Monday, January 26, 2015

The Internet of (Virtual) Things

There has been a lot of buzz about the so-called Internet of Things, which tech gurus like Cisco CEO John Chambers have pegged as a potential $19 trillion market.  You know, a world where everything is connected -- e.g., your car talks to your house, which talks to your refrigerator, which updates the grocery list on your smartphone.  And, of course, we'll all be wearing sensors that track and report our activity and our vitals.

That may all happen, but we may also be paying more attention to things that aren't really there  -- through virtual reality.

It seems like much has happened in virtual reality over the past few months, one of the most notable of which was Facebook putting a price on it through its $2b purchase of tiny Oculus.  Last week Microsoft may have kept itself relevant for the next decade by its announcement of HoloLens, which it describes as "holographic computing" but everyone else is calling virtual reality.

Reviewers who got to try out HoloLens wrote about not just the kinds of experiences one might expect in VR -- like being on Mars or playing Minecraft -- but also overlaying virtual images into the "real world."  For example, using Skype to have a virtually present expert walk a user through changing a light switch, complete with visible pointers and as-needed instructions.

CNET used these examples to contrast HoloLens with Oculus Rift, pointing out that the latter tricks your eyes and mind into thinking you are someplace else, but HoloLens augments the reality you are actually in.  That's a difference between gaming and real life, and it opens up lots of possibilities for when/how one might use virtual reality.

The news and entertainment world is already starting to realize the potential of virtual reality.  Vice News has unveiled a service called Vrze, who hopes to immerse viewers directly into coverage of news stories.  This year's Sundance Film Festival was dominated but by VR efforts such as Birdly, which let viewers experience the world literally through a bird's eye view.  This isn't 3-D trying to win over 2-D efforts; this is potentially a true changing of the guard, like talkies replacing silent films or television relegating radio to an also-run media.

Moviegoers never quite cottoned to wearing those 3-D glasses, and many of us don't relish wearing some clunky VR set-up either, but suddenly Google Glass starts to make a lot more sense (even though Google just pulled its consumer version off the market...for now).  Google has another, cheaper VR solution in the works already -- Cardboard.  As Patrick Buckley, CEO of DODOcase, told NBC News:
Where we are in the whole VR space, consumers need a Model T Ford, they don't need a Lamborghini.  There are 2 billion smartphones in the world that are basically VR devices, and consumers don't realize it.
Meanwhile, Mozilla is trying to bring the VR experience to the browser, further illustrating that we're not as far from using existing kinds of technology to take us to new virtual places.

Right now we're replete with technological options; lots of screens, lots of computing power, lots of keyboards.  The smartphone has gobbled up many functions, including phone, camera, music player, entertainment center, and Internet browser, but we still also have PCs, laptops, tablets, phablets, 2-in-1 computers, various trackers, and standalone gaming consoles like Xbox or Playstation.

My take is that virtual reality will wipe those distinctions away.

We're at a technological point not dissimilar to the 1990s, when we were still walking around with cell phones, music players, cameras, a handheld gaming consoles, maybe a pager or a Blackberry.  It seems archaic now, maybe even a little foolish, with all those subsumed in our smartphones.  In five or ten years the multiple devices that we now use may seem ridiculous as well.

You can already buy a laser projection virtual keyboard, so you don't have to fiddle with the tiny virtual keyboard on your smartphone or tablet.  There are already virtual touchscreens, such as those offered by Displair, that can literally create the screen out of thin air.  If you can have your screen and your keyboard projected anywhere and of any size you want, what do you care what the underlying device is?

So I ask: why would we use multiple devices?  You'll need something with an Internet connection, and at least some modest computing power, but those don't have to come from something we'd recognize as a smartphone or computer.

Indeed, our clothes may be our "device," and anything we're now doing on one of our various devices will get streamed from the cloud and projected for us using virtual reality.  I've said before that apps seem like a very clunky technological solution that will get superseded by a simpler, more consolidated approach, and I'm similarly saying that our devices will go the same way.

That's the kind of thing that virtual reality will be able to do for us.

So what does all this have to do with health care?  Plenty.  Respected experts like Eric Topol have proclaimed that The Future of Medicine Is in Your Smartphone, and it won't be long before whatever we can do (or want to do) on a smartphone, we will be able to do in VR.

