Health data and the fight against Covid-19

There is an ever-increasing number of examples of how health data is being used to benefit the population, and it has particularly come to the fore during the Covid-19 pandemic.

Health data has the potential to revolutionise clinical trials. For example, the Oxford University-led RECOVERY trial, which is testing treatments for Covid-19 patients in hospital, brought in existing datasets such as Covid-19 test results, death records, GP notes, prescription information and hospital admissions data.

Data curated by NHS Digital, the national information and technology partner, and serviced through NHS Digitrials, the health data research hub for clinical trials, has successfully enabled researchers find a “major breakthrough” in the fight against Covid-19. The trial uses data from NHS Digital’s Secondary Uses Service (SUS+) and other data sets, to help assess the effectiveness of a number of potential treatments for Covid-19.

Jem Rashbass, executive director, Data and Analytical Services at NHS Digital says: “Looking at the huge steps that have been taken over the last six months in our data services, I am immensely proud of what we have achieved, in partnership with research colleagues.

“New datasets, innovative linkages and entire services have been set up to meet the urgent needs of the health, care and research communities, to support them to respond to the pandemic, all while keeping patient data secure and ensuring that all uses meet the highest ethical standard.

“Data has a huge role to play in the future of healthcare and we need to keep pushing the boundaries of what we can do as the data engine, driving UK health research.”

Health data: fast-growing areas

Another important area is in the organisation and analysis of electronic health records, says Melissa Lewis-Brown, science manager at Health Data Research UK (HDR UK), the national institute for health data science. Health data scientists working as part of the HDR UK National Text Analytics Resource project have been using natural language processing to analyse text written by doctors in medical records, showing that ACE inhibitor medicines, commonly-prescribed for millions of people in the UK to treat various underlying health conditions such as high blood pressure, don’t increase the risk from Covid-19.

User-generated health data is another fast-growing area. Daily health logging data from millions of users of the Covid Symptom Study app, led by King’s College London and health data science company ZOE, revealed that loss of smell is a key symptom of Covid-19 as well as the classic cough and fever, leading to a change in NHS guidance.

Health data research can show where there are problems in the healthcare system. Ms Lewis-Brown says that analysis of health data has shown that around 5,000 heart attack sufferers might have missed out on life-saving hospital treatment as a result of the pandemic. And there are likely to be 17,000 or more additional deaths from cancer in the coming year due to missed diagnoses and treatment during the first half of 2020.

Understanding the impact of coronavirus on social care

The Health Foundation has used health data to reveal the impact of the coronavirus pandemic on social care in England. “We were able to highlight that between March and June there were more than 30,500 additional deaths among care home residents than would normally be expected and more than 4,500 deaths among people receiving care in their own homes,” says Kathryn Dreyer, a senior analytical manager, data analytics team at the Health Foundation.

Elective admissions and procedures for care home residents dropped to 58% of the historical five-year average in the first part of 2020. “This reduction may have led to a decrease in the risk of transmission of Covid-19 in care homes but may also have led to an increase in unmet care needs in this population,” says Ms Dreyer.

Covid-19 test, track and trace

A national comprehensive testing and contact tracing system is “essential to understanding and controlling disease spread in the context of a virus that has neither vaccine nor effective treatment”, Ms Dreyer says. However, she points out, “there are huge challenges associated with setting up new national programmes of work”.

Reports of the challenges involved in developing, implementing and running a ‘test, track and trace’ system in the UK have been well documented by the media. Six months into the pandemic and news about testing is constantly changing.

Initially the story was about whether to have a centralised or localised approach to testing. Adam Mayer, senior manager at Qlik, says one of the “greatest turning points in the use of data to responds to the Covid crisis” was when the Government took steps to decentralise the data.

“As our own application shows, the infection rate differs significantly across the country, so putting critical health data in the hands of local leaders enables then make more informed decisions in how to respond to crisis in their own local area.

“This decentralised data decision making is not only helping local councils take policy decisions that they hope will help them manage its spread, but it also has a direct impact on the management of stock levels,” he says.

