Sunday, September 6, 2015

Marijuana, Neurosurgery, and Physician Impairment

A brief article several days ago posted the name and picture of a neurosurgery resident accused of smoking marijuana on the job.  Dr. Gunjan Goel, MD is a neurosurgery resident at UC San Diego, and the list of her awards and publications alone is almost as long as my entire residency application.  The article is brief, and rather uninformative.  The only facts that are known are this: the Medical Board of California investigator demanded a hair sample, and in response Dr. Goel acknowledged smoking 3-4 occasions over a 6 month period, on her days off.  This incident touches on a lot of complicated issues- intoxication on the job, surgeon quality and competence, and not least of all anti-drug enforcement efforts.

This piece was picked up by KevinMD, and one of the commentators posted this link to the full court complaint.  The highlights are that no action was taken by the medical board for 8 months after the complaint was received, and that Dr. Goel had a negative urine test and positive hair test- a result consistent with occasional, recreational usage rather than heavy, daily use.

I will not comment further, as this piece is not so much about this individual case, but rather about the more general problem of what regulations and rules should guide the medical profession with regards to marijuana as this country increasingly moves towards legalization.

I should put in a disclaimer here: for personal and philosophical reasons, I have never even tried alcohol, marijuana, tobacco, or any other intoxicant- apart from caffeine of course.  Whether this makes my arguments more or less credible I leave for the reader to decide.  I also feel it is important to state here for any non-medical readers what is obvious to me and physician readers: Should the accusation of daily usage prior to operating be true, Dr. Goel would be guilty of a heinous betrayal of trust, and the entire medical profession would be in overwhelming agreement that she fully deserves all of the consequences that the Medical Board will mete out.  But this article examines the (to me) more interesting question: how do we judge Dr. Goel's acknowledged usage of marijuana a few times per month on her days off?

Intoxication on the Job

To start with, what is the standard expected of a physician with regards to using alcohol and other controlled substances?  The Medical Board of California (not so) helpfully gives us this guidance:

(a) The use or prescribing for or administering to himself or herself, of any controlled substance; or the use of any of the dangerous drugs specified in Section 4022, or of alcoholic beverages, to the extent, or in such a manner as to be dangerous or injurious to the licensee, or to any other person or to the public, or to the extent that such use impairs the ability of the licensee to practice medicine safely or more than one misdemeanor or any felony involving the use, consumption, or self-administration of any of the substances referred to in this section, or any combination thereof, constitutes unprofessional conduct.

So, as long as it doesn't "impair the ability of the licensee to practice medicine safely", it's not necessarily unprofessional.  With this in mind, was Dr. Goel practicing medicine safely, or not?

Our society accepts that if one has a blood alcohol level of .08 or less, they are fit to drive.  Over, and they are not.  If someone brakes late and hits a child crossing the street, they may serve 15 years in prison if they are a .09- and not serve any time at all if they have a .07.  What level of alcohol (or marijauna) use is okay before a physician becomes impaired?  Almost all surgeons would agree that barring an extraordinary emergency, the only acceptable blood alcohol level is zero- which can generally be reliably achieved with a night of abstinence.  But marijuana?  It can be detected in the urine for weeks, and in the hair for months.  So what does it mean to be intoxicated with marijuana while operating?

Now, many online commentators stated that they did not mind if Dr. Goel used marijuana on her own time, as she acknowledged doing.  It is whether she used marijuana before operating, as she is accused of doing, that would be completely unprofessional and appropriate grounds for disciplinary action.  But marijuana's effects are complicated.  While it is true that many of the acute effects dissipate 6 hours after use, attention/executive function (rather important in a neurosurgeon) in heavy marijuana users is significantly affected as compared to light users even after day of abstinence from the drug- with heavy users being defined as smoking a median of 29 of the past 30 days, which granted is inconsistent with Dr. Goel's stated usage of the drug.  And furthermore, few surgeons are ever truly off duty.  I can think of several instances at LA County-USC where a senior surgical resident or fellow arrived at home after a long and difficult day, only to hear that their patient wasn't doing well.  They immediately drove back, and fought to the end for their patients.  The doctor-patient relationship has not been completely slain by duty hour reform.  What would Dr. Goel have done if she had arrived at home, lit up a joint, only to get a call that her patient was coding in the Neurosurgical ICU?  It's difficult to say.  After all, what would she have done if she had arrived home, downed two glasses of wine and got the same call?  One hopes in both scenarios she would have the presence of mind to allow the on-call team to operate on the case without interference.

