Health Should Be the New Tech

I saw that Kaiser Permanente just named its first Chief Digital Officer, and I was reminded of the belief that all companies are now tech companies.  For example, online retailer Stitch Fix 's CEO Katrina Lake recently said in podcast with Kara Switzer:

But if you want to still exist in a decade, I think everybody’s going to have to be a tech company and figure out what is the thing that is going to be lasting about their company.

Now, Ms. Lake isn't exactly breaking new ground with this thought -- a quick search found similar quotes from others going back to at least 2015 --  but the point remains valid.  Still, though, maybe we should be thinking: in ten years, every company is going to be a health company.

I don't mean in the way that employers' healthcare obligations are forcing them to treat health as part of what they do (e.g., the auto manufacturers realizing they paid more for healthcare than for steel, or in the way Amazon/Berkshire Hathaway/JP Morgan Chase founded Haven).  I don't mean in the way that tech companies are looking to see how they can move into health in order to get their piece of the $3 trillion healthcare pie.  I don't even mean in the way that medical marijuana, CBD, or, God help us, Goop are trying to convince us that they can play an important role in our health.

No, I mean in the way that every company would think about, and compete on, how their products contribute to our health and well-being.

Oh, no!

Take our homes.  Fast Company's Katherine Schwab reported on a new study that found that a plurality of traumatic brain injuries among children and teenagers came not from sports or recreation but from things we buy -- especially home furnishings and home features (each 17%), as well as nursery equipment and toys.   There are already numerous safety regulations, and children/teenagers are going to be reckless, but design plays a key role as well. 

How many home furnishing, home construction, or toy companies think of their mission as helping ensure our health?  If they did, we might live in different kinds of houses.  One that is safer and less likely to lead to us getting hurt.

Fast Company also reported on a Chinese house based on a new model of the future of housing: "homes designed not just for able-bodied, nuclear families, but for the different needs that span multiple generations."  It tries to balance privacy with ease of access and mobility for each of the generations, and consciously plans for different stages of our health.

Well, you might say, that's just China, but a related story found that more Americans are living in a household with two or more adult generations.  Although still in the minority, trends like "cohousing" or "intergenerational home-sharing"  may help both address the affordability of both retirement and homeownership, as well as reduce isolation.  One expert told Fast Company: "There’s been so much emphasis on independence and on privacy that we really designed community right out of our lives without knowing it," with all sorts of adverse health consequences. 

Take food.  The food industry has focused for the last 30+ year on making food tastier and cheaper, but usually not healthier, and our obesity/diabetes rates have reflected that.  It's been good for our wallet and terrible for our health, Trader Joe's and Impossible burgers notwithstanding.  We're not all going to stop eating meat or only organic fruits and vegetables, but we could eat better, and the food industry could help.

Suggested supermarket layout Credit: RSPH and Slimming World

In the U.K., the Royal Society for Public Health and Slimming World issued a report critical of how grocery stores are encouraging poor food choices, urging that public health experts help design their stores.  They went further, opening up a pop-up store called Nudge to help illustrate their principles.  Slimming World's Carolyn Pallister asserted:

Supermarkets will argue that they are giving their customers the choice; and we haven’t removed those choices at Nudge, all we’ve done is made it easier for customers to choose healthier alternatives and put less emphasis on promotions of foods likely to cause weight gain.

Or consider, say, airplanes.  They're built to deliver as many people as quickly as possible, and, while they actually have an admirable record for rarely crashing, they're not really designed for our optimal health.  The air is bad and the surfaces are, well, you really don't want to know how dirty things really are.  Where are the airplane manufacturers/airlines that promise not only to get you to where you are going safely but also no less healthy than how you boarded? 

Many companies might argue that they don't directly make products for consumers, but that doesn't mean they don't impact our health.  Think about short and long term health impacts of pollutants and carbon releases, and then think about why those impacts get borne by other entities, such as our health insurance.  More direct fees and taxes on such by-products of production would bring drastic changes, or at least new funds to help address their effects. 

