Monday, December 29, 2014

Great Expectations Aren't Always Good

I didn't post last week, figuring people would be too busy to read it during the holidays, and I probably would have done the same thing this week except that interesting topics just keep coming up.  The one that most caught my eye was a study on patients' unrealistic expectations.

The study, by Hoffman and Del Mar, was a meta-study reviewing the research on patients' expectations about both benefits and potential harm of interventions.  As might have been expected, patients have an overly optimistic view: in 65% of the studies where data was available, the majority of patients overestimated the potential benefits of their treatment.  Similarly, in 67% of the studies where data was available, 67% found that the majority of patients underestimated the potential harm.

The authors noted: "Overly optimistic intervention expectations by patients and the public are undoubtedly contributing to the growing problem of over diagnosis and over treatment."  No kidding. 

In an accompanying commentary,  Deborah Korenstein, MD, calls patient perceptions "the Achilles heel of high value care," cautioning that the study "makes it clear that we must better equip patients to understand the potential benefits and harms of medical care so that they can make the best decisions and improve the value of their own health care."  She concludes that "the effort has to start with us," that is, with physicians.  

None of this is surprising.  For example, a study of cancer patients in 2012 found that 69% of patients with advanced lung cancer, and an even mote astonishing 81% of patients with advanced colorectal cancer, did not understand that the chemotherapy they were receiving was not at all likely to cure their cancer.   Even more disturbing, patients who rated their communication with their physician high were most likely to have over-optimistic expectations.  They may be comfortable talking with their physicians, but perhaps they're not really listening -- or the physicians are not being as blunt as they could be.  

It's not like patients and physicians are talking openly with each other anyway.  The Open Notes Project is getting lots of deserved plaudits for its effort to allow patients to see their own records, but it is dismaying that we need an initiative to do so.  On the other hand, a recent survey found that nearly half of patients don't want their doctors to see all their health information either.   

Lack of openness goes both ways.

In another illustration of the problem, Kenny Lin, MD, wrote a recent post False alarms and unrealistic expectations in preventive care that recaps the dubious data on the value of many common preventive exams and tests, while highlighting the difficulty in trying to change patients' expectations for them.  He fears that the main goal of allowing these unrealistic expectations to persist is financial -- that is, the financial interests of physicians and the associated supply chain.  He acknowledges that "false beliefs meet the psychological needs of patients for hope and safety, as well as for action, agency, and a sense of control.  They enable clinicians to feel they are making a difference; even physicians who know better order unnecessary tests to please their patients."

Sometimes these false beliefs do serve good purposes.  A patient's positive attitude has long been known to impact their response to illness.  Placebos are an age-old part of a clinician's bag of tricks, and science is starting to help us understand them.  Recent findings have shown that they produce measurable changes in the body, even in the brain (e.g., Parkinson's symptoms).  Landmark research by Ted Kaptchuk, et. al,  back in 2010 found evidence not only of the placebo effect, but that placebos had a positive impact even when patients knew they were "only" receiving the placebo.  

In another intriguing study, Kaptchuk gave asthma patients inhalers, two different placebo interventions, or no treatment, and found that only the ones with inhalers actually improved lung function.  However, the patients who received the placebos reported the same subjective improvement as the inhaler group.  They felt just as good, even if their lung measurements said they shouldn't.

Kaptchuk -- an acupuncturist by training yet an associate professor of medicine at the Harvard Medical School -- has made placebos so mainstream that he now heads up the Program in Placebo Studies & Therapeutic Encounters (PiPS), hosted by Beth Israel Deaconess Medical Center.  

One of the most intriguing findings from the placebo research is that, well, presentation matters.  Karen Jensen, also of Harvard and PiPS, has shown that patients look for clues about what to expect, and those clues may be non-verbal -- the physician's demeanor or facial expression, the decorations in the exam room, etc, As she says: "Everything that has to do with the elaborate ritual of delivering care could be a target to enhance the placebo effect,"  E.g., it is not just the pill but how the pill is prescribed.

