Every day, vast amounts of NHS data is collected that could be used to improve care for patients. Whether it’s a worried parent attending A&E with their ill child, a routine GP appointment to review diabetes care, or an urgent admission to hospital for a frail older person, each interaction leaves data that provides information about the care patients received.
However, despite the scale of the data collected by the NHS it is often not linked together in helpful ways, so we don’t always have the full information we need to know how to improve a patient’s care. For example, knowing whether a patient is living in a care home, what support they might have from their family, or whether they are coping with their condition all provides important context to understand what needs a patient might have.
Connecting data sources to find solutions
Linking data sets together from many areas in the NHS and more widely can provide solutions to these problems, and make data more useful to improve care. However, data collection is organised around the divisions in the NHS, with different records of care collected by GPs, community care, hospitals etc. This has important implications for the care patients receive every day. For example, if a patient is discharged from hospital with a new prescription, and this isn’t shared with their GP, the GP may be less able to provide safe and effective care. It also prevents those planning or improving care from tracking the patient over time, and so to develop a full picture of the quality of care patients receive this data must be joined up across health and care organisations.
Supporting patients to manage their health better
For the growing number of patients living with long-term conditions like diabetes or depression, most of their care isn’t delivered by the NHS. In fact, patients and their carers are responsible for the majority of care and behaviours to keep well.
One clinical commissioning group used a survey to ask patients with long-term conditions to assess their confidence in managing their condition, and the results were entered onto their GP record. This information is useful on its own, allowing GPs to tailor care to patients, and refer them to health coaches or peer support initiatives that could improve their confidence. However, once we’d linked this data up with other records, it became even more powerful.
We linked data so that we could follow patients over a number of years, and across primary and secondary care, showing that those better able to manage their health conditions were 29% less likely to attend A&E, and 38% less likely to be admitted in an emergency. This demonstrates the potential value that could be gained through investing in support to help patients manage their conditions across the NHS, and the value of linked datasets to understand the care patients receive.
Enhanced care for care home residents
In many areas of the country, commissioners and providers of health and social care are working together to improve the quality of care provided to care home residents. However, it is difficult for care home managers to know what happens to their residents in hospital, and similarly hospital staff don’t always know whether patients are residents of a care home, or managing their care needs at home.
But GP practices typically know where patients live. So, by comparing this information with a list of addresses of care homes it’s possible to link a list of care home residents to activity recorded in hospital. This linked data can help explain to care home managers for what conditions their residents are admitted, and if some of these admissions might be potentially avoidable.
We carried out analysis for care home residents in Rushcliffe, Nottingham, where enhanced care was provided for residents of both nursing and residential homes. Our analysis showed that the enhanced care package, which includes having an assigned GP practice for each care home that residents are encouraged to switch to when they move in, resulted in a reduction of 29% in the number of A&E attendances for care home residents. These residents were also 23% less likely to be admitted to hospital as an emergency.
Insight into children’s A&E attendance
Like older people, we know from data collected in hospitals that young children can be frequent attenders at A&E. But what is driving this attendance, the health needs of the children, or something more complicated?
Working with one large GP practice we linked GP records into households, allowing us to match parents and their children. Using this linked data we were able to better understand what was driving the demand for GP and emergency care for children. While we found that a child’s health conditions (like asthma) were a predictor of A&E and GP use, we also found that children whose mother or father had a mental health condition were around 40% more likely to be admitted to hospital or attend A&E. Using linked data would help GPs and commissioners identify those in potential need of additional support, and design more holistic care around the family.
Improving patient care by making data linkage part of business as usual
These examples of our work show the importance of data linkage when supporting the improvement of patient care and reducing health care activity where appropriate. In addition, The Health Foundation’s Advancing Applied Analytics programme is funding local teams to develop their own capability to build ambitious linked datasets, and for example, develop insights for their populations.
These projects, and work in exemplar health and care systems are important, and make a difference locally. But there’s still more to do. Too often data sets are only linked together for a specific project or piece of analysis, and often we can’t build a picture of care across the country. Through working to develop new innovations in data linkage, and funding the NHS to build their own capability the Health Foundation is trying to help those working in the health and social care system to make data linkage routine, to maximise the benefit of data for patients and the NHS.