Concern has been raised about the apparent discrepancy between tech's "move fast and break things" mindset and healthcare's "first, do no harm" directive (although the latter is hard to justify with, for example, the industry-inflicted opioid crisis or the well-known level of medical errors).  The truth is that all industries, including healthcare, should be thinking more about the actual health impacts of what they do.  If they did, Facebook and Twitter might not support so many toxic users, and Google would not allow such pervasive health misinformation. 

Don't even get me started on gun manufacturers or the tobacco industry.

Yes, tech is sexy, tech seems futuristic, and tech is loved by investors.  It's no wonder Ms. Lake and others think about being considered tech companies.  But we have to remember that tech is not the goal; tech is never the goal.  Tech is a tool, and what better to use tools for than to make us healthier?

 Almost two years ago, Steve Downs called for us to build a Culture of Health into all we do, and that call is more needed than ever.

Scramble the Healthcare Red Team!

France's Defence Innovation Agency (DIA) -- its equivalent of our DARPA -- wants to set up a team of science fiction writers and futurologists to help them plan for the future.  This so-called "red team" is intended to "propose scenarios of disruption" that the traditional military strategists might not envision, in order to not be caught off-guard. 

As Futurism put it:

Basically, it seems France wants the creatives to imagine how the future could go really wrong so that the nation can stay one step ahead of any trouble — and really, who better to envision dystopia than a team of science-fiction scribes?

It made me wonder: do CMS, United Healthcare, CVS, Kaiser Permanente, HCA, or Pfizer have their red teams of science fiction writers?  If not, why not?

Now, I have to admit that I am a science fiction fan (I am currently in the middle of reading Neal Stephenson's latest opus, Fall; or, Dodge in Hell), so this idea appeals to me.  Science fiction writers have been making some pretty good predictions for a while now, from Jules Verne sending rockets to the moon, H.G. Wells blowing up things with atomic bombs, Arthur Clarke using satellite communication, or William Gibson's cyberspace. 

Of course, they've been wrong much more than they've been right, which is why we're not all using jet cars, teleporting, or taking fast-than-light trips to the stars.  The point is that they imagine a future that is different than what we're used to, and the great science fiction writers imagine the technologies mostly to draw out the implications for what they might mean for society.

The DIA said: "The work of this cell will be to construct valid strategic hypotheses, ie likely to upset the capability plans, " and added that the team would also be "reflecting on the strategic consequences of the arrival of disruptive technologies.”

Bruno Tertrais, deputy director of France's Foundation for Strategic Research, told The Telegraph that the red team was supposed to challenge "any certainties that we may have and hypotheses about the future outside the usual bureaucratic procedures.”  However, he clarified:

It is certainly not the Red Team that will decide France’s military strategy and still less its defence policy. Its role will be to help the Defence Innovation Agency think about future technologies and their impact on strategies.

I can understand not delegating a country's military strategy to a bunch of science fiction writers, although some critics might argue that France could scarcely do worse.  😉

Healthcare organizations aren't about to turn their strategy over to a team of science fiction writers either, nor should they.  But the DIA's notion that science fiction writers might picture possibilities that an organization's leaders and usual consultants might never think of seems very valid, and this is where healthcare should be paying attention. 

We talk a lot of looking for solutions that are out-of-the-box," but we typically have a hard enough time realizing the boundaries of those boxes, much less picturing topologies that don't even include boxes.  That's the kind of thing that science fiction writers try to do. 

Healthcare is already getting into territories that, not that long ago, would have been considered science fiction.  We've got robotic surgery, nanobots, 3D printing for tissues and prescriptions, and an increasing number of applications for artificial intelligence, to name a few.  But we're still trying to graft them onto our existing delivery and financing systems, and, at some point, that is going to fail. 

My favorite example is with artificial intelligence.  We've had telemedicine, with human doctors, for some 20 years now, and we're still trying to make state licensing and reimbursement work.  So when we get A.I. that can fill many roles of those human doctors, and which make decisions in ways that we don't and probably can't understand, how are we going to solve those same state licensing and reimbursement issues?   

That's the kind of thing that science fiction writers may be better at than most healthcare executives. 

It's easy enough for a science fiction writer to just assume really smart healthcare AIs or nanobots that can deliver pinpoint fixes within our bodies.  But I'm not sure that's where we need their kind of imagination the most.  Where we may need it most are in some of the nuts-and-bolts of the healthcare system, where things work, just not very well.

