Transparency data

Vulnerability Working Group - Notes 3rd Meeting (HTML)

Updated 29 July 2024

Attendees

• Firoze Salim (FS) - Central Digital Data Office Chair (CDDO)
• John Olatunji (JO) - Central Digital Data Office (CDDO)
• Emanuel Silva (ES) - Central Digital Data Office (CDDO)
• Nicholas Oughtibridge (NO) - NHS-England
• Paul Davidson (PD) - iStandUK (Tameside MBC)
• Shelley Heckman (SH) -  iNetwork (Tameside MBC)
• Mike Thacker (MT) - Place Group / Porism
• Patrick Kiernan (PK) - Central Digital Data Office (CDDO)
• Emily Sullivan (ES) - (DLUHC)
• Kate Cooper (KC) - Local Government Association (LGA)
• Kirsty Hendry (KH) - (DLUHC)
• George Russell (GR) -(DHSC)
• Jeanette Rycroft (JR) - Wigan Council
• Simon Roberts (SR) - Improvement Service (Scotland)
• David Wright (DW) - (DWP)
• Malcolm Davies (MD) - HM Treasury

Apologies

• Didac Fabregas-Badosa (DF-B) - CDDO
• Suzanne Fry - (SF)

Record of discussions

1. Welcome, introductions and agenda - Firoze Salim (FS) - CDDO

• FS welcomed the attendees and provided an overview of the agenda for the meeting

2. The concept model and use cases - Paul Davidson, IStand UK (facilitator) and WG members

• PD introduced the agenda item and provided background on the Department for Levelling Up, Housing and Communities-funded Scalable Approach to Vulnerability via Interoperability (SAVVI) project and its aims to create a scalable approach to identifying and supporting vulnerable individuals by using data from multiple agencies. PD outlined:
  • the common process followed by various vulnerability scenarios, which includes phases such as finding, assessing, supporting, reporting, and improving, all while ensuring the process is lawful, ethical, and transparent.
  • the concept model to standardise the terms and data used across these processes, facilitating interoperability and collaboration among different public sector entities.
  • key concepts from the model and the definitions and terms that are essential for telling the vulnerability story and linking data across various datasets.
  • the importance of a unified approach to data sharing and the potential for the SAVVI framework to help set government-wide standards.
  • the goal to expand the model to include common public sector datasets and improve the overall efficiency and effectiveness of data sharing for supporting cohorts of vulnerable people
• PD set out that the first “Modelling” workstream under the chair of NO had been considering the SAVVI concept model and had been looking at use cases to validate the model. The following 4 use cases were presented to the working group:.

Additional Notes:

• NO mentioned that they are working towards integrating these concepts with a higher ontology to make the data even more interoperable.

1) Wigan Council - Civil Contingencies - Jeanette Rycroft

• JR described the challenges faced by the civil contingencies team in responding to emergencies before the project began. During events like Storm Christopher in early 2021, there was a scramble to gather data, often resulting in chaotic and inefficient responses.
• The goal was to provide first responders with timely data to identify individuals needing extra assistance during emergencies, such as those with mobility or communication issues, or those posing potential risks to council staff.
• The new system automates data overnight, making it accessible only during emergencies. It allows staff to interact with maps, pinpoint areas, and generate lists of households needing intervention, significantly improving the response process.
• JR highlighted the support from the SAVVI team and the effective use of the SAVVI model and IG, calling it a game-changer for their work.
• PD thanked JR for her input and acknowledged that there was much more to discuss, encouraging further contact for more information.

2) DWP - sharing Universal Credit data - David Wright

• PD introduced the issue of DWP often being asked by councils for Universal Credit (UC) data to help find vulnerable people, prompting a search for a smarter solution.
• DW highlighted the need expressed by local authorities and lobby groups for data sharing to address issues such as child poverty and support for those not in education, employment, or training (NEET).
• DW discussed DWP’s ILOV (Identifying Local Vulnerability) project and the potential shift from point-to-point data sharing for operational purposes to broader, cross-cutting operations. This would enable early identification and intervention for various programs, such as supporting families and addressing homelessness.
• DW explained the ambition to share the entire UC database lawfully, starting with one or two use cases and expanding to include other data like pensions and disability benefits.
• DW emphasised the importance of complying with existing legislation and working towards changing it to lawfully facilitate their goals.
• PD praised the DWP’s efforts and highlighted the potential benefits of a standardised approach to data sharing across departments, using common taxonomies and APIs to simplify data integration and collaboration.
• The discussion underscored the need for a unified approach to data sharing to maximise its utility for identifying and supporting vulnerable individuals.

3) Scottish Councils - Rural Child Poverty - Simon Roberts

PD- Scottish Councils came together to think through how to find families with children who are struggling financially and their remit comes from the Scottish Child Poverty Act. Some of the data is their own, but they also want to access data from DWP and Social Security Scotland. So this work has focused on establishing if there are legal enablers that would give access to that data.

