Previous Submission

Hospitals: Remnants of a Bygone Era

Next Submission

Hospitals would make healthy people sick. They need to be disrupted.

If, just for fun, you admitted a healthy 30-year old male into a hospital, and put him through the ringer as if he were any other ill inpatient, the previously healthy male would likely emerge from the hospital worse than when he entered. Even if he didn’t contract an infectious disease, he would  be de-conditioned from lying in bed most of the day (without appropriate space for physical therapy due to cramped hallways and sharing a double room), eating an unappealing diet, and getting little sleep due to the constant noise and q4h vitals. Yet, somehow, we expect sick patients to improve in precisely the same environment that would make any healthy person ill. At an abstract level, medical treatment of patients is a series of actions that tries to facilitate the body’s ability to heal itself. Unfortunately, the setting in which we practice medicine, the traditional hospital, impedes our ability to do this.

Thankfully, new organizational models of healthcare delivery are beginning to emerge—such as integrated practice units (IPUs) and the CVS Minute Clinic—that shift care delivery out of the traditional hospital into more specialized, integrated, and efficient care delivery settings that provide more appropriate and targeted care for patients. In parallel to this shift in setting, standardization of certain routine processes is occurring. Standardization is a no-brainer improvement in healthcare—it frees physicians from having to stress over routine and mundane tasks, instead allowing physicians to operate at the top of their license by freeing their energy and intellect to focus on more interesting and challenging problems. Nonetheless, as noted by Lucian Leape, standardization has run across opposition in healthcare. As I argue below, however, standardization is not only crucial for current improvement in patient outcomes, but it is also crucial in its ability to facilitate the disruption of behemoth inefficient hospitals by more specialized, maneuverable, alternate delivery organizations.

Enter Prof. Clay Christensen’s theory of modularity and interdependence. Although I am unable to do it justice in this brief space, Christensen’s argument roughly goes that, when the functionality of a given product is sub-par compared to consumer’s demands, then companies with proprietary structures that are integrated across all components required to make the final product will succeed. When a product isn’t good enough, such fully integrated firms are able to harness engineering innovation and creativity across multiple components of the value chain to create a better final product. Such was the state of U.S. medicine 40 years ago. Only highly integrated hospitals had the sophistication to continue innovating in care delivery throughout the value chain, from diagnosis to treatment.

Christensen continues, however, that when product functionality is adequate for the consumer, then the consumer begins to demand improvements in speed, efficiency, and value that can be better achieved by non-integrated, specialized, modular companies. Such is the case for healthcare today—although complicated patients still often find their diagnostic and treatment needs unmet by our scientific capabilities, the pinnacle of our diagnostic and treatment capabilities is more than satisfactory for the average patient, if properly applied. As a result, we are seeing the rise of IPUs, Minute Clinics, and do it yourself diagnostics (ex. what was ultimately the vision of the now unsuccessful Theranos), to name a few. A key step in the shift from full integration to modularity is standardization—once satisfactory functionality is achieved, processes in the value chain can be standardized and a value chain that could previously only be performed by one company can now be split-up into multiple components. Although standardization limits design freedom, it allows for multiple, independent, non-jack-of-all-trades organizations to disrupt incumbent fully integrated firms through flexibility, innovation, and lower fixed costs.

Thus, standardization is powerful not only because it promises to quickly improve patient outcomes, but because it is a necessary catalyst to dissolve archaic hospital structures in exchange for IPUs and other care delivery structures that hold the promise of better value. That said, not everything in healthcare should be standardized—but for those diagnostic and treatment components of the care-delivery value chain that are already excellent, modular structures that break away from traditional hospitals hold a huge value proposition. Modularity allows for more innovative players who can compete to increase the value of healthcare to patients, a necessary disruption in an archaic healthcare system.



Leape LL, “The Checklist Conundrum,” NEJM 3/13/14:

Christensen CM, The Innovator’s Solution, Ch5


5 thoughts on “Hospitals: Remnants of a Bygone Era

  1. Thanks for sharing – your post was very enlightening. I am unfortunately all too familiar with the inpatient stay, and your argument for modularity (vs. archaic hospitals) through standardization in order to achieve better care is thought-provoking.

    Instead of moving medical treatment out of the traditional hospital, I was wondering if there are ways in which we can revolutionize the hospital setting to make it more palatable to patients and their families? For instance, if health care is ultimately a service sector, could hospitals emulate the positive traits of, say, hotels? Is something like a “patient hotel” plausible?

    Related Articles:

  2. This article and analysis were very interesting and reminded me of a lecture that Jeb Bush delivered at HKS this week. Jeb, who is a long time advocate of education reform, was speaking on the inadequacy of the existing education system and advocating that a total revamp as the best and only solution. His rhetorical question to the audience has parallels in your analysis, he asked, “If we were to develop a system from scratch today, would it look anything like our existing system? No.” Applying this question to the health system, if we were to develop it from scratch today, I wonder whether we would retain any large hospital systems, or would be want to replace all health care delivery with nimble value focused IPU-like models?

  3. Thanks for sharing this perspective. The NY Times had an article last week about doctor visits at-home (a la 19th century medicine) that I think is particularly powerful in this context. They basically challenge the norm of making the sick or injured come to an inconvenient, expensive, and foreign location to receive care.

    While I tend to agree with a lot of the thinking here and love your modularity/independence analogy, I am a bit pessimistic about “disrupting” the hospital setting. Christensen’s theory would recognize companies like MinuteClinic, One Medical, Heal, or Doctors On Demand as potential “disruptors” of hospitals – taking on over-served customer segments with basic, convenient services (primary care, diagnostics/evaluations, vaccinations, etc.) that can be standardized, modularized, and carved out from the integrated hospital’s scope. Truly disrupting the hospital like you describe above (with the 30 yr old’s inpatient stay), however, entails moving into surgery and inpatient specialty care. One of my favorite investors, Bill Gurley famously challenged the “physics” of the healthcare industry. In other words, just because something should work a certain way in healthcare, the market forces are too convoluted or entrenched to let it happen.

  4. Thanks all for the thoughtful comments, definitely given me somethings to think about and look into further. Came across an interesting TED talk about designing hospitals to “heal”; the talk focuses on developing countries, but hopefully we can bring those design principles back to the US:

  5. Thanks for sharing this article. I do believe the framework of modularity versus interdependence is a good one here, and appropriately characterized. I wonder what it would take for a hospital to become a setting in which a patient was best positioned to heal — would it be capital investment for larger rooms, wider hallways, physical therapy rooms, or would it be the processes with which we treat a patient, be they clinical or even bedside manner. Would these cost increases be low enough to justify this model in a hospital setting, or should it truly be the job of a smaller clinic?

Leave a Reply