One in four adults in the UK experiences two or more long term health problems (such as stroke, depression or hearing loss), a condition known as multimorbidity. Most stroke survivors have multimorbidity, including those who survived a mini-stroke, a condition similar to stroke but short lasting. International and national organisations such as the NHS agree that we need to design health services that are better arranged to help patients with multimorbidity. This project will describe how multimorbidity occurs among stroke/ mini-stroke survivors in order to help design better health services for them.
We intend to look at anonymised routinely collected health records from 100,000 UK stroke/ mini-stroke survivors. We will distinguish patients with similar groups of long-term health problems and describe how health outcomes, such as number of deaths or hospital admission, differ among these groups. We will also examine the association of individual characteristics including age, sex and ethnicity with multimorbidity.
This study will support researchers to develop better treatments for people who have had a stroke/ mini-stroke and also suffer from additional long-term health problems. Our findings will also help health workers and policy makers to identify and prioritise the most important needs of different groups of stroke/ mini-stroke survivors.
Multimorbidity accounts for 25% of the adult UK population, and over half of hospital admissions and GP appointments. The Academy of Medical Sciences, the NHS in the Long-Term Plan and other organisations have reported multimorbidity as a priority for global health research and call for investigation into patterns of multimorbidity, and for greater understanding of the burden and determinants of common clusters of conditions. Multimorbidity is particularly prevalent in stroke/transient ischaemic attack (TIA; often referred to as mini-stroke) survivors (~94%). However, despite the high prevalence of multimorbidity in stroke/TIA patients, no studies have investigated how multimorbidity is structured within the stroke/TIA survivor population.
Using the Clinical Practice Research Datalink (CPRD) GOLD database, which provides clinical primary care data from UK general practices, this research aims:
1. To classify common clusters of multimorbidity within the stroke/TIA survivor population.
2. To describe how health outcomes differ among different stroke-multimorbidity clusters.
3. To describe how determinants of health differ among different stroke/ TIA-multimorbidity clusters.
In a cross-sectional approach, we will focus on adults surviving a stroke/ TIA. Utilizing a 3-year follow up period we will estimate death, primary care consultation and hospital admission rates. We will link the CPRD GOLD data with area socioeconomic deprivation data and Hospital Episodes Statistics. Multimorbidity is defined as having two or more of 36 long-term conditions. A model-based person-centred cluster analysis, the latent class analysis (LCA) calculating the class membership probabilities among patients will be used. The number of clusters will be defined through statistical criteria (Bayesian Information Criteria (BIC), sample-sized adjusted BIC, log-likelihood ratio test, entropy for classification quality) and clinical input. We will describe the clusters in terms of comorbidities and identify key characteristics and health outcomes (mortality, hospital admission, primary care appointments) associated with them.
We will identify the highest need and commonest clusters so that these can be prioritised in service development.
Multimorbidity defined as having two or more of 36 long-term conditions recorded in patients' medical records (1); number of primary care consultations; number of prescriptions a patient has been issued; number of hospital admissions; mortality.
Efthalia (Lina) Massou - Chief Investigator - University of Cambridge
Efthalia (Lina) Massou - Corresponding Applicant - University of Cambridge
Duncan Edwards - Collaborator - University of Cambridge
Jonathan Mant - Collaborator - University of Cambridge
Steven Kiddle - Collaborator - AstraZeneca Ltd - UK Headquarters
Yajing Zhu - Collaborator - University of Cambridge
Zhirong Yang - Collaborator - University of Cambridge
HES Admitted Patient Care;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation