Uhhh…seriously, HD? You’re an Endocrinologist who writes inflammatory posts that are occasionally funny and sometimes educational. What do you know about fixing a system as massive, convoluted, and broken as the United States’ healthcare system? Did you get an MBA that we don’t know about, perhaps in your spare time?
Spoiler alert: I do not have a masters degree in business, nor do I have a magic bullet for the healthcare debacle we face. All I’ve got is a plethora of somewhat informed opinions based on years of working the front lines. Fortunately, there are people way smarter than I who have spent over a decade researching this unbelievably thorny problem. One of my readers (thanks Emma!) turned me on to a fantastic book called The Innovator’s Prescription, written by lead author and Harvard Business School professor Clayton M. Christensen, along with his accomplished co-authors Jerome H. Grossman, MD and Jason Hwang, MD.
Why do I like this book so much? Sure, it did a stellar job explaining the genesis of our system’s major problems, including unaffordability, unsustainability, fragmented care, suboptimal quality, and poor overall value. But then it laid out a vision for what a better system would look like and presented a road map for how to get there. I’ve spent years sitting in leadership meetings, listening to CEOs, COOs, and medical directors lay out their vague strategies for keeping our organization’s head above water. In all that time, I’ve never heard a cogent plan for how we can transition from our current, dying fee-for-service model to something sustainable. This is where The Innovator’s Prescription really shines.
It’s all about Disruptive Innovation
The primary message of this book is that we need truly disruptive innovation – not incremental tweaks – to effect change on the scale necessary to remake our healthcare system. The authors walk us through plenty of examples of how disruptive innovation in other industries led to transformation of business models in those industries. Based on my experience with the incremental change approach – which is all we’ve been able to do in my organization, given financial and structural constraints – my interest in disruptive innovation has been piqued.
Christensen et al recognize that a major driver of our exorbitant healthcare costs comes from care delivered in general hospitals. Essentially, our hospitals try to be all things to all people, which means that they must be prepared to diagnose and treat absolutely anything that comes in the door. This results in very high overhead and thus very high bills for the healthcare system. It also means that all of their fancy equipment and expensive specialists are arranged in silos, which leads to poor integration; hospitals just aren’t designed from the patient’s perspective of what that patient needs to get done. This tends to result in fragmented care that may not be high-quality care. It’s certainly not good-value care, which is why we are starting to see some insurance companies defining value/quality by how well healthcare organizations do at keeping their patients out of the hospital.
The authors propose three phases of disruptive business model innovation in healthcare that they say have the potential to reduce costs 20-60%, depending on what type of clinical situation we’re discussing. They also claim that these new models will improve quality and efficacy of care received.
Hospitals have to be carved apart into several constituent components. Sound disruptive to you? Yeah, I know, it sounds pretty crazy and quite the challenge. For the past decade around here (and probably where you live), we’ve seen consolidation after consolidation, often with hospitals gobbling up independent practices as well as other hospitals. A cynic might say that this has all been designed to funnel more patients to the mothership (hospital) in order to keep the beds full, thereby perpetuating our expensive system and lining the pockets of overpaid hospital administrators. Of course, there is nuance to this about which I’m choosing to be obtuse – these hospitals need to attract as many of the highest reimbursing problems (like specialized surgeries) they can in order to subsidize the myriad other things they have to do that don’t pay well. The problem is that all of this consolidation has led to the provision of more expensive care* that is taxing the system at unsustainable levels.
OK, so hospitals have to be carved up…but into what? The authors spend much of the book explaining their vision, which boils down to creating three major models for three major types of care:
- Solution shops for thorough medical evaluations and accurate diagnosis
- Value-adding process businesses for focused treatment of an accurately diagnosed condition
- Facilitated network businesses for higher-quality, lower-cost care of chronic diseases
When I first read some of the italicized admin-speak, I was pretty sure I wouldn’t last long with this book. Fortunately, the authors do a great job fleshing out just what it is they mean, and their proposal makes a lot of sense. My only complaint is that they should have used terminology that more clearly conveys their message.
