AI in Healthcare: The lightning strike amidst the perfect storm

The healthcare revolution

The global healthcare market generates $21 trillion in revenues and yet, is still mostly undigitized. Out of the $21 trillion, only 2% is invested in digital technologies, while over ten times that amount is spent on manual administration. The incumbents in this industry, many state-backed, yield incredible power and benefit from its stagnancy. This leaves consumers and providers in this market grid-locked in a massive incentive conflict and archaic processes, ultimately losing out in the long run. This is not news for those who have been following the health tech sector over the past decades. 

Looking back, prior to 2010, VC-backed opportunities within healthcare had been limited to technology that patches up existing broken workflows or digital health, targeting only a small fraction of the market. We all looked to Livongo as the gold standard for exit multiples within this sector while also sober to the fact that there aren’t many like it. 

Now, looking forward, we at Northzone believe that healthtech is finally coming of age. A whole host of digital care delivery companies have successfully scaled to the mass market, paving the way for b2b healthtech startups to infiltrate the rest of the system, and finally the adoption of AI will be a pivotal step forward. AI-powered tools won’t just automate many of the manual processes; they’ll also bring massive amounts of data into the system, empowering both consumers and providers while driving ROI and ultimately aligning incentives with payers. We’ll explore those details more in depth later, but first, we’ve highlighted some of the key changes over the past decade that have paved the way for the paradigm shift we believe is on the horizon – answering the question, Why now? 

Embracing health digitalisation

In the 2010s, the adoption of telemedicine started to demonstrate how digital care models could bypass traditional healthcare structures. We first started to see this in the Nordics in companies like Kry, and soon realized the even bigger opportunity in the US given a much larger market, where we made investments in companies like Spring Health and Thirty Madison, to name a few.

Then came the COVID-19 pandemic, the catalyst that accelerated digital health adoption by an astonishing 38 times. For the first time ever, it was possible to build an entire vertical within healthcare – behavioral health – fully digital-first. To put this into perspective, when we first invested in Spring Health’s Series A round, they had $1 million in revenues, and only five years later they have over half a billion and serve over 10 million people. They’ve become one of the largest mental health platforms in the world and like its peers in this sector – fully online first. 

Today, these now scaled digital health players are looking to re-integrate into existing healthcare systems in order to reach their next stage of growth, which remains challenging as the rest of the value chain is still fairly antiquated. There are dozens of multi-billion dollar opportunities in highly specialized healthcare functions such as, revenue cycle management, credentialing, verticalized EHRs, scribing, coding, record retrieval, intake, prior authentication, eligibility verification, the list goes on. Each of these is dominated by incumbents built in the 80s and 90s. And only recently have digital health companies opted to partner with startups to reestablish these processes, optimizing for seamless customer experience which their end customers have now come to expect. Seeing this adoption pattern, we invested in healthtech SaaS startups Nirvana Health, and Tandem Health among others who are solving convoluted healthcare value chain problems in totally new ways, leveraging AI.

The bigger prize is beyond digital health

While digital health players have been early adopters of new SaaS solutions, driving startups to their first $10-20M in revenues, this is only the tip of the iceberg. We believe the bigger opportunity for the true healthcare revolution lies within the broader market – still largely operating in a brick and mortar model. It’s finally ready to adopt digital technologies, a shift that has been driven by significant changes on both the supply and demand sides simultaneously.

On the supply side, healthcare is under extreme pressure. Private equity roll-ups of healthcare systems are driving down costs, while a global labor crisis is exacerbating shortages of healthcare professionals. Meanwhile, legislative pressures are also pushing for reduced anti-competitive practices among healthcare giants. The modernisation of healthcare data, initiated by the Obama-era legislation, has laid crucial groundwork. The Affordable Care Act of 2010 allocated $30 billion to encourage hospitals and providers to transition from paper to electronic health records, boosting their adoption from just 9% in 2008 to 83% by 2015. Yet again, health systems fell short in embracing the free exchange of information, leading to the 2016 21st Century Cures Act, which mandated interoperability and ensured that, “all electronically accessible health information” could be accessed, exchanged, and used “without special effort on the part of the user”.  

On the demand side, patients are navigating a system that is increasingly difficult to manage. They’re confronted by extraordinarily complex and frustrating consumer experiences, where unexpected medical bills can lead to financial ruin and opaque pricing systems feel impossible to decipher and high health insurance costs. As if that weren’t enough, consumer trends including a rapidly aging population, the GLP-1 proliferation, and the growing prevalence of behavioral health issues, further strain an already overburdened system which can’t meet their needs. Many incumbents in the industry are looking to startups to keep pace with this growing demand, presenting an opportunity for technological solutions. 

Healthtech is standing at the crossroads of a major transformation. A digital transformation is already well underway, with the entire patient-provider experience moving online, streamlining everything from appointment scheduling to insurance verification and payment processes, akin to trends e-commerce and fintech from previous decades.

Video: Snippet of Wendy’s presentation during Northzone’s 2024 AGM

AI’s contribution to the value chain

The proliferation of AI is happening in real time. There are just a few examples of how the technology is already driving ROI across the value chain, beyond the hype. We have been focusing on technologies that help structure the massive piles of unstructured data within healthcare processes and extend their benefits across the value chain to both reduce administrative burden but also improve care. 

For example, Tandem, like other AI scribes, is already becoming indispensable to physicians, not just saving them 40% of their time but also capturing detailed patient data that will be stored in the EMR and visible to other practitioners. Tandem’s data will be used to support clinical decision-making in real time but also inform outpatient follow-ups and even used to train future models. 

Our portfolio company Idoven has collected millions of hours of ECG data across tens of thousands of patients across all different types of devices to help cardiologists better triage, treat, and deliver preventative care to patients. Doctors only need to collect the ECG data across whatever device they use, and Idoven’s algorithms can identify 22 cardiac patterns automatically at cardiologist-level accuracy. They also identify patients that are at increased risk of developing heart conditions within the next 6 months. All of this is integrated directly into EMRs to be accessible at the point of care.

Also, the process of clinical record retrieval at the point of care has already benefited significantly from the 80% digitization of medical records, but the 20% of the long tail offline charts can take providers days and weeks to get ahold of. MeMR uses AI to consolidate and reconcile patient records from a massive pile of unstructured data to complete these requests within hours.

Finally, no other market makes customers guess the cost of essential services, only to get a surprise bill months later that could bankrupt them, forcing them to spend months disputing it over the phone. On the other side of the market, healthcare providers still have to spend $125bn a year out of their revenues (via the RCM process) to figure out how much they’re owed after rendering their services. It is one of the world’s worst data reconciliation problems between some of the most archaic systems. It is also the problem that makes the healthcare market not an actual market. Nirvana uses AI to understand and continually update the price of treatment for every patient within every plan at the point of intake. Nirvana calculates their eligibility, out of pocket expenses, including co-pays and deductibles to eliminate the guesswork, allowing patients to understand the true cost of care and allowing providers to bill the right amount upfront, reducing the RCM burden by 110%. 

There are so many more AI-ready opportunities within healthcare’s multi-trillion dollar budget for manual workflows. Some of the other ones we’re looking into include coding, billing, triaging, intake, amongst others! We are deeply invested in the evolution of healthcare and are enthusiastic about the opportunities in the next five years. The shift from patchwork technology to comprehensive digital solutions is just the beginning. If you’re building in this space please reach out to wendy[@]northzone.com.