Kant vs. daVinci

The other day I was speaking with two gynecologists who both ‘loved’ the surgical robot.  They beamed about the remarkable skills the robot had granted them.  When I asked, “so what kind of cases do you do with the robot,” their reply was remarkable.

Many years ago there lived a quirky man named Immanuel Kant.  He never married and never strayed very far from his home in Prussia.  While many physicians cock their head like my quizzical dog at the name Kant most folks are aware of some of his philosophical positions.  In ethics Kant is most known for the ‘Categorical Imperative.’  There are a few iterations of the imperative but the easiest to digest and most relevant to my discussion goes like, “never use another person as a means to an end.”  Persons are ends.

Hospital credentialing is a screwy thing.  Once you become a licensed physician you still have to convince a hospital to allow you to practice medicine within their facility.  The amount of convincing is not a fixed matter.  In other words two hospitals could have vastly different ‘rules’ about what you have to do to prove you can do what you say you can do.  For instance when I was first asked to fill out my hospital privileges I needed only to mark off what I thought I might want to do.  There was no need for ‘proof’ that I could do the things I marked, I simply needed an interest and a credible residency diploma.  Since that time things have changed but not uniformly; there are still big differences across hospitals.  Thankfully hospitals are getting a bit better at this process but they are not immune to bias.  For example, in some hospital’s enthusiasm to get a robot program going, the ‘rules’ about credentialing have been in many cases too loose.

Back to the two gynecologists and their robot tales.  Citing hospital credentialing rules that stipulated performing a minimum number of cases per month (arguably better than usual credentialing rules but still problematic) these gynecologists admitted at times they utilized the robot in cases they could have otherwise done without it just to satisfy the rules.  What would Kant say?

The enthusiasm for all things minimally invasive has at times bordered on absurd.  Natural orifice surgery (i.e. removing the gallbladder via the vagina) is one brand of minimally invasive surgery that has caught the imagination of some in medicine and perhaps one day it will be mainstream.  The irony amidst this hubbub is that minimally invasive surgery (notwithstanding vaginal surgery) hasn’t always been better than ‘open’ surgery.  For instance, there was one randomized surgical trial with patient and post-operative assessors blinded to the surgical access that documented an open gallbladder removal was superior to the laparoscopic (key-hole) approach.  Go figure.  Insofar as the robot is concerned, one well-done randomized surgical trial comparing it to the traditional laparoscopic approach demonstrated the robot took more time, cost more money and was associated with more pain at every postoperative assessment.  And yet the robot appears on highway billboards promoting a hospital’s commitment to ‘technology.’  Go figure indeed yet the public seems to be receptive to these sorts of inducements.

So now, given what I’ve said so far, what would Kant say about these gynecologists’ reasons for using the robot?  Curiously the argument is echoed in any sentiment that poses the physician benefit as the driver of a technologies adoption.  Better physician comfort, while important but dangerously subjective, dances closely to confusing means and ends.  In the cases of quirky credentialing rules, is the patient’s best outcome foremost in a decision to use the robot (or any technology) that is ultimately based on criteria outside of the clinical scenario?

Sadly as I mingle between academic and nonacademic physicians in diverse settings the reasons to adopt the robot (or any medical innovation) shared among them are not unique.  Indeed some version of this tale is often the central impetus for adoption.  There is abundant evidence that peer perception is at the root of many of our behaviors.  Medical innovators know this and proffer tales of ‘important’ and ‘prestigious’ physicians who have adopted or endorsed their device.  Few have been as successful at this game as Intuitive Surgical the makers of the daVinci robot.  Adopting the robot to gain market share or notoriety or device access because of credentialing rules all disappoint the Categorical Imperative.  Maybe medicine should take a cue for Kant’s isolationism and eschew industry representatives?

What are your thoughts?

Who should cut?

I did a suction dilation and curettage (aka D&C) recently.  A woman had a miscarriage and having failed medical therapy we opted for this procedure.  The procedure went without a hitch.  What struck me about this instance of performing this procedure was that I hadn’t done one in perhaps 3 or 4 years.  This sort of case is typically reserved for junior residents because it’s so ‘easy.’  Furthermore, whether done by a junior or senior resident it is uncommon for me to do most of the case since I mostly just supervise this sort of case.  So was I lucky that despite my relative inexperience things didn’t go awry or are there surgical cases that are so easy a ‘monkey could do it?”  When is a surgical procedure ‘easy’ versus ‘hard?’

