Skip to content
This website uses cookies to help us understand the way visitors use our website. We can't identify you with them and we don't share the data with anyone else. If you click Reject we will set a single cookie to remember your preference. Find out more in our privacy policy.

Linking datasets for better population health management

4 December 2019 | Dr Jenny Shand
Internationally, there is growing use of linked datasets to build more complete understanding of population profiles and population-wide health and care service utilisation.

In the UK, there has been substantial investment at a national level into programmes to enhance data linkage and access for both research and service delivery purposes. Examples include the establishment of Health Data Research UK, the Local Health Care Records Exemplar programme (LHCRE), Digital Innovation Hubs, and the development of population health management (a required function of all Integrated Care Systems by April 2021).

Health problems are complex and in many cases a single health issue may be influenced by interrelated social, environmental, and economic factors. However, progress to link datasets across settings of care is slow, and data linkage with other sectors which contain much of the information on social determinants of health is notoriously challenging.

Over the past three years, we have been working across UCLPartners, Care City, North Thames CLAHRC, BHR CCG, the London Borough of Barking and Dagenham and NELFT to create a unique dataset, the Care City Cohort, of Barking and Dagenham residents from 2011 to 2018.

The dataset brings together information from local government, health providers and health commissioners that has been linked at the individual level. It contains sociodemographic and health information (e.g. age, gender, ethnicity, smoking status and body mass index, prevalence of long-term conditions), information about where individuals live (e.g. levels of deprivation, household occupancy and household tenure), and information about their health and social care service use (e.g. accident and emergency attendances, GP contacts, social care packages, mental health inpatient stays), along with the unique property number to facilitate household level analysis. This dataset is due to be updated on an annual basis and be launched to the wider research community in early 2020.

The dataset is de-identified and hosted in BHR CCG Data Safe Haven, with different datasets linked together using linkage keys in place of NHS numbers. The Care City Community board have a critical role in providing public perspectives on the use of the dataset and supporting us to ensure its findings are relevant and accessible to all.

To date we have been running several research projects to demonstrate the utility of the dataset and its capacity to provide new perspectives on health and service use in the borough. We are delighted to share the outputs of a project investigating costs in the last year of life that has been published in Age and Ageing. The project found that health and care costs were £7,450 higher for patients in their final year of life compared with those not in their final year. The majority of the costs at the end of life related to unplanned hospital care in the final months. As age of death increases the proportion of care costs related to social care relative to healthcare increases.

This paper is the first stage of a three-year research programme being conducted with the North Thames ARC to investigate service use in the last year of life. The next phase will include analysis of additional settings of care (community, mental health, hospice at home, and inpatient hospice care), sub analysis according to place of death, and sub analysis according to different conditions (e.g., dementia). We have a multi-disciplinary research team including palliative care consultants from our local delivery system to ensure actionable insights.

We are excited about the potential this linked data has to inform opportunities for better management of the health of the population. We are even more excited about the potential to deliver real change if the insights from the data can be developed in partnership with the health and care workforce to confirm how and what we need to do to make population health a priority for all.

For further information about the data resource, please contact