If You’re Going to Criticize, Get Your Facts Straight

Author Information (click to view)

Skeptical Scalpel

Skeptical Scalpel is a recently retired surgeon and was a surgical department chairman and residency program director for many years. He is board-certified in general surgery and a surgical sub-specialty and has re-certified in both several times. For the last three years, he has been blogging at and tweeting as @SkepticScalpel. His blog averages over 1200 page views per day, and he has over 7600 followers on Twitter.


Skeptical Scalpel (click to view)

Skeptical Scalpel

Skeptical Scalpel is a recently retired surgeon and was a surgical department chairman and residency program director for many years. He is board-certified in general surgery and a surgical sub-specialty and has re-certified in both several times. For the last three years, he has been blogging at and tweeting as @SkepticScalpel. His blog averages over 1200 page views per day, and he has over 7600 followers on Twitter.

A Forbes article criticized a hospital's postoperative infection rate as being about 100 times higher than most manufacturing plants would tolerate. Of all the possible industrial references, the use of automobile manufacturers as a comparator is almost laughable.

I just found out about an article that appeared in Forbes online 7 months ago. It said that a certain hospital system’s postoperative infection rate was 5% and then said, “For US hospitals, this is not an unusual rate of error—even though it is about 100 times higher than most manufacturing plants would tolerate. No automaker would stay in business if 5% of their cars had a potentially fatal mechanical flaw.”

The piece was written by Leah Binder, who is president of The Leapfrog Group, a well-known patient safety advocacy organization. I’m sure Ms. Binder is a smart woman. After all, she follows me on Twitter. And advocating for patient safety is a good thing.

But the two sentences I quoted are so misguided that it is hard to know where to start. Comparing infections to a manufacturing plant’s error rate is not appropriate, nor is the automaker analogy.

Of all the possible industrial references, the use of automobile manufacturers as a comparator is almost laughable. For example in the year 2012, Toyota, the paragon of Lean methods, sold 9.75 million vehicles worldwide and recalled 10.1 million in the months of October and November of that year alone. That’s a little more than 5%. By the way, this has been going on for years, as I noted back in 2010.

Infections, which are not always preventable, are not all the result of errors. Yes, the incidence of infection can be decreased by good practices, but it is not possible to completely eliminate them. Several recent research papers have shown that even near perfect adherence to the Surgical Care Improvement Project’s standards has not resulted in a significant decrease in the rate of wound infections after surgery.

Now it gets even more interesting. As her source, Ms. Binder cited a paper from JAMA. After reading the paper, it turns out that the figure of 5%, which according to Ms. Binder represented the infection rate, was actually the rate of all surgical complications and not just infections. The real surgical site infection rate was only 0.7% with another 1.1% of patients suffering from sepsis. The remainder were complications such as venous thromboembolism, stroke, pneumonia, and myocardial infarction, which are not necessarily due to errors.

Some surgical complications are preventable, and we should try to prevent them. I have no doubt we can do better.

Reasoned and constructive criticism of the medical profession would be more helpful than taking pot shots.

Skeptical Scalpel is a recently retired surgeon and was a surgical department chairman and residency program director for many years. He is board-certified in general surgery and a surgical sub-specialty and has re-certified in both several times. For the last three years, he has been blogging at and tweeting as @SkepticScalpel. His blog averages over 1200 page views per day, and he has over 7600 followers on Twitter.


  1. This conversation reminds me of the story of the car mechanic and a surgeon.
    The chief difference is one works on the vehicle in the garage with the engine off. The other works on the engine with the engine on. I suppose the error rate would be less than the former and shouldn’t be compared with the latter.
    As an anesthesiologist, we have to make certain trade-offs. When I woman is having a baby and requires a C-section, she has a choice of spinal anesthetic with a 3% headache risk(dependent on age), or a general anesthetic with the associated increased nausea and vomiting, etc. Sometimes it truly is a pick your poison situation. We are trying but we don’t have the technology and/or the biological information to completely eliminate “associated” complications or “errors” yet.

    • I was just notified about this comment today. Of course, I agree with you completely. Complications are going to occur regardless of what treatment option the patient chooses. We aren’t allowed to tell them what to do any longer. So we tell them the details of the various choices, most of which they can’t really understand because they aren’t in the medical profession, and they pick.