Two studies appeared in the April edition of Obstetrics and Gynecology.  Both studies addressed the effect of surgeon factors on surgical outcomes for laparoscopic hysterectomy (a relatively more difficult case than a D&C).  One study was done in the U.S., the other in the Netherlands.  Both studies demonstrated what has been seen in other related studies – more surgeon or hospital volume and better surgeon skills renders better surgical outcomes.  

These findings make sense.  Most of us understand the adage, ‘practice makes perfect.’  The results of one of the two studies stated:

In the multivariable analysis, risk of overall morbidity (relative risk (RR) 0.75), intraoperative complications (RR 0.85), surgical-site complications (RR 0.70), medical complications (RR 0.63), transfusion (RR 0.75), and prolonged hospitalization (RR 0.65) were all reduced when the procedure was performed by a high-volume surgeon.

In this study performing a mere 8 more hysterectomies per year (high volume surgeons were defined as performing 14 operations per year) rendered a minimum of 15% less risk of a negative operative outcome.  Furthermore, high volume surgeons were cheaper spending on average $867 less per case than low volume surgeons.  Do you think most women facing a hysterectomy would want to know if their surgeon was ‘high volume?’  Which type of surgeon do you think health policy organizers would like performing these cases in a day of shrinking health care resources?

The Dutch study adds an additional twist to the volume outcome relationship.  The Dutch researchers added surgical skill that they estimated using a model that included patient, surgeon and environmental factors.  This surgical skills factor led the researchers to conclude that volume alone does not “necessarily guarantee good surgical outcome.”  So if some surgeons possess better skills, either innately or by practice or both, and these skills translate to lower health care costs, why shouldn’t these issues have more sway in health care decision-making?  

The answers to this question are a mix of ethics and statistics.  First of all the overall difference in surgical morbidity is not huge between a low and high volume surgeon (although one does have to wonder if performing just over 1 hysterectomy per month is really ‘high volume’ for comparison among my colleagues we typically perform at least 4-5/month).  For example, ‘any complication’ was 5.8% among low volume surgeons but 4.7% for high volume surgeons.  That sort of difference, while statistically significant and enough to produce some remarkable relative risks, isn’t likely to catch the attention of most women or their potential surgeons.  And the surgical skills model is a bit fuzzy keeping that elusive quality of surgeon assessment still mostly out of investigative reach.  The bottom-line is that while health policy may have incentive to decide who can operate (and thereby save money); poor performers are still likely to be lost in the fog of small effect.

So should surgical training be so egalitarian?  At least in Ob/Gyn there is the view that if you finish your residency or if you do some even less formalized instruction you can operate.  Take a weekend course on how to do procedure ‘x’ wherein you practice on pig parts and poof you’re qualified.  While laparoscopic hysterectomy has grown in popularity a more precipitous rise in minimally invasive hysterectomy can be attributed to the robot.  The surgical robot offers a quirky twist to the problem of surgeon skill and experience.  It is precisely the lack of skill and experience that stimulated the popularity of the surgical robot among gynecologic surgeons.  Given no credible evidence of improved patient outcomes in using the robot for benign gynecologic surgery, it is the devices appeal among surgeons that has propelled this surgical approach.  So were back to – should all surgeons be regarded as the same?  The matter of resources again becomes relevant insofar as the robot exacerbates the cost of low volume surgeons.

All of us want to get the best care possible and in surgery the matter of volume and skill are front and center in defining ‘best.’  Among physicians there is an unspoken unease about our personal skill and experience.  There nearly always seems to be someone better qualified to perform the care expected among the patients we see in our clinics.  If we send everyone out to these ‘better qualified’ clinicians we never improve and done enough we starve.  Access to care can become a problem as physicians drop out of certain kinds of care due to lack of expertise.  This has been one problem seen in subspecialty Female Pelvic Medicine and Reconstructive Surgery.  There are not enough fellowship trained physicians and the training is so variable is it wise to only have subspecialists care for women with pelvic floor problems?  So where is the balance? 

