C2-Ai’s tools could save ‘70,000 lives and £1bn’ across NHS
With entry to the Cambridge Independent Science and Technology Awards 2020 now open, we’re highlighting the work of some of the nominees. This week, we talk to C2-Ai, an incredible Oakington-based company using artificial intelligence-powered tools to help hospitals save lives.
‘If their innovation was scaled across the NHS, I estimate it would save 70,000 lives and £1bn.”
This was the extraordinary assessment of C2-Ai by Hassan Chaudhury, digital health lead at Healthcare UK, part of the Department of International Trade.
From its base on Oakington Business Park, C2-Ai – or Copeland Clinical AI – provides systems to help hospitals reduce avoidable harm and mortality, while helping them to make significant savings.
It employs fewer than 10 people, yet its technology is in use in 11 countries around the world.
Among its array of tools are apps helping to reduce incidence of hospital-acquired acute kidney injury (HA-AKI) and hospital-acquired pneumonia (HAP).
Together, these account for 140,000 deaths annually in the NHS alone, and increase a patient’s typical length of stay by eight days and six days respectively, meaning they also soak up significant resources.
It is believed that 1.6m bed days are lost in the NHS every year to the conditions – 360,000 of those in critical care beds.
“Our AI-backed app enables frontline doctors and nurses to assess patients at point of admission – and change of condition – for their respective risk of developing HA-AKI and/or HAP, and suggests appropriate action accordingly to prevent these serious conditions manifesting themselves,” chief strategy officer Richard Jones tells the Cambridge Independent.
“The clinician selects the type of check they wish to do and they select from the patient type. They select from a list of conditions relevant to the patient and then press for a result. This provides a risk-assessment for the patient and suggests actions that will reduce the chance of the patient acquiring the condition.”
C2-Ai’s tools are built on 30 years of research and a phenomenally large patient data set of 140 million records from 46 countries.
“Our systems have reduced overall AKI levels, and HAP, by 50 per cent or more in hospitals in England, New Zealand and Sweden,” adds Richard.
He said the app takes “two minutes to install and one minute to learn”.
And it takes as little as 30 seconds for assessments to be made, then instantly delivers recommendations tailored to the individual patient.
This can save clinicians significant time, while helping to free up beds – something that has been particularly pertinent during the Covid-19 crisis.
“One in five patients in ICU has HA-AKI in normal times, but reports suggest up to 36 per cent of patients hospitalised with Covid-19 acquire HA-AKI,” says Richard, referring to a study in New York. “That means up to 10 per cent of Covid-19 patients in that situation are dying from HA-AKI.”
In the UK, the typical AKI mortality rate is 28 per cent.
Approved by the Medicine and Healthcare Products Regulatory Agency, C2-Ai’s app has a CE mark and the company is expecting FDA approval shortly in the US.
The company’s technology was named among the 10 essential digital health ideas for a UK Covid-19 national response by Healthcare UK.
And last month in the House of Commons, minister Graham Stuart said: “Companies like Cambridge-based C2-AI … are leading the way in the UK’s cutting edge healthtech sector. C2-AI save lives by predicting avoidable harm and mortality so they free up capacity in intensive care units for Covid-19 patients.”
Richard adds: “Our approach is particularly important for third-world countries suffering badly with poor bed availability – by helping create capacity and saving lives during the pandemic, but also reducing the strain on their limited resources – nephrologists, dialysis etc – caused by the renal damage suffered by patients who acquire acute kidney injury.”
The technology has earned some extremely positive reviews.
Sunjay Kanwar, a consultant general surgeon at St Helens and Knowsley Teaching Hospitals NHS Trust, in Merseyside, who has been trialling the C2-Ai app for AKI and HAP, described it as “phenomenal and incredibly fast”. He said he was “delighted and excited as to how this tool can help us identify these patients early and put in place simple measures, which all have a significant impact”.
Another of C2-Ai’s systems is called CRAB, which helps to pick up the 90 per cent of preventable harms that typically go undetected in hospitals.
“This is a retrospective audit system and it can spot avoidable harm, mortality and variation across pretty much everything a hospital does,” explains Richard. “If you used this across the whole of the NHS, it would lead to 5,000 lives saved annually.”
C2-Ai says in a US hospital it can deliver potential savings of $5m-$10m per year, and reduce clinical negligence claims by about 10 per cent.
“We can save hospitals money and it’s all about them improving. They get a monthly report and an expert system they can access,” says Richard. “It directs people’s attention to where it should be.
“I’d love it to be in Addenbrooke’s.”
The system assesses how a hospital is treating patients by examining a monthly download of data.
“When you leave hospital, there are coders who will translate everything that’s happened to you and every drug you’ve been given – your whole journey – into a series of codes, typically ICD10 coding.
“We’ve looked at those and correlated them very accurately with 146 surgical complications and 32 medical issues to do with blood sepsis, AKI and so on.
“There’s no complex integration – we can be up and running in two to five days.”
In New Zealand, Dr Michael Roberts, chief medical officer of Northland District Health Board, said: “It took two years and a very costly investigation to deal with a competence issue in our organisation some time ago. We set C2-Ai the blind challenge of seeing if they could have found the problem in our historical data. They did so in 20 minutes. Needless to say, we have invested in the system.”
C2-Ai also has an app that enables a pre-operative risk assessment to be carried out by surgeons in about one minute, and which calculates an individual’s risk of complications or death, using AI and a huge patient dataset.
This enables hospitals to make better-informed decisions on when to operate, and prevents avoidable harm.
And it has a new variation on this technology, called COMPASS Surgical List Triage, which is currently on trial in two hospitals and can be used to manage the backlog of elective surgery cases that have built up during the Covid-19 pandemic without having to set up months worth of outpatient appointments.
“We use the same principles and calculate the individual risk of mortality and complications based on physiology. It could take months out of the backlog,” suggests Richard.
C2-Ai was established a decade ago by Graham Copeland, the UK’s former patient safety ‘tsar’, and Steve Mackenney, a former head of quality, safety and clinical governance at the Department of Health.
Richard, who has worked in technology companies from start-ups through to exits, describes their work as “amazing” and says the tools are “transformative but not disruptive”, meaning they can readily be adopted by busy hospital trusts.
The company has entered a number of categories in our Science and Technology Awards, including the new Covid-19 Response Award.
It is also a nominee for AI Company of the Year, a category sponsored by the Entrepreneurship Centre at the Cambridge Judge Business School.
Bruno Cotta, executive director at the Entrepreneurship Centre, said: “We are delighted to be supporting this year’s Cambridge Independent Science & Technology Awards and in particular the AI Company of Year.
“It continues to be a hugely challenging time for those starting and scaling their businesses, but Cambridge is well placed to lead the way with its world-class scientific talent, ground-breaking technology and unique environment for advancing innovation and entrepreneurship.”