  2. ofcourse it will be difficult. Misinformation is easy. Truth is difficult.

    When a “problem” is created that is farther from the truth, there is a greater likelihood of harm to patients from the proposed solutions.

    Our goal should be to redirect the debate towards real preventable problems without any additional harm to the patients.

    When the problem is exagerated to 400 k deaths due to errors, it does harm to the patients by diluting the issue and taking focus away from realy preventable errors, trying to prevent unpreventable outcomes.

  3. SS;

    Number of deaths in USA : 2,468,435/yr

    Trends in Inpatient Hospital Deaths: National Hospital Discharge Survey, 2000–2010

    Key findings

    Data from the National Hospital Discharge Survey, 2000–2010

    The number of inpatient hospital deaths decreased 8%, from 776,000 in 2000 to 715,000 in 2010, while the number of total hospitalizations increased 11%.

    The average age of patients who died during their hospital stay was 72–73 years throughout the period from 2000 to 2010.

    In 2000, 2005, and 2010, about one-quarter of inpatient hospital deaths were for patients aged 85 and over.

    Patients with certain first-listed diagnoses had relatively high hospital death rates, but these rates decreased over the period from 2000 to 2010. Hospital death rates decreased for patients hospitalized for respiratory failure by 35%, for pneumonitis due to solids and liquids by 22%, for kidney disease by 65%, for cancer by 46%, for stroke by 27%, for pneumonia by 33%, and for heart disease by 16%.

    Unlike the above-listed hospital death rates, the septicemia hospital death rate increased 17% from 2000 to 2010. The number of inpatients who died in the hospital with a first-listed diagnosis of septicemia also increased—it tripled from 45,000 in 2000 to 132,000 in 2010 (data not shown).

    The first-listed diagnoses discussed above and shown in the Table account for 66% of total hospital deaths in 2000 and 2005, and for 70% in 2010.

    Hospital death rates declined overall from 2000 to 2010 but increased 17% for septicemia.

    Patients who died in the hospital had longer hospital stays than all patients. In 2010, those who died stayed an average of 7.9 days compared with 4.8 days for all patients.

    So except for septicemia, inpatient deats for every other diseases/complications has decreased drastically.

    Out of the ~ 775 k deaths in the hospitals, the author of that absurd study claims 400 k + die due to medical errors…

    The assumption is > 50 % of the hospital deaths are due to errors and ~ 20 % of the overall deaths in USA as due to Medical errors.

    If these claims are not absure then we need to find a new word to describe it.

    What does the author expect, people to live forever ?

    The only way to reduce the so called “deaths due to medical errors” is reduce hospital admissions and let the people die at their home.

    The average age of the patients who died in the hospital is 73.

    Out of the 775 k inpatient deaths a quarter of them are 85 + years old.

    • Thanks for that comprehensive discussion of hospital deaths. I wish I had thought of those points myself. Very well said indeed.

      • “It’s naive to expect Ms. Binder to see the reality that is not aligned with the best interest of her company.

        Her arguments are rational even if untrue.”

        They will cling on to what ever they can to benefit from this absurd claim of ” 400 k + death due to errors.

        Not jusj Ms. Binder there are other vested interests who are in a pursuit to get fame & $$ in the name of ” Patient Advocacy”.

        Patients should be really careful about these pseudo patient adovocates.

        There is another misleading claim by this Physician

        These people are out there to make $$ by making outrageous claims.

        Physicians should have pushed back when the initial claim of 98 k death from IOM report of ” To Err is Human” came a decade ago.

        It’s never too late.

        We should fight back and identify the real errors and real conflict interest that harm patients in health care.

        We physicians~the real patient advocates should prevent the pseudo advocates from misleading the patients.

        • Thanks for the follow-up comments. Your link went to Dr. Wen’s Twitter account. I couldn’t find the specific claim you referred to.

          • Skeptical Scalpel;

            I apologize for the wrong link. Dr. Wen made a claim about vast majority of physicians having relationship with pharma and device industries but that is not relevant to the topic on hand.

            This is the link, and I don’t think he is a Physician.


            Six Frightening Facts You Need To Know About Healthcare

            1. Up to 400,000 people are killed each year due to preventable medical errors.

            2. $765,000,000,000, or 30% of all U.S. healthcare costs, each year is wasted.

