Data for Change
Why data on household health matters for the NHS
How much health care a person uses may depend on the health of those they live with.
We’ve been counting the number of people in our households a lot recently, to work out who we can safely meet within coronavirus (COVID-19) guidelines. Knowing who is in a household seems straightforward, but it turns out that accessing this information for analysis isn’t always so easy. When it comes to those providing health and social care services, household information could be immensely valuable for organising and improving care, but may not currently be used to its full potential.
Several studies have shown that people who live alone are at greater risk of ill health and earlier death compared with those who live with others, at least during later adulthood. This may be because, for some people, living alone is a marker of social isolation or loneliness. But whatever the underlying explanations for a link between living alone and ill health, public health teams use statistics on single person households for population level planning. This includes planning support for people on the shielded list during the COVID-19 pandemic.
An individual’s health may affect how much health care other people in their household use
But single person households may not be the only type of household with greater care needs. In recent research, we’ve demonstrated that the poor health of an individual may affect the health care use of their household co-residents as well. Our study of over 50s living in two-person households in England found that approximately half of all people who had two or more long-term health conditions (sometimes referred to as having multimorbidity) were living with someone who also had two or more conditions. This proportion was even higher in the most deprived areas. And those in households where both residents had two or more conditions each had more GP consultations and were more likely to use community care.
In previous research, we’ve shown that children and young people living with a parent with depression were more likely to visit the emergency department, be admitted to hospital as an inpatient, and have an outpatient visit. While this study didn’t attempt to show whether parental depression contributes to or is a consequence of a child’s ill health or health care needs, it further highlights the link between health and health care use among members of the same household.
What could the NHS use information about household health for?
This research underlines the potential for us to improve health care and health outcomes by using information about the health of people who live in the same household. For example, we could tailor the kind of care that is provided to include peer support within the household, along the lines of the group care approach, as well as individual patient care. The group care approach facilitates support from the clinician and from other patients (in this case, patients in the same household). There may also be benefits to care providers in terms of efficiency if they are able to systematically schedule care at the household level or facilitate more effective self-management. At the moment, the possible benefits for care providers and for those living with multiple health conditions are unknown and would need robust evaluation.
Information about the household context could also be used by primary care teams and other service providers to help identify and support informal carers, many of whom don’t self-identify or request support. Both the NHS Long Term Plan and Carers Action Plan laid out an ambition for recognising and supporting carers. This should be an even greater priority during the COVID-19 pandemic when access to some services has been restricted and may have led to a greater reliance on informal carers.
How can household information be accessed?
But the starting point to planning and providing care with a household focus is access to data on households and the health of household members. This is possible using the Unique Property Reference Number (UPRN) that accurately identifies addresses and can also be used to link information in different datasets.
Our recent work used data from residents in Barking and Dagenham held by the local authority and linked to records held by local GP practices and NHS trusts. Records were pseudonymised so that we couldn’t identify specific households or individuals and provided us with information about levels of ill health and recent use of health and social care for all household members. At the national level, the master patient index, a dataset based on GP registration data across England, includes the UPRN and can be used to flag people who recently started living alone, or have been bereaved, or are in a care home, for example.
As well as being able to access information about people in the same household, we need to gauge whether patients understand and are comfortable with data analysts and service providers using address matching and information about their households in this way.
The household context matters for health and growing evidence shows it affects use of health care too. We think there is potential to make much better use of the information contained in administrative records to plan, deliver and evaluate care with the household perspective in mind.
This blog was written by Mai Stafford (@stafford_xm) and Kathryn Marszalek (@kathmarszalek). We are part of the Data Analytics team at the Health Foundation. We have built an online repository of analytical resources. Please do get in touch or check out our analytical resources on GitHub.
