Doctor, doctor! Larry Ellison's prescription for an Oracle-led healthcare revolution

Oracle’s $28.3 billion acquisition of healthcare giant Cerner closed last week, leaving Chief Technology Officer Larry Ellison free to drill down further into his vision for revolutionizing healthcare in the US. He pitches:

Better information is the key to transforming healthcare. Better information will allow doctors to deliver better patient outcomes. Better information will allow public health officials to develop much better public health policy. And better information will fundamentally lower health work healthcare costs.

A fundamental problem with today’s health management systems is that they are hospital-centric, not patient-centric, Ellison contends, with every hospital buying and operating its own information systems:

Each hospital system has their own patient Electronic Health Record database. There are thousands of them throughout the United States. When you go to UCLA, your patient records are in the UCLA database. When you go to Cedar Sinai and have an examination, those records, those results, end up in a Cedar Sinai Electronic Health Record database. You go to USC, it's in a USC database. In other words, your data, your electronic health data, is scattered across a dozen or two dozen separate databases.

This data fragmentation causes serious problems for both doctors and patients, he argues, citing an exemplar of a resident of LA having a car crash when visiting Cleveland:

You're rushed to one of the best hospitals in the world, The Cleveland Clinic, where they rush you into the emergency room, and they can't find any of your health records. They cannot get access. They don't know what your blood type is. They don't know what your allergies are. They don't know if you have a stent in your heart. They don't know anything about you. They don't know in terms of your health records.

Another case in point with recent resonance would be if a public health official wanted to find out how many people had been hospitalised with COVID in the past 24 hours. Ellison remarks:

That data is literally in thousands of separate databases. You'd have to go into every single database at [lots of hospitals] and count how many people were hospitalized at that hospital and then add them all up. There's no way to do that. I don't know if you remember when the public health officials in New York City were very worried about running out of available hospital beds during the pandemic. In fact, a hospital ship was dispatched to New York City, just in case they ran out of beds. Turns out the hospital ship was never used. They were worried about running out of beds, but they didn't really have any data and it turns out that they didn't run out of beds. The hospital ship was never used, but, absent data, they were making making very unreliable decisions.

Unified

It’s ironic, he noted, that while a clinic couldn’t get access to health data, it would be able to check out whether you could afford to pay for your treatment because it could check your credit rating via a unified credit card authorization database.

What’s needed is a unified national health records database on top of all the thousands of existing separate hospital databases. That’s where Oracle will come in, says Ellison:

We're building a system where all American citizens health records not only exist at the hospital level, but they also are in a unified national health records database.That national database solves the electronic health record fragmentation problem.

Now, the same example - you're in Cleveland Clinic. You arrived in Cleveland Clinic in an emergency, and Cleveland Clinic can instantaneously access all of your electronic health records. They can get it immediately and they get the very latest version of your health records because the way the system works is we continuously upload electronic health records from the hospital databases into this national database. So that data is up-to-the second accurate so they get all the information they need about you. What what your blood type is, what your allergies are, what drugs you're currently taking, do you have a stent your heart? - they get all of that information instantaneously.

And with that information they can deliver to you much better care and  that information will deliver much better patient outcomes. Doctors are now armed with the latest information so they can make better decisions.

That’s clearly impressive, but big databases full of personal information inevitably raise privacy concerns. Ellison insists that this has been factored into Oracle’s thinking:

There's only one person in the world who can give doctors access to your health records and that's you. You hold the key to all of your health records and the National EHR database….Nobody knows who they belong to until you supply the key, so there's absolute data privacy.

The national EHR database doesn't have your name or any identifier that identifies that you were hospitalized with COVID-19. It knows someone was hospitalised with COVID-19 and they can count all the people that are hospitalized with COVID-19, but they can't identify them. The public health officials can count the people, but not name the people. We ensure that we're not compromising data privacy.

The plan

Cerner’s core offering is Millennium, which Ellison says will be modernized and its capabilities expanded now that the takeover is complete:

First thing we're going to do is make it much easier to use. We're going to add a voice user interface to Millennium that makes it easier for doctors to access data, enter orders. There's an integrated telemedicine module that allows you to consult. If you're living in rural America and you want to consult with a specialist at MD Anderson for cancer, you can you do that via via secure video teleconference.

