The iron triangle of health care is not law, but an observation

A recent commentary in the Journal of the American Medical Association titled, “The Iron Triangle of Health Care: Access, Cost, and Quality” reflected that any health care system can only optimize two of the three elements – quality, access, cost.  A health care system which provides the finest quality and best access cannot do so without raising costs to unaffordable levels. An inexpensive health care system available to all cannot do so without sacrificing quality. The iron triangle of health care was introduced to me during my medical school training in the 1990s. Like many others, I simply assumed it was a fact. An immutable law. A fixed certainty that could not be altered any more than gravity.

What if this iron triangle isn’t a fundamental truth or law? Why don’t other industries have their own iron triangle? Is health care really different than aviation or computing?  Asking this simple yet basic question is something medical students and doctors don’t ask. Fortunately, this was not the case for Harvard Business School Professor Clayton Christensen, author of the Innovator’s Prescription. His book not only details the theory of disruptive technologies, but also how companies who do “disrupt” the incumbent companies and the status quo are the ones that ultimately provide goods and services which are more affordable, more accessible, and of higher quality. Might there be analogies for health care? Professor Christensen highlights aviation and computing as two examples.

Commercial flight for the masses was not a realistic possibility over a century ago when the Wright Brothers navigated their winged contraption in Kitty Hawk. Even decades later, as epitomized by Pan-Am in the 1960s, air travel was for the affluent and a special event. At that point, air travel was not accessible or affordable for the general public. Yet, in the 1970s, a Texas start-up known as Southwest Airlines provided discounted travel to the general public by offering low fares, no amenities, and a point to point service rather than hub and spoke system.

And traveling by air was never the same again.

Now, more people travel by air than anytime in history with unparalleled safety. More accessible. More affordable. Higher quality.

Computers had a similar beginning evolving from a product where only available to a few due to cost and complexity of the systems to now where computers are affordable, ubiquitous, easy to use and of even better quality than the past. One of the first computers in the 1940s, was Eniac, a huge and expensive mainframe computer which was not reliable, extremely complicated, and accessible only to academics. Years later, the general public typically accessed these mainframe computers at work via technicians. Access was limited. It wasn’t until the late 1970s that desktop computers appeared. Hobbyists and others, like Steve Jobs, built computers which were less expensive and underpowered compared to the mainframe computers, but they were more accessible to the general public. It wasn’t until many years later that subsequent computers became more affordable, more powerful, and more accessible in the form of laptops, netbooks, and now smartphones and tablets. More people had access to computing because the products were more affordable and of even higher quality.

And computing was never the same again.

Based on Christensen’s model, we can predict that health care will indeed break the iron triangle and demonstrate it is not a law but an observation. The question is who will lead these changes? Insurers? Doctors? Patients? Entrepreneurs?

Our next generation of doctors must be trained in other disciplines outside of health care. We must collaborate and accept other ways of looking at the same challenges through the lens of other disciplines including business school. Yet, there is a loathing for this. There is the belief that health care is different. Yes, we can continue to talk about the iron triangle of health care and accept that as a reality.

We can also say no. The iron triangle is not a law but an observation. We choose a different path.

The truth is that this is the most exciting time in health care with the intersection of better medical understanding, the availability of technology, and the best and brightest minds working on the issues of better quality, better access, and lower costs. As doctors and educators, it is our job to make sure the next generation is equipped with the right mindset to team with others. If not, others will define the future of health care.

This is what worries me the most.

Davis Liu is a family physician who blogs at Saving Money and Surviving the Healthcare Crisis and is the author of The Thrifty Patient – Vital Insider Tips for Saving Money and Staying Healthy and Stay Healthy, Live Longer, Spend Wisely.

Comments are moderated before they are published. Please read the comment policy.

  • Jimmy_Peanut

    Both of the given examples relied in part on technology advances. Even though Southwest business strategy drove change in the industry, it would not have been possible without newer, more efficient aircraft. In healthcare, new technologies have more often driven up costs, not decreased costs, and that is especially true in the short-term. While that certainly is not true universally, I doubt technology is going to be the answer to costs in healthcare. I would suspect that the iron triangle might still hold true for healthcare for the foreseeable future.

