Tuesday, January 26, 2016

Doing Different Differently

I was all set to write about bacteriophages, both because I hadn't known much about them until recently and because they represent an approach that doesn't just proliferate the antibiotic arms race.  Then I realized that what appealed to me about them was as an example of attacking mainstream problems with non-mainstream solutions, so I decided to write more generally about how organizations are trying to encourage that.

Because, goodness knows, health care needs a lot more of them.

Let's start with IBM.  Yes, I said IBM.  Believe it or not, Big Blue is trying very hard to reinvent itself as a design company, or at least a company that uses "design thinking" to develop products and services.  They're investing over $100 million in the effort, dropping their ratio of designers to coders from 1:80 to 1:15, according to Wired.

Their design principles emphasize "making users your North Star," using collaborative multidisciplinary teams, "restless reinvention," and a continuous loop of "observe/reflect/make."  I particularly like the restless reinvention principle, with its great reminder that "everything is a prototype" and with its explicit admonition to "solve problems in new ways."

So far, about 10,000 employees have gone through the design bootcamp, and around 100 products have been developed using design thinking.  Those are drops in the bucket for IBM, but the approach is an audacious and, one might say, long overdue attempt for IBM to stay relevant in a millennium in which Apple has reminded companies about the importance of design.

Or take Microsoft.  If there is any doubt that Microsoft is well on its way to doing things differently, look at the Surface Book or Surface Pro, each of which has won rave reviews that might make Apple itself envious.  CEO Satya Nadella has been shaking things up ever since he took over two years ago.

One of Mr. Nadella's key actions was to break up Microsoft's Research group, which historically had been kept separate from the day-to-day action.  Bloomberg reports that Mr. Nadella has insisted that the research teams work hand-in-hand with the product teams to get new ideas into actual products quicker.  They focus less on pure research and more on making an impact on products.

Mr. Nadella has emphasized, "we need to be open to new ideas, and Microsoft Research is where they will come from."  This attitude led to Skype Translator becoming an actual product within three months of Mr. Nadella learning about the underlying research, a time frame that was previously unthinkable for Microsoft.

It ain't Bill Gates' Microsoft.

Venture capitalist Anshu Storm has a theory -- "stack fallacy" -- that he believes explains why so many big companies fail to innovate.  The theory posits that many companies suffer from the "mistaken belief that it is trivial to build the layer above yours."

He cites how Apple has built great devices but also has missed on simple apps, or IBM's classic blindspot about letting Microsoft own the OS layer that ran their PCs.  Then there is Google with its efforts to expand into social networks (Google+, anyone?).

Mr. Storm explains that "we (over) value what we know," which can lead to a lack of understanding about end users' needs.  In his view, "Product management is the art of knowing what to build."  The trouble is that too many companies focus on the how and not enough on the "why."

Christopher Mims, writing about Mr. Storm's theory, says that they key to avoiding stack fallacy "is figuring out how to have true, firsthand empathy for the needs of the customer for whatever product you’re trying to build next."  That sounds logical, even obvious, but evidently is hard to actually do well.

For example, think about hospitals.  They're trying hard to position themselves as patient-centered health systems, but no one who has been in a hospital can believe that hospitals see patients as the customer.  Hospital gowns?  Waking patients up in the middle of the night to take vitals?  Corridors upon winding corridors?

Sadly, I could go on, and not just about hospitals, but the point should be clear.

The customers hospitals cater to are physicians, who bring them patients and thus revenue.  That's not to say that the people working in hospitals don't care about patients, but the processes, procedures, and design clearly aren't about putting patients first.  You can't just proclaim you are "patient-centered."

Still, Tom Graham, M.D., the former Chief Innovation Officer at The Cleveland Clinics, thinks health systems and academic research centers will be the "logical crucibles" of innovation in health care.  He emphasizes the importance of collaboration, which hits health systems' and academic research centers' sweet spot, since they already have "interdependent communities with nuclei of creative thought."  His new book, Innovation the Cleveland Clinic Way: Powering Transformation by Putting Ideas to Work, details his thoughts.

