In general practice, approximately 1% of people have intellectual disabilities (IDs) – a reduced ability to
understand new or complex information and to learn and apply new skills. Of these people, around two-thirds
have two or more long-term conditions (MLTCs). Physical ill-health symptoms in this population are often
mistakenly attributed to either a mental health/behavioural problem or as being inherent to the person’s ID
which contributes to health inequalities. A principal way in which support could be improved involves ‘care
coordination, which is defined as joined-up care between health and social services.
However, to date the lack of ability to understand and predict the complex interactions between MLTCs and
care needs of individuals makes it challenging to provide effective coordination of person-centred and holistic
care. In this study, we aim to describe the burden of MLTCs in subjects with IDs and develop methods able to
predict the development of MLTCs in subject with IDs.
Using data from CPRD GOLD, the presence of prevalent MLTCs at or before the diagnosis of the ID (the index date) and over time following the diagnosis of ID (incidence conditions) will be extracted from the records. Prevalence and incidence will be explored for all types of ID and detailed by type of ID. We will pre-process the data and this step includes procedures to characterise associations between variables (e.g. age, sex, geographical locality, genetic cause of intellectual disability, presence of behaviour disorder, autistic spectrum condition, socio-economic status, psychotropic medications) to form a sound understanding of the datasets.
Following the identification of the cohort, statistical analyses and artificial intelligence (AI) algorithms will be initially used to create clusters of MLTCs; statistical methods will allow determining associations of common risk factors in people with ID based on various sociodemographic, lifestyle and other clinical factors (e.g. multiple medications) as well as key events (outcomes) in patients with IDs (i.e., mortality or hospitalisation). AI algorithms will be then used to find significant comorbidity pairs (long-term condition trajectories of length two, e.g., Diabetes Mellitus → Heart failure), temporal patterns in long-term condition trajectories, and to develop tools able to population-level and individual trajectories of MLTCs in a subject with ID. To create the temporal patterns in long-term condition trajectories we will refine and extend existing time-analysis frameworks for large-scale comorbidity studies, where the history of patients are represented as time sequences of ordered events.
Thus, the intended public health benefit of this work includes: define of what clusters of MLTCs exist in people with Intellectual Disability (ID); understanding/predicting risk and trajectories of patients on the ID spectrum; developing interventions and NHS guidelines to inform patients and families, and practice.
The key outcome of this study is to evaluate the prevalence (before the index date, i.e. the date of the diagnosis of ID) and incidence (after index date) of MLTCs: outcomes will be modelled using AI algorithms. The study will also investigate the risk of hospitalisation, all-cause and cause-specific death, and outpatient visits (resource utilisation).
Gyuchan Thomas Jun - Chief Investigator - Loughborough University
Francesco Zaccardi - Corresponding Applicant - University of Leicester
Clare Gillies - Collaborator - University of Leicester
Georgina Cosma - Collaborator - Loughborough University
Kamlesh Khunti - Collaborator - University of Leicester
Navjot Kaur - Collaborator - University of Leicester
Safoora Gharibzadeh - Collaborator - University of Leicester
Satheesh Gangadharan - Collaborator - University of Leicester
Sharmin Shabnam - Collaborator - University of Leicester
Vasa Curcin - Collaborator - King's College London (KCL)
HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Patient Level Index of Multiple Deprivation