary Care Electronic Health Records to Investigate Preclinical Dementia: Development of a Methodological Toolkit
Open access status: An open access version is available from UCL Discovery
Language: English
Additional information: Copyright © The Author 2024. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request.
UCL classification: UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Population Health Sciences > Institute of Health Informatics
UCL
URI: https://discovery.ucl.ac.uk/id/eprint/10196335
Downloads since deposit
44Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item