To add to the confusion I’ll share one more story.  In the book, Switch, the authors Chip and Dan Heath share how in learning Minimally Invasive Cardiac Surgery (MICS) hospital teams, including the surgeons, have to prepare to fail.  The learning curve for this procedure is steep and only those willing to learn amidst the failures will push through to becoming successful at an approach to cardiac surgery with significant recovery benefits over the open-heart technique.  In the context of the point the Heath brothers are trying to make the example of MICS is great; however, what about the patients who are the subjects of being ‘learned’ on?  Innovation is great and clearly medicine is better today for the lessons learned via trial and error on unwitting patients in the past but who wants to be a lesson?  How many women today are undergoing robotic surgeries by surgeons who had acceptable alternative techniques but who are compelled, for reasons having little to do with better patient care, to adopt this surgical innovation?  Indeed the chairman of one of the most prestigious Ob/Gyn departments in the world confessed that the robot today isn’t a net patient good but it is a necessary step to tomorrow’s robot.

To this mess I offer a few suggestions:

1. Provide basic volume and outcome performance data for all gynecologic surgeons. 
2. Tighten the hospital credentialing process and add an independent skills assessments piece to be part of that process.
3. Develop third-party CME-like surgeon skills education opportunities that independent of industry teach surgical technique not just ‘as needed’ but as part of regular scheduled skills training.  Use these training sessions as part of the credentialing process.
4. Link advancement in residency to achievement of year-specific surgical skills and volume.  Do not make graduation from residency something only based on time on the job – particularly since that time seems to be getting less and less with residency hour restrictions.
5. Interest in surgery does not qualify one in it.  It is OK to not be good at surgery; there are other great options in and outside of medicine.  Skill can be both an innate and time-limited feature of a surgeon.  Once a great surgeon doesn’t mean always a great surgeon.  Referenced feedback will help clarify these matters.

Ultimately any balance to be found in this inexorably messy process will not come from professional organizations or evidence or government regulation; even if all these oversight mechanisms endorsed my admittedly challenging suggestions.  In the end the balance is found within the individual surgeon.  The moral character of the surgeon is the best defense to finding the right balance between honoring the patient and the profession.  Being ‘other-oriented’ is tantamount to being a true physician and surgeon.  Thankfully the D&C went well presumably due to translated surgical skills from other procedures I more regularly perform and from perhaps some innate sense of how to do what is mostly an easy procedure.  Would some feedback have helped me better assess my performance?  Sure but perhaps we should start with the robot and work backwards.

Unhappy Medicine?

This past week I became aware of the results of a recent survey conducted by the Doctors Company. The Doctors Company is the nations largest medical liability insurer. The survey was done among >5000 U.S. physicians with good representation across specialties, age, sex, etc. One result particularly stood out:  Only 11% of physicians today would recommend medicine to a friend or family member.


Contrast the 11% figure with the 39% of Harvard's 2010 graduating class marching onto Wall Street. It is true there is an apple and oranges quality to this comparison but if the nations best and brightest college graduates become more aware of the discontent in medicine and even law (see http://tinyurl.com/2g5ahq4) the number heading to Wall Street may climb even higher.


Why all the unhappiness? To be fair there is no single answer to this question. A struggling economy doesn't help. Uncertainty about the future of health care doesn't either. I would, however, like to suggest one possible answer that relates to bioethics.


Anne Fadiman’s book The Spirit Catches You and You Fall Down has been widely praised as a fine example of medical anthropology that emphasizes the need for greater cultural competence among U.S. physicians. The book centers on the life of Lia Lee – a Hmong girl with severe epilepsy born in the U.S. to native Hmong parents. Despite both cultures seeking to do what was best for Lia neither succeeds. The immiscibility of the two cultures made a difficult situation worse. While the book is particularly sympathetic to the Hmong people, the struggles felt by the involved health care workers is also featured. There is a sense of defeat among the health care workers as there was no effective communication between themselves and the Hmong people. The clinicians in time may have become more cultural competent among the Hmong but there is no doubt which view of medical science they saw as 'correct.'