Oracle also plans to add disease specific AI modules, with Ellison stating that the firm will be working with medical professionals to develop these:

The idea is the people at Oracle are not going to be developing these AI models. Our platform, Cerner Millennium, is an open system and allows medical professionals at MD Anderson…if  they're experts in treating cancer , it allows them to add these AI modules to help other docs treat cancer. So if you go to MD Anderson, you have lots of cancer specialists; if you're 50 miles outside of Boise, Idaho in a community clinic being treated for cancer, now you get the benefit of these AI modules.

We're integrating clinical trials with Millennium to speed them up. And we're putting all the diagnostic devices on a single Internet of Things network. We're keeping all the diagnostic device results, all the images and all the other data in a database that we use for machine learning. Again, we guarantee data privacy - all of that data is anonymized, the only one who can look at that data in its identified form is you, the only one who can give that that permission.

Administration

Oracle already has a substantial footprint in the healthcare sector with Ellison pointing to non-clinical administrative duties that health professionals have to deal with:

A doc wants to recommend a better drug for you. They can't do that until the insurer agrees to reimburse them for that drug. So they have to communicate, 'I'd like to prescribe this drug' and get approval. That takes time. We want to automate all of that. Ordering drugs, ordering medical supplies, managing the inventory on the inside of hospitals is very complicated. The inventory is distributed throughout the hospital. Some of it's in inventory closets, but some of it's in the pharmacy, some of it's in the operating room. some of the inventory is in intensive care unit, some of this on a nurse's station. So just managing that inventory is enormously complicated. We're making major improvements on all with all of that.

A hospital is a very complicated workforce, much more complicated than, say, a company like Oracle or a bank like JP Morgan Chase. Docs often are not employees of Stanford Hospital or employees of UCLA or Cedar Sinai. They're often contract workers and they work a couple of days at UCLA, they work a couple of days someplace else, they have their own practice. So just scheduling the docs, recruiting the docs, paying the docs is one of the most complicated workforces in the world. Another thing we are dramatically improving is our HR system for hospitals. Of course, hospitals have to do a lot of regulatory reporting and financial accounting. We are automating all of that. It's interesting that Oracle Fusion in the cloud ERP and HR applications are already dominant amongst healthcare providers. It's a long list. We are now the dominant provider of ERP and HR for health care providers.

He concludes:

In summary, this new health management system will deliver much better information to healthcare professionals. It will deliver better patient outcomes, help docs deliver better patient outcomes, help public health officials improve public health policy, and lower overall costs.

That is now our primary mission here at Oracle.

My take

Over the 32 years that I’ve been tracking Oracle, I’ve seen Ellison fired up about various topics. When he’s on a mission, you really know he’s on a mission. Sometimes these work out, sometimes they take longer to come through. The Network Computer in every US school kid’s satchel,  which he pitched to President Bill Clinton in the back of a limo touring Silicon Valley, didn’t really come to pass in the way he’d imagined, but it can be argued that Apple delivered on the core vision several years later. And while we all had a chuckle at Ellison’s ‘just water vapor’ comments about cloud computing, he’s realigned Oracle around the cloud in a major way and made it work. Basic message - you never underestimate Larry Ellison. So when he declares he’s on mission to overhaul healthcare, I’m more than ready to believe him.

What he’s proposing with the unified database is something that has been tried before in other places. It was interesting to note former British Prime Minister Tony Blair popping up to add his support for the Oracle strategy. When in office, Blair commissioned a similar exercise for the UK’s National Health Service (NHS), the National Programme for IT, based, it is said, on a one hour briefing that caught his attention. That turned out to be perhaps the most notorious public sector IT failure in UK government history, resulting in massive cost over-runs, legal battles with tech suppliers and precious little to show at the end of it. The Oracle plan, I hasten to add, is self-evidently a lot more robust than that ill-planned pursuit of modernity.

There are still a lot of questions that will be answered in due course. In particular, I’d be keen to know how much of this applies to a non-US healthcare model. There are clearly a lot of common functions between, say, NHS and US hospitals, but in the UK, the insurance element, for example, is still largely absent (for now, at any rate!). But overall, this was prime Ellison - bit between his teeth and ready to make waves. What happens next is going to be something to keep a close eye on.