For example, here are some of the grubbier aspects of healthcare that need reimagination:

• Eligibility: how does everyone in healthcare agree that Person A is Person A?  Right now, we have NPIs that change seemingly willy-nilly, health insurance IDs that change with every change of health plan, and often multiple IDs for the same person.  It's amazing we ever get it right.
• Billing: we spend way too much time and effort trying to get billing "right," the definition of which healthcare professionals and health plans would not agree on.  With ICD-10, we're taking an already incomprehensible coding system and ramping it up still further.  We mix billing with diagnosis/treatment in ways that don't serve either purpose well.  Billing isn't supposed to be the point, but, in many ways, it is.
• Claims payment: Few of us understand our health plan benefits.  Few of us ever really know how a specific medical encounter is going to get paid.  Payments are often less than we expected, and many denials get reversed upon appeal.  Despite health plans' best efforts to automate, many claims end up needing a human to adjuticate.  It is to health plans as billing is to healthcare organizations -- vital to get right, hard to explain, and hugely expensive.
• Financing:  Despite just about every variation experts can think of, no country that I know of is happy with how health care is financed, nor how health care services are paid.  People grumble about taxes, complain about health insurance premiums, and worry that we're incenting the wrong kinds of health behaviors and treatments.  As we spend ever more on health care, this will all get worse.  

Imagine healthcare systems that solves those, and I'll be impressed.

A lot of very earnest, knowledgeable people are working on these and other healthcare problems.  More power to them, but it's not clear to me that they can see over their own biases.  This is where we need fundamental rethinking.  This is where we need daring speculations.  This is where we need wild imaginations to describe a very different future.

This is where healthcare could use science fiction writers.   

The Other Amazon Effect

I was keen to write about how Amazon Prime Day related to healthcare, but my esteemed colleague Jane Sarasohn-Kahn beat me to it, so I'll turn instead to another set of headlines that Amazon has been generating lately: their "Upskilling 2025 initiative to retrain a third of its workforce.  The company plans to spend some $700 million to retrain 100,000 workers by 2025. 

I wonder if healthcare is prepared to make the same kind of upskilling/retraining investments.  Because it is going to need to. 

Not everyone is wowed by Amazon's initiative.  As Peter Cappelli, a professor at Wharton, told The Wall Street Journal: "It’s not altruistic.  There’s some hard-nosed business-decision-making behind this.”  Or, as Marc Perrone, the president of the United Food and Commercial Workers Union said: 

Jeff Bezos’s vision is clear – he wants to automate every good job out of existence, regardless of whether it’s at Whole Foods, Amazon warehouses, or competing retail and grocery stores.

Let that charge sink in: "he wants to automate every good job out of existence."

As it happened, on the same day that Amazon made its announcement, the McKinsey Global Institute released its report The future of work in America: People and places, today and tomorrow. The report illustrates the trends that Amazon is trying to address, including its finding that: "Nearly 40 percent of US jobs are currently in occupational categories that could shrink between now and 2030." 

Equally important, they point out:

Previous MGI research has found that less than 5 percent of occupations can be automated in their entirety, but within 60 percent of jobs, at least 30 percent of activities could be automated by adapting currently demonstrated technologies.¹ What lies ahead is not a sudden robot takeover but a period of ongoing, and perhaps accelerated, change in how work is organized and the mix of jobs in the economy.

 McKinsey rated health care professionals as among the lowest whose jobs faced displacement (but still around 10%), while "health aides, technicians, and wellness" had higher but still below average displacement risks.  Still, "Office support" workers had the highest displacement likelihood, and I wonder how many "healthcare" jobs would truly fit into that category. 

Credit: McKinsey Global Institute

If your healthcare job does not entail being directly involved in actual patient care, you should probably be thinking about that upskilling (and even those giving direct care may need some).

Let's face it: for all the expensive technology, our healthcare spending mostly goes to people.  U.S. healthcare's administrative costs are rightly criticized as being much higher than in other healthcare systems, and those administrative jobs seem to be growing much faster than clinical ones.  We have complex processes that often don't make sense to patients, and that require constant intervention from people to execute them.  We have many clinical decisions based on personal preferences, not empirical evidence.

