Improving cancer mortality and tackling waitlists with AI technology

The beginning of February marks World Cancer Day. In the UK, there are

around 1,000 new cases of cancer every day[1]. As we know all too well,

the NHS has been battling long waiting lists since before the COVID-19

pandemic. But the consequence of that pandemic is especially serious for

people with cancer. Macmillan estimates that, as of March 2022, the

number of people starting treatment for cancer in the UK was still at least

37,000 lower than expected[2].

NHS England would need to work at 110% capacity

for 7 months to catch up on missing cancer treatments

because of the pandemic[3].

Currently, only 3 in 10 NHS Trusts in England are meeting their referral

targets[4] - a worryingly low number for patients facing a cancer diagnosis.

Later stage diagnosis has serious implications for a person’s ability to

survive their cancer and to their after-effects following treatment. That’s

where we come in.

Figures show that there are around 7.21 million people waiting for

treatment[5] and that, last year, 7.8 million appointments were missed

across NHS services[6]. We’ve created a way for NHS trusts to manage

their backlogs, meet their cancer targets, and improve the health of the

nation. This means that we can help NHS trusts to get people diagnosed

with cancer onto the right, life-saving treatment pathways sooner. And we

can get more people into important screening appointments.

New evidence suggests that black women routinely suffer late-stage

diagnosis for women with breast, ovarian, uterine, non-small cell lung

cancer and colon cancer, and that black men are diagnosed with prostate

cancer at a later stage than white men[7].

Jon Shelton, head of cancer intelligence at Cancer Research UK, says

“[t]hat’s why tackling known barriers to help seeking, whether that’s fear or

difficulty accessing a GP, is so important – so more people come forward

with symptoms.”

We have a plan to address this, incorporating our unique AI insights with

brand new opportunities across social media. When we speak to people

directly, from the palm of their hand and in their own language, we can

address factors that stop them seeking help. Things like not knowing what

symptoms they should be reporting to their GP, something that twice as

many women from an ethnic minority background reported compared to

white women (23% vs 12%)[8].

In fact, we’re already working with 5 hospitals across

2 trusts with over 2 million outpatient appointments.

Our pilot is currently live across Mid and South Essex NHS Trust. We’re

empowering their administrative staff to know which appointments will be

missed, why, and how to do something about it. Our products that mean

clinics can rely on a steady stream of appointments to give all patients

access to healthcare.

Because not everyone does. Improving engagement with healthcare

providers is a major factor for reducing health inequalities[8]. And stories

like Cheryl’s [hyperlink to blog post: World Cancer Day blog] demonstrate

that there can be multiple considerations for patients from minority

backgrounds when it comes to engaging with their healthcare provider.

When hospital teams know more about a patient’s accessibility needs,

cultural or religious needs, or other commitments, they can personalise a

patient’s care with smarter appointment times. That includes switching

people to virtual appointments, when it makes sense, and reserving high-

value appointments for the most in-need.

“Every appointment freed up could be used for other

patients, especially those who have been waiting the

longest, helping us continue to make progress to

reduce the longest waits.”

That’s what Steve Barclay, Health and Social Care Secretary, had to say

about initiatives that directly tackle the NHS backlog[9]. Initiatives like ours.

We were founded in 2019 with a clear mission: to use AI to understand

human behaviour, create efficient services, and challenge health

inequality. And we’ve achieved over 90% Area Under the Curve accuracy

in predicting non-attendance with over 2 weeks notice – which could save

the NHS ￡1.2 billion a year.

More than 15 million general practice (GP) appointments are missed each

year. Around 7.2 million missed appointments are with busy family

doctors, which adds up to more than 1.2 million GP hours wasted each

year – the equivalent of over 600 GPs working full time for a year10].

The combined cost of these missed appointments could pay for:

? The annual salary of 2,325 full time GPs

? 224, 640 cataract operations

? 58,320 hip replacement operations

? 216,000 drug treatment courses for Alzheimer’s

? The annual salary of 8, 424 full time community nurses[11]

We’re at the cutting edge of something truly game-

changing for the NHS. And we’re looking for forward-

thinking trusts to get involved.

At Deep Medical, we’re deeply innovative, deeply perceptive, deeply

curious, deeply collaborative, and deeply principled. We’re looking to work

with like-minded Trusts to expand our pilot and demonstrate clearly the

benefits of our products - for patients, and for healthcare providers.

Because when everyone has access to healthcare, and the NHS can save

on unnecessary cost burdens, everybody in society wins.

We want to help NHS trusts make the best use of every available slot they

have, especially in light of the COVID backlog and cancer waitlists. So

we’re opening the call to hospitals and GP surgeries who’d like to get

involved. You can learn more about us and our products on our website

[hyperlink https://deep-medical.ai/] and get in touch with us here [hyperlink

to relevant inbox].

[1]. https://www.cancerresearchuk.org/health-professional/cancer-statistics-for-the-uk#heading-Zero

[2]. https://www.macmillan.org.uk/dfsmedia/1a6f23537f7f4519bb0cf14c45b2a629/9468-10061/2022-

cancer-statistics-factsheet

[3]. https://medium.com/macmillan-press-releases-and-statements/macmillan-responds-to-nhse-

november-2022-cancer-waiting-times-data-1ac6eb4b65ac

[4]. https://www.theguardian.com/society/2022/oct/26/seven-in-10-nhs-trusts-in-england-failing-to-hit-

cancer-referrals-target

[5].https://www.bma.org.uk/advice-and-support/nhs-delivery-and-workforce/pressures/nhs-backlog-

data-analysis

[6]. https://www.england.nhs.uk/2023/01/nhs-drive-to-reduce-no-shows-to-help-tackle-long-waits-for-

care/

[7]. https://news.cancerresearchuk.org/2023/01/27/new-analysis-reveals-black-women-in-england-

more-likely-to-be-diagnosed-with-late-stage-cancer/

[8]. https://news.cancerresearchuk.org/2023/01/27/new-analysis-reveals-black-women-in-england-

more-likely-to-be-diagnosed-with-late-stage-cancer/

[9]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5725414/

[10]. https://www.england.nhs.uk/2023/01/nhs-drive-to-reduce-no-shows-to-help-tackle-long-waits-for-

care/

[11]. https://www.england.nhs.uk/2019/01/missed-gp-appointments-costing-nhs-millions/

[12]. https://www.england.nhs.uk/2019/01/missed-gp-appointments-costing-nhs-millions/