            3. 33% of hospital patients suffer some form of preventable harm during their hospital stay.

            4. 58% of clinicians felt unsafe about speaking up about a problem they observed or were unable to get others to listen.

            5. Critical care patients each experience nearly 2 medical errors per day.

            6. 92% of U.S. physicians admitted to making some medical decisions based on avoiding lawsuits, as opposed to the best interest of their patients. ”

            He consludes with this paragraph

            “Being able to accurately report preventable errors and waste in the closed culture of the healthcare industry is obviously a difficult challenge. The studies that generated these statistics certainly aren’t without their own controversies. Some may argue that the real numbers are impossible to accurately determine or that the reported data is either over- or under-estimated. Regardless of the methodologies used, when you look at these reports in totality, there is too much information to ignore. To put it bluntly, too many people are suffering preventable harm and the skyrocketing costs are not proportional with the quality of care being provided. Altering this paradigm should be our primary concern.”

          • That 400,000 number is out on the street and will be difficult, if not impossible to refute–even with facts. And item #6 is a sad commentary on the situation, but how can physicians be blamed?

  4. We surgeons would be happy with patients that never get infected, devices that worked perfectly, and practice error free medicine. When we find a system that works- we adopt it, and we are far from perfect.
    A surgical site infection is competition between bacteria and normal tissue competing for nutrients – bacteria have the advantage because they can reproduce every 20 minutes. Our standards to minimize them over time has decreased surgical site infection, but has not eliminated them. We continue to look for better ways to minimize them.
    Epidemiology does tell us that some hospitals have higher rates, and now hospitals are being held more accountable for those rates with payments being withheld by insurance companies as well as medicare.
    I don’t think any of us are satisfied with poor patient outcomes, and all of us have had them. But the automobile industry does not have the rich academic studies that surgery does- and has the benefit of tossing away steel if they get a bad idea.
    We don’t like mistakes- or errors – or bad outcomes, and to say that we tolerate it is both incorrect and ignores our constant surveillance (which gives some of those inflated numbers) and the volumes of information and studies showing that we do indeed care- and have for decades.

  5. This conversation is a good example of what’s called “group think,” where people are so accustomed to talking with each other and agreeing wholeheartedly with their own perspective that they are blinded to any point of view that contradicts long-held group consensus. But let’s focus on fixing what we agree on: the complication rate in surgery in the United States is far too high, and patients pay the highest price for that. Those who don’t like the word “error” shouldn’t use it.

    • Ms. Binder: My goodness, but you are persistent. Doctors and nurses will address a poor performing physician or nurse, but you have to have the support of administration or the limb you stand on will be sawed from the tree. Hubris and stupidity are a lethal combination in medicine, and that includes upper management and CEOs. I have a lot more I would like to say, but I am weary of restraint.

    • I find it interesting that you dismiss the opinions above as “group think,” instead of pondering the fact that if many learned physicians point out that not all poor outcomes are due to “error,” then perhaps it is your conception or use of the term that could be the real error. While communication in this way is prone to misunderstanding, I can’t help say that this comes across as pretty arrogant on your part.

      As for complication rates of surgery, I would challenge your assertion that they are “far too high.” Obviously, we would all like to decrease any preventable complications related to surgery. As physicians, we take great pride in the care we provide to patients, and take poor outcomes personally. We also realize that medicine today is very litigious, and that any “error” made by a surgeon or hospital could easily result in a lawsuit. Thus, as physicians, I would think that we would have as much or more incentive to minimize complications as would you (a CEO of a company whose basic premise is that there are too many medical errors and hospitals should pay for and use your services to address this problem). For this reason, I find it disconcerting that such a self-professed advocate for patient safety would be so dismissive of the actual surgeons and physicians who would appear to have the best perspective regarding medical errors and the ways to reduce them.

      • It’s naive to expect Ms. Binder to see the reality that is not aligned with the best interest of her company.

        Her arguments are rational even if untrue.

        • Ms. Binder could you please provide us the data ?

          Please do not give us the guestimate that you love to give like the 400 k deaths in hospital due to errors.

          Also please give us data how your company has made any difference when you are hired ?