Given the calamities between Lia Lee's family and western medicine there is hope for positive change; western clinicians becoming more accepting of different and sometimes divergent cultures and lifestyles. Indeed this sort of perspective aligns well with a growing sense for patient-centered care. There is, however, the possibility that lots of different cultures and/or lifestyles could render a sort of paradox of choice. Just as too many options for a given type of product may tip at some point to induce consumer dissatisfaction, perhaps too many cultures contribute to clinician dissatisfaction?


Among the cited reasons for physician dissatisfaction is disparate expectations of patients. There must be settings where the expectation gap arises from differences between cultures. Lia Lee’s parents had expectations of Western medicine that could not be realized. While health care may attempt to accommodate different cultures insofar as they interface Western medicine per bioethicist Tristram Engelhardt there is ultimately no basis to define which culture is ‘right.’ Should I consult the shaman or another physician in this difficult cross-cultural case? Reacting to this view some might say, ‘but it is not important to know which is correct’ this is a setting of ‘both’ ‘and.’ Herein is the appeal of choice but does western science, to which medicine draws its strength, accommodate more than one 'right' answer? What moral guidance can be gained without a referent? Some aspects of choice between cultures can be celebrated but not all. Lo Mein and wienerschnitzel can appear on the same menu but there is not ordinarily a buffet for the answers to scientific or moral questions.


Facing any given patient, a clinician today cannot be sure of the merit of his or her wares. Is longer life a good thing or a bad thing? What is normal human function to which medicine aspires to restore? Do we 'fix' the seizures or just better appreciate the families perspective that the seizures portend a future in shamanism? The clinician today may be unhappy because at the end of the day the same act in two different patients may in the one be heroic and the other failure and in both cases divergent from the physician's personal view on the 'correct' outcome. Practicing medicine has become less linear. It is an unreasonable expectation that as a clinician becomes more adept at his or her care, patients will inexorably ‘benefit.’ Benefit can become so nebulous an outcome that short of a stable referent clinicians cannot help but at times become despondent.


Wall Street does often look more appealing. Money is a much more universally and easily understood good...even if love of it is the root of all evils.

What is Informed Consent?

Life is full of decisions.  For most us that means weighing options and making a choice that best fits with our values and the ‘facts’ related to the decision.  Facts and values are in many respects analogous to our heart and mind or ‘what we want’ versus ‘what we know.’  The less we know before making a decision the greater the chance we will make a choice that is not consistent with the facts, our values or both.   Medical decisions are no different than any other kind of decision.  Accessing medical facts relevant to a medical decision, however, can be difficult.  Likewise understanding our values can be unclear for health states we cannot realistically imagine.  Physicians are the primary conduits of medical information for most patients facing a medical decision.  Curiously physicians are also important in helping a patient sort out their values.  Informed consent is that process whereby a physician imparts medical facts and clarifies a patient values.

Informed consent is a modern feature of western medicine.  The philosophical underpinnings of informed consent are diverse.  Clearly the principle of autonomy comes to mind but beneficence and justice are also important.  More broadly understood, informed consent draws from deontological ethics wherein physician duties to patients and society are recognized.  Simplistically informed consent reduces to disclosure, competency and choice.  Physicians as purveyors of health information have a duty to disclose to those in their care the health information necessary to ensure the patient makes an informed choice.  From an ethical perspective, competency is that state wherein the patient understands and can manipulate the disclosed health information so as to make a choice that reflects the facts of the decision and how those facts are personally valued.  Reflecting on these elements should give us all pause.  While informed consent law and litigation has focused on what is to be disclosed, the ethical principles upheld in informed consent are better captured by what the patient comprehends since this is the basis upon which choice is made.  Informed consent, therefore, is not about dumping voluminous information on the patient and letting them sort it out.

Among members of the Navajo Nation it is a common belief that to speak ill of something increases the likelihood of that adverse event happening.  In the context of informed consent this can mean to disclose treatment risks will provoke these events to occur.  This is an example of the tension between disclosure and competency.  Members of the Navajo Nation have declined clearly beneficial treatments because the informed consent disclosure was seen to induce the uncommon adverse events associated with the treatment.  If autonomy were the only principle honored in informed consent then disclosure of risks in this context would undermine beneficence.  Do such situations then warrant paternalism?  No.  What should be understood is the complex and inexorably time-consuming process of understanding the context of a disease in the life of an individual and how treatment decisions are made within that context.