With respect to the need for coherent solution shops, Christensen’s theory is that healthcare is often both expensive and ineffective because the original diagnosis is wrong or incomplete. He cites one example of a friend whose asthma remained uncontrolled despite seeing multiple different specialists and spending thousands of dollars in the process. This friend then went to National Jewish Medical and Research Center in Denver, where he underwent a battery of tests. After testing, he sat down with an Allergist, a Pulmonologist, and an Otolaryngologist, who integrated their perspectives on his history and test results; they then explained what was causing his symptoms, and they prescribed a course of therapy that finally solved his problem.
In the general hospital systems where this patient had previously received care, he saw the same types of specialists, but they weren’t integrated in a way that allowed accurate diagnosis. This tends to happen with diseases that cross specialty lines; unfortunately, a huge number of diseases have this trait.
There are other medical centers where this type of care exists – Texas Heart Institute, Cleveland Clinic, Mayo Clinic – and the authors believe that this should become the dominant model for accurate diagnosis:
Patients there are processed through solution shops whose specialists, equipment, and procedures are knitted together across each of the potentially relevant organ system specialties, in order to provide the best possible diagnosis as fast and at as low a cost as possible.
Of course, the savvy physicians (and possibly laypeople) reading this are yelling at their screens right now, “How the heck are you going to pay for this level of care?” The authors have ideas, which I’ll get to in a bit, but first I want to talk about the other two models for care.
Value-adding Process Businesses
At the risk of sounding like a broken record, I so hate this term. Employing my author-to-English dictionary, a VAP business is one that specializes in treating a very narrow scope of disease. For example, the Shouldice Hospital, near Toronto, repairs only external abdominal wall hernias. The process entails a four-day visit for preparation, surgery, and rehabilitation in a country-club-like setting. In the US, this procedure is done as an outpatient, yet the entire cost at Shouldice is still 30% lower. How can this be? Well, a specialty hospital can integrate and optimize everything for performing one specific job, whereas a general hospital cannot. Oh, and the complication rate at Shouldice is 0.5%, compared to 5-10% at US hospitals.
Christensen and his colleagues acknowledge that we will always need tertiary care hospitals that can provide the broadest array of care for the sickest, most complicated patients. But younger, healthier, less-complicated patients could easily be treated at VAP clinics and hospitals with lower cost and higher quality.
Facilitated Network Businesses
This term is even worse than the last one. To translate: the problem the authors are trying to solve is how to provide cost-effective, high-quality care for chronic diseases that require long-term behavioral change and adherence to therapy for years, if not lifelong. It may be great to have a solution shop that helps make an accurate diagnosis, but that doesn’t help with the long-term followup and care of these patients.
In our current model, it’s very hard to get paid for so-called “high touch care,” in which a patient is in frequent contact with some member of the care team (nurse, case manager, physician, etc). But we know that this high touch care works to keep patients healthier, which presumably keeps them out of very expensive emergency departments and the hospitals. I have previously written about a great example of disruptive innovation in the chronic care space – Virta Health. This private, for-profit company is pioneering an app- and phone-based, high-touch model of care for type 2 diabetes. So far, their clinical results are pretty awesome, with incredible patient retention rates to boot. This is the type of care that needs to be rolled out with other diseases, as it has the potential to keep people healthier for long periods of time, thereby keeping them out of EDs and hospitals.
The reason why the authors call these businesses facilitated networks is because they envision the participants (patients) exchanging user-generated content. For example, Alcoholics Anonymous is an organization in which the patients teach each other how to overcome the disease of alcoholism, providing support in the process. Weight management centers bring patients together to discuss the challenges and successes in managing their obesity. There are various non-profit organizations out there as well, designed to bring patients and their families together to better manage a given chronic condition. These are the types of networks that will need to step into the void where solution shops and VAPs just can’t viably provide care.