          I have also read some how miraculously your company has improved safety in rural hospitals, we would love to hear that , how you cooked up those numbers.

          • She based the 400,000 deaths on a recent paper that I blogged about. Here’s the link: Needless to say, I found some problems with it. Read the comments too.

  6. If more thoughtful physicians of conscience, like those who have posted here, took steps to call out/censure their outlier colleagues (the small percentage of HODADS, as described in Marty Makary’s book Unaccountable, known to cause harm), the conversation about reducing harm would be much less urgent. And yes, patients need to understand the risks, the roles they can play, and take responsibility for potential negative surgical outcomes about which they were (or should be) warned. A much more holistic system would be one where MDeities acknowledged they are fallible, and patients understand how to stay healthy and work toward that goal. If the AMA et al truly want to work toward reduction in poor medical outcomes, they could use their considerable clout to urge patient education be taught in middle school health classes.

    • You are right about calling out those who are outliers. But even when we do, it takes a long time for medical boards to take action. See this post I wrote about a Texas neurosurgeon
      I have no faith in the AMA accomplishing anything, so I doubt it will take the lead in educating people to become more involved with their care. Also, I am sorry to say that many patients have no interest in taking charge of their health.

  7. Ms Binder,
    I really thought I’d stay out of this, but your incessant use of the word “error” and your bundling it with the statement: “From the point of view of the patient, it’s a bad outcome, which is an error no matter who made it” is giving me heartburn. Really. I’ll be as nice as I can, but really think you are petulantly clinging to defense of a point of view others have demonstrated to be lacking in merit. I think you confuse the quest most hospitals engage in through quality assurance committees or morbidity and mortality conferences in which they examine bad outcomes in an attempt to uncover error, with the notion that all bad outcomes result through error. Most physicians understand that errors need to be avoided and that there are processes available to minimize them. Most physicians also know that bad outcomes can occur when everything is done carefully and with skill. Your insistence that error is inherent in every bad outcome (regardless of the patient’s opinion…which, like your opinion, does not make it so) I believe has many of us shaking our heads at your lack of understanding. I hope i said this nicely enough.

    • Well said, and very politely. Your frustration was palpable, but you kept it under control.

  8. I don’t think a non-physician can fully understand the complexity of the human organism or the variability of the body’s reaction to stress, whether physical or emotional. The same stressor (surgery) can have very different outcomes depending on the body’s response. Surgical errors can certainly cause bad outcomes but there can be bad outcomes without errors. We should do everyting we can to avoid avoidable errors and the patient safety movement has appropriately brought this need to the attention of the medical community, which has responded in many ways, though often (unfortunately) inadequately. More needs to be done to improve patient safety and avoid real errors. But comparing human bodies to cars and post op complications to bad welds is an invalid comparison. Robotic welders directed by computer programs should never make errors but when a surgeon works with his hands and sharp instruments on living tissues in a complex organism the results just will never be so certain. So, yes, let’s reduce medical errors every way possible, but let’s not use inappropriate analogies to make the point.

    • Well said, and with admirable restraint.

    • Finally!! I am reading this blog, with my morning coffee and thinking: until human bodies can be operated on in the same manner that cars are built, there is absolutely no comparison and the analogy is ridiculous. So, Ms Binder, unless you are willing to be the first human subject to have your gall bladder removed on a moving assembly line, I think it’s time for you to take your soap box and slip quietly out the side door, unless you are willing to concede you have erred. I will ask Santa not to hold this against you, but please do not come back with any more lame arguments. Thank you, thank you so much.

  9. Leah Binder seems to believe that being treated in a hospital means the whole body has be remanufactured to original specifications and would hold a car’s tune-up mechanic responsible for later transmission failure.

    My mother had a painful hip from age 60 that continued to worsen for 23 years before she went in for a hip replacement at 83. She died a couple of months later without really coming out of the anesthesia . I would like to blame someone at the hospital but Mom was in much better health when Dr’s were pushing a hip replacement at age 67 . I have to think that her daily decision to put off surgery for 15 years as the hip pain increased and her general health failed was a poor one in hindsight. That she could convince the Dr to do surgery was really a function of how much pain she was in and the possibility ( not guarantee) of a good outcome.

