The Prisoner’s Dilemma

For this blog, I wanted to run a thought experiment.  Like everyone else, the issues of physician burnout and well-being have been weighing on my mind.  There is so much that has been written about this but based on what I have seen, the discussion tends to fall into three buckets.  The first bucket, from a physician perspective, is that of exit strategies.  The argument assumes that the situation will only get worse and so it is in the best interest of those who can to find alternatives to their chosen career.  The second bucket comes from “the organization” where the discussion centers around the risk of burnout in terms of patient safety and quality and how to stabilize the physician work force so that they are productive and safe.  Sitting in between these buckets, is the academic bucket that seeks to understand not only the signs and symptoms of burnout but countermeasures to combat it. 

To me, this is the fundamental challenge.  We have not gone any further than countermeasures to address a problem that is decreasing the wellbeing of an already strained workforce.  People point to the electronic medical record as the source for burnout. At my institution, efforts have focused on scribes, classes, and other tools to help make physicians more facile with the system so they spend less time outside of clinic documenting and completing clinical tasks. But this is just one aspect of the “burnout” picture.  Issues of work-life balance brought on by changes in priority of the physician work-force (and generational differences), lack of meaningful time to pursue other aspects of our careers, increased expectations from the organization, payers, and patients, barriers to delivering care by regulations, guidelines, insurance policies, and resource constraints are just the tip of the iceberg in terms of contributors to “burnout”.  Similarly, wellness officers have only scratched the surface when it comes to addressing these issues- resilience training, templates that include slots for charting, mindfulness, physician lunches, recognition are in my humble opinion window dressing to tackle the problem. 

The tension between physicians and health systems is an age old professional struggle.  It is the conflict between the physicians need for autonomy and the health systems need for efficiency, productivity, and growth to continue to fund the clinical enterprise.  What has fueled this to the point of burnout being an epidemic is how rapidly physicians have transitioned from small business owners to employees giving up their autonomy.  A non-employed physician would previously use the hospital as a resource and the administration would view good clinicians as customers that provide their patients good services.  Currently, physicians (their salaries) are viewed as losses on the P&L and productivity is the only way to balance the cost. 

The prisoner’s dilemma is an example of game theory to help explain how individuals or groups might cooperate in a particular circumstance.  In the classic experiment, two people are brought in and charged for a crime.  There is enough evidence to convict both on a lower level crime but they believe they are responsible for something worse.  They are each separated from one another so that they cannot cooperate.  If the two people remain quiet, they both receive the lower penalty.  If one of them betrays the other but one remains silent, one person walks free while the other gets the maximal penalty.  Finally, if they both betray each other, they both get the maximal sentence.  The scenario is set up for people to select the self-interested choice because this gives them an opportunity to walk away.  Assumptions built into this model are that there is no downstream consequence to the action of betrayal.  Interestingly, the data suggests that there is a bias towards cooperation which seems to maximize the utility of both individuals. 

In my search of the literature, I actually found very few scholarly articles applying the prisoner’s dilemma to health care and most of it was focused on the patient-physician relationship.  None of the studies or other writing I found on this model examined how the prisoner’s dilemma would apply to the relationship between physicians (or other employed clinicians) and healthcare systems (the employer).  So let’s try to run the model. 

First, we must understand the charge we as physicians and our academic healthcare employers are facing. The lesser charge is operating under a fee for service model that encourages clinical activity as healthcare tries to transition to value based care. To simplify this discussion, I want to focus on the employed physician model since it is rapidly becoming the most dominant model.

In the first scenario, the hospitals ratchet up demand while doctors remain silent. This is largely accomplished by carrots and sticks. More RVUs mean greater financial rewards while failure to comply with increasing expectations around clinical productivity, timely documentation, access, patient satisfaction, national quality measures, and a laundry list of other requirements puts the doctors financial and resource wellbeing at risk. This is the heart of the burnout issue as it attacks the autonomy of the physicians time and decision making. While health systems argue that margin supports mission, the need to fuel growth can seem like a treadmill that has no end. Because the goal is unclear, buy in from physicians and other members of the healthcare team is forced by carrots and sticks rather than believed. Ultimately this perpetuates the crime of the productivity model and leads to backlash.