Although I don’t do a whole lot of diabetes care anymore, I still see enough to know that the facilitated network model is the right model for the future. It is not possible to take great care of the typical diabetic patient, seeing her 3-4 times per year. Highly motivated self-starters can do great checking in a few times per year, but most of my diabetic patients do not fit that mold. If the typical patient had a health coach, case manager, and support group, all checking in regularly and helping keep her on track, she would be much more likely to take control of her diabetes.
Phases Two and Three
Remember that Christensen and his co-authors proposed three phases of disruptive business model innovation; we’ve only worked our way through Phase One, above. Phase Two involves the emergence of lower-cost business models within each of the business models I described in Phase One (solution shops, VAPs, and facilitated networks). For example, among solution shops, a telemedicine-based institution would be a lower-cost disruptor of hospitals and outpatient practices. Among VAP businesses, ambulatory and mobile clinics could disrupt specialty hospitals.
Phase Three will occur across business model types:
Retail clinics, for example, transfer the ability to care for rules-based disorders from solution shops to value-adding process businesses. Firms like SimulConsult, through which the published research of thousands of specialist physicians is integrated on the computers of primary care physicians, disrupts the solution shop of specialists with a professional network business model.
As an illustration of what the three phases of disruption might look like, the authors use the example of angioplasty. Currently, the type of doctor you see for your angioplasty depends on which blood vessel needs opening. Need your coronary artery blown up and stented? See a cardiologist. Need your carotids done? See a vascular surgeon. Kidneys? Go to interventional radiology. Peripheral arterial disease in the leg? You might see any of the above three specialties.
Wouldn’t it make sense to deliver this care in highly specialized environments that are optimized to do one thing – angioplasty/stenting of blood vessels? Specialty heart hospitals have already disrupted general hospitals in this way, but there is room for further disruption. In other words, does it really make sense to organize this care around the medical specialty of the physician?
[T]he competitive category is really the same for all of these patients – blood vessels. Patients undergoing this procedure, beginning with those with stable disease and low risk of complications, would be much better served by seeing a “vascular interventionalist” who does nothing but angioplasties day in and day out in a value-adding process business, surrounded by support staff who are experts in managing all the processes of angioplasty care, no matter which organ system happens to be involved. By reframing the categories of competition, the value-adding process business model allows you to bring scale and lower overhead costs to areas where you didn’t have it before.
The authors don’t think the disruption stops there. They postulate that, eventually, the interventionalist may not even need to be a physician (gonna get some hate mail from professional societies for that assertion). They believe that a technician with good video-game skills could be trained to do these cases, starting with simple ones, of course. Although my initial reaction was that’s one step too far, upon reflection, I do think it’s possible. And it would certainly lower costs. We all agree, however, that experienced physicians should continue to manage the most complex vascular diseases.
Barriers to Disruption
Christensen et al have clearly dedicated much thought to barriers to disruption of the hospital business model:
Achieving these disruptive changes to our hospital system will be extraordinarily complicated, because interdependent changes in pricing and payment methods, regulation, and certification, will all be required. Pricing, performance, and quality data need to be coupled with incentives for patients and physicians alike to make optimal decisions and trade-offs. The tremendous political clout of academic medical centers and general hospitals, often among the largest employers in any community, will prove formidable.
They also recognize that their proposal for disrupting the physician’s practice business model increases fragmentation of care:
…it is disconcerting that the disruptive value network we see emerging is a more fragmented system than the one we have today. Instead of a generalist physician and a general hospital being “responsible” for all our care, we foresee nurse practitioners working in retail clinics caring for our everyday, rules-based disorders; facilitated networks helping us with the care for our behavior-dependent chronic ailments; coherent solution shops to sort out disorders that are still in the realm of intuitive medicine should they arise; and value-adding process clinics for procedures that need to be performed after definitive diagnosis. Given that the coordination of care is already being handled in a hit-and-miss fashion, when we receive care from so many independent, focused providers, won’t the problems of coordination be exacerbated?