Physicians often manage illnesses more so than disease.  The distinction being drawn here is that a disease references the medical facts present in a given person but an illness considers how those fact manifest in a given persons life.  Consider two women who void 15 times a day.  One woman is a rural mail carrier who comes from a very traditional American home.  The other woman is an immigrant office worker.  Access to a bathroom is vastly different between these two women.  The cultural expectations of adult life are also different.  These differences will very likely impact which of these two women will present for treatment for what is precisely the same biomedical facts.  Acknowledging the distinction being made here highlights again the need to understand the patient’s illness when rendering an informed consent.  The biomedical facts of a disease will not necessarily precipitate the necessary disclosure to ensure a patient is competent to make an informed choice regarding treatment.

There are many books and articles written on the topic of informed consent.  The topic can become overwhelming to consider and many of the issues alluded to above cannot be made formulaic yet this can be the impression of informed consent for many clinicians; just get the patient to sign the informed consent form.  Physicians have an ethical duty to construct the decisional environment necessary to allow a patient an informed choice.  This process is inevitably imperfect.  This process should inevitably call on the clinician to know treatment specifics and the patient to whom these specifics will be disclosed.  This process is a disposition for ethical care that honors the best traditions of medicine.

The Mesh Mess

Announced recently,  Johnson and Johnson will no longer manufacturer four mesh kits designed to repair vaginal prolapse (Prolift, Prolift M, Prosima) and urinary incontinence (TVT-S).  While not a recall, this action highlights the trouble these kinds products have had since the FDA’s July 2011 safety communication (http://tinyurl.com/3kzd6zd).  In case you missed that report, the FDA announced that two mesh complications - mesh erosion and mesh contraction - had persistently high rates of occurrence and were associated with “severe pelvic pain, painful sexual intercourse or an inability to engage in sexual intercourse.”  Furthermore, relative to traditional repairs (that ironically could include the use of mesh, although placed abdominally) the FDA did not identify globally improved treatment outcomes associated with the use of these transvaginally placed mesh kits designed to repair vaginal prolapse.  Herein lies a treasure trove of topics to discuss for anyone interested in bioethics, medical innovation and health care delivery.  One such topic is the implications of medical innovation on surgeon supply.

According to researchers at The Dartmouth Institute, all health care treatments can be lumped into one of three buckets.  One bucket is care that is evidence-based to demonstrate improved therapy outcomes.  Using a beta-blocker after a heart attack would be an example.  Another bucket is preference sensitive care where no singular best therapy has been identified as superior to another.  An example of this sort of care would be lumpectomy or mastectomy for the treatment of breast cancer.  In this situation either therapy has been evidenced to render equivalent outcomes and the choice of one over another hinges on patient preference.  The last therapy bucket is supply sensitive care.  This sort of care is the great majority what Medicare and Medicaid spends money on and it is a kind of care that medical innovation might just accelerate.  Supply sensitive care is where the supply of the care, not the burden of disease, defines the amount of care delivered.  More cardiologists are associated with more cardiology visits even if the burden of heart disease is held constant.

Drooping vaginal organs is a curious disease.  The FDA cites that while the prevalence of this disease is high (upwards of 50%) attributable symptoms are only found in a handful of affected women (one study estimates this rate at 2%).  Consistent with this fact is the curious reality U.S. women living in the South are nearly twice as likely to have surgery for lost vaginal support relative to women living in the Northeast.  This sort variation is even more disturbing when examined over smaller geographic regions.  Variations in care can suggest that features outside of a patient’s symptom burden drive the care delivered.  If industry surgical devices, like a vaginal prolapse repair kit, increase the supply of physicians ‘capable’ of treating a given disease then per a supply sensitive paradigm surgery for that disease increases even if the disease burden stays the same.  This sort of arrangement challenges the notion of informed consent in care delivery.  Is the woman who has an asymptomatic vaginal bulge truly giving informed consent for a surgery that is being propelled by the availability of a care option?  Coercion, albeit subtle and likely innocently unrecognized by the surgeon, is the ethical concern with supply driven care.

