In England, males and females with learning disabilities die more than a decade earlier than their counterparts in the general public. Among people with learning disabilities, cancer is the third leading cause of death (excluding COVID-19). It is unclear whether individuals with learning disabilities are more or less likely to develop cancer due to genetic or lifestyle factors. It has previously been shown that people with learning disabilities are less likely to participate in national cancer screening programmes, which aim to detect cancer early, increasing the chances for successful treatment. People with learning disabilities may also be less likely to consult their doctors about cancer symptoms until the disease has advanced, which could reduce treatment options and life expectancy.
This research project aims to explore the numbers of new cases of cancer each year in individuals with learning disabilities and their life expectancy following diagnosis. It also aims to explore uptake of vaccination against human papillomavirus (HPV), which causes several cancers. It will measure use of cancer screening tests and timeliness of cancer diagnoses. To investigate possible inequalities, we will make comparisons between those with and without learning disabilities, taking into account other important factors, such as age, sex, ethnicity and deprivation. We will conduct this study using a large dataset of general practice, hospital and death records in order to include a large number of people. This study will enhance our understanding of the impact of cancer on people with learning disabilities and help guide future healthcare strategies.
Life expectancies of males and females with learning disabilities are, respectively, 14 and 17 years lower than the general population. Cancer in this group is a leading cause of death but has not been well-studied. Differences may exist in exposure to cancer risk factors, use of cancer prevention measures (e.g. HPV vaccination) and use of cancer risk assessments (e.g. QCancer). Screening uptake may be lower and there may be inequalities in accessing diagnostic procedures and treatments.
This study will use CPRD Aurum, Hospital Episode Statistics (HES) and Office of National Statistics (ONS) mortality data to investigate cancer diagnoses and survival among individuals with learning disabilities. Incidence rates will be estimated for cancer overall and the 20 commonest UK cancers. Incidence rate ratios will be estimated using Poisson regression and survival using Kaplan-Meier and Cox regression. Cancer screening participation (i.e. for breast, colorectal, and cervical cancer, and prostate specific antigen [PSA] testing) will be assessed along with screening incidence rate ratios, relative risks of test positivity and times to diagnoses.
Emergency presentations of cancer will be examined by comparing dates of diagnoses in primary and secondary care records. Participation in the national HPV vaccination program will also be evaluated. The use and accuracy of primary care cancer risk assessment tools, particularly QCancer and QCancer-10yr, will be investigated using measures of calibration and discrimination. Random forest modelling, which has been used previously with CPRD data for risk prediction, will be assessed for its cancer risk prediction accuracy in this group.
This study has the potential to inform targeted interventions and healthcare strategies to address disparities in cancer incidence, survival, and screening among people with learning disabilities. Insights will enable stakeholders to develop and implement evidence-based approaches to improve cancer-related outcomes in this population.
Cancer incidence (for any cancer and the 20 commonest primary cancers); place of first presentation, overall survival; cancer specific survival; 1-year survival; 5-year survival; 10-year survival; HPV vaccination, cervical cancer screening performed; positive cervical cancer screening test result; cervical cancer following positive screening result; breast cancer screening performed; positive breast cancer screening test result; breast cancer diagnosis following positive screening result; bowel cancer screening performed; positive bowel cancer screening test result; bowel cancer diagnosis following positive screening test result; PSA test performed; positive (i.e. abnormal) prostate specific antigen test result; prostate cancer following positive (i.e. abnormal) prostate specific antigen test result; recording of QCancer and QCancer-10y risk scores; predicted risk of current and future cancer by QCancer and QCancer-10y risk scores; predicted risk of current and future cancer by random forest model; observed current and future cancer risk.
Darren Ashcroft - Chief Investigator - University of Manchester
Oliver Kennedy - Corresponding Applicant - University of Manchester
Alison Wright - Collaborator - University of Manchester
Louise Gorman - Collaborator - University of Manchester
Paul Lorigan - Collaborator - University of Manchester
Tjeerd van Staa - Collaborator - University of Manchester
Umesh Chauhan - Collaborator - University Of Central Lancashire
Alex Trafford - Collaborator - University of Manchester
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation;CPRD Aurum Ethnicity Record