Quality and Competence

But, there are many, many factors that affect attention and executive function, not to mention all kinds of cognitive performance tests.  While a recent, large study actually found no difference in outcomes for patients of surgeons who happened to be on call and were up all night, it is well known that sleep deprivation negatively affects cognitive performance of doctors, though these effects are complicated with some studies finding no or even positive effects.  A large number of fields- from long distance truck driving to piloting aircraft have adopted rest rules in an attempt to improve the cognitive performance of the individuals in these fields in light of significant data like studies that show that sleep deprived drivers are far more likely to crash.

Let's compare two scenarios: 1)  A surgeon smoked pot once upon getting home, and promptly and slept 5-6 hours before waking up and going to work.  2)  A surgeon got home, and promptly went out and got smashed at a bar, getting home late and sleeping only 2 or 3 hours before waking up and getting to work- both sleep deprived AND hungover.  (Or alternately, stayed up all night taking care of her hypothetical sick kids- same sleep deprivation, minus the hangover).

There is zero data comparing the two scenarios.  But if I were a patient I would strongly prefer scenario 1, even though scenario 2 is the only one that is completely legal.  That preference would change if I knew that the surgeon was smoking marijuana every day- since I think there is mounting evidence that long-term use of marijuana results in cognitive deficits that do not ameliorate.

And here we come to the crux of the matter: how long should Dr. Goel have abstained from marijuana prior to operating?  Without solid data, the kind which is really difficult to obtain in a country where marijuana is technically illegal, there is no accepted standard we can use to say how long one must wait after using marijuana before one can be considered non-impaired.

Should patients be required to choose between an exhausted and sleep deprived surgeon and the pot-smoking surgeon?  No.  Physicians should understand their limits, and get the rest they need to operate.  The evidence, as shown in that article cited above, suggests that overwhelmingly they do- canceling cases when they feel they are too exhausted to operate properly.

But how are we to decide what constitutes physician impairment either by chemical substances, sleep deprivation, diseases like the flu (shown to impair driving just like intoxication), and otherwise?

It's worth going back in time to the dawn of modern surgery.  Dr. William Halsted is a surgeon so revered in American medicine that every academic surgical department in the country tries to emphasize the direct passage of knowledge and technique from him to themselves.  He also struggled with life-long addictions to cocaine and later morphine.  His addiction to cocaine developed as a result of his experiments on himself to develop it as a form of local anesthetic- a purpose for which it is still used today.  When his addiction proved to be untenable, he later went to a sanitarium (an early version of rehab) where his addiction was converted to morphine- it was thought better to be addicted to morphine than cocaine.  He would remain dependent on morphine for the rest of his life: from the late 1880s through the late 1910s, during which he would train all of the founders of modern surgery and continue to develop new operations.  His life is so extraordinary it was recently made into a TV show called, The Knick, with no less than Clive Owen playing an analogue of Dr. Halsted.

And here is the rub, isn't it?  Despite the fact that he was addicted to morphine, any woman with breast cancer (or man with a hernia) in 1910 America would rather have an addicted Halsted doing their operation than practically any other surgeon in the country- he was that skilled, despite his addiction.  But how is anyone to know?  Foolish and near-criminally negligent attempts by ProPublica to the contrary, there is still no good way to truly measure a surgeon's skill that is repeatable on a wide scale- and how much that skill decreases in the event of marijuana intoxication.  Perhaps surgical videos will eventually get us there, and it is in that kind of individualized (but highly resource intensive) methods that I see some hope for progress- having one's ability to operate after sleep deprivation, while sick, and in other conditions evaluated repeatedly over time should allow surgeons to understand and make better decisions about their fitness to operate.  But what about bad actors who do not act responsibly?  This brings us to the final issue raised by this incident: what kind of measures must we take to protect patients from physicians who are intoxicated?

