Diseases of ageing are increasingly important for our healthcare system and impact society as a whole. Two of the most important diseases in older age are dementia and cancer. For these conditions, factors such as age, poverty, ethnic background, previous education, and having other health conditions may influence how quickly patients get diagnosed, and how effective their care and treatment is. This may lead to worse outcomes in disadvantaged groups.
In this study, we aim to study dementia and cancer to:
1. Describe who is likely to get a delayed diagnosis, and identify the risk factors for delay in diagnosis, such as age, frailty, or having several other diseases (comorbidities).
2. Describe who is likely to get treatment or access to services which is below standard or does not meet national guidelines, and identify risk factors for under-treatment such as age, frailty or comorbidities.
3. Examine whether delayed or missed diagnosis, and inequalities in treatment affect outcomes such as quality and length of life.
4. Investigate the interaction between dementia and cancer, particularly in terms of how they affect recognition, management and outcomes of each other. As an example, memory symptoms may delay diagnosis of cancer and may hamper treatment plans (e.g. for pain), which will go on to have a further damaging impact on memory symptoms.
We hope that by identifying patterns and causes of delays and under-treatment we can contribute to improvements in health services that will enhance quality of life for our ageing population.
We plan a cohort study using primary care data, with linked hospital records and cancer registry data. We will study two key diseases of ageing for which delays in diagnosis and treatment in the elderly are known to be a problem. Cancer and dementia are both of increasing public health concern, affect mainly the elderly population, and improving the timely diagnosis of each has been targeted in the NHS long term plan.
In our cohort, we will identify cases of each condition, by triangulating evidence from 2-3 sources of data (primary care, HES, and cancer registry). We will identify evidence of delay in diagnosis for both conditions. We will identify evidence of under-treatment for cancer cases, and examine services used following diagnosis of dementia. We will then look at the consequences of delays, different types of service used and adequacy of care or treatment on outcomes including prognosis, quality of life and pathways of care.
We will assess all possible risk factors, social inequalities, and other known confounders and comorbidities as potential contributors to under-diagnosis and under-treatment. We are interested in factors which may reflect inequalities in society, such minority ethnic status, or multi-morbidity. Comorbidities which may impact on recognition and treatment include diabetes, respiratory disease, heart disease and frailty, while social factors include age, sex, ethnic background, local economic deprivation, smoking and drinking. We will examine the relationship between cancer and all types of dementia, to generate hypotheses about the mechanisms which underlie their negative association.
We will use conventional epidemiological methods such as logistic and Cox proportional hazards regression to identify key risk factors and patterns, and will employ data-driven learning methods such as LASSO penalised logistic regression and random forest algorithms to identify the best set of risk factors which explain the patterns in the data.
Objective 1: Cancer delay, stage and setting at diagnosis; Dementia delay, severity and health-care setting at diagnosis.
Objective 2: The adequacy of treatment for Cancer; any differences in treatments or support received for Dementia.
Objective 3: Cancer and Dementia outcomes e.g. quality of life, prognosis, pathways of care (all defined in section N).
Objective 4:
1) Incidence of dementia and subtypes of dementia following cancer
2) Incidence of cancer and specific types of cancer following dementia diagnosis
3) Delay to diagnosis as above
4) Adequacy of treatments or services provided, as above
5) Cancer and dementia prognosis, quality of life and pathways of care outcomes, as above.
Elizabeth Ford - Chief Investigator - Brighton and Sussex Medical School
Elizabeth Ford - Corresponding Applicant - Brighton and Sussex Medical School
Agnieszka Lemanska - Collaborator - University of Surrey
Agnieszka Michael - Collaborator - University of Surrey
Anjum Memon - Collaborator - Brighton and Sussex Medical School
Anotida Madzvamuse - Collaborator - University of Sussex
Bria Browne - Collaborator - Brighton and Sussex Medical School
Dorina Cadar - Collaborator - Brighton and Sussex Medical School
Gayasha Batheegama Gamarachchige - Collaborator - University of Surrey
Harm Van Marwijk - Collaborator - Brighton and Sussex Medical School
Imogen Rogers - Collaborator - Brighton and Sussex Medical School
Jackie Cassell - Collaborator - Brighton and Sussex Medical School
Louisa Gagatsis - Collaborator - Brighton and Sussex Medical School
Man-Chie Chow - Collaborator - Royal Surrey NHS Foundation Trust
Seb Oliver - Collaborator - University of Sussex
Stephen Bremner - Collaborator - Brighton and Sussex Medical School
Sube Banerjee - Collaborator - Brighton and Sussex Medical School
Agnieszka Lemanska - Collaborator - University of Surrey
Priyamvada Paudyal - Collaborator - Brighton and Sussex Medical School
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;Mental Health Services Data Set (MHSDS);NCRAS Cancer Patient Experience Survey (CPES) data;NCRAS Cancer Registration Data;NCRAS National Radiotherapy Dataset (RTDS) data;NCRAS Systemic Anti-Cancer Treatment (SACT) data;ONS Death Registration Data;Practice Level Index of Multiple Deprivation