Interview, News

Interview: John Klepper, CEO and co-founder of Swiss company PIPRA [Pre-Interventional Preventive Risk Assessment]

Whilst at Web Summit, we met up with John Klepper, co-founder and CEO of PIPRA (Pre-Interventional Preventive Risk Assessment), a Zurich-based medtech company that has developed its first product; an AI-based surgical tool to assess a patients’ risk of suffering from postoperative delirium.

Hi John! Thanks for meeting us. Can you tell us a bit about yourself and PIPRA?

I’m the co-founder and CEO of PIPRA, which I founded in 2019 along with my two brilliant co-founders; Nayeli Schutz, an anaesthesiologist and Ben Dodsworth, a cell biologist.

Our first product is focused on delirium, a neurocognitive disease area which is extremely prevalent and under-addressed. Research indicates that delirium affects almost one in four patients over 60 and it can lead to serious negative outcomes; for example, it is linked to a six-fold increase in nursing home admissions, and 38 percent of patients who experience delirium remain with long-term cognitive complications, such as dementia. It also doubles mortality risk, and it is 60 percent more expensive to healthcare than diabetes. It affects 60 million people per year, and that figure is growing.

Essentially, it means that patients – especially elderly ones – are coming to the hospital for a procedure like a surgery and they are leaving in a worse condition than they arrived.

I used to work in Alzheimer’s disease, using artificial intelligence for prediction and prevention purposes. When discussing dementia with clinicians during stakeholder interviews, in an attempt to uncover a clinical need in hospitals, the problem of delirium kept cropping up in conversation. Unlike dementia, delirium is much more acute; a patient can seem fine, and then suddenly they may experience complications that can include hallucinations, confusion, memory loss and disorientation. It can be very extreme – a lot of patients describe delirium as the worst experience of their life. For many patients it’s terrifying.

During these interviews, I remembered that a few of years earlier, my mother had gone through something similar. When I later called her and asked her to explain her experience in more detail, it was clear that she had had delirium and that it was a horrific experience. To this day, while she was lucky to have fully recovered, the memory of that event resulted in her refusing to go near a hospital again. That, coupled with the huge unmet need, gave me the drive to explore this further. It was around this time that I had the great fortunate to meet my co-founders and the three of us, as a multidisciplinary team, created PIPRA.

The role of tech in PIPRA

Delirium, much like dementia, is very multifactorial. There are lots of different risk factors and what may contribute to a high risk score for one person may not have an influence on another. Predicting delirium at present tends to rely on a subjective guess by the clinician; even experts in delirium admit that they can get predictions wrong, because of the multifactorial nature.

As such, AI can be used effectively in calculating the different risk factors resulting in an individual risk score for that patient. PIPRA’s algorithm runs in the background, extracts routinely collected data from electronic health records and stratifies patients into risk groups, highlighting which patients are at risk for developing delirium. PIPRA’s solution helps support the effective management of delirium throughout the perioperative setting to ensure that targeted prevention, compliance on screening and diagnosing is correctly implemented.

PIPRA’s mission

We want to ensure that patients go home in a better state than before interventions and that hospitals save money through reduced length-of-stay. In the UK, for example, we are talking to hospitals about reducing length-of-stay, helping the NHS tackle the enormous waiting list burden the NHS faces today.

We also want to help raise awareness – delirium really does have devastating consequences for patients and their loved ones, and has a huge impact on healthcare systems. Ultimately, in ten years’ time, I’d like to see delirium reduced on a global scale.

If we are very successful in our journey, then we may spread out into other types of neurocognitive diseases like dementia, but for now we want to stay focused on this problem because it is so huge and so unmet.

What do people say about PIPRA?

We’ve had some great comments on PIPRA from key opinion leaders from highly respected hospitals. For example, Robert D Stevens is the Director Division of Informatics and Precision Medicine and Co-director Precision Medicine Center of Excellence in Neurocritical Care at Johns Hopkins University in Baltimore. He said that there is a “perception that elective surgeries have become extremely safe, thanks to advances in operative technique and anesthesia care. This view overlooks the fact that a substantial number of patients, in particular older ones, experience adverse cognitive outcomes after surgery. Postoperative delirium is a potentially devastating complication which might be preventable or treatable if appropriate predictive information became available.”

He called PIPRA “a game-changing solution which squarely addresses this need. It estimates delirium risk accurately and reliably, thanks to an AI algorithm trained on routinely collected clinical data. This is an extraordinarily valuable tool which will have a positive impact on choices made by patients undergoing surgery and their families, as well as surgeons, anesthesiologists, nurses, and other professionals who have a stake in the perioperative enterprise.”

We’ve also heard from José R. Maldonado, John & Terry Levin Family Professor of Medicine and Professor of Psychiatry at Stanford University School of Medicine. José noted that “as clinicians consider and discuss surgical risks for their patients, we should similarly consider the risk for post-operative delirium” and added: “The use of PIPRA can personalize this risk calculation and provide both patients and clinicians with actionable information, with the purpose of instituting a delirium mitigation plan for that specific individual.”

From Switzerland itself, Prof. Dr. med. Luzius Steiner (Head of Anaesthesiology, Universitätsspital Basel) commented on PIPRA, saying that it “proposes a new tool that allows rapid determination of an individual patients risk to develop postoperative delirium after an operation. This is a very exciting and important development which can improve current clinical practice. We will finally be able to not only target resource intensive non-pharmacological preventive interventions to those patients who benefit the most but we will also be able to design clinical trials regarding delirium based on risk groups.”

The digital health landscape in Switzerland

In terms of innovation, I think Switzerland is one of the leaders in digital health. The country places tremendous focus on healthcare and on improving digital health.

I would say that the key challenges lie in adoption – healthcare payers and providers tend to be slow-moving and need an incredible amount of evidence. Of course, a reasonable amount of evidence is definitely needed, because we are dealing with patient lives and data privacy issues. But sometimes innovation can be stifled by too much emphasis on further studies and evidence. For example, once your software has been certified as a medical device and it has been validated with publications in respected journals, I believe patients would be better served if hospitals would accept these qualifications and move faster to adopt innovation.  

We have found that other geographies – such as Scandinavia – tend to be much faster at trying out new technologies and seeing how things like AI can help support the clinician in their professional decision-making and how it can improve patient outcomes. Delivering value-based healthcare is what drives us and defines us as a company.

Many thanks to John for taking the time to chat with us.

Explore HTN Interviews