I worry that putting a bunch of smart health care people together may generate more innovation, but that the resulting innovation may look like, well, more health care ideas.  That's not what we need.

We need the health care experience to be less like health care and more like things we actually like.  Nick de la Mare suggests that hospitals (and schools) "should be more like theme parks," and that designers should be aiming for "magical experiences."  As he asks, why shouldn't health care settings be the "happiest places on earth?"

That's the attitude we need to be taking as we try to innovate; it's not just doing more, but really rethinking the overall consumer experience.  I was particularly struck by Mr. de la Mare's caution:
The trick is to deploy technology strategically and sparingly, since new tools tend to introduce unintended complexities....A hospital patient may feel similarly overwhelmed by impersonal and bureaucratic processes that seem to serve the health care provider at their expense. Just because we have the technology to do something, doesn’t mean we should.
I especially love that last sentence.

There is cool innovation going on within health care.  David Chase, for example, raves about how Zoom+ (which I've written about before) has revamped the ER experience, and there is no shortage of other health care companies hoping to be disruptive (e.g., Becker's list of 30).  These may not reinvent the consumer experience, but at least they may improve it.

There is plenty of incremental innovation going on, and health care sure can use it, but I continue to be on the lookout for breath-taking innovation -- innovations that surprise, excite, and delight.

Sunday, January 17, 2016

My Digital Twin Is Worth More Than I Am

It used to be said that the human body was only worth pennies, based on its minerals (although that has been updated to $160),  If one counts the value of our organs, the figure is several hundred thousand dollars.  And, according to other estimates, if one includes the value of our DNA and bone marrow, a human body is actually worth close to $46 million.

The funny thing is, other people seem to get all the benefit.

What started me thinking about this was a report from MobiHealthNews on the FTC's PrivacyCon, particularly a discussion about the monetization of consumer health data.  We talk a lot about Big Data, but it is worth remembering that Big Data is made up of lots of little data from people like you and me.

Companies are keen to make money from the data they collect about/from us.  The Economist Intelligence Unit just released The Business of Data on this topic.  Some of its key findings
  • Over 80% of surveyed organizations are taking steps to generate more value from their data.
  • 60% are already generating revenue from their data.
  • Over 50% are selling or buying data.
  • 86% say their customers trust them with their personal data, although only 34% think they are very effective in being transparent about their use of it.
  • 82% believe they are effective about keeping their data secure, although 34% admit to having had a significant breach within the past 12 months.
The apparent discontinuities in those last two bullets would be amusing if they weren't so scary.

The Pew Research Center also just released a new report, Privacy and Information Sharing.  They found consumers have mixed attitudes towards sharing their data; "it depends" was the prevailing attitude when evaluating various sharing scenarios.  They often see benefit in giving up some of their privacy -- we love "free"! -- but worry about how their data will be used and by whom, especially given the risks of security breaches.

One of the scenarios Pew presented was having your doctor upload your personal health data to a secure third party site, in return for access to your medical record and easier appointment scheduling.  Somewhat surprisingly, more respondents (52%) found this sharing acceptable than in any of the other, non-health scenarios,  Those who found it unacceptable tended to cite doubt about how secure the data really would be, noting the consequences if their data was hacked.

This concern is certainly valid.  Being hacked is bad enough, but even having your data de-identified may not be enough, since whether de-identified data can stay de-identified is very much in question, as datasets get larger and analytic techniques get better.  

As a researcher at PrivacyCon said:
What we have to realize is that genetic data is the most personal data out there...We also know this data is inherently identifiable.  There's growing recognition that it is not possible to de-identify this data in a way that is not possible to re-identify later.  The other thing is, this data is irrevocable.  If there's a privacy breach, you can't change it.  It's not like your iTunes password.
We're increasingly in a world of biometric screening and DNA sequencing (including DTC sequencing).  We are starting to use biometrics as a form of identification, as DNA has been used for many years, most notably in law enforcement.  We've gotten used to the prospect of identity theft when it comes to our financial information, but it takes on even more ominous connotations when it comes to our genetic and health information.

