As pressures on emergency care within the NHS continue to rise, research has focused on the demand attributed to our growing older population. However, we know far less about what is driving children and young people’s demand for emergency care. Over a single year between 2014/15 and 2015/16 there was a 5% increase in accident and emergency (A&E) attendances for children and young people.

Many factors influence a child’s need to go to A&E and their use of other health care services. These might go beyond medical need, such as who they live with and the health of their parents, particularly their mental health. We thought these factors could be important drivers of a child’s use of health care services, but there was limited evidence about the extent of the association – so we designed a data linkage study to tackle this complex question, now published in BMJ Paediatrics Open.

From individuals to households: a new way of exploring data

The NHS collects a wealth of data, but the majority of these data sets record medical need (such as the conditions a patient might have) and their use of health care (such as procedures or prescriptions). Very rarely do the data sets contain information on the other factors that might impact a child’s health, like what health needs their parents might have.

As part of our data analytics work at the Health Foundation, we want to use existing data in new ways to dig deeper into the context that surrounds patients. This research is the first in a series of data analyses looking at the impact social and economic context have on people’s use of the NHS and outcomes for patients.

Working in collaboration with Valentine Health Partnership, a large multi-site GP practice in London, we used anonymised data from 25,252 patients registered at the practice, including 6,124 children and young people (under the age of 16).

We restructured this data so we could look at households. We used an address-matching algorithm to create anonymised household groups – a novel approach to using this routinely available data – and then removed variables that could identify individuals before carrying out the analysis. By using this technique, we could group parents or carers with their children and then model the impact of parents’ long-term conditions, for example mental health, on children’s number of GP appointments, A&E attendances, outpatient appointments and hospital admissions.

Grouping people living together in this way meant we could take into account the influence of having siblings, the overall health care use patterns of the household, plus the influence of socio-economic deprivation and what long term conditions a child might have.

The link between parents’ health care use and health and their child’s A&E attendance

We found that children’s health care use mirrored their parents’ across all health care services. For example, after adjusting for other factors, if a parent goes to A&E one additional time, their child is 27% more likely to attend A&E.

The most striking finding was that a child with a parent who has depression is 41% more likely to have an A&E attendance than a child whose parent does not have depression. This finding highlights the potentially broader impact of mental health conditions and the importance of understanding the wider context of a household. Our results are consistent with findings from a study conducted in Wales which found a 17% increased risk of emergency admissions for children and young people living with an adult with a mental health disorder.

Our research is an observational study and did not explore the reason for increased demand. There are several possible explanations for the findings. For example, it could be that parental depression might be a consequence of prolonged illness in a child. While we controlled for many factors in our analysis, the association we detected may also be confounded by other factors such as the wider social support available to a family, or their household income.

How can our findings help clinicians improve care for children and young people?

We found that a child’s family and social context, in particular their parents’ mental health, are closely linked to their use of health care services. This highlights that GPs and other clinicians need to understand the impact of wider factors on the use of health care services in order to provide appropriate and effective care.

Dr Rebecca Rosen, a GP Partner at Valentine Health Partnership and co-author of the research, highlighted the value of these findings to GPs:

‘This study provides evidence about the influence that parental health can have on children’s use of GP services and highlights the importance of providing personal and family continuity for some patients. We need to develop simple ways for practices to identify which families would most benefit from this type of relational holistic care.’

With health care services under increasing pressure, it’s essential to understand the different factors influencing the way in which children and young people use the NHS in England, so approaches can be designed that are holistic, appropriate, effective and have good outcomes for both children and their parents.

 

Kathryn Dreyer (@kathrynadreyer) is a Principal Data Analyst at the Health Foundation.

 

Find out more

Read our news release

View the paper in BMJ Paediatrics Open