|
Key issues: Data
The event then considered some of the key issues in more detail, namely:
Using Data to support the model
What data did you require to run the model?
The data used to refresh the high-risk, was provided to PCTs by Acute Trusts:
The criteria was
- patients aged 65+ with 2 or more acute admissions in the last year
The data items collected included:
- patient identifier information
- diagnosis
- GP identified
- LOS
top
Where, how and when did you get the data?
PCTs then needed to quality check the data, for example, to remove deceased patients and those no longer resident in the area.
The cohort identified in this way was then cross-referenced with Social Services & Mental Health Trust data who were also asked for an at-risk list. The list was then supplemented with nominations of high-risk patients from GPs. It was important to take account of what local partners criteria for high-risk would be.
This process is repeated on a regular basis to 'top up' caseload, identify new high-risk patients as part of spread and scale-up.
Generally data issues have been addressed differently according to the local circumstances/environment.
top
What problems have you faced getting the data?
There have been a number of challenges in terms of extracting data. These have mostly related to primary care data, for example the range of GP systems.
There is also a capacity issue in extracting data. Requirements for nGMS have made it difficult to assign resource to the data requirements for this project. PCTs also found that they had a skills gap in terms of analysing data and a sense that they were data rich but information poor.
Coding was a common issue for data from both the primary care and acute settings.
top
How have you overcome those problems?
The group felt that the key to overcoming this problem was viewing data as a catalyst for change and something that allows you to identify the problems and measure change.
top
If you were starting the project again without the support of United/Evercare, what resources/tools would help you to meet the data requirements of the project?
Some of this was based around tools. For example, data checklists and standardised information such as read codes/ICD. Analysts would also need to be up-skilled to turn data into information.
It was also very important to generate data for evidence of positive change. The APN contact sheets, Hospital Admission Tool and a retrospective look at the high-risk cohort were all important but a comparison with those patients that met the criteria, but were not part of the project would be really powerful.
There is a critical need to show reduced admission, LOS and readmission.
top
Click here to return to the Agenda
|
