Using linked health and social care data to understand service delivery and planning and improve outcomes

Key Points

• The COVID-19 pandemic has highlighted the need for reliable, routinely collected, shared care data.

• Existing linked data sets are not comprehensive enough to accurately predict demand for long-term care in England.

• The DACHA study will pilot linking health and social care data for service planning and delivery.

 

Adult social care provides short or long-term support with many essential activities of daily living, such as washing, dressing, getting enough food and drink and feeling safe. It provides a vital safety net for many hundreds of thousands of people and their unpaid carers across the UK. However, use of long-term, publicly funded adult social care is determined by more than individual’s needs. Access to long-term care is also associated with the level of informal support that the individual receives, the constraints of local service capacity (supply) and, unlike the National Health Service (NHS), a financial means test. Demand far exceeds supply, in 2019–20, local authorities in England, who have a legal responsibility to assess people’s needs and eligibility for support, received the equivalent of 5,290 requests for support per day [1]. Despite expenditure on adult social care increasing, the number of older people (aged 65 and over) receiving long-term care has decreased year on year [1], because those who are eligible, require intensive packages of support. Publicly funded care is now reserved for those with the highest needs and lowest assets.

Nakubulwa et al. [2] argue that there is a need to better predict the demand for long-term care to inform service planning and development and note that existing linked health and social care data for this purpose in England are not easily accessible. Using a longitudinal, retrospective cohort design, [2] used de-identified, routinely available data derived from clinical electronic health records to explore factors associated with accessing publicly funded, long-term social care amongst older adults (age 75 and over) and build a predictive risk model forecasting future service use.

The Northwest London Discover Database, which is used in the research [2], links data from primary, secondary and tertiary care, community and mental health care, emergency departments and social care. The linked data contain rich information of participants’ sociodemographic characteristics and health conditions. However, its power to accurately predict demand and access to adult social care is limited by a lack of data on key indicators, such as availability of informal support (whether or not the person lived alone was missing for 82% of the sample), the individual’s socioeconomic status and if they were funding their own social care. Unsurprisingly, the study found that individuals were more likely to receive long-term, publicly funded adult social care if they were older, lived in areas with higher socioeconomic deprivation and had a preexisting mental health or neurological condition (which are likely to be associated with loss of function and independence in old age). The lack of high quality shared care data on this population meant that accurate individual-level prediction was not possible.

Their study completed in December 2019, just 2 months before the start of the COVID-19 pandemic. Since then, the need for reliable, routinely collected shared-care data has become both obvious and urgent [3, 4]. The Department of Health and Social Care’s Shared Care Records programme is committed to bringing individuals’ health and social care information together in one digital record. To do this, they first have to accelerate the uptake of digital care records in the adult social care sector. Approximately 30% of social care providers are still using entirely paper based systems, with another 30% using a combination of approaches [4]. However, this landscape is rapidly changing and there is now growing recognition amongst providers of the potential of digital care records to reduce duplication of recording, capture real-time information and ultimately improve the needs and outcomes of the people using their services. Given that social care providers are often the only organisations holding information about the needs and characteristics of self-funders (those who do not meet the criteria for public funding), being able to match these data to NHS data will address an important and long-standing evidence gap.

The utility of anonymised data drawn from linked NHS and social care records, alongside local authority and regulatory data, to inform care delivery and improve outcomes is being explored in an ongoing study to develop and test a minimum data set (MDS) for care homes in England [5, 6]. The DACHA study (Developing research resources and minimum dataset for care homes’ adoption and use) is working with digitally enabled care homes to collect resident-level data from care records and match this to native data held about them (e.g. in hospital and GP data) and the care home in which they live (e.g. Care Quality Commission (CQC) data). It is also exploring the acceptability and feasibility of collecting comparable information about older people in receipt of domiciliary care services.

Clearly, for an minimum dataset (MDS) to have predictive power and be used to project demand for future long-term care on an individual, regional or national level, we need a comprehensive MDS for health and social care. Any data collected for the purposes of populating such a data set must also be of importance to individuals and serve to ultimately improve their outcomes, including their quality of life. Successful implementation requires that care providers benefit from the information collected and are able to use that data to demonstrate their impact, improve quality and inform people’s care [6].

Declaration of Conflicts of Interest

None.

Declaration of Sources of Funding

This paper draws on work undertaken by the author as part of the DACHA study. The DACHA Study is funded by the National Institute for Health Research (NIHR) Health and Social Care Delivery programme (HS&DR NIHR127234). The author is supported by the NIHR Applied Research Collaboration Kent, Surrey and Sussex. The views expressed are those of the author and not necessarily those of the NIHR or the Department of Health and Social Care.