What if, however, women suffer in silence and these industry options are the only realistic treatment for women in a given community?  Fair question.  This sort of question raises the issue of justice.  Communities with surgeons more familiar with complex benign gynecologic surgery may have an unfair advantage over communities without such a surgeon population.  This thinking has some legitimacy and it is a poor assumption to believe that surgeons skill (that is at least a product of inborn skill and experience) is uniform across geographic region or age or year of training or any of the host of other factors that have been linked to this quality.  So how should surgeons be trained?  How does industry develop needed health technologies yet not feed into the care supply trap?  How does formal surgeon training accommodate new technologies since to introduce such necessarily impinges on training for already accepted treatment options?  How do surgeons who have completed their training maintain their skills and learn new technologies?  I could go on but the happy days when what a surgeon learned in residency would be all they ever needed to know are gone.  Gone too are those days when medical innovation primarily arose from within the clinician community avoiding the pressures that can propel vexing ethical challenges.  The mesh mess is just part of larger problems in modern healthcare that strains the familiar tensions between respecting persons and distributive justice and between what is innovation and what is reckless.

Therapeutic Lemons?

 Various professional sources estimate that:

Surgeons in the United States perform two-thirds of hysterectomies through a large incision in the abdominal wall;

40 percent of women are unaware that they can choose a less-invasive alternative;

98 percent of women expect their physician to disclose the existence of such alternatives, even if the physician must refer the patient to another surgeon;

90 percent of women would pursue a second opinion if they knew of other options;

8 percent of obstetrician/gynecologists would choose the more invasive hysterectomy route for themselves or their spouse.

What’s going on here?  Why are Ob/Gyns doing unto others what they would not have done to themselves?  Why the breach of the ‘Golden Rule?’

The truth is sometimes medicine unwittingly markets what, for lack of a better metaphor, amounts to a therapeutic lemon. 

As with a car whose defects show up after we buy it, the therapeutic lemon fails to produce the best overall health outcome.  The therapeutic lemon persists because consumers/patients can only know so much about care alternatives.

Borrowing again from the more familiar consumer setting of buying a used car, before resources sprang up – particularly on the Internet – for checking the reliability of used cars, potential buyers looking over the car lot struggled to distinguish the 'cherries' from the lemons.  The dealer had a better, albeit imperfect, sense of this distinction. According to economist George Ackerlof, the difference between what the buyer and the dealer knew about the available cars tended to drive out the good ones. So to make a profit, dealers sold the lemons because that was all they could afford to sell; they weren’t necessarily being dishonest (as with those Ob/Gyns who perform more invasive hysterectomies), but the dealers incentives worked against helping the buyer acquire a cherry – a car with a minimum of hidden, ticking time-bombs.

Fueling the market for lemons is consumer ignorance. Buyers would pay more if they only knew they were getting a cherry.  Trusting the dealer to sell buyers what they pay for could help fix the problem.

Admittedly, medical care is more complicated.  And yet patients still trust health care professionals to 'sell' the therapeutic cherry and not the lemon.  Information asymmetry is the fancy economics term describing a setting where clinicians know more than patients.  Curiously a fiduciary relationship, the sort of relationship that describes what is between a patient and clinician, acknowledges a setting of information asymmetry and the peculiar need for trust to bridge the information divide.

Trust isn’t the sort of thing that you find in a medical school curriculum.  Being trustworthy is a virtue and virtues are a matter of ethics.  Professional ethics can be a remedy for the mischief information asymmetry can breed.  Narrowing the information divide between patients and physicians – helping them understand their treatment options and outcomes are important steps toward finding the therapeutic cherry.  Likewise making clinicians more aware of their personal treatment outcomes is also a powerful referent.  In the end better-informed patients and clinicians establishes the referents needed for better professional ethics and we need good ethics, lest we all, sooner or later, buy a therapeutic lemon.

Postscript:

Some of you may be thinking, “The robot will fix the hysterectomy problem.”  Nope.  While the surgical robot may change the numbers of minimally invasive hysterectomies, it also could evolve into a lemon ‘swap’ relative to the open most invasive technique: costing more, and with growing evidence that it yields no differences in the pain or the mobility that the patient experiences after open surgery, at least in the setting of benign gynecologic surgery.