100 Ways of Using Data to Make Lives Better

Unlocking Data to Improve Hospital Discharge Arrangements for Homeless People in England

Published on: 22nd August 2017

Download case study

Public Health
Case Study 72

Project Lead: Robert Aldridge, University College London

The Challenge

Most people being discharged from hospital can recover in a stable home environment with support from family and friends. Homeless people, by definition have no stable home environment to go to and this is likely to substantially affect their recovery and lead to high levels of hospital readmission. In response to this problem a number of projects have been set up around the country to help plan for hospital discharge in homeless patients. Some have clinical input, some do not and most rely on housing support workers, but there are also many parts of England that do not have any support in place. If homeless discharge services are to become more widely available we need to understand which approaches work best and show whether they provide good value for money. To do this, researchers need to be able to access the health records of homeless people using these services.

The Research

A team from The Farr Institute at University College London (UCL) hosted a workshop with people with lived experience of homelessness and social exclusion. A range of activities were developed and views were obtained on collecting operational data and health records. 85% of homeless people surveyed were happy for data on homelessness, clinical records and criminal records be linked together in order to understand the underlying causes of ill health in this group and the effectiveness of homeless hospital discharge services.

Participants agreed that it would be impractical to ask permission from all homeless patients to link their data as many have difficulties engaging with health services and this would be likely to exclude the most vulnerable from analyses. Participants agreed that it would be acceptable to use unconsented individual level identifiable data as long adequate data security safeguards are in place and studies have passed ethical and legal review.

The Results

Linkage of different datasets requires patient identifiers such as name and date of birth to be available during the linkage. It is not legal for researchers to have access to this personal information without patient permission unless strong safeguards are put in place to protect personal information and there is a strong health justification for this, and it would not be practical to seek permission from all patients involved. However, the researchers have now got all the necessary ethical and legal permissions.

The Impact

Farr Institute researchers at University College London and other research partners will use data from electronic health records from all regions in England to estimate hospitalisation rates, A&E attendances and deaths in homeless people admitted to the discharge services involved in this research. The aim is to find out what happened to the people after they were seen by a hospital discharge team. The team will use innovative methods to use linked health records from the homeless hospital discharge services, a health screening register of homeless people in the community and hospital death records.

Using linked data from discharge services and national health records will allow researchers to understand the impact hospital discharge services have on the quality of care of homeless people and whether it reduces hospital re-admission and which types of discharge services work best. This will provide the evidence needed for investment in these services.

The researchers hope that the findings from their study will provide new insights that policymakers and commissioners will be able to utilise to make recommendations and effect changes in practice.

Enquiries to Natalie Fitzpatrick, Data Facilitator, The Farr Institute of Health Informatics Research, n.fitzpatrick@ucl.ac.uk

Back to top+