Predicting and detecting abuse in registered care homes and supported living for people with a learning disability and/or autistic people (CQC LD AP Data Study)
Final report
Chester, H., Martineau, S., Manthorpe, J., Hickman, B., Cooper, V., & Yuan, Z. (2024) Predicting and detecting abuse in registered care homes and supported living for people with a learning disability and autistic people. London: NIHR Policy Research Unit in Health and Social Care Workforce, The Policy Institute, King's College London. https://doi.org/10.18742/pub01-166
Easy Read Summary Report
Chester, H., Martineau, S., Manthorpe, J., Hickman, B., Cooper, V., & Yuan, Z. (2024) A study about making the most of Care Quality Commission data (Predicting and detecting abuse in registered care homes and supported living for people with a learning disability and autistic people): Easy Read Summary Report. London: NIHR Policy Research Unit in Health and Social Care Workforce, The Policy Institute, King's College London. https://doi.org/10.18742/pub01-168
KCL news item about the publication of these reports (9 May 2024).
Background
The abuse of people with a learning disability and autistic people sadly features in health and care services despite legal safeguards, as highlighted by an independent review of the Care Quality Commission’s (CQC) regulation of Whorlton Hall, an independent hospital in County Durham. This review explored whether abuse could have been recognised earlier by CQC’s regulatory or inspection process and made recommendations about how CQC could improve its regulation of similar services and the use of data that it holds and collects.
Most people with learning disabilities in England live in their own homes, but some live in care homes, where personal care and accommodation are provided together. Some also live in supported living where they live in their own home and receive care and support. This may be individual or shared with others needing support. Care homes and supported living providers who provide personal care have to be registered, meet set standards and are inspected by the regulator, the CQC. The CQC collects a lot of information about these services and visits them to talk to residents, their families and staff.
In this study funded by the National Institute for Health and Care Research, we want to find out how information collected by the CQC could be used to detect and predict (and therefore prevent) abuse of people with a learning disability and autistic people living in care homes and supported living environments.
Aims and research questions
This study aims to inform decisions about the feasibility of future research to focus on the use of data to predict and detect abuse in registered care homes and supported living settings for people with a learning disability and autistic people. The research questions are:
- What are the main issues/concerns and priorities for stakeholders, public and the DHSC and CQC relating to use of data in the prevention and detection of abuse in these settings?
- What data and intelligence are currently held within CQC that could be used and accessed by the research team to produce measures of abuse and early warning signs/flags that might be used in the prevention and detection of abuse?
- How might Natural Language Processing (for example sentiment analysis) be used to analyse CQC data to identify risk factors for abuse?
- Based on findings from 1-3 and the policy and CQC context, what should plans be for phase 2 of the project?
These will be addressed in the four work packages below.
Methods
There are four work packages:
- Work Package 1: Consultation with a broad range of stakeholders. This will focus on the main concerns and priorities for stakeholders, public and the DHSC/CQC relating to use of data in the prevention and detection of abuse in registered care homes and supported living settings to inform subsequent work packages.
- Work Package 2: Mapping the extent, range and nature of data available within the CQC that could be used and accessed by the research team to produce measures of abuse or early warning signs/flags that could be used in the prevention and detection of abuse. This will include an assessment of the size and scope of each dataset, format, completeness, potential for linkage and exploration of data preparation and access issues.
- Work Package 3: This will explore how Natural Language Processing (NLP) and Machine Learning (ML) (for example sentiment analysis) could potentially be used to analyse unstructured CQC data to identify risk factors for abuse. We will work with colleagues from CQC to establish what data is available, its quality and the data preparation required to use a suitable NLP and ML process.
- Work Package 4: Findings from work packages 1-3 will be synthesised and presented to stakeholders and public contributors in two sensemaking exercises. They will be asked for their views on the findings and reflections on the research process to inform publications from the study and plans for future research.
Public Involvement
Public contributors will be involved from the start and throughout the study through two specially convened public involvement groups: people with learning disabilities and autistic people; and family carers. They will help us understand the main issues that are important to them, to make sense of the study’s findings and to develop the future research study.
Timescale
1 January 2023 – 31 December 2023
Funding
NIHR
Ethical approval
LRS/DP-22/23-35181
Expected outcome and impact
This 12-month study is research to see if this is something that needs a larger future research project. A research summary and a peer-reviewed publication will be produced as well as plans for future research.
ORCID ID NUMBERS
Principal Investigator
Dr Helen Chester: https://orcid.org/0000-0001-6587-6618
Team members
Professor Jill Manthorpe: https://orcid.org/0000-0001-9006-1410
Stephen Martineau: https://orcid.org/0000-0002-3562-8290
Dr Zheng Yuan: https://orcid.org/0000-0003-2406-1708
