No one wants to hear Odysseus go to the corner store.

– Alex Blumberg

Recently, I was sent an interesting article about how “Cowboy(/girl)” doctors are driving up healthcare costs in medicine.  The authors cite a study that argues that, when it comes to dying, certain doctors “reject ‘evidence-based professional guidelines for appropriate care’ and… order invasive and costly procedures despite little chance their efforts would delay the inevitable”.  These “Cowboy” doctors, as they are labeled, contribute up to 35% of end-of-life fee-for-service Medicare expenditures, and up to 12% of overall Medicare expenditures.  To give you an eye-popping dollar figure, that would amount to $84.5 billion[1].

Beyond questions of morality and what constitutes “little chance” in the study, this raises another interesting question for patients and providers: should we drive toward standardization of care plans?

On the one hand, the word standardization itself implies, at the very least, a reduction in inequality for all.  But furthermore, recent studies of standardization of care across patient populations shows it cuts costs and raises quality, two front and center goals of the IHI’s Triple Aim.  In business school, we had a saying in strategy: “No customization without differentiation”.  Basically, I took this to mean that unless you could prove beyond a doubt that your special snowflake was far superior to the rest, you should not customize the product/company/service you were offering.  Bundled payments programs like CMS’s Comprehensive Care for Joint Replacement (CJR) aim to use standardization to drive improved outcomes and cost, perhaps implying that they believe current customizations between providers are not differentiated.

However, standardization of care (often through the specter of “socialized medicine”) is a common boogeyman evoking images of “death panels”, rationing, and the removal of physician judgment from the medical process[2].  Obviously, depending on how the problem is framed public opinion shifts, and there are different care delivery models that can make these problems better or worst.

As we enter HIMSS19 this week, and we start to see more intelligent offerings that aim to bring standardized, evidence-based practices to global healthcare delivery, I can’t help but hearken back to when the human genome was first sequenced.  The attitude back then, growing up in a family of physicians, was that from that day on every single treatment would be individualized and targeted.  No longer would your Advil bottle say “Adults over 12 take two pills per day”, and gone were the days of treating every infection with Amoxicillin.  But now, we have done a complete U-turn.  HIMSS vendors and attendees are focused on treating populations of thousands or millions of patients with reduced variation, reduced cost, and greater predictability.  Not greater customization.

In fact, I was in a conversation at Cerner last week where we began to touch on some of our advanced genomics offerings.  One business leader remarked that there were well-set standards in pharmacogenomics, but fewer in oncological genomic medicine.  This focus on standards led me to think that there may be a new two-tiered approach emerging in medicine.  First, advanced medical practices are studied and honed at leading specialty centers (for example, MD Anderson for cancer).  Then, they are formulated into practices that are proven to scale to thousands or millions of patients in a repeatable manner.

It’s a far cry from the precision offered by gene sequencing and individualized treatment, and more akin to the University -> Industry pipeline that has existed in the tech industry for years (think Stanford PageRank -> Google Search).  I don’t know what to make of this yet, but I think there are implications that we need to think through:

• Will the drive toward standardization compound the problem of ignoring rare illnesses, which the U.S. had to combat with special Orphan Drug regulations?
• Will the rise in standardization cause a loss of provider autonomy and worsen the problems of burnout and disengagement U.S. clinicians face?
• Is the human body and its millions of variations even “standardize-able” in such a way that allows this approach to work?

Either way, technology has a role to play.  But I always worry about pendulum swings, and my take is that competing approaches should be balanced to find an optimal path.

---

[1] See the HHS 2018 budget in brief for more details.

[2] I’m looking for a comprehensive article or journal paper discussing the systemic effects of rationing in single-payer, integrated delivery systems, and different competitive models but I can’t find anything good.  Let me know of any good links.

Trust is the coin of the realm.

– George Shultz

This week, most people are talking about all the news coming out of JP Morgan’s Annual Healthcare extravaganza, and rightfully so.  However, I don’t think there’s much I can add to the discussion, so I want to talk about another big event that happened in the past couple of weeks: CMS’s implementation of the much-anticipated Price Transparency rule for hospitals.

Unfortunately, over the break I was in a hospital for a family-related medical event for which I would be financially responsible.  Once I knew the issue would be resolved without any major harm or loss of life, my next question turned to: would I be able to afford this? Cerner starts its new plan year along with the calendar year, so being in the hospital in early January means incurring medical expenses without the added benefit of having time to fill your HRA/HSA.

Having knowledge of the new price transparency rule, I turned to Google and looked for the hospital’s chargemaster.  The first hiccup: it wasn’t there.  Being a determined consumer, I politely called the billing department to note that they had not released their prices and therefore were non-compliant with CMS’s new regulation.  To this organization’s great credit, they told me they were working on it and within a week got back to me with the link.

Next, once I downloaded the chargemaster as an Excel spreadsheet, I started to comb through it on my phone.  This was the second hiccup: although the pricing information was machine-readable, it was not human-readable.  Searching through abbrevations like INF, HOSP, etc. was time-consuming and not straightforward, especially as a non-medical professional.

So, lastly I decided to buzz my friend (who happens to be an RN at the hospital, convenient!) and asked her to explain to me what a typical set of codes/services might be for the stay that occurred.  Her response: no clue man.  And that is fair, because she is a healthcare provider and knows nothing about billing.

