What the Khan Academy Teaches Us About What Medical Education Will Look Like Ten Years From Now

From SFO, I carefully followed my Droid Navigator’s directions off Highway 101 into a warren of non-descript low-slung office buildings—non-descript except for the telltale proliferation of Google signs and young adults riding colorful Google bikes.  I drove around to the back of several of those complexes and finally found the correct numbered grouping.  It really could have been any office or doctors’ office complex in the U.S.  The Khan suite is on the second floor.  There’s a simple brass plate saying “Khan Academy” on what looked like oak double doors. I let myself in and immediately encountered a large, central open space—with long dining tables, food, an ample sitting area with couches conducive for group discussions—and a friendly greeting by programmers and staff.  Oh, and computers—there were lots of computers.  As far as I could tell, nobody had their own office—though maybe Sal does.  Everyone was also open, friendly and passionate about the great work happening there.

After some trial and error, Rishi and I found an unused office and huddled around his Mac for a Google Hangout interview with a Bay Area reporter about the Khan/RWJF health care education project.  Later, I met with Shantanu, the Khan COO and former “math jock” high school friend of Sal, as well as Charlotte, external relations, and Matt, software engineer. They’re all long termers at Khan—that means they’ve been there for about two years.  Overall, the energy was pretty electric.  One other small thing—do not be fooled—these incredible people are, how should I put it—ferociously—intense and focused.

Pioneers in flipping the med school classroom

The next morning, Rishi and I met at Stanford Medical School—in the Li Ka Shing Center for Learning and Knowledge—an enormous and beautiful building off Campus Drive near the hospital that did not exist back in my days as an earnest Stanford law student.   We were there to observe some pioneers in medical education attempt to use Khan-like videos to flip the medical school classroom.  This work at Stanford is part of the current Khan Academy and RWJF collaboration. We’re trying to understand what happens when a medical school attempts to use the Khan-style videos to change the classroom interaction.

The class we observed was an ECG Cardiology course.  The professor, Dr. Paul Wang, was everything you’d want from a teacher—smart, informed, compassionate, patient—and popular with his students.  The students were great—also smart—and empowered.  Several noted that they liked viewing the video before class.  In fact, at one point, a number of female students grouped around Rishi—once they discovered he was the man behind the Khan medical school curriculum videos—and gushed.  I momentarily flashed to black and white footage of Beatle-mania.

This Stanford class-flipping experience is new—in fact, it’s just a week old—so I got to see it at its beginning.  In the first hour, Dr. Wang essentially gave the video lecture again.  In the subsequent two- hour small group sessions, the students worked with teachers on ECG problem-solving and games.  They engaged well with the teachers and each other—and seemed to be using new vocabulary and identifying ECG patterns pretty adeptly by the end.

The dean, Dr. Charles Prober—a Stanford Medical School champion of moving medical education content into YouTube format and onto the Khan platform— was there.  Another local champion of this work, Dr. Drew Patterson, associate professor of anesthesiology, was there as well.  These leaders, Drs. Prober, Patterson and Wang, are creative and brave people.  They are trying to bring the first ripples of widening care transformation to fortress academia, and no doubt the status quo will not adjust quietly.  Both Drs. Prober and Patterson spoke passionately about the enormous potential of this technology along with changing attitudes about medical education.  They are trying hard to get their medical school to embrace that change and help lead it.

Facing massive changes and challenges

This new work is not without problems, of course.  In fact, it’s pretty challenging.  Rishi and I witnessed early baby steps.  Drs. Prober and Patterson readily admit that they would like to rely more on videos, better empower students to teach themselves based on those videos and more quickly change the role of teachers to be more like coaches.   Rishi and I also, though, wondered about waves of transformation hitting health care now—around, for example, patient empowerment and professional accountability for results and decreased cost.  Those massive changes include efforts to alter the dynamic both between the professional and the patients and among various health care professionals.  Those challenges are enormous and could swamp fledgling incremental efforts to help a few medical students learn well and efficiently.  Right now—in this interesting experiment—these teachers are not yet training for that new day.  I say, though, give them time.