None of that can last.  None of that is good for our health.  And none of that should reassure anyone about their healthcare job. 

Three things are clear to me:

1. Administrative processes are going to continue to become more automated, as we seek efficiencies and as technologies continue to facilitate them;  
2. Robots are going to help care for our aging population, as demographic trends overwhelm our labor force;
3. Artificial intelligence will become an integral, even primary, part of clinical decisions and care, as Big Data and A.I. grow more powerful.

All of those spell fewer healthcare jobs.  The bull market in healthcare employment that we've seen for the last 20 years is not going to last much longer.  

Amazon's mantra has always been to focus on the customer, not on profits, but that mantra often seems more observed in words but not actions in healthcare.  Healthcare organizations are certainly looking at their costs, but patients often feel that budget cuts seem to adversely impact the people dealing directly with them instead of all those mysterious people in the back offices.

Technology is going to help change that.  For example, an article in blockdelta describes 15 trends in "medical robots disrupting healthcare," including robotic surgery, endoscopy bots, companion robots, nanobots, and AI diagnostics.  Similarly, John Nostra predicts that "AI will be smarter than the clinician. It will even become compassionate."   He believes clinicians need to focus on owning the compassionate domain rather than the cognitive one.

Whenever I am in a care setting, I try to talk to whomever I'm dealing with about the technology they're using, usually the EHR.  The stereotype often holds true: the older the person, the more they grumble about it, the harder they seem to find it to use.  The thing is, the technology is not going to go away (although hopefully it continues to get better); their job might be.  

In its press release, Amazon noted that its fastest growing jobs have been data mapping specialist (832% growth), data scientist (505%), solutions architect (454%), security engineer (229%), and business analyst (229%).  Few believe they're going to upskill many warehouse workers to data scientists, but, as Ardine Williams, Amazon's vice president of people operations, told The New York Times: 

When automation comes in, it changes the nature of work, but there are still pieces of work that will be done by people.  You have the opportunity to up-skill that population so they can, for example, work with the robots.

So needs to happen in healthcare.  

Healthcare has been focused on finding enough people, and is belatedly realizing it needs to finds more of the right people, with new skills and approaches.  The harder part, though, will be to upskill its current workers to the healthcare system that we're going to soon have.   As Ryan Carson, founder and CEO of Treehouse, told The Journal, 

The big secret is there is no lack of talent.  We just haven’t been looking in the right spots. That talent is often at your own company. They literally already work for you.

Maybe med techs or nursing aides aren't going to become data scientists either, but healthcare needs to figure out its upskilling strategy before it faces a jobs apocalypse. 

Our Dunning-Kruger Healthcare System

Psychologist David Dunning, originator of the eponymous Dunning-Kruger effect, recently gave an interview to Vox's Brian Resnick.  For those of you not familiar with the Dunning-Kruger effect, it refers to the cognitive bias that leads people to overestimate their knowledge or expertise.  More importantly, those with low knowledge/ability are most likely to overestimate it. 

Does this make anyone else think of the U.S. healthcare system?

Credit: drschat

Professor Dunning proposed the effect in 1999, in a paper in Journal of Personality and Social Psychology, along with then-graduate student Justin Kruger.  Since then, it has become widely known and broadly applied (not always accurately, as Dr. Dunning explains in the interview). 

Their paper had a chart that summarized their findings:

Credit: Dunning-Kruger, Journal of Personality and Social Psychology

Dr. Dunning believes that we tend to think that this effect only applies to others, or only to "stupid people," when, in fact, it is something that impacts each of us   As Dr. Dunning told Mr. Resnick, “The first rule of the Dunning-Kruger club is you don’t know you’re a member of the Dunning-Kruger club.  People miss that.” 

Or, as Dr. Dunning characterized it in a 2014 Pacific Standard article, "We are all confident idiots."

So, how does this relate to our healthcare system? 

We brag about our excellent care, our great hospitals and doctors, and all those healthcare jobs powering local economies.  Yet we have by far the most expensive healthcare system in the world, which is expensive not because it delivers better care or to more of its population than health systems in other countries, but because it feels it is justified in charging much higher prices.  Our actual outcomes, quality of care, and equity are all woefully mediocre on a number of measures. 

