Internet Explorer is no longer supported by Microsoft. To browse the NIHR site please use a modern, secure browser like Google Chrome, Mozilla Firefox, or Microsoft Edge.

Policy Research Programme - Detecting abuse in residential settings for people with learning disability and/or autism


10 May 2022


1.0 - May 2022


Timetable and Budget



Deadline for Stage 1 Applications

14 June 2022, 1 PM

Notification of outcome of Stage 1 Application

August 2022

Deadline for Stage 2 application

28 September 2022, 1 PM

Notification of Stage 2 Application

January 2023

Project Start

February 2023


Up to £700,000


This research specification provides details for applicants to the NIHR Policy Research Programme (PRP) call for a single research project to support detecting abuse in residential care settings for people with a learning disability and/or autistic people.


In 2019 the BBC Panorama programme uncovered abuse and mistreatment of people with learning disabilities and / or autism at Whorlton Hall, an independent hospital in County Durham [1]. Professor Glynis Murphy undertook an independent review of CQC’s regulation of Whorlton Hall between 2015 and 2019. This was designed to identify whether abuse could have been recognised earlier by CQC’s regulatory or inspection process and to make recommendations for how CQC could improve regulation of similar services in the future.

Professor Murphy produced two reports from the review, the second of which was published in December 2020 [2]. At the request of CQC’s Board this included a systematic review of the international research evidence in relation to the detection and prevention of abuse in services for adults with a learning disability and autistic people. The review revealed a distinct lack of recent or large-scale research into the subject matter. This research is intended to help address the gap identified in the review.

Research priorities

The work currently being commissioned is intended to be the first phase of research to identify better ways to predict and detect abuse in residential care settings for people with a learning disability and/or autistic people.

The overarching research questions (to be answered by this and subsequent research) are:

  1. What are possible risk factors for abuse in residential services for people with learning disabilities and autistic people, including what tips a poor culture into an abusive one?

  2. How can the current regulatory system be improved to more reliably:
  • identify where a poor culture might tip into an abusive one
  • detect abuse.

Central to answering these questions will involve identifying:

  • the wider factors that can lead abuse developing in a service (e.g. closed settings, small teams, unsuitable physical environment etc)

  • trigger points or flags that care is deteriorating and an abusive culture is at risk of developing.

This research project should mainly focus on data currently available (within CQC and possibly external to CQC) although may include some primary methods. This is intended to maximise exploitation of data that are already gathered before moving on to substantial new primary research, subsequent to this proposed project.

An important outcome of the research project will therefore be to specify key remaining uncertainties to be addressed by further primary research beyond and after the current project. This should include the identification of specific research questions and research specifications.

This research is intended to inform CQC’s regulatory approach, particularly in relation to Regulation 13: Safeguarding service users from abuse and improper treatment. It will provide essential insight for CQC’s regulation of services to ensure poor cultures are identified early so that action can be taken to protect people from harm before abuse occurs.

This research is also intended to support positive care models that are free of abuse, including in helping services to put appropriate safeguards in place themselves, informing commissioning decisions and oversight, and informing protective factors that might be put in place to help prevent abuse and reliance on restrictive practices.

Technical requirements

Two strands of work in the research project are expected to help understand possible risk factors for abuse in residential services, and through this identify areas CQC should consider as part of its monitoring and assessment of residential services for people with learning disabilities and autistic people.

The first strand should largely draw on quantitative methods, and the second on qualitative methods, but with some capacity needed to apply mixed methods across strands.

Strand 1. Analysis of CQC-held structured data

CQC holds a range of data sets and has developed sets of indicators (available for location of service, or provider or corporate brand):

  1. Information reported to CQC – including concerns raised by staff working in services, feedback from people’s experiences of care, information about events that providers are required to report to CQC (e.g., deaths of people using services, abuse and allegations of abuse, events that stop services), Mental Health Act complaints reported to CQC.

  2. CQC’s judgements of quality in registered providers – including ratings, enforcement actions and inspection reports (NB. some of this information is qualitative).

  3. Information from CQC’s data collections from providers – e.g., the provider information request completed by residential care homes which includes information about staffing levels.

  4. Indicators and metrics that CQC has developed from third party datasets – e.g., the mental health services data set, Skills for Care information about staffing in adult social care settings, location characteristics drawn from ONS datasets.

CQC has developed a new dashboard that pulls together many of these data sources and indicators, but this is not yet embedded (and as a consequence it has not yet been evaluated). This information could also be supplemented by additional indicators about care providers, including further staffing data and provider characteristics.

The exact analytical approach is not prescribed. There is some uncertainty about the most fruitful analytical approach and exploration of a number of possible options is expected.

This may include but should not be limited to:

  1. Examining datasets and indicators with episodes of abuse as the dependent variable and all available other variables as independent variables and potential candidate risk factors for abuse (e.g. using regression analysis). However, many cases of actual abuse are not recorded but some proxy dependent variables may be agreed; for example, repeated allegations or expressed staff concerns.