Rishi and I also talked about an even more worrisome point.  What if in the near future much of this learning becomes anachronistically analog?  Imagine the coming proliferation of Watson-like artificial intelligence in health care.  On our visit we observed bright minds learning how to “read ECGs.”  That’s what medical students have done since ECGs came to medicine—that’s part of what medical students must learn.  What about when things change, though?  It’s not too much of a leap to imagine that a device will simply inform teams—including the patient, by the way—of the definitive ECG reading.  All this learning about how to read ECGs would then be superfluous—an “FYI.”  What then?  My guess is that we’ll need professionals who are very adept at taking that knowledge and working together with patients do the actual healing—you know like Bones on Star Trek.

We have a lot of work to do.

Michael W. Painter, JD, MD is the senior program officer at the Robert Wood Johnson Foundation. This post originally appeared on the RWJF Pioneer Blog.

12 replies »

  1. I am extremely impressed with your writing skills as well as with the layout on your blog.
    Is this a paid theme or did you modify it yourself?
    Either way keep up the excellent quality writing, it is rare
    to see a great blog like this one these days.

  2. OK: In my situation as an anesthesiologist looking at ECG’s in the PACU (not the EP lab) the machine readout is immediate and usually tells me that I don’t have to rely on waiting for a cardiology reading to pros=ceed with treatment.

  3. platon20–yes, I agree with you…now. My point was that at some point machines will be better than humans at reading these sorts of results–but they’re not there yet.

  4. You’re kidding right? The “regular” ECG machines absolutely suck at interpreting anything meaningful. Sure, there’s a few advanced computer algorithms out there that are supposedly pretty good, but those are NOT available commercially, they exist only in research labs.

    A run of the mill EKG machine will alert you to a possible “S-T segment elevation” but any cardiologist will tell that at best it is a 50/50 guess on whether it is truly a meaningful interpretation that will change management or decisions.

    Certainly no cardiologists that I know feel that the EKG machines are accurate enough to bypass the actual human reading of the recording.

  5. As we think more and more about the role of the future physician (and in my own practice), one important theme that seems to emerge over and over is having excellent communication skills. Communicating risk might mean helping a family comprehend the difference between 1/1,000 vs. 1/1,000,000 chances of an adverse event. Communicating a willingness to listen might help a patient divulge their drug addiction to you. Communicating a differential might help a patient understand why they’re being sent off to see 2 other specialists. We don’t usually teach/learn how to develop good written, verbal, and non-verbal communication skills in medical school, and yet it seems that the role of a doctor is rapidly changing from conveying technical information, to helping a patient interpret and contextualize technical information – in many ways, it’s like being a good coach!

  6. I completely agree! We are collectively realizing how important it is to start addressing these fundamental aspects of health and medicine through all educational channels: in-person and on-line.

  7. “the real revolution will be in educational content”

    From the Weeds’ “Medicine in Denial”:

    …it is useful to look back a century ago to Abraham Flexner’s famous 1910 report on medical education. At that time, many physicians were educated outside of universities. They attended trade schools with low admissions standards, and much of their learning occurred through apprenticeship. Their training did not keep up with scientific advances. Rejecting this model, Flexner advocated the Johns Hopkins, post-graduate model of education, founded in basic science, conducted at universities, and oriented towards research, not practice. As described by Paul Starr, Flexner saw that “a great discrepancy had opened up between medical science and medical education. While science had progressed, education had lagged behind. ‘Society reaps at this moment but a small fraction of the advantage which current knowledge has the power to confer.’“

    Were Flexner to return today, he would find that current knowledge has the power to confer vastly greater advantage than it did a century ago. But he would not find that society reaps a greater fraction of that advantage. “Between the health care that we have and the care we could have lies not just a gap but a chasm,” the Institute of Medicine has found.234 Failings in medical education and credentialing are a central reason the chasm exists.