We're already seeing virtual reality being used to train physicians (e.g., surgical or other procedures) and even for PT/rehab.  If I was WebMD or any of the other health content providers I'd be rapidly retooling my patient education materials to include VR; why read a paragraph or watch a video when you can see the problem and its treatments in VR?

The other obvious use for VR is in telemedicine.  Instead of a video chat between a provider and patient, the consultation could take place in a virtual exam room, or in the patient's augmented living room, for that matter.  Think of the HoloLens demo where the expert guided the user to replace a light switch, and imagine the possibilities for patient education and treatment.  All those remote monitoring and smartphone-based diagnostic options that Dr. Topol evangelizes about should still be available in VR, giving the patient and the clinician a powerful set of tools to work with.  

You can bet that Microsoft is furiously working to upgrade Skype to try to be the preeminent VR tool.

With virtual reality, we have to recognize that interactions will be different, the concept of place will be different, and what we can do with and for information will be different.  It will require new approaches and new flexibility, and I suspect most organizations are already behind the curve.

I've previously discussed my fondness for holographic medical records and virtual assistants, and, as a result, I've gotten some teasing for the Star Trek-like similarities (e.g., TNG's holodeck or  Voyager's virtual doctor).  The future almost never happens in quite the way we predict, but that doesn't mean we're still not going to be amazed...and with VR it is happening now.

Monday, January 19, 2015

Making the Old New Again

I always love it when someone looks at something familiar in a completely new way.  I only wish health care had more examples of that.

The example of this kind of totally fresh thinking that caught my eye concerns traffic lights.  Yes, traffic lights.  We all know traffic lights.  Most of us have stopped at traffic lights (and sped through some yellows!), probably more often than we'd like.  They've been around in virtually the same format for over a hundred years now, and are indelibly part of the urban/suburban landscape.  Cities without them would seem like chaotic third world cities.

Well, if researchers from Carnegie Mellon University, led by Professor Ozan Tonguz, have their way, those familiar yellow boxes with the lights could become unnecessary.

The CMU researchers have developed "virtual traffic lights" (not to be confused with the separate CMU "smart traffic signals" project).  Instead of using physical traffic lights, lights would show up on the driver's dashboard as needed.  As Professor Tonguz told CNN: "With this technology, traffic lights will be created on demand when [two cars] are trying to cross this intersection, and they will be turned down as soon as we don't need it,"

The researchers claim the virtual, on-demand signal could reduce commuting times by 40%, as well as reduce carbon emissions and accidents.  And, of course, we wouldn't need all those physical lights; think of the savings on new lights, poles, and wires, plus on ongoing maintenance.

All that would be required is that every car -- and that means, every car -- is equipped with the required vehicle-to-vehicle communications technology.  No small task!  Some think this could happen in a year or two, others a decade or two.  Either way, it's mind-blowing to think that such a familiar part of our driving experience could be so utterly transformed by what seems, in retrospect, such an obvious solution.

Of course, with driverless cars coming on quickly, we may not even need the visible virtual lights.

Let's contrast this kind of thinking with health care.  You go to your doctor's office, and chances are he/she will come into the exam room wearing the usual white lab coat, a stethoscope around his/her neck, and with your medical record.  The lab coat goes back a hundred years, originally intended as a symbol that you really are dealing with a physician and yet now serving more as an incubator of germs; the stethoscope goes back 200 years with only minor changes since then; and, while the medical record may be kept in an electronic format now, the fundamental content and presentation hasn't really changed from the traditional paper versions.

Yes, I know -- health care has plenty of new technology and many kinds of improved treatments, but I'm not sure we're getting a lot of reinventing.  Where are our virtual traffic lights?

One small -- well, maybe not so small at that -- health care example is a new patient tracking system called PatientStormTracker, developed by Lyntek Medical.  As the name suggests, PatientStormTracker borrows from weather tracking to present patient monitoring data as systemic color monitoring.  Instead of trying to follow the usual rows and rows of data, clinicians can actually see a patient's status -- color-coded -- and watch it progress in real time, including which body systems are currently being impacted and how much.  

The video is pretty cool.  

Lyntek's founder and CEO, Dr. Laurence Lynn, told The Columbus Dispatch that traditional patient monitoring is like a fire alarm -- either on or off.  As he said: "We have this simple fire alarm idea that existed from the 1980s, and it didn’t evolve, it didn’t improve."  I think saying it only dates to the 1980's is being generous.