SR - highlighted the main issue in UK local government regarding the use of data, such as DWP data. Local authorities often approach data usage in an unclear manner, leading to rejection.

• The SAVVI framework has been utilised to bring together data, policy, and legal experts in local government to create clear data propositions. These propositions specify the data needed, legal gateways, and purposes for use within local government.
• SR mentioned they have written to the ICO in Scotland for approval of this approach and expect a response by the end of the week. The response will help in approaching entities like DWP and Social Security Scotland for data access.
• SR’s team manages unique property reference numbers and address data in Scotland. They aim to use this data to identify households in rural areas and provide targeted interventions.
• The joint approach in accessing and utilising data, such as early learning childcare, shows promise for improved data sharing practices in Scotland.

4) BOLD - Homelessness Prevention - Kirsty Hendry

• PD explained that councils have identified various data items that could help in identifying individuals at risk of homelessness, allowing for proactive support to prevent homelessness.
• KR discussed the complexity of risk factors for homelessness, noting that many risk factors are not readily available to local authorities.
• KR highlighted domestic violence as a significant risk factor for homelessness. Victims and perpetrators often need new housing, but the data must be handled sensitively to ensure victims’ safety.
• The conversation underscored the importance of making data accessible and shareable across local authorities, DWP, police, and justice services, while also ensuring it is done securely to prevent further risks to individuals.
• PD briefly introduced SAVVI’s logical model, which outlines how data and messages could look based on the concept model. This includes defining fields and values for key entities.
• PD mentioned that the next stage of work in the “Modelling” workstream would look to develop/agree the logical model and a common understanding of attributes and primitives to standardise data descriptions, like those of persons or places.
• NO provided feedback, suggesting that the definition of “person” should be changed to “individual” to avoid confusion and to ensure clear concept definitions.
• NO emphasised the importance of master data and reference data management, which are critical for the SAVVI project but fall under the remit of the Central Digital and Data Office (CDDO). NO also mentioned that the health sector has technology for master data and reference data management, suggesting its potential extension beyond health and adult social care.
• NO highlighted the necessity of a common method for sharing master data and reference data, and keeping roles and identifiers separate to avoid confusion.
  • PD responded by acknowledging the importance of using common identifiers, like the Unique Property Reference Number (UPRN), and discussed incorporating data matching in SAVVI.
  • FS agreed with NO’s points and indicated that key members of the working group are leading these efforts at CDDO.
• FS asked the group if (subject to the change requested by NO) they were content with the concept model and for it to be submitted to the Peer Review Group and DSA Steering Board for endorsement. The working group approved the concept model.

3. Next steps and deliverables - Firoze Salim (FS) , Emanuel Silva (ES) , Nicholas Oughtibridge, NHS,

Master Data and Reference Data:

• NO highlighted the critical need for master data and reference data sharing, emphasising it as a CDDO responsibility, not a SAVVI problem. The necessity for a common way of sharing and identifying data was discussed.

Concept Model and Logical Model:

• PD and NO discussed the importance of common terms and identifiers, such as UPRNs. NO suggested the definition of Person  be changed to refer to ‘Individual’ instead of ‘Person’ PD agreed that maintaining consistency in identifiers is essential for data matching and interoperability.

Taxonomy and Terminology:

• FS advised the group that SF would lead the 2nd workstream on ‘Terminology’ and that work would commence shortly. SF was unable to attend the meeting but set out initial thinking and a number of questions for the group to consider. The group emphasised the importance of having a common vocabulary, handling evolving vocabularies, and ensuring clarity in defining risk factors.
• ES mentioned the need for subgroups to handle different aspects, such as content and technical sharing.

Legal Interoperability:

• PD stressed the importance of understanding legal gateways for data reuse and sharing. Establishing a common process for information governance is crucial for both local and wider public sectors. FS agreed and advised that the 3rd ‘Enabling’ workstream would cover this important area but the workstream would start at a later date given the potential pressures on people’s time for the two workstreams.

AOB and close - Firoze Salim (FS)

Next Steps:

• FS outlined that NO, PD, SH, SF, ES and FS would be meeting to discuss next steps on the Modelling and Terminology workstreams and that another meeting of the working group would be arranged shortly. FS said that SF and EM would welcome any  feedback on the proposed taxonomy work in the meantime.

Comments and Suggestions:

• NO pointed out that maintaining updated vocabularies and handling semantic qualifications like certainty and negation are important.
• ES proposed forming subgroups focused on content and technical sharing to handle different aspects of the taxonomy development.

Wrap-Up and Thanks:

• FS thanked everyone for their contributions and committed to keep the group updated on the DSA approval for the concept model.

Additional Notes:

• The importance of ensuring that all terms are relevant to everyone and specific to no one was reiterated by PD.
• The need to establish a common understanding across local and public sectors regarding what data can be used for what purposes was highlighted.
• Suzanne Fry’s absence was noted, but her colleague ES provided insights on her behalf regarding the next steps for the taxonomy work stream.