Still, placebos and positive thinking only go so far.  As Kaptchuk says: "Sham treatment won’t shrink tumors or cure viruses."

Or can they?  The hottest area in oncology these days is immunotherapy -- getting the body's own immune system to attack the tumors, rather than relying on radiation or chemotherapy.  A new class of drugs are, in essence, releasing the natural brakes on the immune system.  Right now, we need drugs to accomplish this, but recent research has demonstrated that, amazing as it sounds, a person's mind can actually trigger gene expression.  So far it has only been tested triggering genes in mice, but perhaps we're not so far away from releasing those immune system brakes in ourselves.

Advocates of biofeedback must already be celebrating.

It's fun to speculate about literally thinking ourselves to good health, just as it is nice to imagine nanobots doing all the hard work.  Those days may come, and sooner than we think, but for now we're stuck with our existing health care system, and that system is built on physician-patient interactions.  

It's easy to see why those interactions are fraught with inaccurate patient expectations.  One of the biggest problems, of course, is that for most interventions, the supporting data is woefully sparse or inconclusive, especially in the context of a specific provider's performance with specific set of patient characteristics.  Even if a physician takes the time to share the data, and the patient has the aptitude and inclination to understand it, they may not be much better advised than uninformed patients.  

Whenever I think about shortcomings in our health care system, I often think of Congress.  Its approval ratings are at all-time lows, but the fact is we've elected them, or through our apathy allowed others to do so.  Similarly, we've allowed our health care system to be a medical care system, and abdicated much of our own responsibility for our own health to our health care providers.  As a result, when something happens to us, we have these unrealistic expectations about what they can do for us.

It's on us to take better care of ourselves.  It's on us to do better research on our treatment and provider options, and to demand better data.  It's on us to insist on frank and open communication with our physicians and other providers  -- even if that doesn't fit within the 15 minute appointment window they might have slotted for us.  

We should never take hope out of care, but we can do better about making it informed hope.

Wednesday, December 17, 2014

The Convenience Truth

The U.S.  Mint reports that it now costs 1.7 cents to make a penny; nickels are slightly better, costing "only" 8 cents to make the 5 cent coin.  This is economics the way health care practices it.

According to Christopher Ingraham of the Washington Post, we could save $100 million annually by eliminating both coins.  Or we could change the metal composition of the coins to make them cheaper, but that would create havoc with vending machines.  So we just blithely chug along, using coins that few of us would miss and whose costs exceed their value, mostly because we've always had them and the businesses that revolve around them don't want to change.

See how this is like health care?

What made me think about this was a recommendation from Britain's National Institute of Health and Care Excellence (NICE).   They now say that midwife-led birthing units are safest, and advised more women to consider them for low risk pregnancies.  They believe this could account for as many as 45% of live births.  Moreover, they think home births are just as safe as births in a hospital.  The Netherlands is considered the leader in home births, at a little under 25%, although that rate has been dropping in recent years.  The U.S. has 1.36% home births

Although I don't have any first hand evidence, I'm willing to bet that Dutch women aren't built any different than U.S. women, and don't love their children any less.  The non-hospital birth rate in the Netherlands versus the U.S. must be due to cultural and structural differences, not medical ones.

The literature -- often drawn on Netherland's data -- generally supports the NICE recommendation, but not everyone is convinced.  A lot of factors come into play, not just the specifics of the pregnancy itself but also how midwives are trained, how far the home or birthing center is from emergency services, and so on.  It would be very easy to weigh all the factors, and conclude that even a relatively small increase in risk is not something you'd want to take for your own baby, and opt for the "traditional" OB/hospital delivery.

This is where the penny analogy starts to really apply.  These decisions on risk reduction are not without financial consequence.  A vaginal delivery with no complications averages about $10,000, whereas a birthing center costs under $2,500, according to Childbirth Connection.  I assume home delivery is less expensive.  As NICE's Dr. Mark Baker said about hospital births: "Yes, it's a very expensive way to deliver healthy babies to healthy women.  Saving money is not a crime."