    These failings are rooted in Flexner’s embrace of the university model of formal education. This model was seen as the only way to bring scientific advances to medical practice. Scientific advances were viewed as advances in knowledge, overlooking the advances in intellectual behavior that engendered modern science (as Bacon envisioned). And knowledge was seen as residing in the human mind (Karl Popper’s World 2), rather than as objective content existing independently of the mind (World 3) (see the discussion at note 108 above). Flexner thus missed the crucial insight of his contemporary, Whitehead, who saw that civilization advances by lessening dependence on human thought — an insight that Hayek applied to the domain of commerce (see our discussion at notes 103 and 133 above). By missing this point, Flexner helped erect a barrier to quality care in medically underserved communities. The barrier is dependence on highly educated, expensive physicians who do not come from those communities. Affordable training in medicine for local inhabitants must have become more difficult to find after the Flexner reforms. The loss to local communities from this phenomenon is not just reduced access to affordable care. The deeper loss is a decline in quality of care, resulting from cultural disconnect between physician outsiders and local patients. As compared to outsiders, practitioners drawn from the patient population can deliver better care by reason of their personal relationships with patients and personal knowledge of the community. No medical training can create these connections.

    Now, at the centennial of the Flexner report, its basic perspective remains in place. The health care reform legislation does not contemplate fundamental change in medical education and credentialing, and that issue is not even on the agenda for the future. In particular, the concept of the highly educated physician at the top of the hierarchy of practitioners is still accepted as inherent in advanced medical practice.

    This point of view ignores what Francis Bacon and modern cognitive psychology have shown about the limits of the mind. It ignores Karl Popper’s distinction between World 2 and World 3. It ignores John Dewey’s insight that education must be tied to experience, that learning depends upon doing (discussed below). It ignores powerful critiques of credentialing systems based on formal education. It ignores the turning point in medicine’s history that modern information technology represents. It ignores the reality that Flexner’s approach led practitioners away from using information technology for what should be its core function—combinatorial analysis. And it ignores the experience of graduate medical education—for many, an experience of disillusionment.

    2. The medical school experience

    According to the Institute of Medicine, “many believe that, in general, the current curriculum is overcrowded and relies too much on memorizing facts” and that “the fundamental approach to clinical education has not changed since 1910.” Even though the issue is largely absent from the health care reform agenda, many involved in medical education recognize that this stagnation is unacceptable.236 Consider the following 2003 commentary on Harvard’s New Pathway curriculum. After reciting that this reform “reinvigorated the educational experience” and “served as a national model for similar reforms,” Dr. Joseph Martin described the sense of futility felt by many:

    “But despite all the New Pathway has accomplished, one of its central aims— the true integration of clinical and basic science learning throughout four years of medical school—remains a largely unfulfilled promise.

    … There is a pervasive and growing sense—not only at Harvard but around the country—that current approaches are no longer working.”

    … The core purpose of Flexner’s reforms was to bring scientific knowledge and rigor to medical practice. A core justification for the enormous time and expense of physician training, and for the legal monopoly and high compensation conferred on physicians, is their scientific training. Presumably that training enables physicians to apply medical science to patient needs with scientific rigor. Yet, one of the leading medical schools in the world here describes itself as failing to provide adequate experience in the elements of clinical medicine, failing to provide good learning conditions in either hospital and ambulatory settings, failing to provide uniformity of content, failing to enforce educational rigor, failing to reliably evaluate students’ core competency and failing to integrate basic science and clinical medicine.

    Failure to integrate the two is predictable, given what happens in the medical school curriculum. At the beginning, faculties overload students with abstract knowledge—textbook answers to questions they never asked about observations they never made. Learning of this kind is the antithesis of scientific inquiry. Students who undergo this process can easily become doctors who “quote what is in the book and deny what is in the bed.”… [pp 195 – 199]

  8. This would have amazing implications for CME.
    ECG’s are already pretty much read by the machine that records them.

  9. The revolution in medical education will utilize these innovative technologies- But the real revolution will be in educational content,

    Look for much more emphasis on prevention, public health, nutrition, exercise,behavioral science, aging, death and dying ,sociology and spirituality etc

    Dr. Rick Lippin

  10. Are these Video medical Lectures mentioned in the article available online for viewing? I am a Medical Student and would find them very helpful in my studies. Anyone aware on this issue, please do let me know if they are available or not, and if available, where. Thank You!