  2. Examining a wider set of indicators from the range of datasets identified above, against CQC judgements. One option may be to look at ratings under the Safe key question and its subsections. This would require qualitative extraction and a model for this would need to be developed (through topic modelling, or training terms for searches, etc). There is some uncertainty about this type of analysis and similar types of reviews undertaken by CQC previously have not proved fruitful. However, those have been of much broader data sets, and a more focused analysis such as this is yet to be tested and may be beneficial.

Strand 2. Analysis of un-structured data

Exploration of unstructured data from CQC regulation to help identify risk factors for abuse in residential services for people with learning disabilities and autistic people, including what tips a poor culture into an abusive one is expected. Desk based research should also be undertaken to identify other relevant data, such as any existing and available data from CCGs or local government departments of social care.

As with strand one, the exact analytical approach is not prescribed and should not be limited to what is described here. However, it may include both qualitative methods and quantitative analysis of un-structured data, such as:

Qualitative methods

A focused investigation by qualitative methods of a sample of CQC records of services, perhaps stratified by ratings. The focus may include consideration of factors that influence inspector decision making, what information is accessed and used, confidence in judging locations, inspector heuristics, emphases in reports and whether patterns of judgments emerge in relation to overall final ratings.

At the same time as seeking to understand how different regulatory interactions work and how judgments are made, other sources could be studied qualitatively. A likely beneficial option is to review formal enquiries such as Whorlton Hall (which generate substantial documentation and recommendations) and cases that have resulted in Safeguarding Adult Reviews. Reviewing services rated Inadequate may also be a suitable starting point.

Quantitative analysis of un-structured data

One approach may be to transform qualitative data from large samples of inspections into quantitative data by, for example sentiment analysis and machine learning. This may be based on areas that qualitative analyses have flagged up as potential areas for quantitative analysis.
Expertise required

Successful applicants will be able to demonstrate:

  1. A range of relevant mixed-methods expertise, including a track record in delivering analysis of complex, national level and large-scale data;

  2. Expert understanding of relevant data sources, tools and systems;

  3. Experience of undertaking research in social care, including broad content knowledge of relevant social care data and contexts

  4. Experience and expertise of working in sensitive areas of research

  5. Experience of involving people who use services, their families and representatives in the research design and seeking their views as part of research

  6. Experience in developing a range of research outputs, including for policy customers and service-facing findings.


Applicants are asked to consider the timing and nature of deliverables in their proposals to inform CQC regulation of residential services for people with learning disabilities and autistic people, as well as wider learning for services and commissioners.

Recommendations regarding, and specifications of, new primary research that may be needed to address outstanding uncertainties, with prioritisation of questions based on consultation with relevant stakeholders within CQC and beyond should be a defined output.

A meeting to discuss policy needs with CQC and DHSC officials will be convened as a matter of priority following contracting to inform outputs.

Budget and duration

Reflecting the requirements set out in this document, and our planning assumptions, a budget of up to £700,000 is available for this research. Outputs are ideally required by March 2024.

Costings can include up to 100% full economic costing (FEC) but should exclude output VAT. Applicants are advised that value for money is one of the key criteria that peer reviewers and commissioning panel members will assess applications against.

Management arrangements

A research advisory group including, but not limited to, representatives of CQC, DHSC, people who use services, and the successful applicants for the research should be established. The advisory group will provide guidance, meeting regularly over the lifetime of the research. The successful applicants should be prepared to review research objectives with the advisory group, and to share emerging findings on an ongoing basis. They will be expected to:

  1. Provide regular feedback on progress

  2. Produce timely reports to the advisory group

  3. Produce a final report for sign off

Research contractors will be expected to work with nominated officials in CQC, DHSC, and the NIHR Central Commissioning Facility. Key documents including, for example, research protocols, research instruments, reports and publications must be provided to DHSC and CQC in draft form allowing sufficient time for review.

The contract between the successful applicants and NIHR will require a data processing agreement (DPA) to be in place for the processing of personal data on behalf of the Care Quality Commission for the purpose of this research.

Under the agreement, the researchers will be given restricted access (as necessary for the purposes of the research) to personal data (including special category personal data) on CQC systems and using IT equipment provided by CQC. The researchers will not be permitted to extract, copy or share personal data outside of CQC systems without explicit written permission from CQC. Aggregated, anonymised and other non-personal data may be transferred to the researchers and held outside of CQC systems as necessary for the purposes of the research and subject to confidentiality agreements within the contract.

References and key documents

  1. Triggle (2019).Whorlton Hall: Hospital 'abused' vulnerable adults. [Accessed 10 April 2022].
  2. Murphy (2020). CQC inspections and regulation of Whorlton Hall: second independent report. [Accessed 10 April 2022].
  3. Health and Social Care Act 2008 (Regulated Activities) Regulations 2014: Regulation 13: Regulation 13: Safeguarding service users from abuse and improper treatment | Care Quality Commission. [Accessed April 2022]