Looking at Administrator Verma’s take on the new rule, I think she is absolutely right: this is a first step.  It is a great first step and I applaud it, but the major issue I take with the data is not inaccuracy for those with insurance.  Healthcare stakeholders need to understand that data is only as good as the tools that use them.  So, I would propose several next steps for those implementing and administrating price transparency rules:

1. Campaign for Awareness – I was only aware of this new data source because I happen to work in Digital Health.  99.9% of Americans don’t even know this data exists.  Well-placed TV or radio dollars could help get the word out, increasing patient utilization.
2. Encourage Sophisticated Tooling – By the time I got to the chargemaster it was in Excel format and highly complex medical jargon.  There are some great Price Transparency experiments that CMS is running to help modernize federally-run insurance programs, but private enterprise should be encouraged to build direct-to-consumer tools on top of the new information.
3. Integrate Pricing with Workflows – The fact that my friend, who is an RN at the health system, could not tell me how specific clinical procedures mapped back to billing is troublesome.   Better integration between the EHR and revenue cycle solutions will help clinicians guide consumers as healthcare becomes more and more consumer-driven.
4. Create Consumer-Friendly Price Hotlines –  Many major health systems are moving toward employing in-house billing and coding specialists to maximize revenues and improve cost efficiencies across their facilities and networks.  As this happens, health systems should also create “consumer helplines” that answer simple questions about pricing and services ahead of time.  Lots of startups like ZenDesk, LiveChat, and Intercom provide a front-door chat interface that integrates natively with websites and could solve this issue elegantly.

All-in-all, I was not able to get a sense of what the cost for my visit would be and had to wait for the bill to arrive.  This begs a question though: if I had been able to find out what the visit cost, would it have changed my behavior? The answer is almost certainly no.  My main concern was making sure that my family was healthy.  What amount of price savings would I trade off for that? That thinking alone is probably more likely to continue driving healthcare costs upward faster than any downward pressure this Price Transparency effort could provide.

“Why do you rob banks?”

“Because that’s where the money is.”

– Conversation with Willie Sutton

Last month I posted about joining Cerner in their executive development program.  Since then, I have spent six weeks immersing myself in all things healthcare.  I think it’s fair to say that I’m now an expert.

Actually, as you might expect, after spending six weeks reading every possible news source or book I could find, the only thing I know now is how much I really do not know.  It will come as no shock to anyone that healthcare is highly regulated, very complex, blah blah blah.

But one potential cause of this chaos that I wanted to write about today is what I am starting to call the client/user dichotomy.  The reason healthcare is different is because the clients doing the paying are not the people using the products.

Let’s take, by comparison, a simple retail transaction.  In a simple retail transaction I walk into a store to buy some candy.  The cashier (who represents the owners of the company) takes my money and hands me the candy.  I both buy and eat the candy.  The cashier collects my money and provides me the goods.

Healthcare is different.  When you or I go to the doctor to get healthcare services (like vaccines or antibiotics), we are not paying them for their service.  Analogous to most insurance companies, we are submitting a “claim” against the service we got.  So if we got an antibiotic, we are saying to our insurance company, “Hey my body was broken so I had to go fix it, can you pay for this?”

The major thing that’s different here is that usually in an insurance setting, the claimant is responsible for collecting the reimbursement.  When you get into an accident in a car wreck, you pay for the repair before submitting a claim.

In healthcare, you receive the service and the provider (doctor, nurse, etc.) has to go through your insurance to get paid[1].  In this sense healthcare insurance is a bit of a misnomer.  It’s more like a proxy paying service.

But this odd situation is compounded by another complication: the healthcare providers are not tied to the payment from their hospital services and systems[2].  In the retail example we gave above, the cashier is paid by the owner of the store to handle transactions.  But there are preset prices for all goods, the cashier does not make this up on the fly.  With healthcare, the amount that the provider and health system is actually able to collect from a patient will depend on their health insurance, which is not known ahead of time.  In addition, payee models are changing from a fee-for-service modality (similar to paying for the candy you buy) to value-based care.  Each of these specific topics deserves its own post to dive into, but for the sake of explaining the dichotomy, this simply means that the provider is essentially no longer an agent acting on behalf of the person collecting the money, but an independent operator whose work is leveraged by the health system to collect revenue at a later date.

So we get the resultant flow above.  Which means that there are six potential interactions between the key players in the space[3].  For those building businesses in and around the healthcare industry this leads to interesting questions:

1. Who is really your client: the agent that is paying (insurance), the agent that is being paid (health systems), the provider and/or the patient?
2. How can you improve the provider/patient relationship with the presence of these middlemen?
3. Are there ways to completely disrupt the middle layers to bring efficiency?
4. How can this landscape be navigated in a highly regulated environment?
5. How does the fact that 37% of all dollars in this system are controlled by the government shape innovation and delivery?

If you have an analogous industry that would provide color on business models or potential improvements I would love to hear them.  I still have found this model unparalleled, for better and worse.

—

[1] This does not even cover co-pays and deductibles which are their own strange things.  This is just about the rates that providers get directly from insurance companies as reimbursements for their goods and services.

[2] This is a little different in private practice because the doctor owns the practice and therefore has “skin in the game”.  But recent trends show a major decline in private practice from 62% in 2008 to 35% in 2014 and a massive consolidation of major healthcare systems.  Both of these trends exacerbate the payee/provider dichotomy.

[3] Patient to Insurance, Patient to Health System, Patient to Provider, Insurance to Health System, Insurance to Provider, and Health System to Provider

Credit to John Gallemore for introducing me to the “Willie Sutton Rule” in cost accounting.