Sadly, I can't even imagine a public body in the U.S. making such a bold recommendation.  I can't even imagine having a productive discussion on this topic in the U.S.  Just try and see how quickly it leads back to "death panels."

In a piece for The New York Times, professor/physician Aaron Carroll notes that the ACA-created Patient Centered Outcomes Research Center is explicitly prohibited from considering cost effectiveness.  Its website says: "We don’t consider cost effectiveness to be an outcome of direct importance to patients."  


Dr. Carroll uses the clever example of a pill that would extend your life one day, but cost $1 billion, versus a pill that cost $1 million but might extend your life 10 years.  We think we know how to decide about the former -- I hope -- but decisions on choices like the latter get murkier (e.g., Sovaldi...).  And it's always easier to keep paying for things that we've historically paid for, even when their value becomes more questionable.

In theory, value-based purchasing will help us address these decisions.  In practice, though, most of the value-based purchasing arrangements I am aware of -- and that certainly is not an exhaustive list -- reward providers whose outcomes are simply what we'd hope for, may penalize them slightly for disappointing results, and are indifferent about if the care could have safely been done elsewhere for less.  Does anyone think there is a health plan in the country which would dare reward midwives for the savings they provide through home births, or penalize OBs for hospital deliveries?  The data say we should, but I think I can safely say we won't.

There is a way out of this mess, but it may not be what you think.

I'm beginning to think that trying to reshape our health care system through value-based purchasing, cost-effectiveness, or even greater transparency may not work.  The "killer app" may not prove to be any of those high-minded strategies but rather a much more basic one: convenience.

Indeed, one of the earliest urgent care chains attributes its inspiration to the example of McDonald's.  We are, after all, the nation that invented fast-food, decided even that wasn't fast enough and so invented drive-throughs, which we use for over half of our fast food.  We liked the convenience of them so much that we've extended the approach to banks, car washes, pharmacies, even weddings and  funeral homes.  The concept of drive-throughs itself is rapidly being supplemented and even superseded by mobile apps, allowing consumers not to even have to get in their car.

Health care cannot ignore these consumer demands for more convenience.

Walgreens' chief medical officer recently noted that: "The idea of convenience ... is really becoming a dominant theme in health care."  It's no coincidence that Walgreens has been investing in in-store clinics, has a 24/7 Pharmacy Chat option, and just rolled out a direct-to-consumer physician virtual visit app, similar to American Well's Amwell service.  Not to be topped, Kaiser is now offering EMT home visits, in addition to its array of in-office and virtual visit options.  

Our traditional approaches to care delivery have revolved around convenience for the providers, not the consumers.  Many people still like to see their doctor in person, and to go to the hospital when they have serious health issues.  Thus, hospital births.  The rise of urgent care and  ambulatory surgical centers illustrate consumers' appetite for lower cost, more convenient options -- even though not everyone is convinced these options are as safe.  Perhaps home births will end up following this pattern as well.

Many consumers, especially younger ones, find ridiculous the notion that they have to call for an appointment that may end up weeks away, go to an office or facility that may not be close, only to wait there with sick people, and perhaps be sent to some other office or facility for more services.  They'd rather get their care via their mobile devices and/or in their home, and the technology is increasingly allowing that for many health concerns.  Not everything, mind you, but more than we realize now.

We've come to recognize that health care is one of the few industries where technology typically not only doesn't lower costs but usually adds to them.  Maybe, though, expecting providers whose revenue is at stake to focus on cost-effectiveness is asking too much of them.  Focusing on convenience shouldn't be.

It may not initially be about cost -- smartphones are more expensive than the mobile phones they are rapidly displacing -- but once consumers start opting for more convenient options, it's hard to see that costs won't plummet, as the traditional care silos start to fall.