Their plan for overcoming obstacles is multifaceted and difficult to do justice by summarizing it; I highly recommend reading the entire book. Nonetheless, I do want to further explore one facet of the plan here.
One major component of their plan is implementation of a universal electronic medical record (EMR) standard. Those of us who are intimately (and often painfully) familiar with EMRs should have significant insight here, so here’s my two cents. The current state of EMRs is: there are a few major players in the space, and they don’t interface with each other. For example, if you get your care within Medical System A that uses EMR 1, but then you get referred to a specialist in Medical System B that uses EMR 2, your care teams at each institution cannot easily view your data. Further, if you drop in to Retail Clinic C for a bacterial sinus infection, they may use EMR 3, which means they are relying on you to tell them that you’re on a drug that could interact negatively with their choice of antibiotic.
The authors say that we need to move in the direction of open source EMRs, such that the EMR really becomes a PHR – Personal Health Record. The data in the PHR would be downloadable/cloud-accessible by any care team or pharmacy, wherever the patient goes, providing the glue that this fragmented network needs.
Unfortunately, because major, proprietary EMRs have already become entrenched in almost all medium-large medical systems, it is unlikely that we are going to see these systems jump to a completely different standard. Although smaller, disruptive medical organizations could theoretically agree on a new standard for EMRs and build a better system, we would still run into the brick wall of the existing players who have invested millions in their systems and don’t want to change. Therefore, the route forward would appear to be technology that translates “‘foreign languages’ into a common one that allows previously incompatible formats to work seamlessly together.”
I agree that having a universally accessible PHR would go a long way toward integrating fragmented care. But, if the route forward for the major players is going to be the use of software to reconcile incompatible systems, I’d rather be working for the smaller disruptive organizations that build a new EMR on a new platform, from the ground up. While our current EMR is great for being able to access large amounts of data quickly, it is absolutely terrible for inputting that data in an efficient fashion. Each new update rolled out has incremental improvements that are often offset by increasing complexity in pathways for entering data. I am tired, tired, tired of using software designed by engineers with no insight into my workflow. I would love to see savvy physicians be part of a technology team that builds an entirely new system that is much more intuitive and predictive.
I realize that I’ve hijacked the argument here by ranting about the poor usability of existing EMRs, but I believe it is highly relevant to any attempt at healthcare reform. When politicians and other influential people discuss these issues, they seem to view EMR adoption as an end unto itself. It is critical that we keep the conversation focused on making these instruments of the devil easier to use, as the downstream effects of poor EMRs are myriad: decreased doctor-patient face-to-face time; decreased patient satisfaction; decreased physician job satisfaction; increased data entry for doctors; decreased physician productivity; increased physician burnout; and the scariest – the potential for missing an important diagnosis because the visit had to be cut short to fulfill documentation requirements.
The Million Dollar Question
How are we going to pay for any of this? The authors admit that challenges in disrupting the reimbursement system have historically been tremendous:
Most discussions about reforming health care run into a dead end when the participants realize that the regulatory system we call reimbursement will not allow it…Those who wish to disrupt the system by changing the very architecture of care, however, are often stymied by the specter that there is literally no money to be made from doing it.
Christensen and his colleagues go on to recount the history of health insurance and reimbursement, to give the reader a sense of how we got to the current state. Then, they comment on the pros/cons of systems around the world, explaining how the different systems have created all kinds of problems. Finally, they offer their “recommendations for two reimbursement systems – integrated capitation and a pairing of high-deductible insurance and health savings accounts – that can overcome the problems created by existing products.”