Anti-Drug Enforcement

One part of this story that may be easily overlooked is the investigator's odd request: for a hair sample.  From one drug testing company's website, a hair sample may typically show as positive for marijuana or other drug use over three months ago.  And the best part?  Recent use isn't detected- it must have been used more than 5-10 days ago before marijuana shows up in the hair.  A single use may not even show up in hair at all.  This is for a typical hair, tested 1.5 inches from the root.  From the picture, Dr. Goel like many women has far more than 1.5 inches of hair- she could have conceivably have tested positive had her last period of use been over a year ago.

Marijuana testing is quite controversial.  The typically employed urine test would be positive even if the last use was 2-5 days ago, or longer.  Blood tests are a little more short-term, coming up as positive only 12-24 hours after the last use.

So, what was the investigator going for?  Clearly not to prove that Dr. Goel was intoxicated at that point in time- which is interesting, in light of the California Medical Board's assertion that "she uses it daily, while at work, and on call".  Perhaps the real goal was an implicit threat- if the hair tested positive at all, the medical board would have grounds to go after her license.  Cooperate now and enter a rehab program, or you will never practice as a doctor.

The medical board would not necessarily be wrong in taking this stance.  As previously stated, there is no hard data on how long one has to remain abstinent from marijuana in order to operate safely.  In the absence of any data, perhaps the most careful and ethical course of action for a surgeon would be to abstain from marijuana entirely except possibly on the first day of a two week vacation.  While I personally wouldn't be affected in any way, I also don't think it is particularly realistic, and would make criminals out of a lot of otherwise quite capable surgeons who use marijuana to relax after extraordinary taxing days.  After all, neurosurgeons probably rank behind only air traffic controllers in terms of being the most stressful profession imaginable.  But, the military, the airline industry (often held up as the gold standard of safety culture) and many other professions do indeed maintain an apparently successful zero-tolerance policy for marijuana even in light of the liberalization of society with regards to the drug.  Who is to say that such an approach is not best?

A less strict, but still valid and empirical approach may be a simple policy: no one may operate with detectable levels of marijuana in their blood.  Blood tests have the advantage of actually testing for the psychoactive substance in marijuana- tetrahydrocanabinol, unlike the inactive metabolite that the urine test looks for.  As marijuana is no longer detectable in the blood 12-24 hours after the last use (longer depending on body fat percentage), and most acute effects of marijuana apparently dissipates within 6 hours, this seems to be a reasonable and empiric policy that mirrors the 0 blood level standard of the alcohol test that we use.

If more research is required, then as marijuana becomes decriminalized in more states research to fine tune the limits should become easier to do.  It should be a relatively simple affair to take groups of attending and resident surgeon volunteers, give them marijuana, and measure their abilities at set intervals- along with the detectable levels of marijuana in blood and urine.  When the performance of the surgeons in question is statistically indistinguishable from the baseline, that blood level of marijuana should be taken as the legal limit for surgeons.  Perhaps it will be 7, 8, or 10 hours.  Or perhaps the data will show that it takes 2 or 3 days for the effects to fully dissipate- particularly for heavy users.

And here is the sticking point- someone who uses marijuana once or twice a month is likely to escape any long-term side effects.  But someone who lights up every day immediately after work may still successfully show a blood level of 0 by the next morning- while putting themselves at risk for long term cognitive impairment.  It is here that the strongest case for a zero-tolerance policy can be made.  But perhaps another way would be to adapt a similar approach that physician groups have been adapting for elderly physicians with cognitive impairments- through specialized and comprehensive testing that determines if any general declines in competency and decision making have been noted.  If they have, the physician would be put under heavy pressure to either quit their marijuana usage, or give up their license.

Ultimately, as always, our profession's oath is to do no harm.  Physicians must not put their patients at risk by impairing themselves unnecessarily, especially surgeons who must make complex decisions under extraordinary stress.  But there is no widely accepted definition of what constitutes impairment with marijuana, especially considering that the most commonly used test for it will stay positive for much longer than the time it may reasonably impair someone.  In light of the increasing prevalence and acceptance of marijuana use in the population, it may be time to define an acceptable level of use by physicians that is considered safe for patients.  Perhaps that level of use is never, or perhaps that level is one that results blood level of zero whenever one is in the hospital.  But regardless, without clear guidance that goes beyond an incredibly vague 'not impair the ability of the licensee to practice medicine safely', it is difficult at this time to judge the acknowledged actions of Dr. Goel or physicians like her.