Focusing on convenience is simply a way to make sure we're focusing on the consumer (AKA "patient").  Isn't that supposed to be the point?

Monday, December 8, 2014

Borrow From the Best

If imitation is, as they say, the sincerest form of flattery, health care should be doing a lot of sweet talking.

I was amused -- no, make that intrigued -- to see that the Defense Department recently released an RFI, asking both defense contractors and non-traditional vendors to help them dream up new defense technologies.  "I'm game for anyone," deputy assistant secretary Stephen Welby claimed at the press briefing.

Mind you, this is the Defense Department that already has DARPA, whose track record of technological innovation includes stealth planes, GPS, and -- in case anyone had forgotten -- the Internet.  But DOD still thinks they need help in developing "...novel or unconventional applications of technology in ways that would provide significant, enduring capability advantage..." to quote their RFI.

Health care needs more RFIs like that.

I was primed to react to the DOD RFI by an article in the Harvard Business Review: "Health Care Needs Less Innovation and More Imitation," by Anna Ross and Thomas Lee, M.D.   As they say:
Health care is infatuated with innovation.  We're awash in innovation conferences, organizations proclaiming innovation as a core value, newly minted Chief Innovation Officers, prizes for best innovation.  We think innovation is great, but there's a downside.  When organizations overemphasize innovation, they can miss out on the power of imitation -- copying existing approaches that actually work.  Providers need to actively seek out good ideas that have been tried and refined, bring those ideas home, and adapt them for local use.
They propose -- "our tongues only partially in cheek" -- an International Institute for Imitation, where Chief Imitation Officers could find good ideas to borrow.  They cite several examples of successful imitation, although the examples where health care organizations have borrowed ideas from other health care organizations are more convincing than the ones borrowing from outside health care.

I've written numerous posts over the past few months about how technology companies are becoming interested in health care -- e.g., AmazonFacebookGoogle -- but also expressed my concern that many of these outsiders are less interested in revamping health care than getting their share of the $3 trillion health spending.   If we're going to imitate, we need to make sure we're picking the right things.

One example of imitation is health organizations bringing in experts from retail industries, hoping to improve their patient/member experience.  Disney is often cited as one of the gold standards.  After all, Disney manages to take a hot, crowded, expensive experience and make people love it.  No wonder health care organizations think they can learn something.

The online version of customer experience that health organizations are now aspiring to imitate is Amazon.  Whenever I read about health care organizations wanting to create an "Amazon-like experience" (and President Obama helped get that particular ball rolling, bragging about before its disastrous initial roll-out last year), I think about why that is unlikely to happen.

Here's an example of the problem.  If my car was damaged I'd probably get a estimates from a few auto body shops, hopefully at least one of which my auto insurance will agree to pay in full.  I'd expect my car to get fixed on schedule and for the quoted price.  If it happened like it happens in health care, it's unlikely I could get an estimate in the first place, and when it came time to pay I'd probably get lots of bills -- from the auto body shop, from each of the mechanics, some for any diagnostics performed, and so on.  My health insurance might decide not to pay for some of these, claiming they weren't "necessary," or make me pay extra for anything it didn't consider part of its network.  What a mess.

When even auto repair looks consumer-friendly by comparison, you know health care is badly in need of reinvention.

The influx of different technologies and customer orientations certainly is welcome, but at best they paper over some of the underlying major problems, and I'll highlight a few:

  • Billing: Transparency is heralded as a way to help consumers make better choices.  There are no shortage of vendors in the space, and even some research suggesting it can help.   More transparency is needed, but the problem is that the underlying pricing structures make no sense.  They're based on wildly inflated charges that bear no relation to either actual costs nor the numerous negotiated rates.  Even worse, they grow ever more granular, whether based on CPT codes, DRGs, or other mechanisms.  There are thousands -- tens of thousands! -- of them, few of which carry any real meaning to consumers and which are rarely disclosed until long after the service is rendered.  Even if a provider's fees were made perfectly transparent, consumers still would be at a loss to understand them.  
  • Comparability: Let's pick on health plans, although I could do the same for providers' services.  It's not so hard to compare cost-sharing provisions like deductibles, copays, coinsurance, out-of-pocket limits, and premiums, and if you are relatively healthy that may be enough.  But if you have a chronic illness or suffer some major acute episode, things like how health plans define "medically necessary" or "experimental treatment," what's on their formulary, even which providers are in their provider network may be even more important.  It is hard to find and compare health plans on any of these prior to enrollment -- and health plans can change them with no warning during the plan year.  
  • Episodes of care: It's pretty rare that an encounter with the health system results in seeing only one provider, or only once.  A doctor's visit may led to the pharmacy or lab, an ER visit may include out-of-network physicians, and hospital stays end up with bills from a surprising number of providers, both during and after the stay.  Patients rarely know exactly what they are "buying" or from whom they are buying it, much less what it will all cost.     
We have armies of billing experts on the providers' side trying to maximize their revenues, countered by armies of claim examiners on the health plan side trying to combat those efforts, while experts on both sides fight over contractual language like Talmudic scholars.  None of this helps the consumer -- the patient/member -- and very little of it is understandable to them.

Yes, I know smart people on both sides are working on approaches like bundled payments and value-based purchasing, but I fear that even these may end up so technical that only those experts can understand them.

Health care is very complex; I get that.  But so are lots of things, such as computers or cars, and buying them is much easier than buying health care.  We shouldn't accept that this has to be the case. I think the winners in the health care wars will not necessarily be the ones with the coolest technology or the friendliest customer experience, but those who have the easiest-to-understand products and pricing.

If we want to borrow from other industries, technology is nice but simplicity would be better.

Tuesday, December 2, 2014

Cutting More But Not Saving More

There's an epidemic in American health care, and I don't mean the commonly lamented ones like obesity, diabetes, or even Ebola.  It's surgery.

It would be easy to think I am referring to cosmetic surgery.  After all, according to the American Society of Plastic Surgeons, there were some 15.1 million cosmetic procedures in 2013, more than double the number in 2000.  One can question what all those cosmetic procedures say about our society's values, especially with some of the stranger procedures being done or the reasons for wanting them (selfies!) but at least cosmetic surgery is usually patient-driven and paid for out-of-pocket.

No, I'm concerned about the increase in supposedly medically necessary surgeries.

What started me thinking about this was an article in The Wall Street Journal detailing the increase in women with early stage cancer in one breast who are having both breasts removed.  The proportion rose from 2% in 1988 to 11% in 2011.  For most women -- unless they have a BRCA genetic mutation -- the double mastectomy doesn't appear to improve chances of survival.

Experts speculate that women may be making the decision out of fear or anxiety rather than objectively looking at their risk, but it still make one wonder why that percentage is rising so fast.  Is breast cancer more scary than it used to be?