    • davisliumd

      It depends on the technology advances you are referring to. In terms of medical technology, then yes what you describe is true.

      If you talk about other technology, teledermatology, the understanding of precision medicine and using protocols and physician extenders to do the simpler tasks (, skin scanner app, or retail clinics run by nurse practitioners and very focused on specific diagnosis), and technology / solution available in both the US and rural areas (handheld ultrasounds, iPhone EKG machines), then it is extremely possible to lower cost, improve quality, and access.

  • AuthenticBioethics

    I think most industries have a similar “iron triangle” of cost, speed, and quality. And it is predicated on the present state of affairs. Yes, a disruptive technology can change the equation to a new equilibrium. The computer has made many things faster than 15 or 20 years ago. Today’s “fast, good, or cheap, pick two” is different from yesteryear’s, but the rule holds today as much as it did before.

    But you seem to imply that market dynamics such as competition and profit motive, which is often the driver of disruptive technologies, might play a role in health care. Today’s computers are better and cheaper and more easily obtained than ever before; but that state of affairs, it could be argued, is also the result of natural market forces. If that might apply to healthcare, then the present stat of affairs could be seen as an artificial rather than natural market, and the solution might be less intrusive tinkering rather than more. That may not be what you have in mind, but I would tend to agree with it.

    Still, I think when healthcare is seen primarily under a business model — and market dynamics do not necessarily mean that the thing in question is merely a business — that it will also suffer. It is not about economies of scale in dealing with disease, but helping a human being regain or preserve his health. Medicine is also an art, and art does not care about market dynamics but is its own end.

    • davisliumd

      My concern is that the way medical students are taught about the iron triangle of health care that it is immutable. I agree with you that perhaps a better way of framing it is that disruptive technology can change the equation to a new equilibrium.
      Not sure I meant to imply that market dynamics like competition and profit motive might play a role in health care. What I do know, however, that there are plenty of entrepreneurs who see tremendous opportunity to “think different” about health care and are actively trying to create new solutions.

      The question I have is whether this group will be best equipped to solve it alone or perhaps better solutions would be created if doctors and entrepreneurs teamed up?

  • Jack Cain

    The “iron triangle” *is* an immutable law for every discipline. Being in IT, that is understood by anyone achieving a certain level of competency.

    What is variable is the *size of the triangle*. At this point, there is a minimum size, but that size – as Jimmy_Peanut points out – is dependent on the specifics of a time and place.

    The danger and opportunity lie in what you accept as “normal”. I have made a living beating “normal”, but only once was that advance in capability understood and acknowledged (at IBM). Even in the best management chain I have ever had, with a team of extremely capable individual contributors, the advancement of the state-of-the-art was never truly realized.

    The key to breaking down the size of the iron triangle is not innovation. Innovation and “disruptive
    technologies” happen all the time. The key is having people who can translate “tech” or “medical” or “whatever” into terms that can be easily recognized and understood by the business managers who drive what the rest of us have to use at the macro level.

    A genetics researcher understands genetic diversity and individual variation. A Hospitalist understands that patients must be monitored to make sure an individual treatment is effective
    because different patients can react differently to the same medication. A project manager understands “there is way too much variation for us to deal with because we can’t spend ourselves to perfection”. That’s the human side of the “iron triangle”.

    You know the most useful characteristic of a triangle? Only two of three sides meet, and then only in one place. Great for structural integrity. Not the best configuration for interaction and collaboration.

    • davisliumd

      Certainly there will be possibility of variation in outcomes given the same diagnosis and using the same medication. However, increasingly we are finding that patients have more in common than we thought, yet doctors haven’t be willing to standardize protocols. Standardization of protocols can only occur if the understanding of illness is at the level of “precision medicine” a term coined by Professor Christensen.

      Certainly award winning writer and Harvard surgeon Dr. Atul Gawande as spoken much about this

      The hospital system, Intermountain, is one of the national leaders in understanding standardization to provide consistently high quality care as well as balancing the individual variation that will inevitably occur, learn from it, and then further standardize the exception. Dr. Brent James notes in this NY Times piece –

  • Margalit Gur-Arie

    Applying Prof. Christensen’s innovation theory to business is brilliant. Applying the same to health care is disastrous in my opinion.