Dr. Lynn wants to monitor patterns and detect trends earlier, when interventions are more likely to be effective.  "People don’t just die of some instantaneous thing,” Dr. Lynn said. “There’s an evolution of instability that is often unrecognized.”

PatientStormTracker is in clinical trials.  

One proponent of radical changes in health care has long been Dr. Eric Topol, who happens to have a new book out (The Patient Will See You Now: The Future of Medicine Is In Your Hands).  I have not yet read his book, but I did read his related op-ed in The Wall Street Journal.  His version of virtual traffic lights, if you will, is the smartphone.

Dr. Topol outlines not just increasingly common functions like virtual visits or monitoring using a smartphone, but also apps that assist with testing and even diagnosis.  He gives the example of an app that allows you to take a picture of a worrisome rash, and have a computer algorithm generate a message that suggests what you should do about it -- instead of worrying for weeks while you wait to consult a dermatologist.

I especially like his prediction that wearable sensors will make it possible that "...except for ICUs, operating rooms and emergency rooms, hospitals of the future are likely to be roomless data surveillance centers for remote patient monitoring."  That would certainly upend how we view hospitals...finally.

Perhaps those remote patient monitors will use something like PatientStormTracker.

Dr. Topol is not alone in proclaiming this DIT (do-it-yourself) movement; PwC listed it as one of their top trend healthcare trends for 2015.  As their report says: "“Apps formularies,” smartphone plug-ins and intuitive devices may become as important to clinicians as the prescription pad was to an MD in 1960."

Of course, even that prescription pad is likely to be smartphone-based now too, or at least electronic.

The smartphone technology options are cool, but what Dr. Topol sees as an even more important trend in putting all the newly-captured data in the cloud, mining it, and using it to target interventions. thereby "transforming it [medicine] from a weakly evidence-based practice to a data science, with empowered individuals at center stage."  Sadly, both the data science and truly having individuals at center stage would be pretty radical changes.

Changes are going to come at us from seemingly left field.  We can never be quite sure where they will lead.  Apple wanting to sell music online led to the iTunes store which led to selling apps in it which led to smartphones exploding in popularity which is now leading to using Apple Pay instead of credit cards, and now ATM cards and drivers licenses are also starting to follow the smartphone trail. Who'd have guessed -- and who can tell where it will lead next?

It just takes some innovator to see the familiar in a different way -- and then manage to convince us, and the medical-industrial complex, to change.  I fear the convincing is harder than the seeing.

Hey, I'm still waiting -- and hoping -- to see the holographic medical record I've suggested before...

Monday, January 12, 2015

The Right to Make Bad Choices

We talk a lot about making our health care system "patient-centered" and "empowering patients," but sometimes it seems that means only when they make choices we like.

Take, as Exhibit 1, the case of Cassandra C.

In case you have not been following the story, Cassandra C is a 17-year-old who was diagnosed with Hodgkin's lymphoma last September, and chemotherapy was recommended.  According to her recount of the saga, she and her mother wanted to get a second opinion, but the time spent doing so triggered her mother being reported to the Department of Children and Families.  Cassandra was placed in a foster home, and was only returned to her mother with the promise to undergo chemotherapy, which she did for two days.  That was enough for her; she ran away from home to escape the treatment.

Cassandra returned home after a week, and was hospitalized -- confined to a room with limited freedom.  The hospital subsequently resumed chemotherapy against her wishes.  As she writes: "I was strapped to a bed by my wrists and ankles and sedated. I woke up in the recovery room with a port surgically placed in my chest. I was outraged and felt completely violated."

Last week the Connecticut Supreme Court ruled that she had failed to prove she was a "mature minor" able to make her own medical decisions (Connecticut doesn't even have a "mature minor" law, as is also true of 32 other states), thus allowing the chemo to continue.

I'd probably feel different about Cassandra's situation if her mother was urging her to get the chemo, but she's not.  It's got to be impossibly hard to support your daughter when she is willing to risk her life on principle, but when they both agree on this painful choice, who are we to disagree?

The good news, such as it is, is that experts say her lymphoma is highly treatable.  They claim the chemo gives her an 85% chance of survival, versus a life expectancy of only two years without it.  Certainly the state, and the medical professionals involved in her care, are acting in her best interests, right?

I wonder where the court and her doctors would draw the line.  What if, say, instead of the 85% success rate the chemo only had a 50/50 chance of saving her?  Or 25%?  Or what if, instead of curing her, it "only" offered her an extra six months of life?  Would they still insist on a painful course of action Cassandra didn't want?

