A pioneering report published today by a collaborative team based in Population Data Science at Swansea University explores the barriers of linking police data with health data and sets out recommendations for future work.
Why is linking this data important?
Government legislation (the 1998 Crime and Disorder Act) requires Police, local government, and the NHS to collaborate on joint crime reduction strategies that include data sharing to inform targeted responses.
However, data sharing is not widely adopted. Cross-organisational information sharing is vital as date linkage can provide insights and a much broader picture, leading to more efficient decision-making.
Linking police data to healthcare data is not straightforward. Researchers found three main challenges.
- Data in police systems are often written in a narrative format containing highly-sensitive information. Although text data is rich descriptively, anonymising and analysing these data is challenging.
- Police services in different parts of the country use other software systems, making harmonising data difficult.
- Linking entire police datasets with healthcare data has not yet been attempted.
Assessing the benefits of linking data
The work supported by the National Centre for Population Health & Wellbeing Research and ADR Wales and funded by the NIHR Unlocking Data Grant looked to assess whether the benefits of linking police and healthcare data outweigh the effort required and any risks involved.
As part of the work, the research team carried out a pilot study to determine:
- Whether text mining, which uses artificial intelligence to transform the written text into structured data, could be used to code the text data fields in police data systems.
- The views and opinions of police staff on using a single system and sharing data with other agencies.
- What learning was possible from linking police data with health data.
The Pilot study
The pilot study explored using text mining methods to extract police data sources, enabling textual data to be transformed into a coded format that could be subsequently anonymised and shared.
The team conducted interviews with 36 individuals from the four police services in Wales. Interviewees felt that it would be a positive development if all forces moved to the same system. The main barrier prohibiting sharing data is a lack of knowledge on what could be shared. However, this issue could be overcome by a framework endorsed at a high level agreeing on what data can be shared.
The research team undertook a case study linking Public Protection Notification (PPN) data with GP and hospital data. A PPN is an information-sharing document that allows police officers to record safeguarding concerns. The case study looked at the data of 8,709 people.
Recommendations for future work
The findings from the pilot study were presented at two workshops, which had representation from academia, Welsh and English police services, Welsh Government, the Home Office, and the third sector. Attendees concluded recommendations and discussed the next steps, which included:
- That future direction should focus on quick wins and show the benefits of data sharing.
- There needs to be consultation with the public to ensure public trust in any data sharing proposed.
- Future work should be undertaken to agree on what information should be shared, the agreed standards for this data and in what format it should be collected, and how data can be shared while meeting data protection regulations.
Professor Sinead Brophy, Lead researcher on the report, said:
“This work demonstrated that the main barriers to sharing data are not physical barriers. Linking data can significantly help improve outcomes and assist the Police and NHS in directing and targeting resources. This report is the first step in driving forward this program of work, and as a Centre, we look forward to continuing to work with the Police, health services and other partners to achieve this.”
This study is funded by the NIHR Public Health Research programme (NIHR133680/NIHR Unlocking Data to Inform Public Health Policy and Practice Grant). The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care.
The National Centre for Population Health and Wellbeing Research and ADR Wales are part of the 13 Centres of Excellence based in Population Data Science at Swansea University Medical School.