Credit: Kaiser Health News

How many of you live in an area that has at least one hospital system claiming to be one of the "best" hospitals in the country?  I've lived a lot of places, and each of them had one or more hospitals making that claim.  Your local hospital(s) may as well.  I hate to break it to you, but, in most cases, that claim is not likely to be true. 

Similarly, how many of us like to believe that our doctors are "the best"?  Perhaps they even have "best doctors" plaques in their offices to support this claim.  Again, it's possible that they are, but, in most cases, those beliefs are not likely to be true. 

Statistically speaking, most of us receive average care, and some of us receive sub-standard care.  We don't live in Lake Wobegon.  We can't all be getting the best care, or even above-average care. 

Just look at how few hospitals earn high ratings from The Leapfrog Group.  Look at nationally or even internationally known hospitals like UNC Children's or Houston Methodist, both of which had embarrassing revelations about the quality of some of their programs uncovered by investigative journalism.  Look at how even problem doctors often evade our vaunted medical licensing system. 

They all probably thought they were much more capable than they were. 

In the Vox interview, Dr. Dunning referred to a 2018 paper he co-authored, which found that beginners don't start out displaying the Dunning-Kruger effect, but often soon manifest it:

Although beginners did not start out overconfident in their judgments, they rapidly surged to a "beginner's bubble" of overconfidence."...Hence, when it comes to overconfident judgment, a little learning does appear to be a dangerous thing.  

I think about how uncertain medical school students turn into nervous residents and ultimately become the uber-confident physicians we're used to.  Dr. Dunning discussed how we need to do better distinguish facts and opinions, and be more willing to admit "I don't know."  How often does your physician admit they don't know something -- and would that give you more, or less, confidence in them if they did?  

In The Atlantic, Olga Khazan reported on a new study that suggests that, despite all their supposed superior knowledge, doctors don't really make better patients than the rest of us.  They get C-sections about as often, and about as unnecessarily as we do, they get about the same amount of unnecessary/low value tests, they're not better at taking needed prescriptions.  

As Michael Frakes, one of the authors told Ms. Khazan, the doctors "went through internships, residencies, fellowships. They’re super informed.  And even then, they’re not doing that much better."  Professor Frakes speculated that even physicians tended to be "super deferential" to their own physicians, despite their own training and experience.  

They're underestimating their own knowledge, and their doctors may be overestimating theirs.  A new meta-study found almost 400 "medical reversals" -- common tests or procedures that are not supported by research data.  As one of the co-authors told The New York Times: "You come away with a sense of humility.  Very smart and well-intentioned people came to practice these things for many, many years. But they were wrong."  

It is widely accepted that as much as a third of our healthcare services are unnecessary or inappropriate -- even physicians admit that -- but, of course, it is other physicians doing all that.  No one likes to believe it is their doctor, and few doctors will admit that they are the problem.  

Dunning-Kruger, indeed.

Much as they'd like us to, it is not enough for us to always assume that our healthcare professionals and institutions are qualified, much less "the best."  It is not enough for us to trust that their opinions are enough to base our care recommendations on.  It is not enough to believe that local practice patterns are right for our care, even when they are at variance with national norms or best practices.

"Trust" is seen as essential to the patient-physician relationship, the supposed cornerstone of our healthcare system, but trust needs to be earned.  We need facts.  We need data.  We need empirically-validated care.  We need accountability.  

Otherwise, we just fall victim to healthcare's Dunning-Kruger effect.  

In Defence of Ignorance

You may not realize it, but there is something wrong with your body.  Perhaps you have a mutated gene.  Maybe some cells are misbehaving.  Your body chemistry may be out of whack.  Some of your microbiota could be changing from a mutual benefit, or at least a stalemate, with your body to battling it.  It could be lots of things.

If you had a computer program as sophisticated as your body's code, thrown together over as long a time by as many programmers, you'd expect some bugs.  If you had a piece of machinery as complex, and used as hard as your body, you'd expect a few things to be constantly breaking down.  Yet somehow we seem surprised when we're told something is wrong with us.

Let's face it: past your early/mid-twenties, what we politely call "aging" should more realistically be called "deteriorating."   We just don't always realize it, and that's not necessarily a bad thing. 

