Using data to identify good-quality care for older people

There is a large amount of variation in the quality of care that is currently delivered to older people across England. Numerous initiatives have been set up with the aim of improving care, but much of the evidence of their effectiveness remains anecdotal. Furthermore, when multiple improvement activities are in place in one area, it is not always clear which parts improve outcomes and which do not.

 In the absence of useful data to show that an activity is working to improve care, it may be useful to approach from the other direction by starting with the data and asking whether certain patterns visible in routine data are good at identifying activity that improves quality of care. If they are, then we may have a way of identifying activity that might otherwise be overlooked.

This report describes the results of some pilot analysis to ascertain the usefulness of using this approach, using a few indicators that were mainly derived from acute emergency hospital use. The aims were to test out ways of using data to identify potential success; to understand the challenges of attributing data findings to real-world activity; to judge the feasibility of extending into wider work; and to inform what this wider work should look like.

This study shows that there is scope for using more sophisticated analytical methods for identifying improvements in care quality, and that they have advantages in improving specificity and as continuous monitoring tools. This may be particularly true at the local level, or even at a lower level, such as individual GP practices. While we applied these techniques retrospectively, there are likely to be advantages in using these methods for prospective monitoring and evaluation.