    The Southwest airline legendary model was to enter the fray under market, with a service that was admittedly not as good as the existing services, but was “good enough” to attract non-consumers of commercial flights because it was cheap by comparison. This is most likely the model for all the “cheap” health care we are seeing now, such as minute clinics, tele-health, nurse practitioners instead of doctors, etc.

    While standing in line and fighting for seats and getting peanuts for service and ending up in an airport miles away from your destination did not kill anybody and was certainly better than driving, getting second rate health care, when compared to the first rate health care available through more upscale institutions, could very well kill you. Not to mention that, at least in this country, we don’t really have non-consumers of health care that can be courted and acquired by a cheap service (unless they are forced to accept a cheap service). Perhaps globally, cheap medicine is better than nothing, but I find it morally reprehensible to offer this type of second rate medicine to the masses and call it innovation.

    The end result of the Southwest experiment was that the company became very successful and in the process it dragged the entire commercial airline market to its lower service standards. Except of course those who can and do continue to fly first class with other airlines, and those who have their own private accommodations for flying, contrasted by the still very large population at the bottom who is unable to buy plane tickets even at the new Southwest rates, now available at all bare bones airlines.

    As to that iron triangle, Southwest like every business operated within the triangle. It released the cost constraint a bit, and balanced the equation by driving quality into the ground, which drove up volume to make the money add up. This is just fine for most industries, but not for medicine.

    We need a different way of thinking if we want to find a truly innovative health care solution that doesn’t require different triangles for different people based on ability to pay..

    • davisliumd

      As you may know, Professor Christensen in his book, the Innovator’s Prescription, didn’t actually attempt to draft on this theory of disruptive technology on health care, but rather he tried to understand the issues plaguing it. As a consequence, he spent a decade writing the book, which had he known it would have taken him that long, he probably wouldn’t have started.

      The first rate health care available in this country isn’t available to millions of Americans for a variety of reasons. Would “second rate health care” be better than no health care? You are correct in that – “cheap” health care we are seeing now, such as minute clinics, tele-health, nurse practitioners instead of doctors, etc. Specifically to Christensen’s case, he describes these areas as precision medicine, where illness and disease can be protocoled and treated easily so more complex problems are directed to more expensive resources (doctors, hospitals). is a good example.

      I suppose we can disagree about whether Southwest lowered the entire industries service standards. The simply went to the market and asked – if we offer multiple flights that were inexpensive, and offered no inter-airline transfers, nothing but “peanuts”, would people choose that over traditional airlines?

      In health care, would patients prefer care via technology like zipnosis, teledermatology (without seeing a dermatologist in person –, using apps (skin scanner apps), and using computer analytics from IBM Watson instead of seeking one on one an oncologist at Sloan-Kettering ( At this time, it isn’t clear, though Vinod Khosla, co-founder of Sun Microsystems as well as a partner in a couple venture capital firms is betting on YES –

      Whether global health care trails the US and people have to fly far to get “second rate care”, may no longer be the case with Dr. Devi Shetty, famed cardiothoracic surgeon, breaking ground on a new hospital in the Cayman Island, a short flight from Florida.

      Dr. Shetty was profiled in Fast Company and the Wall Street Journal in the article titled – The Henry Ford of Heart Surgery – – and it is clear that their outcomes are far from second rate.

      Looks like the Patriots and Rams will be playing this weekend!

      Always appreciate your thoughtful comments.

      • Margalit Gur-Arie

        Dr. Liu, I think we should be careful about creating brand new and much bigger problems in the process of addressing the health care issue. Health care in this country is expensive, but by all measures the excess costs are not due to the expense of diagnosing disease. It isn’t the primary care profession take home income that is bankrupting the system. So why do we need websites or machines to replace the diagnosing primary care doctor?

        Sure, it seems very convenient to go online and have your cold diagnosed for $25. I would argue that everything zipnosis has to offer is already available during a stroll through a Walgreens aisle. Perhaps the “worried well” will find such service cute and reassuring, but folks in poverty will never spend $25 for someone to tell them they have a cold. This service, just like Mr. Khosla’s other machinery, is not solving a problem. Instead it is creating a new venue for extracting some more money from the system, and in addition, these things are beginning to transition medicine from an anchored profession to one that is geographically insensitive, which brings me to the cheap excellence centers in India.

