Counting on data for better health and care

Health data has been vital for national and local policy and decision makers during the pandemic and is helping us to understand the long-term impact on people’s health.  

New data collections have helped us understand the changing pattern of infections, which reached their highest ever level in the first week of April this year, when 1 in 13 people in England had COVID-19. Data told us about the consequences of the pandemic – that so far over 175,000 people in the UK have died from COVID-19 and that in May 2022, 1 in 36 of us were reporting long-COVID symptoms. And data told us about progress with vaccinations – with more than 9 in 10 of those eligible for a vaccination now having had at least one. Data was extensively used in the No. 10 pandemic briefings to reassure the public that the government’s response was being informed by the latest trends.  

There have also been novel uses of existing data to answer pressing questions. At the Health Foundation we’ve worked with a major provider of digital primary care systems to look at how GP practices used technology to engage with patients online during the pandemic. And through our Networked Data Lab we found ways of looking at how the health care use of clinically extremely vulnerable patients who were asked to shield was affected.  

The availability of large COVID-related datasets has also stimulated people from outside the health field to analyse data, produce insights and visualise the data in innovative ways. Overall, the pandemic has shown how data, when used well, can help us understand and address the challenges faced in health and social care, and improve services. This is an opportunity the Government is keen to exploit with its ambitiously titled draft data strategy, published last summer, Data Saves Lives, Reshaping Health and Social Care with Data. The Health Foundation, which has long championed the power of data and has invested heavily in supporting data and analytics, welcomes this new interest, and the priority being given to creating and using new data assets.  

But data can also be misused and cause harm. Health data has been used disingenuously to promote misleading narratives about NHS performance (such as waiting times for elective care) when in reality the issues are much more complex.  

And neither is more data a panacea to the challenges health and care faces. If it were, the UK, with its world class data assets, would have had a better pandemic performance relative to other countries. In addition to more data, several other things are needed. 

First, rather than starting with the data we have, we need to start with an understanding of the things that matter to people. This may require the collection of new data on those things.  

Although technology and automated systems are generating thousands of gigabytes of new data, there are still major gaps in what we know. Take mental health for example. Over the last nine months, through our Networked Data Lab, we have been working with colleagues in five areas across the UK to use local data and analysis to understand and improve the performance of mental health services for children and young people. But what became clear was that, despite making the most of linked data assets available within areas, key data needed for good policymaking or services is lacking. For example, no data were available on services accessed in schools or the community, or services funded by local government or the voluntary sector. There is also a lack of data on use of privately funded mental health services, which may be growing as NHS services have struggled.  

Second, related to this, there needs to be a greater awareness of who is missing from the data that’s collected and the biases this may introduce. For example, the adult social care survey is used to assess how effectively local authority funded services help service users to live safely and independently in their own homes. But because it is only a survey of users, it excludes the increasing number of people who don’t receive services because their needs aren’t deemed to be high enough. As a result, it tells us nothing about unmet need.  

Gaps in data can have more insidious effects too. If data driven tools, powered by algorithms and artificial intelligence, are developed using only data from people in certain population groups, they may not help those from outside these groups – potentially increasing inequality. 

Third, once we’ve got the data, we need to analyse it properly to avoid drawing false conclusions. There are a variety of mistakes that can be made. Confusing correlation with causation is a common one. Another is not comparing like-with-like. For example, data may show that patients treated by one service have worse outcomes than those treated by another; but this may be solely because the first service has been treating sicker patients. The Improvement Analytics Unit has been set up to help with precisely these sorts of issues, ensuring that the effect of service changes are robustly evaluated. My colleague Arne talks more about this in his interview.  

And fourth, if analysis of data is to save lives and improve health it needs to influence decision makers and the people they listen to or take advice from.  

Some of this is in the hands of data analysts. How they communicate their work is important. At the Health Foundation we spend a lot of time curating data and analysis about key issues, for example through our What drives health inequalities? evidence hub, which brings together data, insights and analysis exploring how the circumstances in which we live shape our health.  

But, perhaps more importantly, there needs to be a culture of evidence-informed policymaking and a demand for good data-driven analysis, including that which challenges existing beliefs.  

We all have roles to play here. Decision makers and advisers need to stay curious and open to challenge. And as citizens we must all do our bit to hold decision makers to account and demand openness about the data and evidence on which decisions are based. 

Charles Tallack (@CharlesTTHF) is Interim Director of Data Analytics at the Health Foundation.

This content originally featured in our email newsletter, which explores perspectives and expert opinion on a different health or health care topic each month.