The reason why I think everyone should sit up and take notice at this point is because what the authors proposed back in 2009 (date of publication) is now happening, at least to some extent. With respect to integrated capitation, they do a great job distinguishing it from non-integrated capitation, the 1980s-era experiment that failed miserably. Integrated capitation involves self-contained systems – like HMOs (Health Maintenance Organizations) that are an insurance company and medical care system, rolled into one – allocating resources to keep people well. This stands in stark contrast to the predominantly fee-for-service system in which the rest of us toil, where we can only get paid for “sick care.” Remember earlier, when I discussed high-touch medical care? Well, in a closed, HMO-like system, this can be profitable if the end result is saving millions of dollars by keeping people out of the ED and hospital. The HMO can then pocket the difference between their members’ premiums and the organization’s medical expenditures.
Over the last 5-10 years, some of these larger closed systems have been expanding their reach; I fully expect that these systems will grow large enough to be able to drive healthcare reform forward, in exactly the way the authors propose. If I may editorialize for a moment, though, I have experience taking care of patients who belong to an HMO. These patients come to me because the HMO doesn’t yet have an Endocrinologist in our area (this will change soon, I’m sure). I hate taking care of these patients. I have no beef with the patients, but the system they’ve bought into is incredibly restrictive, to the point where I cannot prescribe the standard of care for many conditions. The HMO controls costs with an iron fist, which is great if you’re healthy or don’t need expensive medications or imaging. Once you’re sick, good luck getting what you need if it costs money.
So, although I see the merits of a closed system, I also see an incredible amount of pain in our future. American patients will need to be socialized to lower their expectations for what kind of care they are entitled to and how much they can receive. How do you think that’ll go? Not well, perhaps?
Christensen’s proposal for how to disrupt reimbursement when integrated health systems are unavailable is for employers to contract directly with healthcare providers to manage their employees, and pair this with high-deductible health plans (HDHPs) and health savings accounts (HSAs). We are already seeing this with some of the largest employers in our area. I’m not sure how well it’s actually working with respect to cost containment, as we have not yet seen the level of business model disruption the authors claim is needed to lower costs substantially.
Christensen does think that patients need to have more skin in the game when it comes to purchasing healthcare, and I agree with that. The amount of unnecessary testing – sometimes patient-driven – and other waste in the system is legion. But, even if shifting costs to patients to drive more reasonable consumption of care is a necessary evil, that must occur simultaneously with drastic decreases in prices. That clearly has not happened. At all. Have you seen how much branded drugs cost nowadays?!
The truth about our current state of reimbursement is: HDHPs have become quite popular among employers, but patients of modest means who have medical problems hate them. Office visits, lab tests, imaging, and medications are too darn expensive; a patient with a chronic disease and a high-deductible of $6000 is going to max that out every year and still have additional care expenses beyond that. I’m worried that it will take many more years to see costs come down, while the incidence of medical bankruptcy increases for the middle class and below.
If healthcare reform had obvious answers, it would have been done already. However we move forward, there will clearly be many failures along the road. My hope is that passionate, sensible people with the power to drive reform (I’m looking at you, Jeff Bezos, Warren Buffett, Jamie Dimon!) will base this reform on plans like those elucidated in The Innovator’s Prescription. I encourage everyone with an interest in this subject to read the book, as there is so much gold in there that I could not possibly cover in a blog post. For now, let’s just all try real hard to stay healthy – being sick costs too much.
*One example of how the same care can be more expensive at a hospital than at an outpatient clinic: after being acquired by a hospital, your cardiologist’s group practice has the same doctors, the same staff, and the same equipment. But when you get your next bill, you notice that there is a $450 “facility fee” tacked on, which insurance has declined to reimburse. That facility fee is something that the hospital charges to help cover its massive overhead. Sound fair?
It’s your turn now. What are your thoughts about healthcare reform in the US? Do you live outside the US? If so, what challenges do you face in the provision (as a healthcare provider) or reception (as a patient) of care? Comment below!
By interacting with me in the Comments and reading this site, you agree to abide by my Disclaimer.