Earlier this September, the Government announced plans to expand coronavirus testing to 10 million tests a day by early 2021. But Dr Chaand Nagpaul, council chairman of the British Medical Association, said it was unclear how the so-called ‘Operation Moonshot’ would work – given the “huge problems” currently seen with lab capacity.

Just days later, the Government was considering rationing coronavirus tests for the general public and prioritising NHS and care home workers amid continuing testing shortages. In parliament, prime minister Boris Johnson said the Government was trying to meet a “colossal spike” in demand at “record speed”, but health minister Matt Hancock told MPs it would likely take weeks to resolve the country’s testing issues.

Dr Tempest says that the UK needs a high number of tests “because we have a high population”.

“We need to have the infrastructure, the diagnostic capabilities and the human capital to be able to do the swabs, the retests, and then the localised tracking to prevent further outbreaks.

“We’re on a journey to get there with test, track and trace. But we’re only at the start of that journey, and we’re six months into the pandemic”, she says.

The health service “still has great strides to make” in the use of data sharing to help them roll-out population health programmes that will accurately identify at risk individuals and health trends, says Mr Mayer.

For example, he says investing in data sharing and analysis between multi-agency organisations to better understanding population health “should be a key priority for the government in helping public services provide more at-home and community care to the over 50s throughout the pandemic”.

The most commonly heard health statistics currently relating to the over 50s is the greater risk of complications from Covid-19 as age increases. “This is an important example of where population health programmes can benefit this age group, as identifying early interventions in the community that would help avoid more serious illnesses that require hospitalisation will significantly reduce those individuals’ risk of contracting the virus while undergoing healthcare,” Mr Mayer says.

Understanding the health needs of over 50s

Ms Dreyer says health data can be “incredibly useful” in helping to understand the health and care needs of different groups of the population, including those over 50.

Recent analytical projects conducted at the Health Foundation, using health data, have allowed the organisation to explore a number of different issues. One project showed that the risk of having multiple conditions rises with age and is greater in socioeconomically deprived areas.

And another study supported by the Foundation found that living alone has an impact on health service use for patients over the age of 65. Older patients who live alone were found to have a higher probability of going to the A&E and visiting their GP, after adjusting for patient demographics and clinical characteristics.

“The richer the data, the better the care”

NHS Digital’s Jem Rashbass says that during the pandemic, “the importance of understanding the health and care needs of different groups of the population has never been more apparent”.

“Whether it is our work with Cambridge on the impact of the virus on different ethnicities and age groups, or our new adult social care dataset that captures information like the types of care provision and demographics of care receivers: the richer the data, the better the care that can be planned and delivered for patients with many different needs,” he says.

One ‘big idea’ that has the potential to improve quality and efficiency of ageing research is ‘big data’. This type of data research is routinely collected from health, social care and society, and has the advantage of ensuring that multi-morbidity, frailty and those lacking capacity are not excluded, explains Dr Oliver Todd, a registrar in geriatric and internal medicine at Bradford Teaching Hospitals NHS Foundation Trust.

Big data could improve care for older people

In a recent study, Dr Todd and colleagues found this routine data has the potential to improve the quality and efficiency of ageing research (Todd et al, 2020).¹ Dr Todd says: “Big data could improve care for older people. Examples include the involvement of more ‘hard-to-reach’ older people in clinical research for better treatments and therapies, as well as to provide transparent and objective means to better monitor quality of care of older people across health and social care settings.”

However, he cautions that if misused, “big data also has the potential for harm – particularly, to further entrench existing inequalities and inequities of the health care system,” he says.

But for routine data to benefit care this patient population, “Engaging carers and clinicians of older people in big data is crucial – both to reap the rewards for older people, and to mitigate potential harms.”

Part one of this article looked at how health data can improve patient care and save lives.

References

Age and Ageing, Volume 49, Issue 5, September 2020, Pages 716 to 722, https://doi.org/10.1093/ageing/afaa018