    I don’t blame my mother but her delaying the hip replacement for 15 years was likely the error .

    • You are right. In this instance, the patient made the error.

  10. Thanks everyone for the interesting and thoughtful comments. I appreciate the support that most of you have given me in my quest to try to define complications and errors.
    I as stated above, I am not denying that doctors and nurses make errors, nor am I against efforts to try to prevent them. I also agree that there are important variations in the numbers and types of errors among hospitals and providers (I hate that term). But let’s focus on the real errors and stop trying to blame every complication on someone or the “culture” of a hospital.

  11. Ms. Binder,
    You say, “for what it’s worth, I think of errors as institutional phenomena”. You need to take a step back and educate yourself until you understand how fundamentally in error you are, and how harmful propounding your point of view can be. What brought the patient to the attention of the physician in the first place. Is their disease an “error”? A mistake? An institutional phenomenon? It is an adverse outcome of their existence, to be sure, but it is a consequence of the imperfection of the human condition, not an error. You would do better to quit defending your point of view because you hold it and makes sense to you in the context of you. It’s a rather limited perspective.
    There are medical errors. Some are episodic. Some are negligent. Some are systemic. Penalizing (true) negligence and identifying and correcting systematic issues are worthwhile endeavors.
    There are complications. Some arise through error (although not all errors result in a complication). Some arise through the imperfection of the human condition. Treating them as equivalent events dilutes the effectiveness of efforts to prevent errors, since as far as I know we don’t really want to eliminate the human condition. Following your “logic” confounds understanding.
    In time, some adverse events that are currently not preventable will become preventable as we learn to understand and manipulate the underlying pathophysiology. They may then join the ranks of preventable complications.
    Some adverse events simply occur contemporaneously with medical care, and would happen irregardless. They are not complications and not errors but are adverse events from the patient’s (or anyone else’s) point of view.
    And finally, as has been already pointed out to you, not all contexts of care are equal. Variations occur for an incredible number of reasons, many of which are under no one’s control, any more than the weather. It snows more in Minnesota than California. Are the weathermen there making more errors?

  12. I have read most of the comments. I have been a surgeon here in the USA and elsewhere in the world virtually 50 years of my practicing life. M/s Binder appears incapable of admitting that she is wrong in calling a complication an error in spite of very reasoned arguments to the contrary. It is not that I have never made mistakes in surgery, but only a surgeon knows how mortifying it is to have a complication following any surgery whether a mistake has been made or not. Just remember that virtually everybody accepted the Institute of Medicine report
    which manipulated statistics to come up with 90,000 or more deaths in the USA due to medical errors. Since then I have noticed that any figure is acceptable, and, there are reports that the incidence of medical errors is rising. I must admit that one good outcome of the Institute of Medicine report is the
    involvement of the pharmacist in helping physicians with dosing of antibiotics and other medications, but insisting that all 90,000 complications–even if true–are the result of errors is unconscionable as shown by the above discussion. It is time to look at this issue with the proper perpective.

  13. Our system is created in a way that rewards unbalanced motives.

    If we want to understand people’s behavior(why people criticize without having accurate facts or blatanly mislead), then we have to go back and see if the behavior/critcism reflects a mismatch between who the behavior is costing and who the behavior is benefiting.

    If benefit(potential business) and cost(accountability for misleading information) are not both shared to some degree, we will have potential troublesome motives.

    What is the benefit for Ms. Leah Binder/Leap Frog ? It is obvious, Isn’t it ?

    The Leapfrog Group
    Employers and other purchasers of health benefits driving a market for hospital safety and quality in the United States.

    Washington, D.C. ·

    What is the cost for her/her organization ? Nothing .

    What would be the rational thing for her and her organization to do ? To maginfy the errors

    When they have the solutions, all they have to do is create or magnify a perceived problem.

  14. Rich,
    Very interesting post. I do understand how physicians would take the word “error” as a personal attack. But for what it’s worth I think of errors as institutional phenomena that may or may not be related to the physician’s behavior. From the point of view of the patient, it’s a bad outcome, which is an error no matter who made it and despite the fact the world is an imperfect place to have surgery in. I like the “just culture” response, to identify where problems occurred (including Acts of God) without laying blame on individuals.

