Website of the Department of Health

For further information, or if you are interested in helping to take the work forward in any way, contact:

Roger Staton,
PSS Information Development,
Department of Health
Tel: 020 7972 5790
Email: Roger.Staton@doh.gsi.gov.uk

Purpose

This document provides an update of progress on work to define a core data set to reflect the information needs relating to the provision of social care services and outlines the next stages.

Background

A proposal for a project to define a core data set to capture the information needs relating to social care was put to the Personal Social Services Information Strategy Steering Committee within the Department of Health last October. Key extracts from the proposal are attached at Annex 2. Approval was given for the work to go ahead and this paper provides an update on progress so far.

The Core Data project is one of three projects looking to map social care information from different perspectives. The other two are the Quality Protects MIS project and the Management Information for Social Care (MISC) project. The QPMIS project has focused specifically on children's services and has looked in detail at the processes involved at the local level in the provision of care services for children and the information required to support those processes. The MISC project is again looking at the information issues from a local social services perspective but this time focusing on the information required by senior managers to "run the business". Both projects aim to come up with a statement of the information requirements.

Progress on the core data project

The Core Data project has taken as the starting point the information supplied by social services authorities to the Department of Health on the current annual statistical returns together with proposed new returns.

The work focused initially on the returns relating to children's services, providing an overview of the children's data submitted to the Department to link with the main QPMIS project. The model developed on the basis of the children's returns was 'validated' with the statistical section handling the returns and subsequently with one local authority - Norfolk. Minor changes were made to the model as a result. Encouragingly, the work on the QPMIS, carried out independently by Parity Consulting, has come up with a model that is almost identical. Since that model has been developed from a local perspective, there are one or two additional aspects to it but the core elements are the same.

The statistical returns relating to adult services, staffing and finance were subsequently analysed and the model updated. Very little change was necessary to the core model derived from children's returns.

During the course of this project, the team has identified some work done elsewhere around modelling of social care information. Of particular relevance is that of the Australian Institute of Health and Welfare (AIHW). They have produced a model which is very similar to our model, with 'party' (which includes both persons and agencies/service providers) and 'event' as the key components. (National Community Services Information Model Version 1; National Community Services Data Dictionary, both prepared by the National Information Development Unit, AIHW). Their work is further advanced than ours in that they have also started on the development of a data dictionary for community services. We have also looked at similar work done for the government in Hong Kong.

The basic model

The high level data model resulting from the above activity is attached as Annex 1. In essence, the model comprises three core elements - person, event and service/service provider - each of which has a number of associated data items and attributes. All the basic information collected centrally through statistical returns can be mapped against this structure.

Further details on different aspects of the model are given in Annex 1.

Next steps

The modelling work so far has in many ways only scratched the surface. It has enabled us to produce a high level representation of the data currently collected but to be of any use, the work needs to be taken to a lower level of detail, with all the data items and attributes identified and defined.

An important aspect of the work will be to look at the links with health care information and existing health care data models. This will be particularly important in relation to the development of information strategies planned as part of the implementation of the National Service Frameworks for mental health and older people.

One possible way forward would be to focus the further development of the model on either mental health or older people, so as to address health/social care interface issues and at the same time meet the emerging needs of the mental health and older peoples information strategies.

The eventual goal should be to produce a list of data items, clearly and precisely defined, which could form the basis for a social care data dictionary, the main purposes of which would be to:

March 2000

ANNEX 1

CORE DATA PROJECT - HIGH LEVEL DATA MODEL

The diagram on the following page is a high level representation of the data currently collected on the annual statistical returns and the information which underpins that data. For example, some of the information on "person" is not actually collected centrally but is required to be held locally in order to derive the information which is sent to the Department.

The diagram identifies the key components of information and presents them in the form of a logical data structure, showing the links between different components. It is not meant to show how the information should be stored physically. Further work would need to be done to turn this into a physical data model which could be implemented as part of an IT system. In fact, there could be any number of different versions of the physical model reflecting the different hardware and software configurations and local practice.

The model identifies some data items within the main entities but further work is necessary to complete the list. This applies particularly to the "event" entity. Some work has begun on the children's side to flesh out these details.

The model explained

Boxes represent the main entities and attributes. Lines represent relationships, the dotted part indicating an optional relationship. Relationships are shown as one to many (indicated by the 'crows feet').  Person can be - a client, an informal carer, a care worker, a relative, a service provider. Event can be - a referral, an assessment, start of service, home care visit, exam results, change in legal status.

ANNEX 2

DEFINING A SOCIAL CARE CORE DATA SET

(Extract from Project proposal put to the PSSISSC)

Purpose

1. This paper sets out proposals for a project to define a core data set to capture the information needs relating to social care. Once defined and implemented, this core data set would enable local authority social services departments to meet local and central information needs.

Background

2. Local authority social service departments are currently bombarded with requests from central government for a range of different information, which are met in a variety of ways. At a local level, senior managers and elected members are becoming increasingly interested in information and its importance for the effective management of services. There are increasing demands for performance measures that can be used as indicators of the levels and quality of services within an authority and as comparators between authorities.

3. The introduction of social services performance measures is turning the spotlight on local authority information and information systems and there are increasing demands from authorities for the development of a core data set which would provide a foundation for more reliable and comparable information.

Project scope

4. In summary, this project aims to:

5. The primary focus of the project is to define the core data items that authorities need to hold in order to meet central information requirements. The initial focus would be on information required by the Department of Health but the work could be extended to cover all central social care information requirements, including those of the Audit Commission and CIPFA. It is recognised that these may only be a subset of the local management information requirements.

6. There are inevitably some overlaps between health and social care information requirements and this project would aim to identify those data items of common interest to health and social services that might eventually form part of a joint shared record.

7. As well as identifying the data items, the project would seek to come up with agreed definitions for each of the data items that were acceptable to authorities. Experiences on the Referrals, Assessments and Packages of Care (RAP) project and on the Children in Need (CIN) project have demonstrated that this is not an easy task. The project would seek to build on the RAP and CIN work and pick up some of the work on social terms carried out under the NHS Information Management and Technology Strategy.

8. Although information requested by the centre is generally at an aggregate level, this is usually built up from information relating to individual clients. It is these individual client records which are seen as the essential building blocks for good data. This project would seek to develop a data model that would identify the key components of a social care record that are necessary to meet central information requirements.

9. Putting the data items, data definitions and data model together would effectively result in a data dictionary for social care that could sit alongside and complement the NHS data dictionary. Creation of the data dictionary however is probably outside the scope of this particular project.

Links to other work

10. The project is not starting from scratch. A significant amount of information and knowledge is already available and a number of other projects are already starting to address some of the issues raised above. In some respects an overarching aim of the proposed project is to pull together some of the many strands of work that have already been carried out or are underway at present. Recent work commissioned by the Department of Health Information Sharing Working Group (with representatives from DH and local authorities) identified in excess of 100 'projects' looking at information issues. The attached diagram at Annex A identifies some of the projects that are currently addressing core data issues and which would form the starting point for identifying key data items.

11. The following ongoing pieces of work are particularly relevant to this project:

This project would in effect start to address the first three stages of the ADSS IMG proposals for development of an electronic social care record (ESCR).

12. This project would also link into the wider programme of work on Information Management in PSS which has been discussed by the PSS Performance Assessment Steering Group.

Roger StatonOctober 1999