Jem Rashbass, director of master registries and data at NHS Digital, explores why data has been a key role in the fight against coronavirus.
When one thinks about the gigantic NHS effort during the coronavirus (Covid-19) pandemic, it’s only right that our first thoughts go to the staff on the frontline.
Thousands of health and social care workers are doing everything in their power to manage this once in a lifetime health threat.
But behind the scenes, there is another less-visible line of defence that is also playing a key role in the fight against coronavirus, and that’s data.
Dreams of integration
I was studying molecular biology at the University of Cambridge when it first dawned on me that at some point in the future personalised medicine would be faced with the challenge of ‘how do we take genomic and molecular data on an individual patient and understand the clinical dataset associated with it?’
In order to do that you need large, population-level, high-resolution clinical datasets that underpin personalised medicine.
Having worked for Public Health England for seven years, I moved into a joint role with NHS Digital in November last year, where I was delighted to be focusing on how we could build an all-disease registration service.
The plan was to move away from isolated single collections for each registry and instead create a harmonised integrated clinical dataset within the national safe haven that is NHSD from which we could deliver the data and insights to improve care.
But then came coronavirus and the world changed radically.
Data in the fight against COVID-19
The first time I significantly got involved in coronavirus work was in early March – following a call late Friday evening asking me to draft a paper to go to a COBRA meeting the following Monday about how we could best identify patients potentially at risk of adverse complications from COVID-19.
This was far from being a straightforward task. For anyone unfamiliar with NHS datasets, this information does not lie neatly in one easily accessible place.
If we’d had had a primary care prescription service that included information on both the prescription AND the condition that it was treating, we would have been at a great advantage, as this would have given us instant access to a population-level dataset with actual diagnoses.
For example – individual medicines are often used to treat very different conditions – so you need to know if someone on a beta blocker is being treated for anxiety, tremor, hypertension or ischaemic heart disease.
In the absence of such a dataset, we spent the weekend trying to establish how best we could pull this information together.
Questions were asked such as “How do we identify cancer patients?” “What data is available that might give us a handle on patient with renal disease, rare conditions, cystic fibrosis, recent organ transplants?” “What data is available in primary care?”
This was a piece of investigative work that ordinarily would have taken several months. We did the initial work in the space of a weekend.
It was that weekend that brought home to me that we were moving into a very different world.
New ways of working:
We’re now seeing changes in the public sector being completed in weeks or even days, in some cases, that would have previously taken years.
Of course, no-one’s comparing this to the life of the teams fighting COVID-19 on the frontline, but these secondary support services at a national level are vital to deliver what’s needed.
Coronavirus has completely changed our way of working. From a data perspective, it’s proved to be an opportunity to look at things differently and to modernise at speed.
Take rolling out a Nightingale centre, for example. How do you set up what’s potentially a 4,000-bed resource in London in a week?
Behind the scenes, there are so many different components from smartcards and authentication processes to laptops and networks. Even registering appropriate GPs so that they can work in that environment.
How do you get the equipment in there? How do you get the networks in there? Then there’s a huge list of trivial things like setting up an organisational data code to make that organisation become part of the NHS.
Thanks to the flexibility of NHS Digital staff, we’ve been able to create a volunteer army that can be deployed to the next major challenge…at one point we had people at home working at their kitchen tables printing smartcards for Nightingale centres.
We had a huge upsurge in the number of people registering to use the NHS App, so we’ve trained up additional staff members on how to authenticate these applications – a function they’d previously never done before.
One of the things I’m hugely proud of at NHS Digital is the way that the organisation, at a system level, has stepped up to deliver some of the most extraordinary resources across the service.
Our Secondary Uses Service (more commonly known as SUS+) is a healthcare data collection in England, which is typically used for healthcare planning, supporting payments and commissioning policy, development and research.
It is now also being used by scientists at the University of Oxford to help assess the effectiveness of a number of potential treatments for coronavirus.
In time, data from our Hospital Episode Statistics (HES) will also be used to help understand the overall effectiveness of the treatments, particularly in patients with underlying health conditions.
This is a marvellous example of using an existing robust dataset to try and solve a relatively new and evolving problem.
Over at the University of Cambridge, trials have begun on a system that will use machine learning to help predict the upcoming demand for intensive care (ICU) beds and ventilators needed to treat patients with coronavirus.
This COVID 19 Capacity Planning and Analysis System (CPAS) has been developed by NHS Digital data scientists and a team of Cambridge researchers to support hospitals to plan more accurately and help ensure that resources are deployed to best effect.
We’re trying to deliver real services in real time that will help to make a real difference.
Supporting primary care:
It’s no surprise that many frontline workers have reported feeling overwhelmed. But for GPs, this was exacerbated by an increase in requests for data to support the coronavirus response.
Having been contacted by the British Medical Association (BMA) and the Royal College of General Practitioners (RCGP) about the issue, we put in place a central service to collect and disseminate data from general practice to support vital planning and research during the pandemic.
It is hoped that this central collection will significantly reduce the administrative burden on GPs, while also providing a more robust and controlled access to GP data (via NHS Digital).
Front page news:
One of the positives to take from this challenging time is the fact that data is now front-page news.
You don’t make decisions without good data or evidence, so it’s been quite refreshing to see data take centre stage in recent weeks.
The fact that we all wake up on a morning and try to understand the latest coronavirus figures is a real shift in people’s view of the centrality of data.
However, it does also highlight the difficulty of interpreting data. Take the tragic number of coronavirus deaths, for example.
If you look at the curves produced by countries across the world, nearly every one reports a dip at the weekend.
Now, I don’t believe it’s a real dip. I certainly know that in the UK it isn’t a real dip. It’s a reporting issue on the flow of data in the system.
It shows where the response to the pandemic is pushing our systems to the absolute limit of where we have previously built them.
We’ve never been asked for national daily death feeds that are consistent and accurate. It’s actually a very hard thing to do.
The nearer you move to near-real time data at scale, the more difficult that becomes, but I do think it’s wonderful that as a nation we are looking at graphs and thinking about data.
We’ve all dreamed of a data-driven NHS for a long time. If one good thing can come from the coronavirus pandemic, I would hope it’s the fact that we are now starting to see significant changes. We’re moving at speed and we’re using this very challenging period as an opportunity.
Public recognition of the importance of data has come to the fore. Let’s make sure we don’t slip back again.