    • How can you use the word error and not understand that it is an attack? Error is defined as

      1) a mistake. “spelling errors”
      synonyms: mistake, inaccuracy, miscalculation, blunder,

      2) oversight; More the state or condition of being wrong in conduct or judgment.
      “the money had been paid in error”
      synonyms: wrongly, by mistake, mistakenly, incorrectly;

      3) in BASEBALL: a misplay by a fielder that allows a batter to reach base or a runner to advance.

      4) technical:
      a measure of the estimated difference between the observed or calculated value of a quantity and its true value.

      5) in LAW: a mistake of fact or of law in a court’s opinion, judgment or order

      Of those five definitions i found, only one is somewhat neutral in judgment, and it does not apply to your scenarios you discuss in the manner in which you use it (it should be a statistical term as in concepts like standard error of the mean, variance, etc.). The other 4 all have an element of judging someone’s actions as insufficient or incorrect.

    • “From the point of view of the patient, it’s a bad outcome, which is an error no matter who made it and despite the fact the world is an imperfect place to have surgery in.” Like others, I would respectfully disagree with your equating bad outcomes with errors. As has been stated above, there are bad outcomes that occur even if the physician, nurses and hospital system have done all the correct and appropriate things. Sure, physicians, nurses and hospitals can indeed make errors that result in bad outcomes (and I agree that we all should be trying to minimize these), but it is certainly possible to have bad medical outcomes when there has been no error made in treatment.

    • The word “error” certainly does have a negative connotation, much more so than, say, “bad outcome.”

      Here’s a real-life example for you. Last week a close family member had a catastrophic aortic dissection. The mortality for this time of event starts out around 30-50% and goes up with every hour until it is repaired.

      His heroic surgeons operated for 10 hours straight. He came through the surgery. He’s off the vent and is walking and talking.

      But. He still is needing dialysis. After a massive dissection and 10 hours of open heart surgery, his kidneys took a beating. It’s yet to be determined if the damage is permanent.

      I believe you would call this an “error.”

      Personally, I call it a freaking medical miracle. From the first onset of his chest pain, he was a dead man. Somehow, his surgeons, anesthesiologists, and amazing critical team pulled him through. Not perfect, not 100%, but more than anyone could have expected.

      So. Error? Mistake? Bad outcome? Less-than-optimal outcome?


      A miracle.

  15. This has been a terrific (and extremely respectful!) discussion by folks who are are genuinely concerned about improving our healthcare system. Such passion–and willingness to sincerely consider different perspectives–is inspiring.

    Words matter, and I personally think that the common ground to first seek is terminology that is simultaneously meaningful but not pejorative. There are lots of ways that “error” is defined, but the first that pops in my mind is that which Merriam-Webster online lists first: “an act or condition of ignorant or imprudent deviation from a code of behavior .” When most folks hear “error” their first reaction is “someone screwed up” which–for right or for wrong–only perpetuates our current culture of blame.

    That said, what we may call today an “Act of God” is something that down the road we may attribute to a genetic marker, an environmental factor, or something which our finite mortal minds cannot consider.

    Like all other humans, I’ve made errors. But, I’ve had a lot more “adverse events” that, given the current science and technology, were completely unavoidable (and, on some statistical level, anticipated).

    Tell any human he’s erred, and the most likely initial response (enhanced by our litigious culture) is to circle the wagons. Whether he initially was or wasn’t, we’ve now made him part of the problem.

    Tell him that we’re all committed to reducing adverse events–which can be either classified as avoidable or (at least at present, from a statistical perspective) unavoidable–and then we potentially engage him as part of the solution.

    The terminology herein is admittedly imperfect. Smarter minds than mine–through consensus–would need to identify the right nomenclature to strike that delicate but important balance between transparency and engagement.

    If we can transcend individual words, I think we’re all much closer than at first glance it may seem.

  16. The two examples given are easily explainable using basic medical concepts. In #1, skin bacteria already present at the time of incision were not completely killed by the skin prep used (no drug works 100% of the time) and the patient got a surgical site infection. In #2, inflammatory changes which are a completely normal response to a knife going through skin caused an existing coronary plaque to rupture, causing an MI. No way to prevent these things completely by avoiding “errors”.

