Investigating links between socio-environmental factors and multimorbidity patterns in patients with severe mental illness
Multimorbidity, the co-existence of two or more conditions, is more prevalent in socio-economically disadvantaged people irrespective of age and there is an urgent need for research that takes into account socio-environmental factors. The syndemics approach consider the social context contribution to the clustering and interaction of chronic health conditions at individual and population level. The interplay of physical health and severe mental illness with broader socio-environmental factors remains unclear, within which the intersection between factors related to socioeconomic position and other markers of marginalized status such as homelessness, unemployment and racial/ ethnic minority status need further investigation. This project aims to detangle these complex inter-relationships in order to address an important gap in research. In the present PhD studentship, the student will for the first time be able to tackle this important issue through using ,electronic health records from a large secondary mental health provider linked to census data at the individual-level to understand the social risks for onset and outcomes in multimorbidities in people with mental illnesses. This linkage is important as detailed measures of social/ economic circumstances alongside race/ ethnicity are incompletely recorded or missing in standard electronic health records. The studentship will allow the applicant to learn about and develop statistical and machine learning models to cluster individuals by their patterns of multimorbidity and examine their association these key socio-environmental factors. The studentship will give the applicant skills in working with large-scale linked records and would suit a candidate with a good postgraduate degree in epidemiology, demography, computer science or statistics.
Novelty and Importance:
From a clinical point of view, this project would allow to identify individuals at higher risk of disability and mortality from one of Europes largest secondary mental healthcare providers, South London & Maudsley Trust (SLaM), which could allow the development of targeted interventions earlier in time. This studentship will also have a basis in a novel data linkage- to our knowledge one of the first of its kind, bring mental health together with detailed social/ demographic information. From a methodological perspective, tools developed could be directly relevant and applicable to electronic health records collected by CRIS SLaM and other CRIS resources.
To understand the relationships between the most common patterns of multimorbidity and key socio-environmental factors and investigate their association with disability and mortality.
Planned research methods and training provided: BHI training and KCL early career training opportunities. Specific training on NLP and predictive statistical modelling (including machine learning techniques). The candidate will also be able to attend advanced statistical training as programme.
Objectives / project plan:
- Year 1 (Objective 1): Brief literature reviews. CRIS data retrieval and preparation (including NLP techniques). Validation work of NLP outputs. Begin data cleaning and analysis. Ethical/ access approvals. Upgrade at 9 months.
- Year 2 (Objective 2/3): Data analysis to identify the most common patterns of multimorbidity and associations with socio-environmental factors. Paper submissions.
- Year 3 (Objective 3): Data analysis to identify individuals at higher risk of disability or mortality. Thesis and BRC report on potential clinical implications write up.
Scientific themes : 1) Learning from Big Data for Health. This project uses large, distributed, heterogeneous data sources such as CRIS EHRs to address major public health challenges such as multimorbidity.
All access to CRIS and its Linkage has been approved.
Socio-Environmental Factors; Multimorbidity: Severe Mental Illess, Electronic Health Records.