The court ruling on Cassandra's dilemma provides new opportunities for forced interventions.  Why shouldn't the authorities be able to prescribe bariatric surgery for morbidly obese teens?  Why not unilaterally put teen drug addicts -- or smokers -- in rehab?  Why not force the hospitalization of high-risk pregnant teens to help reduce infant mortality?

Yes, there are lots of interventions that the court ruling would seem to allow.  And they'd be wrong too.

While I'm sure all the parties are doing what they think is best for Cassandra, the point is that it shouldn't be anyone's choice other than hers.  Yes, 17 year-olds do a lot of immature things, some of which can be life-threatening (e.g., texting while driving).  Cassandra probably didn't help demonstrate her maturity by running away.  But here's her take on the situation:
This experience has been a continuous nightmare. I want the right to make my medical decisions. It's disgusting that I'm fighting for a right that I and anyone in my situation should already have. This is my life and my body, not DCF's and not the state's. I am a human — I should be able to decide if I do or don't want chemotherapy. Whether I live 17 years or 100 years should not be anyone's choice but mine.
As Cassandra also says, "I care about the quality of my life, not just the quantity."

If any of that sounds like an immature person not able to make and articulate her own choices, I know a number of so-called adults that the Connecticut Supreme Court should take some rights away from.

The problem isn't just about Cassandra and Connecticut, nor about teenagers below the legal age of informed consent.  It's much broader than that.  As Exhibit 2, consider advanced directives.

Last summer The New York Times wrote a disturbing piece about how advanced directives are often ignored.  It cited a national study that concluded that having an advanced directive has little effect on whether or how often people were hospitalized, or whether they died in the hospital.  Whether your end-of-life preferences will be honored has more to do with how the physicians in your area practice, a problem the Dartmouth Atlas has been hammering home on more generally for the past 30 years.

Many health care providers worry about being sued for failing to do all that they can for patients, and there may be some truth to that -- although probably not enough to justify the scale on which patient wishes are ignored.  It's more likely to be an artifact of our expectations, both on the part of patients and health care providers. Modern medicine gives us many powerful weapons, but -- as any Spiderman fan knows -- with great power comes great responsibility.

We all see people every day who make what we think are bad choices, and we may wish we had the power to decide for them.  But, fortunately, we don't live in a society where we can usually do that.  Reason with them, cajole them, make sure they are fully informed about the consequences of their choices -- sure, we should do all those, and I hope all that has been done for Cassandra.  But the bottom line for me is that giving people choice means sometimes they will make bad choices, or at least choices with which we don't agree.

If Cassandra gets through the chemo and beats the Hodgkin's, she may go on to live a happy life.  Decades from now, she may look back on all this and think, gosh, if I'd gotten my way, I'd be long dead and missed so much.  Proponents of her forced treatment might then feel vindicated.  But you don't get do-overs in life; if we take the road of taking health decisions away from people, it's hard to see where it will head.  Nowhere I want to go, anyway.

In Greek mythology, Cassandra had the gift of prophecy, but was cursed in that no one would believe her predictions.  Cassandra C. may be warning us about our future, and I hope we pay attention.

Monday, January 5, 2015

The Adjacent Possible Is Closer Than You Think

Here we are at the start of 2015, with various recaps of 2014 and predictions for 2015.  Those are always interesting to read, but what I usually have the most fun writing about is the "adjacent possible" for health care.

The term is one from biology, usually attributed to Stuart Kaufman, and popularized by Stephen Johnson in his 2011 book on innovation Where Good Ideas Come From.  At the risk of oversimplifying both men's work, the concept suggests that change is constrained by the available "neighborhood" -- of ideas, materials, etc.  Johnson illustrates, in that book and again in his most recent How We Got to Now, how new ideas are often borrowed from other fields and applied with unexpected results.

However, if the idea is too new, and/or doesn't have the necessary infrastructure or mind-set to nourish it, it can quickly die a forgotten death.  E.g., Edison was far from the first to invent the electric light, but he was the first to also provide the supporting electricity grid.