In a very real sense, we are our data, with DNA a fantastically effective means of storage.  Having this data stolen is especially troubling.  I wouldn't want to suddenly run into my clone.

One way or another, Big Data is posed to become a huge market.  Pharma in particular stands to benefit, with Big Data potentially revolutionizing drug development and targeting.  All of this relies on our data.  We may directly benefit from our sharing, or the benefits may accrue to other patients, particularly future ones.

Many people (including President Obama) believe precision/personalized medicine is the future of health care, tailoring treatments and even diet to a person's specific make-up.  The success of this depends on Big Data.  The President has also pushed for a "moonshot" to cure cancer, and many feel that not only does this need Big Data but also that lack of sharing data has been one of the big problems in cancer research (and, indeed, in all clinical trials).

GE's Digital CEO Bill Ruh believes that a detailed software model of a person, which he calls a "digital twin," will become integral to health care (as well as to products in other fields).  As he said,
I believe we we will have a digital twin at birth, and it will take data off of the sensors everybody is running, and that digital twin will predict things for us about disease and cancer and other things.  I believe we will end up with health care being the ultimate digital twin.  Without it, I believe we will have data but with no outcome, or value.
Whether we like it or not, whether we realize it or not, monetization of our data is happening,  While research suggests we are overwhelmingly willing to share our health information for research, once the implicit value of our contributions becomes clearer, that willingness may be more conditioned upon an explicit return.

Take, for example, Datacoup.  They are one of the first companies to allow consumers sell their data.  Consumers upload their data to Datacoup, which then markets it to data purchasers, with the consumers getting paid based on how much those purchasers pay.  Datacoup's CEO told the EIU: "If merchants are willing to provide value in exchange for more or better consumer consented data, then you’ll see a vibrant and massive marketplace spawned for the direct exchange of data, and in effect a more direct relationship between consumers and merchants."

I'm all for pooling our data to figure out better ways of helping people, but I'm less sure that I should be doing that for free, especially given the risks of my data being stolen or used for purposes I didn't intend.

If our digital twins are going to have economic value -- as they no doubt will -- shouldn't we share in that?

Monday, January 11, 2016

Self-Driving Cars As The Future of Health Care

I'm a sucker for driverless cars.  I've only owned cars with manual transmissions, and in recent years I've been telling people that I'm more likely to jump straight to a self-driving car than to simply get an automatic transmission.  There's lots going on in the field these days, but what caught my eye was a new report from KPMG that forecast the number of auto accidents could drop 80% due to autonomous cars, shrinking the auto insurance industry by some 60%.  They are not alone in these kind of predictions.

We should be talking about those kinds of changes for health insurance, and health care more generally.

Most of us drive, and most of us probably think we're good drivers -- it's all those other bad drivers! -- but over 90% of auto accidents are caused by human error.  No wonder; there are so many other things we'd rather be doing, like texting or talking on our phone.  Maybe even checking our destination on Yelp or Google Maps.  They even have a name for this -- distracted driving -- with plenty of grim statistics that you'd think would deter us, but which do not.

Google has gotten lots of press for its driverless car project, but we've been heading this direction for some time.  For example, cruise control and anti-lock brakes have been routine for decades, but now there are also self-parking cars and ones that can alert you when other cars or objects are too close.  Tesla is in beta with a feature to let your car essentially valet itself (although they've already had to put further restrictions on it), and Elton Musk thinks fully autonomous cars will be ready in 2-3 years.

All this is in line with KPMG's prediction that by 2025 we'll be in an autonomous car world, with consumers shifting over to it by 2040.  That's when the auto insurance industry will be in big trouble, at least in its current form.

The industry executives KPMG surveyed aren't worried yet.  Only 29% felt they were very knowledgeable about autonomous cars, with 23% claiming to know nothing about them.  Most felt that their business wouldn't be impacted for at least ten years.  When the changes come, the executives thought that their focus would move more towards commercial, instead of personal, auto lines, and they expected new competitors would include tech companies like Google and the auto manufacturers themselves.