    It is not “natural” to have any type of surgery. Surgery is, by definition, an artificial, invasive procedure, and is a trade-off of hoped-for benefits with an expected incidence of complications. If someone is so concerned about surgical errors, their decision should be to forego the operation and hope for the best. People, however, consent to surgery hoping for an improvement in their condition and understanding there is a risk of complications.

    It is absurd to state that complications occurring after surgery are forcibly related to an error by somebody. To state that errors are involved in every complication demonstrates ignorance of the complexity of the human body and the medical system.

    More broadly, all of medicine (with the possible exception of preventive medicine) incurs some risk. Nothing is risk-free, even when nobody commits an error.

    (All this talk of “God” is ridiculous BTW.)

    • You took the words right out of my mouth–or off my fingers, rather–right down to the whole “act of god” nonsense

  17. In both of your examples, where you’ve nicely excised any possibility of physician error, there could have also been an error by nurses or aides or the general culture in the hospital.

    Or maybe these are Acts of God. I totally agree that sometimes a complication comes from bad luck. But God doesn’t generally concentrate His acts in one hospital and not so much in another, so we know it’s not always a natural disaster.

    But fundamentally for the patient, it’s a mistaken outcome regardless of whether the mistake is made by a human being or a deity.

    • Well, I feel that I have accomplished something. I got you to agree that sometimes a complication may be due to bad luck. Now I’ve got to work on your terminology. Mistakes and errors are pretty much the same things no matter what James Reason says. Some complications are neither errors nor mistakes. Let’s just call them complications.
      In an email to me, Judge Ralph Adam Fine of the Wisconsin Court of Appeals said the following: “Sadly, the law too often uses ‘bad results’ as per se indications of negligence. But, as you point out, the myriad interactions between things about which we still know very little, makes the bad result inexplicable. But the common view is that someone ‘must’ be at fault.

    • Dear Ms. Binder,

      Skeptical Scalpel pointed out that your figure of 5% for postoperative infections was incorrect as the JAMA paper cites the actual figure as 0.7%. He also pointed out that when evaluating the number of automobiles manufactured versus those recalled (granted, for one company) that the “error rate” would be greater than 5%. Could you address these two points?

  18. We agree complications are not intended outcomes from surgery. So if they aren’t errors, then, ummm, what are they? Natural disasters? Acts of God?

    By using the term errors, I’m not pointing a finger at one person for making the error. Hospitals are complex systems, and complications could result from a series of unfortunate events involving many people. Complications might also result from the patient’s actions. But I don’t think they are natural disasters that just happen, and physician leadership and skill is an extraordinary resource to prevent them.

    • Let me give you two hypothetical, but common, examples of complications. You tell me who made the error.
      A patient is admitted for a colon resection. He receives the correct prophylactic antibiotics at the appropriate time. The instruments have been verified as properly sterilized. The OR checklist and time out are executed to perfection. The case goes smoothly. There were no breaks in technique. Six days later, the wound is red and painful. An infection is drained at the bedside. Who made the error?
      A 65-year-old man is scheduled for an elective laparoscopy cholecystectomy. Since it is being done in the United States, he is sent for a cardiology consult to be cleared for surgery. He has a normal EKG and a normal echocardiogram. The anesthesia induction is flawless, and the case is completed in 45 minutes without any incident. He was not hypotensive and lost no blood. On the second postop day he returns to the ED with chest pain. A myocardial infarction is diagnosed. Who made the error?
      I could give you many more examples. I don’t know how to tell you this, but some complications are in fact acts of God.

      • Very well stated. I have enjoyed the “back and forth” commentaries on this topic and certainly see both sides. As an emergency physician I see many illnesses that have to be classified as an “act of God”. I usually term it as back luck. I see many patients with apparently unprovoked venous thromboembolism. While we know certain patient factors such as immobility and smoking can lead to this, I don’t universally blame the patient and rather provide information and say the cause is “bad luck”. Perhaps one day we will know more about the exact cause of apparently idiopathic DVT, but for now that’s the best I can do. Acute gout is another example. Most patients deny excess of alcohol and red meat and most have no identifiable reason for an acute episode. Rather than accusing a patient that it “must have been his diet “, I educate about the risks and treatment. This seems to create a better solution than blanket accusation as what is being suggested regarding post-op infections.

