Skip to content

2018 Conference Presentation

Service evaluation United Kingdom

11 September 2018

Understanding long-term care needs from administrative data

Javiera Cartagena-Farias, Personal Social Services Research Unit, LSE, United Kingdom
Sanna Read,

Tom Snell Personal Social Services Research Unit, LSE, United Kingdon
Francesco D’Amico Personal Social Services Research Unit, LSE, United Kingdon
Jose-Luis Fernandez Personal Social Services Research Unit, LSE, United Kingdon


An ageing population has increased the demand to provide services that support older adults to maintain their independence and enhance their quality of life. This increase in the number of individuals with care needs, and financial pressures, have led to a growing interest in how to target resources in the most cost-effective way. Using assessment records from three Local Authorities in England, we assessed the potential to use administrative records to achieve a better understanding of individual care needs and their changes over time. We constructed a longitudinal user-level information database including social care assessments, user socio-demographic characteristics and service use for the period 2010-2017. Around 30,000 older users living in community and in institutional settings in each Local Authority were analysed. When the reliability of these administrative records was tested, some discrepancies were found when general social care need patterns were compared against survey data. This may have been caused by differences in the formulation of assessment questions and the challenges of comparing national trends to specific local trends. Social care users were categorised in groups that represent underlying patterns and combinations of needs using Latent Class Analysis techniques. Four main types of social care recipients by needs were identified. The content of the four classes varied somewhat between the three LAs, but some common patterns were evident. Class 1 was characterised by relatively low levels of needs except for specific activities of daily living (ADLs) associated with physical disability (about 10% of the sample). Class 2 included older people with particularly high levels of needs in instrumental activities of daily living (IADL) and was associated with dementia (about a third of the sample). Class 3 presented moderate to substantial levels of needs (about a fifth to half of the sample). Class 4 was characterised by very high levels of both ADL and IADL needs (about a quarter to third of the sample). Making use of the longitudinal nature of the available data, it was found that after two years, ten percent of social care recipients experienced an improvement while around forty percent of community experienced an increase in their care needs – finding also that higher levels of local authority support were associated with lower levels of deterioration.

We observed in the use of administrative data a potential monitoring tool that could be used by local authorities to target their resources and to develop local strategies according to their own population characteristics. Despite the challenges associated with the use of administrative data, we have reasons to believe that they are a rich and unique source of information.

Keywords: ageing, social care needs, changes in care needs, social care, latent class analysis, administrative data

Skip to toolbar