Perhaps it is no surprise, then, that GM just invested $500 million in Lyft, hedging its bet not only on a self-driving world but one where car ownership shifts to the "Uber" (sorry, Lyft!) model where cars are available on demand, with someone else doing the driving.  That model really disrupts the traditional auto industry, as it removes much of the personal attachment most of us have historically felt about our cars.

Once it isn't our car and we're relying on another driver anyway, do we really care if that driver is the car itself?  We just want to get where we're going safely, and, oh-by-the-way, more cost-effectively.

It won't just be the auto insurance industry that will be impacted by these trends.  The whole ecosystem will be: car manufacturers, parts manufacturers, repair shops, rental companies, gas stations and oil companies, and advertising agencies. to name a few.  The changes will be bigger than we realize and sooner than we expect.

All that is very interesting, but what does it have to do with health care or health insurance?

When it comes to managing our health, we're not much better than we are with our driving.  We weigh too much, we eat the wrong things, we don't get enough exercise or sleep, we have too much stress, and when we do get medical care, we often don't listen to or understand our doctors, or simply don't follow their instructions.  If any area of our lives seems like it needs someone/something to take over for our (poor) judgement, it would seem to be in regard to our health.

We might think those auto insurance executives who believe they have ten years to get ready are being short-sighted, but no one in health care should be feeling any better about its existing model.  The seeds for similar disruption are already here for health care as well.

Telemedicine, AI advice, digital tracking, and consumer diagnostic tests -- all these are the kinds of things that have the potential to take the locus away from your friendly local physician/hospital/lab/pharmacy/imaging center.  They are slowly working their way through the laborious regulatory system and the consumer acceptance stage, just as they are for autonomous cars and the ride-sharing industry.  But it would be foolish to believe that they won't become mainstream.

Of course, short of uploading our minds to live inside an android or a robot, it would be hard to fully delegate our health to some third party.  I've previously made a pitch for an omnipresent digital health assistant that would help guide us to better health choices, but we're not quite there yet.

That's why perhaps wearables/digital health are the most likely bet in the short term.  We're seeing traditional tools like stethoscopes being replaced by digital optionslab testing being ordered by, and results delivered to, consumers, and an array of other self-monitoring options.  Two-thirds of Americans say they are willing to use digital health tools.

Products like Apple Watch, Samsung Gear, or Fitbit are all doing their best to be a ubiquitous and value-added presence in our lives.  Whether they can actually help us improve our health remains to be seen, with early results indicating...maybe.

Health care professionals think that their revenue can only go up -- here's looking at you, drug companies! -- but I'm convinced that well before the auto insurance industry suffers its big decline in revenues, health care will have already done so.

We've used "consumer-directed" as shorthand for making consumers responsible for more of the bill and "patient-centered" as shorthand for getting all the various health care professionals involved in delivering a patient's care to talk to each other, but in both cases without truly changing any of the fundamental dynamics.  Perhaps the real parallel to driving is not handing management of our health over to some third party -- which we've done too much of already -- but in actually taking control of it.

When it comes to health, the "self-driver" is us.

Monday, January 4, 2016

What If They Are Right?

Let's say you have a product that you just can't get the public to want.  You market the heck out of it, you sell a bunch of different versions of it, and you warn people about how their lives could be ruined if they don't buy it, which, in fact, isn't untrue.  You even manage to get the government to require people to buy it, and also to provide subsidies for them to purchase it.  Yet still the customers you want most remain stubbornly uninterested.  At some point most people would conclude it isn't a marketing problem but a product problem.

When the product is health insurance, though, we tend to think those who won't buy it are short-sighted and even irresponsible -- but what if they're right?

The New York Times just reported that many Americans find that it is cheaper for them to pay the fine for not having health insurance than it is to actually pay for health insurance, despite increasing fines, subsidies for the purchase, and the presumed financial benefits of having health insurance.  It cites a recent Kaiser Family Foundation analysis which found that there are some seven million such uninsured Americans.  As one resister told the Times, "I don’t see the logic behind that [buying coverage], and I’m just not going to do it."