Credit: Shutterstock

A few weeks ago I saw a primary care doctor.  Almost as an afterthought, she suggested one more scan.  I suspected I was being upsold, but I didn't want to be difficult and it wasn't particularly expensive, so I agreed.  Much to my surprise, the results indicated I might have a serious condition -- not life-threatening, but certainly life-altering.   There were treatments for it, but they were lengthy, unpleasant, and expensive.  Moreover, they could only slow the progression of the disease, not cure or even halt it.

I was not a happy camper.

She sent me to a specialist.  More tests, more scans, and he confirmed her suspicions.  He sent me to an even more expert specialist, who did more tests and scans, with some surprising conclusions.  There was some damage, he agreed, but nothing too serious, just something to watch.  I didn't yet have the disease in question.  Come back in a year just so we can keep an eye on it.

All this took several weeks, which I'm sure is very short compared to what others have had to endure, but which seemed interminable to me.  I kept thinking about all those treatments, all that money, and how my life would be different as the disease had progressed.  All for naught, as it turned out.

I'm kind of wishing I had turned down that original scan. 

We're detecting problems ever earlier.  CTs, MRIs, PET scans can pick up cancer cells at early stages.  We're already using "liquid biopsies" or fecal samples to find tiny traces of cancers, in minimally invasive ways, and are on the cusp of detecting early stage cancers and other problems through even more refined blood tests. 

The trouble is that not all the findings are important.  For example, many scans pick up abnormalities that should be considered "incidentalomas," researchers found last year.  “We know that the diagnosis of incidentalomas can cause patient anxiety and is likely to lead to further investigations and treatment, some of which will be unnecessary, and some of which will cause harm,” said lead study author Dr. Jack O’Sullivan of the University of Oxford. 

Credit: MedPage Today

As H. Gilbert Welch, MD, a professor at the Dartmouth Institute for Health Policy and Clinical Practice and author of a related study, told MedPage Today: "The human body is full of small anatomical abnormalities.  The more we look, the more we find, and the more we intervene. And that’s generally not in our patients’ interest."

Similarly, those new blood tests may have only a 1% false positive rate, but: "One percent in a patient population of 5,000 is manageable.  With 1 percent in hundreds of thousands or millions, you quickly get into different dynamics,” Joydeep Goswami of Thermo Fischer Scientific said in an interview with MedCity News at the recent annual meeting of the American Society of Clinical Oncology.  

It's cool that the Apple Watch can give us 24/7, real-time monitoring of irregular heart beats, and who knows what kinds of such monitoring we'll be able to do next.  But we're also going to turn up apparent issues that aren't really issues, getting lots of people needlessly worried and subject to subsequent tests/procedures/expenses.  

We can get findings.  Lots of them.  More all the time.  We just don't always know what to make of them.  As Dr. Welch said, "the more we look, the more we find."  Sometimes it might be better not to look. 

I think back to my early days in health policy, when there was much discussion about the supposed rationing in the U.K.'s National Health Service.  For example, it was widely believed in the U.S. that no one over 65 could get dialysis through the NHS.  A study -- and I can't recall the authors -- found that, in fact, there were no explicit prohibitions or even guidelines about it.  It was just that physicians were reluctant to order it for seniors.  As one physician put it, "well, at that age we're all a bit crumbly, aren't we?"

Indeed, we are all bit crumbly, and the trouble is we don't always know which of our many flaws are "normal" and which need action.  

We don't have the right baselines.  We don't know what "normal" should look like, for which people, at which ages.  We don't know which of what Dr.Welch called our "small anatomical abnormalities" will never manifest as real issues, and can be safely ignored.  No, give us a finding, and someone will want to take action on it. 

Should we stop trying to find ways to find and diagnosis potential issues at earlier stages?  No, of course not.  But we need to spend at least as much time and money understanding what they really mean.  Sometimes a flaw is just part of who we are.

Should we stop trying to treat issues once identified?  Well, maybe.  Most of us freak out when given a serious diagnosis, but part of physicians' job is to help us understand when action is necessary and when waiting or even ignoring something is the best course of action.  Part of their job may be not to put us through that at all.

Sometimes ignorance is, in fact, bliss.