        The largest cost component in a hospital, as I am sure you know, is payroll. These places can afford to price their services so low because everybody from janitors to physicians in India get paid a fraction of what they get paid here. They can buy their supplies from places where the line workers are paid practically nothing more than “peanuts”.
        What this solution does is to provide starving wages to locals and drive US wages down into the ground and/or move all these previously good jobs in the US overseas where exploitation is a way of life. This is what globalization has done so far, and if you guys are not careful the medical profession is next in line.
        And guess where all this disappearing wealth is going? This is the Walmart business model on a global scale and it’s no innovation that I want to see occur.

        Back to the US, if we could just find a way to use computers and advanced technology to get rid of administrative complexity, we would shave a significant unproductive expense off the books of our health care system. Computers and technology are perfect for the task with no need for X-prizes, and yet nobody is simplifying administrative tasks. Why? Why isn’t Mr. Khosla building a Watson based administrator that can conduct all payer transactions without bothering the doctor and her staff? You have to wonder……
        Why aren’t we throwing away all this antiquated bureaucracy and make everything clear and simple? Probably for the same reason: no one individual or corporation stands to make billions from it. And the opposite may also be true.

        Americans don’t go to doctors more than other developed nations and they don’t go to hospitals more often either. However, Americans are the only developed country where profit seeking entities are running the system. This is, in my opinion, what has to change if we want better and more affordable care for all. Otherwise, we will just get a vicious cycle of cheaper care for increasingly poorer people. Doctors included.

        I hope the London air has miraculous effects on the Rams, otherwise… maybe Prof. Christensen can suggest some much needed innovation here too… :-)

        A pleasure as usual….

        • davisliumd

          The example of zipnosis was not meant to imply that precision medicine only applies to primary care or that the brunt of the health care costs in the US is from primary care. As medical care becomes more precise, the more expensive parts of the system are avoided. For example, better understanding of heart disease and use of statin medication has decreased the need for open heart surgery and stent placement.

          Agree globalization will affect medicine they way you describe.

          The Watson article (or another similar one) indicated that insurers plan on using it to speed authorizations. After all, if the advice suggested is from the experts at Sloan-Kettering, that is the right care, right treatment, and hence no need to delay approval. Certainly, athenahealth is another example of using technology to decrease administrative hassles and costs.

  • buzzkillersmith

    I see that Dr. Liu has been smoking the theory again. Hard for him to stay away from the stuff.
    As others have noted below, applying the experiences of airlines and computers to medicine is absurd. Let us dispense with that.
    Medicine, unlike computing, for example, is a personal service industry in which the physician ideally spends time with the patient, gets to know the patient and his or her preferences, and helps the patient. It’s the time and personal attention, Dr. Liu, at least in primary care. Perhaps not so much in cataract surgery, but you’d have to ask an ophtho about that. Commodity industries like computing and banking (although not personal banking for the wealthy) are not as time-constrained. The time constraint, based on the need for personal attention, is the basis of the iron law in medicine.

    Prostitution is really a better model here than air travel. Sure pornography can be commoditized, but prostitution is still a personal service. If you want high quality and enough time, it is going to cost, and you might have to wait a bit. At least that’s what they tell me.

    • davisliumd

      Good to hear from you! I didn’t provide the framework of Professor Christensen’s model of how he was able to apply learnings from aviation and computers to medicine. A more detailed summary of his book is here –

      Have you read it?

      I completely understand primary care. That is what I do everyday. How do we use the understanding of precision medicine (where particular diagnosis and treatments are well understood and clear) that we can provide more convenience, better quality, and service? For hernia surgical repair, the work at Shouldice Hospital is a good example

      Your example of cataract surgery is a clear example of precision medicine. Over time, it became better understood the best surgical approach and technique and which were the right patients to have this procedure. As doctors got better and understood more, more of the procedure could be standardized. Over time, prices started to fall as computers become more involved in assisting and planning the surgery as a consequence of understanding these protocols. Result – better quality, more accessible, lower cost.