The article gave examples of several consumers who were surprised -- in one case, "offended" -- that the plans featured large deductibles that would require them to pay the initial several thousand dollars of medical expenses.  If something truly catastrophic happened, one claimed: "I feel like it’s better just to die."

Maybe not such a great strategy.

It's not that there aren't some positive signs.  NCHS data indicate that the number of uninsured has dropped from 45 million in 2013 to 28.5 million for the first half of 2015.  Sarah Kliff reports that people under 35 account for 35% of healthcare.gov's open enrollment sign-ups, versus 33% in 2014, with close to million new enrollments from that demographic, versus under 700,000 a year prior.  Not wanting to pay the penalty motivated their enrollment more than older enrollees.

The NCHS also reported that the number of people under 65 who had problems paying medical bills in the past 12 months decreased from 56.5 million in 2011 to 44 million in the first half of 2015.  The percentage varies by insurance status (29.8% for uninsured, 21.8% for those with public coverage, and 12.2% for those with private coverage) and income (24.5% for the poor, 27.1% for the near-poor, and 12.2% for the not-poor).

Still, 28.5 million people remain without coverage, 44 million are having trouble paying medical bills, even among those with private insurance or "not-poor"...gosh, that doesn't seem like a raging success.

A couple other articles have deepened my belief that we've gone down the wrong path:

  • A new study, as reported by The New York Times and with some additional observations in Health Affairs, confirms the long-known wide ranging pricing and utilization in the U.S.  What's new is that Medicare spending is not necessarily linked to private insurance spending -- "lower cost" markets for one does not automatically result in lower costs for the other.  It also reaffirmed that prices vary wildly not just across markets but within markets, even for negotiated insurance rates.    
  • A new study found that our convoluted patent/FDA approval system makes it more financially advantageous to invest in cancer drugs that extend cancer patients' lives by a few months instead of on drugs that might prevent cancer.  Huh?

This is not all the fault of health insurance, but it has had a role.  We entrusted public and private payors with negotiating on our behalf with health care professionals and industries.  As it turned out, it was easiest for them to do within some arbitrarily narrow definitions, limiting who can be paid how much for what.  Our "health" system is a medical one.

The end result is this system where prices vary irrationally even when one has insurance, where the care you get is a function of which provider you get it from rather than what is best, and where we spend more than any other country and certainly more than many of us can afford.  The end result is that the same kinds of entities that were getting the biggest share of our health care dollar in 1960 are still getting about the same shares now (as I discussed previously), which sure seems like some kind of a sweetheart deal to me.  

It seems foolhardy to self-fund oneself in a world where a lengthy hospital stay or drug regime can easily bankrupt anyone other than the so-called 1%, yet health insurance allows and, to some extent, exacerbates that kind of expensive treatment.  It has become part of the problem.

Look, I worked in the health insurance industry for a long time.  I helped introduce consumer-driven/high deductible plans to help foster cost-awareness.  I bought into the protection-against-big-expenses meme.  I personally have never not had health insurance.  So, by most standards, I should be biased in its favor.  But I'm beginning to wonder if health insurance itself is the problem, or at least a big part of the problem.  

I've written before about some of the new entrants into health insurance; more power to them, and the more the merrier.  What I continue to be disappointed by is that we're not really seeing fundamentally new approaches to what health insurance is.  Going back to the product analogy above, we're just iterating versions of the iPod instead of introducing the iPhone.  We probably would have continued to love our iPods if the iPhone hadn't come along, but not many of us would opt for a world that had better iPods but no iPhones.

I wish I knew what the alternative to health insurance was.  I wish I even had a good idea for an idea for it.  The closest I've come is a thought experiment I posed some time ago, where we crowdsourced any help we wanted for our health spending, which would be subjective and unpredictable, but which would at least open our eyes about who is paying for our care.

Here's a radical suggestion: instead of thinking of more ways to force people to buy health insurance they don't want, we should be developing products to protect their heath that they actually want to buy.