Painting by numbers: Using data to understand social care

Information about social care need and provision in England is routinely collected by local authorities, but could we be doing more to make better use of this data to improve the care people receive? We asked Dr José-Luis Fernández, Associate Professorial Research Fellow and Deputy Director of the Personal Social Services Research Unit at The London School of Economics and Political Science (LSE), for his thoughts on the picture presented by social care data.

What does current data tell us about social care in England?

‘There has been a huge contraction in levels of support, in particular for older people, at a level that is unique internationally. In five years, the likelihood of an older person receiving state-supported social care has been reduced by an estimated 40%. I haven’t seen that in any other comparable country.’

Why do we need to use data to improve social care?

‘First, we need evidence about what works. What is the impact of social care on the outcomes we are trying to achieve? Are social care interventions improving the quality of life of people with social care needs and their carers?

‘Second, we need to look at whether and how people who could potentially benefit from social care are accessing services, and particularly what the state and local authorities are doing to allocate that care. We also need broader evidence about the extent to which doing something in social care releases pressure on other parts of the health and social care system, such as hospitals.’

What data is routinely collected and why is it useful?

‘Local authorities in England provide data returns, which describe how much money they spend on social care, and what they spend the money on. We can use that data to improve the social care system by comparing how local authorities are supporting people in their patches.

‘There are two important caveats to that. Firstly, in contrast with the NHS, social care is a means-tested system and secondly, it is a system which allows local variability. Local authorities are encouraged to reflect local preferences and their local environment in deciding what sort of support they provide to people. That makes it difficult to judge the differences we see in patterns of the way services are allocated in different parts of the country. Some variability is to be expected on the grounds that, for example, unit costs are different. It might be more expensive to invest in community services as opposed to residential care in different areas. But we are seeing a huge amount of variability which is unlikely to be explained solely by local preferences and the local environment.’

What are the gaps in the data we have?

‘We still don’t have a lot of information about how social care support is allocated to people individually in England, although that is improving. Internationally speaking, we are in a privileged position, as local authorities collect data from individuals in their user and carer surveys. That data could be better, but it’s still a really interesting source of evidence we need to make more use of, to understand how well local authorities are planning social care services.

‘But the evidence is particularly limited about the needs of and services being provided to people in England who do not meet the means-test for local authority funded support – and therefore are not captured in local authority data.

‘One way of painting the picture across private and public service users is to look into proxies for social care need. This means looking for equivalent data that can tell us what social care needs people might have, even if they are not accessing care through local authorities. One option is to look at health data from primary care. Linked data sets across health and social care would help us get an idea of the overall levels of potential need for social care. Nobody is recording on a systematic basis if somebody buys their own social care – that would be a really useful data set.’

What other sources of data are there?

‘There are other useful datasets, but no one dataset paints the whole picture. For example, the English Longitudinal Survey of Ageing  and the Health Survey for England give us information to understand social care needs and support. But the sample sizes are small, partly because they cover a lot of different groups and partly because people with a high level of need find it more difficult to participate in those sorts of surveys. You can use something else as a proxy for total activity which includes state and private provision, such as social care workforce data. That’s not going to tell you who is getting what care, but it is telling you something about the volume of care – and that’s been going up. The data suggests that private consumption has been going up as state activity goes down.

‘What we don’t know about are those people who are falling through the cracks. Not everyone who isn’t eligible for local authority support will be able to afford to buy their own social care. There will be people who don’t have family or friends who can step in and provide the care that is no longer provided by local authority services. We don’t know how to paint that picture. There will be groups of people who are likely to be suffering as a result of these changes, but we need better data to understand that.

‘Ultimately, more complete data will help us paint a picture of how to best use social care resources to improve people’s quality of life.’