Expert to Expert: Ginni Rometty, Alex Gorsky & Clay Johnston
CEOs from IBM and Johnson & Johnson talk with the dean of the Dell Medical School about the convergence of health and IT. Know it or not, it’s changing the way you interact with your doctor.
During a panel at 2017’s South by Southwest Interactive Festival, IBM CEO Ginni Rometty, Johnson & Johnson CEO Alex Gorsky and Clay Johnston, dean of the Dell Medical School at The University of Texas at Austin, discussed cross-industry partnerships and the convergence of health and IT delivering improved health outcomes and transforming the patient experience. This is an excerpt from their conversation, edited for length and clarity.
What Do Band-Aids & Computers Have to Do With Health?
Clay Johnston: To start, tell me a little about how consumer wellness and big data are working together these days.
Alex Gorsky: In today’s environment, where there are just so many opportunities in health care and technology coming together in new and innovative ways, it’s almost impossible to not be collaborating. We’re working with IBM and Watson [IBM’s cognitive computing system] on how we incorporate artificial intelligence and other tools to improve early research and development of pharmaceutical products. How do we use these things to better detect adverse events and give early signals when we see them happening? How do we use them to better identify patient populations where they could be applied?
We’re also doing work around orthopaedics — how can we take a more holistic, patient-centered approach versus the typical approach where you go and you see the doctor just for the immediate problem? How can you intervene from the very beginning, all the way through to when someone’s on their way to recovery?
Ginni Rometty: I have a very strong belief that data will be at the center of helping solve many of health care’s issues. We’re going to need technology. Alex used the word Watson — for us this is cognitive technologies that understand, reason and learn. With the volume of medical data growing at its current rate — I don’t know the latest stat, but last I saw, it doubles every 75 days — you have no chance with traditional technology to ever understand what it means because you can’t program around it. You need a technology that understands.
What Alex and I are doing together is something called the Watson Health Cloud. The idea is for a platform where we send lots of data. You put that together with artificial intelligence and other technologies and you can really do something, like use the data to marry the behavioral and the clinical together in predicting what’s going to give the individual the best chance of success over a condition. I don’t know, Alex, if you think of it this way: It changes what we’re offering from a product to a service.
AG: Absolutely. It’s all predicated on having the right data, being able to collect it, being able to analyze it, being able to transform that data into information that can ultimately be put into terms that can help the patient understand the outcome, can help the system better integrate the care and can even develop new reimbursement models. So I think it’s pretty exciting.
CJ: In that transition from product to service, you’re starting to mix two different areas — the provider area, with physicians at the center, and the product area. Providers are going to determine how products are ultimately introduced into the system and used.
AG: In many ways, health care has been procedure-based, has been volume-based, has been focused on the particular product versus an outcome- or episode-of-care-based approach. I think that what’s going to result are not only product companies but also insurers, reimbursements, large provider systems all working together and creating new kinds of partnerships.
CJ: So a movement to value-based care and a movement to population health models of care are what you're talking about, fundamentally.
GR: But also you’re going to solve problems you’ve never solved before.
CEO, Johnson & Johnson
Why Aren’t Doctors More Like ATMs?
CJ: You know, I can use my ATM card anywhere in the world and when I do that, it accesses private, personal, confidential information about me.
To make the kinds of changes that we’re talking about in population health — to be able to measure outcomes — we have to know where patients are treated, and to adequately care for people we need to know what allergies they have, what medicines they’re on, what’s been tried previously. The excuses are always the same — that policies are hard to get around, the data is complex, providers want to control the data, health record companies want to control the data. Where is the solution?
GR: We have to start with the question of privacy and security. I would offer three ideas about that. First, a strong public cloud actually makes data more safe. Any time you can have all your standards in one place, you can create something that is safer.
I use a health analogy for how you have to approach security and privacy in this industry. Think of data like an immune system. You have to assume there are going to be germs. And if they act up, what stops them from getting too crazy is your immune system. It’s a simple but, I think, a really good analogy for the security that you need — big data and analytics that are constantly looking for just the smallest thing out of place. So you start with a strong public cloud, you put in these very sophisticated analytics systems to watch it and then you’ve got to be able to have control of data — of who can access it physically, where it’s located and the ability to isolate what happens with it. If you’re a company or if you’re an individual, your data and medical record are yours. You have to be able to control that.
CJ: You said start with strong public clouds, but my question is how do we get to that strong public cloud.
GR: I have one for you.
CJ: But does your cloud include my health data?
GR: Only if you choose to put it there.
CJ: See, there needs to be a cloud that includes my health data unless I don’t want it there.
GR: How would you do that, though?
CJ: That’s what I’m asking you.
GR: Do you want information there without your permission? I don’t think you do.
CJ: As a physician, I see the mistakes that are made when we care for patients without complete information about their health, and in fact we make far more mistakes than we do any harm in having access to information.
GR: With respect to what you just described, I would agree.
AG: To build on that — I think from a patient point of view having a strong, flexible and clearly secure environment is absolutely essential. But how do we do that in a way that allows access to the right people? Ultimately, you have two extremes here. One is population-based health and the other is individualization and personalization.
CJ: Yeah, I guess we’re getting to the two different missions that we’re all talking about. One is using anonymized data to develop new products, new tools, new insights. The second is actually using data to drive population health, where we’re paying for outcomes — recognizing that the individual data that we have on patients is quite useful to improving those very same outcomes.
GR: I think that’s already happening. If I’m a diabetic and I have my monitor, it can predict hypoglycemia 3 to 4 hours in advance. I know what to do then, what to eat. The health care provider becomes confident that — you know what, I can keep you out of the hospital because I now use this data and I see a patient changing behavior.
I’m confident enough that I’ll therefore price care at a certain level because of that. That’s where we’re headed — that mixture between very personalized and value-based care.
When Do the Robots Take Over?
CJ: Broadly, artificial intelligence is seen as facilitating the role of the physician. Obviously, it’s not the robot treating a patient, it’s somebody controlling the robot.
But just as we’re hearing about autonomously driving vehicles, might there be a future in which more and more is automated?
GR: I wish it hadn’t been called artificial intelligence. I wish it had been called augmented intelligence. Every doctor I know wants to spend more time “doctoring.” Instead, they are spending time gathering data, trying to put it together. I believe we’re going toward an era where you will be able to do a better job — you will be able to do more things — with technology that assists. We are far, far off from a time when technologies are self-aware. They are taught.
AG: We’re living in an age where there is remarkable technology. But at the end of the day, the personal touch is what matters. The more we can help make the system more efficient, more effective — so that we can have more one-on-one time with patients — the better. As you know as a physician, even the best science doesn’t always lead to the best outcome at a personal level. Making sure that we never ever lose sight of that amid all this great technology is essential.
CJ: The reality is that physicians can’t do everything. In order to be more effective in their roles, they can’t be distracted by technologies, and that’s what they are today. Electronic health records are a total disaster. Physicians want to spend more meaningful time with patients.
The other reality is the placebo effect is very real — and it is driven by the relationship between the patient and the doc. By that, what I mean is the emotional state that the patient is in determines the success of surgery, even more than the operational skills of the surgeon.
AG: It’s very personal, as it should be.