Breast cancer is a very emotional diagnosis, and I can't really put myself in those women's places.  If it was just this trend, I'd leave the topic alone, but there are sizable increases in many other types of surgery as well.  Here are some examples:
  • Cesarean sections:  In 1996, the C-section rate in the U.S. was 20.7% of all births; in 2011, it had risen to 31.3%, after reaching a high point of 32.9% in 2009.  In 1965 it was 4.5%,  The WHO recommends that the rate should be no more than 15%.  Many blame the dramatic rise on mother's requesting C-sections, more medical need for them, or physicians performing them out of malpractice concerns, but Childbirth Connection disputes all those arguments.  
  • Spinal surgery for spinal stenosis:  Use of spinal fusion increased 67% in Medicare patients from 2001 to 2011, according to a report from the Dartmouth Atlas Project.  Even more telling, rates for spinal decompression vary eight-fold across the U.S., while rates for spinal fusion vary fourteen-fold.  It would appear something other than medical necessity is in play.  
  • Knee replacements: These have soared from 250,000 annually fifteen years ago to over 600,000 in 2012.  The fastest growing population is not the elderly but those 45-64, whose rates have increased 205%, versus "only" 95% for the 65+.  According to a study by researchers at Virginia Commonwealth University, as many as one-third of knee replacements may not be appropriate.
  • Heart Stents: Over the past 25 years, annual use of stents to help narrowed coronary arteries has grown from virtually none to over 500,000.  Even the AMA and the Joint Commission think that is too many, reporting that 10% may be inappropriate and another one-third are questionable. Nortin Hadler, a professor of medicine at UNC, told Bloomberg News: "Stenting belongs to one of the bleakest chapters in the history of Western medicine,"  
  • Bariatric surgery: In 2000, there were only 36,700 bariatric surgeries, versus an estimated 179,000 in 2013 (although that number may have peaked at 200,000 in 2008 and 2009).  Evidence for its effectiveness is mixed, with some studies indicating that the surgery may not have the long term positive impact on costs and longevity that many expected. 
I'm sure more clinically oriented readers can come up with other examples; the list is meant to be illustrative, not exhaustive.

Why is this happening?  There are lots of possible explanations, starting with the fact that we're simply not very good about evaluating risk.  Our brains evolved to deal with a different set of risks than surgery.  Those same brains also evolved to focus on immediate gratification over long term benefits, so when faced with a choice of surgery versus lifestyle changes or treatments that might take longer, well, it's easy to go for the perceived quick fix.

It doesn't help that there rarely are meaningful data on the risks/benefits of the surgery, the specific surgeon, and the non-surgical options.  It's hard to make informed decisions without the right information.

Another reason for the increase in surgeries could be simply because we're developing new and better ways of doing them.  Cataract surgeries are on the rise, but would anyone want to go back to how they were done in the 1960's?  Anyone not want to have minimally invasive options for a host of procedures?  There are truly marvelous surgical approaches available, requiring incredible skills from our surgeons, and it almost seems like a waste not to use them.


Finally -- and there's no avoiding it -- there is the money.  Surgeons tend to be among the highest paid physicians.  The Wall Street Journal exposed significant financial conflicts of interest for spine surgeons back in 2010, which the recent Open Payments data released by CMS have only made more troubling -- and not limited to spine surgeons.  We like to think we're not getting surgery because the surgeon stands to make a lot of money from it, but we'd be foolish to think it doesn't factor into the recommendation.

In a previous post (Mistaking Failure for Success), I had suggested that we should look at hospitalizations as a sign of failure, arguing that each admission should be accompanied by a review of what could have been done to avoid it.  I propose that we should view surgeries in the same way.

Certainly some surgeries are inevitable -- if you are shot or are born with a hole in your heart, chances are you better have surgery -- but I'd be very curious about what percentage a careful analysis might find actually are truly necessary.  A 2013 USA Today study estimated 10-20% of surgeries in some specialties might be unnecessary, and I'd bet a more in-depth review would produce an even larger share.

Any way you look at it, there's more cutting going on than there should be.

Mayo Clinic co-founder Charles H. Mayo once (supposedly) said: "It is the surgeon who benefits most from elective surgery."  That sure sounds like a recipe for an epidemic of surgeries.  More and better data on risks and benefits would be a good start to addressing the issue, so that we could at least try to make rational decisions.

I suspect, though, that we won't make real progress on avoiding unnecessary surgeries until they go from being a revenue source for providers to being a cost center.  If we were paying providers to keep us well, not to do more things to us, surgery might not be as attractive to them.

As surgeon Norman M. Kenyon once said: "The hardest thing about being a surgeon is knowing when not to do something."

To be fair, I could have equally discussed the large increases in procedures (e,.g., colonoscopies) or imaging (e.g., CT scans), for much the same reasons.  Doing more, not necessarily better, seems to be the hallmark of our health system.

Maybe we should spend less time worrying about Ebola and more time worrying about these health system-induced epidemics.