      I would just conclude by noting you say – “applying the experiences of airlines and computers to medicine is absurd” as an analogy. Yet others, like award winning writer and Harvard surgeon Dr. Atul Gawande offers analogies to health care via cowboys and pit crews ( and the restaurant industry (Cheesecake Factory) –

      You offer – “Prostitution is really a better model here than air travel.”

      • buzzkillersmith

        Gawande ?! Need I say more?

        Cataract surgery, come on. We’re talking about the entire system here.

        At bottom, you are focusing on trivia and completely missing 90% of the triangle. Sure, anesthesia is better than it was, and so is cataract surgery. But this has very little to do with the iron triangle, which mainly results from the fact that good diagnosis, thinking deeply about the internal pathologic physiology of our fellow humans, requires a lot of physician time, and physician time costs a lot. Yet it is the core of medicine. This is where the systemic problem lies– because such deep thinking is not adequately paid for, it is not engaged in as much as it should be.

        All the systems tinkering you advocate has nothing to do with this stuff, and this, sir, is the reason for the primary care shortage that sends people to the ER, that causes diabetes and hypertension not to be treated, that allows for preventable hospitalization, that costs this country an enormous amount of money (high cost), still with poor access. And any one who is paying attention knows that the US has only one side of the triangle now–good quality for the relative few. And you’re talking about advances in cheesecake and airline peanuts technology.
        I say it again, Dr. Liu. Your analysis makes no sense at all to those who actually know about medicine.

        • davisliumd

          Agree that the current health care system does not reimburse cognitive medicine (another term Professor Christensen uses) which is what you described.

          I would disagree however in your belief that doctors’ ability to get good diagnoses is as good as you suggest. Dr. Jerome Groopman’s book, How Doctors Think, demonstrates that we have plenty of cognitive errors and our ability to get accurate diagnoses is far from the level you imply.

          Many of our understanding of internal pathology physiology is well understood and able to be put into a protocol. For example, HIV which the pathophysiology was not known in 1981, is now so well known that doctors have very precise and specific medications which could be managed by a care manager and not a doctor. Is this case, would not the care be better quality, more precise, and cost less?

          You example of anesthesia isn’t quite correct. Yes it is better and allows us to many great things. However, did you know that there were two versions of anesthesia equipment which was widely available in the US a couple decades ago and each had designed valves which turned in opposite directions of the other machine for critical tasks? As a result there were many cases of OR fires as doctors inadvertently thought gases when closed when they were open. This tinkering and standardization between vendors resulted in cutting down fire and improving safety and quality.

          Medicine still has much to do as it is no where near the high reliability found in aviation safety or consistency in service as Gawande argues in his two New Yorker pieces.

          I would disagree that the systems tinkering does not make an impact. Note the recent NY Times oped – Simple Treatments, Ignored –

          My analysis and perspective are from experts, writers, thought leaders, and policy makers on health care and I’ve referenced all of them in both the opinion piece and in the comments.

          It is fine for you to disagree with my analysis.

          • buzzkillersmith

            The death rate from anesthesia has gone down over time. Output, Dr. Liu, output.

            Indeed, diagnosis is uncertain, but one thing that is certain is that lack of attention to diagnosis, mainly because of lack of time, just makes things worse.
            Care management is indeed a good idea for routine diseases, but you are once again missing a huge point.
            The main thrust of your argument is that routinization will break the iron triangle, yet you are now giving examples of routinization improving quality. I never disputed that it can. Of course it can. Checklists and so on, used wisely, are a good idea.

            But we’re talking about the triangle, and you are focused on one side of it, The simple counterargument, which most people at this site believe so implicitly that they don’t even argue with you but rather simply shake their heads, is that such routinization will do little to lower costs and increase access in the long run. Hello, it’s the labor costs! Hello, it’s the drug costs! Hello, it’s the expanding pool of old sickies! While HIV gets cheaper from a labor standpoint, something else, like dementia, takes its place. It was ever thus, doctor. It is the nature of medical care. Enough food gives way to clean water gives way to infectious disease gives way to heart disease and cancer, and on and on. You seem to have little understanding of the obvious truth that health is a moving target and that technological improvement requires more labor and more capital to keep up with that–in the medical field. You seem to have little understanding of the difference between labor-saving capital and capital that uses labor as a complement. This is econ 101 (or maybe 102). So you give us crazyhappytalk about how innovation in processes with save us and tell us about the books you read and the thought leaders you know about.