Take the auto industry.  It is terrified that millennials appear to care less about driving, car ownership, and even getting drivers licenses, and so is trying to keep itself relevant by making vehicles more technologically sophisticated.  "Cars are transforming into digital devices," according to Joe White of The Wall Street Journal, citing new infotainment and collision avoidance systems.   The New York Times profiled how automakers are furiously trying to make cars more tablet-like.  Rob Csonger of technology company Nvidia tells Fortune:
"The car is rapidly going to go from the most stupid electronic device a consumer owns to the most powerful supercomputer a consumer will ever own -- way more powerful and sophisticated than your phone, tablet, or PC."
Still, in a world of driverless cars, as-needed services like Zipcar or  Car2Go, and wirelessly connected cars that always know your preferences, the entire paradigm of owning your own car suddenly becomes at risk.  And auto makers know it.

Then there's health care.  While automakers are busy trying to turn their products into digital devices, we're still trying to get consumers access to their own health records.  According to a recent survey from Xerox, 64% of Americans don't have online access to their records, although most would like to.

With our hospital-oriented system, we're still selling consumers the equivalent of those clunky 1970's Detroit gas guzzlers.  Sure, we're doing some cool things with apps and remote monitoring, but, with a few exceptions (e.g., cardiac patients or diabetes monitoring) they have as much to do with mainstream care as putting a hula girl on the dashboard did for driving those old cars.  As evidence, a recent survey of 20,000 physicians found that only 15% had been asked by patients about incorporating data from wearable trackers or other health apps into their record.

The auto industry has also seen transparency totally change the buying experience for consumers, leveling out the information asymmetry dealers used to enjoy.  Health care?  Umm, we're still working on that...if we could ever agree what the "product" really was, how we should measure quality, or what the true price might be for a given consumer. 
I don't mean to single out the auto industry as the best one one to emulate in searching for the adjacent possible, nor do I limit that adjacent possible to technology.  I use these examples because I worry that health care isn't even looking in the right places for its adjacent possible.

Here's what scares me: 87% of enrollees in the federal exchanges are getting subsidies.  That sounds like good news, that ACA is helping make insurance more affordable, but it reminds me that care is so expensive that our financing system for it is not affordable for most people.  Add to that the 48% of the population who get coverage through their employer (and thus benefit from the tax preference) or the 15% of the population with Medicare (and thus are subsidized by the inter-generational tax funding), and it turns out that very few people actually know, much less pay, the full price of their own coverage (coverage numbers from Kaiser Family Foundation).

Our model is an expensive, intervention-oriented, medical approach.  As examples, a recent study in JAMA Internal Medicine found that high risk heart patients actually seem to fare better when their cardiologists aren't around to treat them, or Dr. Laura Esserman's assertion that we are grossly overtreating breast cancers.

Meanwhile, health care is spending a lot of effort borrowing ideas for improving customer service/the patient experience -- e.g., Sue Schade's ideas for the inpatient experience or Micah Solomon's suggestion to focus on timeliness -- but I'm less sure we're looking to reinvent the underlying model.  We continue to make progress in new treatment options, but not so much in business models.

Many believe that payment reform -- the so-called value-based purchasing -- is the solution.  I worry that payment reform at best will simply prove to redistribute the existing money differently (see Alan Weil's partly-tongue-in-cheek Why I Oppose Payment Reform).  

Others see the future in direct primary care, sometimes known as concierge medicine.  I like the concept, especially if divorced from health insurance, but primary care capitation in the 1990's showed that the trouble is what care "leaks" outside primary care and ends up in the vastly more expensive hospital/specialty care worlds.  

The always interesting Margalit Gur-Arie argues that what direct primary care really does is convince people that primary care is a cheap commodity, thus accelerating primary care's decline.  As she says: "Strangely enough, I don’t see too many dermatologists stepping all over each other to convince us that what they do is worth very little money."  She sees physician informaticists as the new general practitioners, advising that: "modern physicians, who want to become rich and famous rapidly, should find something to do with computers."

We need new paradigms, and they're probably not going to come from within health care.  We should be asking radical questions like, ok, if health care spending had to be limited to 50% of what it is now, what would we do differently?  As I wrote about in Who Ya Gonna Call, we need to be open to adjacent possibles where care looks very different and where the training and practice of the people giving that care also are very different.  It could be a whole new ball game.

And, honestly, I don't see much future for a middle ground between a system that is directly consumer-purchased or one that is totally government financed.  That's the middle ground we've been trying to stand on, and some adjacent possible is eventually going to generate an earthquake that swallows it up.

We need to keep in mind that the adjacent possible is already here, somewhere, in some form; we just haven't realized it yet.  The question is, who will see it first?