The backlash is where physicians and other who deliver care hold health systems hostage. While physicians as a group do this poorly, there are scenarios where physicians placing the blame on health systems can have a negative impact. I’ve already mentioned one strategy above- physicians leaving medicine altogether. This is a bigger cost than the lost FTE. It is the cost of the years of investment to train that individual. More subtle is the choice of decreased productivity that physicians are making in the interest of work-life balance. While this might help burnout, health systems incur high fixed costs of recruitment and benefits to bring in additional supplemental FTEs. In addition, it is unclear how “part time”physicians perform from a quality standpoint- does it just take longer or do they ever achieve the level of clinical expertise as a full time clinician. Is this part time model even possible for some specialties where clinical demands and patient expectations do not fit neatly into 36 hours/week?

There are other ways that physicians place the blame. In the most dramatic scenarios, health systems have been subjected to the loss of entire service lines leaving a hole in clinical expertise that ripples through the system. For smaller systems without a deep bench, even the loss of a single physician can have an impact as employed physicians become mercenaries willing to move for a better RVU rate or income guarantee when their contract is over. In fact, many physicians have entered the “gig economy” providing locums services at high cost to systems who must pay more than salary to bring physicians in on a temporary basis (housing, licensing, malpractice,etc). For less common and high need specialties, this mercenary approach is costing healthcare dearly.

More subtle but no less problematic is the literal adherence that some physicians take to their contract. Call coverage, providing care at new sites built or purchased by a health system can be ways in which physicians leverage better pay for very strict adherence to the wording of a contract. In academics, this plays out as the percent effort. Even in the setting of acute clinical need, funded physician scientists may block days on their schedule based on time that is really only partially funded by grants or other sources. True, they may get penalized for decrease productivity but the excess work merely gets transferred to less funded more clinical colleagues.

When physicians and systems both blame each other, the real loser is the patient. Our greater crime is that we have focused on throughput over the health and wellbeing of our patients. We all understand that the healthcare model is transitioning to value based care but we are all trying to grab what we can until we get there. In the circumstance where both physicians and health systems turn on each other, ultimately we have a complete breakdown of culture. Patients can sense this immediately and in markets where patients have choices shy away from systems where animosity runs high.

Which leads back full-circle to where we started. In the prisoners dilemma, the scenario in which both parties keep silent results in a lighter sentence for both minimizing the losses of both. In our healthcare model, cooperation requires both physicians and health care systems to give up something in both groups’ interest and more importantly in the interest of patient care. I believe that many systems have taken this cooperative approach by becoming clinician run organizations and this helps to an extent. The problem no matter how much you try as a physician leader to remain engaged is that you rely on an increasingly narrow and sometimes distant experience to provide insight on the administrative level. This is easier for physician leaders who maintain some clinical practice. However, the further a clinician gets from clinical care, the more likely they are to rely on numbers that can be taken out of context.

Is there a solution? I believe there is. We have models that already exist to guide us. The idea of a salaried model with minimal to no incentive would allow physicians to focus on a practice that is high value and patient-centered over productivity. Health systems would somehow need to make this worth it by providing physicians pensions or by incentivizing people with time for other activities. Physicians would need to give up individual autonomy but embrace the transition to leadership of the clinical team rather the lone wolf model we were trained in. Collaborative practices in which responsibility sat with a person based on their license and expertise allows everyone to focus on what they are trained for. There is no doubt that systems would have some decrease in productivity in the current model but it would prepare them well for the value based care model of the future that would allow physicians to engage in activities that benefit patients without worrying about the RVUs it might generate.

There are problems with this model. Lack of incentives will result in some either leaving to pursue economic fortune elsewhere as salaries would likely need to reflect lower productivity. Other may use this as an opportunity to fly below the radar and try to do no more than necessary. While some would argue that the lack of incentive discourages innovation, I would argue with the right culture, the model could be liberating allowing clinicians to focus more on health and wellbeing over throughput.

The other cooperative model would be for employers to become owners. I think back to the Saturn car commercials of the 1980s that showed a highly motivated work force because they owned part of the company. There is nothing like shared financial risk to bring groups together. If you allowed every employee of a health system to place some of their salary at risk each year and either share in the margin if it is positive or take a financial loss if negative, this could encourage cooperation. Those who wanted no risk could opt out but I suspect the majority would choose to participate. I can only imagine that this violates some laws and for those who are smarter on these topics, let me know. But I could envision a situation where everyone would be interested in conserving resources, designing processes that manage patients in lower cost settings, etc. Here the cooperations minimizes risk to both parties while offering the potential for an upside.

I would really like to learn your thoughts. There is no easy answer but I have become convinced that while we are all prisoners in a complicated system, there is a way out and that requires doctors and other healthcare providers to collaborate with the health system in the interest of patients’ health and wellbeing. Our models of blame do not work and while cooperation will still require sacrifice, it will be better for us, the system, and our patients in the end.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s