Dr Chris Tackaberry, CEO of healthtech AI pioneers Clinithink, highlights the crisis facing the NHS and the importance of Diabetes Week from June 12-18.
Every week, diabetes leads to around 180 amputations, 590 heart attacks and 2,300 cases of heart failure. The financial impact of this on the NHS is eye watering. The total cost of diabetes to the health service is over £10 billion a year, or more than £25,000 a minute – equivalent to 10% of its entire budget.
Over 5 million people in the UK have diabetes, and it’s estimated that millions more are at risk of developing the disease due to the growing prevalence of risk factors such as obesity and physical inactivity. Diabetes is a life-changing condition, putting sufferers at a high risk of serious health complications including kidney failure, stroke, amputation and blindness.
Early diagnosis and careful management can control the disease and even put it into remission. However, there are estimated to be around 850,000 people in the UK living unknowingly with type 2 diabetes – the most common form of the disease – leaving them without access to vital medical care and support, and putting them at significant risk of major health issues.
Furthermore, even those with a diagnosis are often unaware of the red flags – such as a loss of sensation – that could signify a major complication, and therefore don’t adapt their behaviours or seek medical attention when it is needed.
The health complications caused by diabetes are largely preventable through disease management – controlling blood sugar levels, blood pressure and cholesterol – and increasing numbers of sufferers are turning to technology to achieve this.
‘Diabetes tech’ includes everything from smart insulin pens, which automatically record when and how much insulin a person has injected, through to continuous glucose monitors that allow people to check their blood sugar levels without drawing blood. An ever-growing number of apps and smart tools enable diabetes sufferers to keep a close eye on their weight, food consumption and physical activity to help them keep a better handle on the disease.
Finding the needle in the haystack
Technology is also being used to enable the early diagnosis of patients with diabetes complications. For example, diabetic foot disease (DFD) is a common health issue that affects around 15-20% of diabetes sufferers and which, without early intervention and careful management, can lead to amputation. Concerningly there are significant numbers of people living with undiagnosed DFD in the UK. Without earlier diagnosis, the condition could damage patient health outcomes and increase the burden on the NHS at a time when the health service is already under immense strain.
The problem is that the valuable healthcare data needed to help clinicians identify whether a patient has DFD – or is at risk of developing the condition – is hidden away within dense, unstructured medical data, which is not computable by traditional methods and therefore typically goes unanalysed.
So how can technology help to solve this data challenge? Well, AI-driven data analysis tools can now review and interpret medical records, including the unstructured data, with radical speed and precision, identifying vulnerable groups of patients who are at risk of getting DFD.
The technology has already been rolled out in Barts Health NHS Trust, one of the largest NHS trusts in the UK, which generates millions of documents a year containing vital patient insights, but lacks the clinical time needed to sit down and read them all. By applying AI, Barts Health has been able to find 350% more at-risk patients than previous methods allowed. This enables interventions – such as an urgent referral to a podiatrist or vascular surgeon – to take place while a patient is in hospital, thus reducing the risk of an amputation or other serious health problems, and saving the trust precious clinical time.
Diabetic retinopathy is another major complication, which can lead to blindness if left untreated. Recent research done in a collaboration between DeepMind (now Google Health) and Moorfields Eye Hospital and reported in Nature Medicine demonstrated the potential for AI to accurately identify Age-related macular degeneration (AMD) from retinal scans, demonstrating that AI can also be applied to improve diagnostic speed and accuracy for diabetic retinopathy.
A growing health emergency
Cases of type 2 diabetes are on the rise as the numbers of people who are overweight or obese across the UK grows. As we approach this year’s Diabetes Week, a UK-wide initiative to increase awareness of the disease and boost funding for future research, it’s vital that we communicate the urgent need for more NHS organisations to explore the benefits of adopting AI technologies to speed up the pathway to diabetes diagnosis.
Using AI to flag up patients with diabetes complications, or at risk of developing complications, is essential for the NHS to combat this growing health crisis – before it becomes an out-and-out emergency.