            My advice: Elevate your game or stop blogging. Don’t wind up like the

            Wall Street Journal editorial page.

          • davisliumd

            You are making a different assumption. HIV patients doing better and their costs are higher because they live longer and also others develop dementia and also add costs, this is true. But is it possible to care for them at lower cost than would otherwise be true if we do the things that are described?

            The point of the piece was not to make false assumptions or mindsets regarding healthcare, particularly for those in training.

            I suppose I don’t understand your point of view. Perhaps I missed it, but I don’t see you offering any solutions. I see mainly personal attacks and interesting analogies.

            “I see that Dr. Liu has been smoking the theory again. Hard for him to stay away from the stuff.”

            “Prostitution is really a better model here than air travel. Sure pornography can be commoditized, but prostitution is still a personal service. If you want high quality and enough time, it is going to cost, and you might have to wait a bit.”

            Professor Christensen’s book the Innovator’s Prescription – have you read it?

            So, what are your solutions? Have you submitted a blog post to Kevin so we all can see your ideas?

      • rswmd

        Isn’t the take-home message from Shouldice and cataract surgery centers is that there are big bucks to be made (even in Canada!) by skimming off the lucrative, easy procedures and leaving the grunt-work to the local docs?

        • buzzkillersmith

          Pretty much.

          • rswmd

            Experts, writers, thought leaders and policy makers would agree.

          • buzzkillersmith

            You are wise beyond your years.

        • davisliumd

          Being able to do procedures in situations which are high volume will be better quality and lower cost. This is occurring whether we like it or not.

          If this is the future reality, what can local doctors do?

          Had the UAW thought about the possibility of work moving overseas, how might they have responded differently?

  • rswmd

    Once again, the OP shows himself to be clueless.

  • Dorothygreen

    “Americans don’t go to doctors more than other developed nations and they don’t go to hospitals more often either. However, Americans are the only developed country where profit seeking entities are running the system. This is, in my opinion, what has to change if we want better and more affordable care for all. Otherwise, we will just get a vicious cycle of cheaper care for increasingly poorer people. Doctors included.”

    Bravo! you are right

    And please add:

    1. A VAT (RISK) Reduction In SicKness tax on processed sugar.,non-essential and heat and chemically processed fat and excess sodium (other countries are doing) on all packaged foods such as was imposed on tobacco – with a STRONG messag on the packages of the RISK for chronic prevenatable disease – over half our health care costs. We will not reduce costs or waste until we reform our eating culture. Call it the Creative Destruction of SAD – Standard American Diet.

    2. Get rid of the $750 B in waste (IOM study). Changing health ID and FBI and computers should take care of fraud. Standardization of the medical record and a better design, say, from the the output summary one can see at a glance on paper or the computer working backwards to the input should take are of the adminstrative waste.

    3. Design a Healh SYSTEM similar to Switizerland such that for profit insurance for essential services is outlawed – let there befor profit for supplemental. A SYSTEM where employers are not mandated to provide insurance. Their costs are half ours and the next highest in the world. 1000 insurance companies in that small country Patients usually pay up front (to a point) and then are reimbursed within a week), obviously a good computer system, probably standardized. Participation is mandatory and subsidies are about the same as US.

    They don’t have a Medicare. They are talking about one thought – I don’t know where I would be without Medicare AND I haven’t even used (and will not) the amount of money I contributed over the years. And, BTW Switzerland has 9% obesity and is concerned.

    Seems to be the role of government in Switzerland is to keep physicians from performing unnecesary services and insurance from overchargng people – not involved in administrative details.

    Do all these things and we can have our access, quality and and lower costs as well. It seems to be a win-win for both political parties and any others that might come along. Lots fo private supplemental insurance and enough not-for profit to keep jobs in the private sector and out of government, job mobility, decreased cost and of course – no other country even talks about the possibility of not having universal coverage.

Most Popular