        • Thanks for commenting. Bad luck certainly is a force at times. I like your approach to the problem. I was going to give an example of a postop DVT. The interesting thing about postop DVTs is that they occur even when the patients are given optimal prophylaxis. In no study that I am aware of does a prophylactic regimen result in a zero incidence of DVT.

    • Complications are, in fact, quite often “natural” and not preventable. Your comments belie an extraordinary naiveté for normal biology and the response to injury, and for the way in which health care providers and hospital do their best to prevent and treat disease, and stave off the inevitability of death. Some adverse outcomes indeed may be preventable, but your flippant characterization of all complications as “errors” is just a manifestation of the typical bravado of the non-combatant journalist.

      • Jeffrey, very well put. Thanks.

    • It seems that everyone is making the case for hospital error and leaving out the one variable that cannot be controlled – the patient. Everyone is different. you can start with the variability to how drugs effect them – some no effect, some profound effect. Take for example – Benedryl – I can take 2 pills to control my post nasal drip with no effect other than my nose dries up. Everyone else I know would be asleep. Everyone’s intestinal biome is different as well. I do believe that no one is the same as anyone else and that cannot be left out of the equation.

  19. I do enjoy following you on Twitter Skeptical Scalpel. But you’re wrong here.

    I stand by my statement that automakers do not stay in business if 5% of their cars with potentially fatal mechanical flaws, even Toyota in its worst year ever.

    Manufacturing has come a long way in reducing errors and health care can learn from it. With Lean engineering principles in place manufacturing expects 3.4 defective parts per million; the JAMA study cites 5%, or roughly 50,000 defects per million. So my estimate that the error rate in surgery is 100 times was indeed a math error, because the actual comparison is far worse for health care, even acknowledging not all manufacturers achieve the highest performance.

    You don’t like my use of the term “error” to describe surgical complications and infections. But from the point of view of the patient, complications and infections are indeed errors, certainly not intended outcomes. A mistake was made somewhere–even if the patient herself made the mistake.

    And mistakes can be minimized. Given the variation in complication rates among providers, it stands to reason that many complications are preventable since some providers do better than others. I don’t think we’ve reached the point where we know just how many can be prevented.

    • Thanks for commenting. It looks like Toyota has quite a way to go before it reaches the mythical 3.4 defective parts per million. I can’t speak for all physicians, but I think most of us would reject your blanket characterization of all complications as errors. That is a very simplistic take on such a complex system as a human being. Humans cannot be compared to cars or airplanes or widgets. I agree that there are variations in complication rates, and some complications are indeed preventable. My issue is that the message is unclear and unnecessarily inflammatory if stated in your terms. You are much more likely to get buy in from the medical profession if your criticisms are more realistic.

      • Having read through the comments below, I think your article (Dr. Scalpel) is correct that exaggerations of human error and broken processes are unhelpful. Unfortunately, most of the commentators are back to defensive position that human error is minimal.

        Using the car analogy, accidents happen when roads are icy. This is due to several factors and people usually blame the environment and weather. However, the liklihood of accident on an icy road in Alaska, Montana, Kansas or Florida is probably very different. Focusing on the weather and ignoring the driver is too easy. Furthermore, people who regularly drive in icy conditions take steps to avoid accidents including better tires, sanding, and even road construction. This is in addition to driver training.

        Getting back to healthcare, until the majority of physicians start agreeing that outcome variation (obviously standardized) is far too great to be explained by environmental factors the “human error” of not bothering to improve will persist.

    • “Given the variation in complication rates among providers, it stands to reason that many complications are preventable since some providers do better than others.”
      While on the face of this there is a certain sense to this, I would say that this logic does not always hold true. When comparing numbers between “providers” one must also take into account other variables, particularly the patient population. Some surgeons (especially in academic centers) routinely are referred and operate on patients who are considered “high risk” (i.e., they are sicker, morbidly obese, have more comorbidities, etc.), and it would be expected that these patients would have poorer outcomes or an increased risk of complications than those patients who are less sick at the time of surgery. Thus, looking at just the complication rate will not necessarily tell us anything about quality of the care provided, nor can we make any conclusions about “errors” when looking just at the metric of complications.


Submit a Comment

Your email address will not be published. Required fields are marked *

eighteen − twelve =