Upstream Learning & Downstream Dysfunction: How to Train Leaders for a New Future


“The arrogance of success is to think that what you did yesterday is sufficient for tomorrow.” – William Pollard

In health care right now, lots of people are talking about innovation as a generic way to describe the new holy grail to some of medicine’s biggest problems. Of course this line of logic might lead some to think medicine is in some state of “innovation-deficiency,”¬† a condition for which innovation replacement therapy is the clear and sole treatment. While this obviously isn’t always true, introducing new technologies and processes, if done the right way, can certainly help minimize waste, control cost, improve efficiency, and introduce equity into existing public systems.

So if medicine needs to up its innovation game, as many are suggesting, who’s it going to turn to? Where is the pipeline of leaders to fill the holy grail and how do you turn on the faucet of new and better ideas?

I’ve been thinking a lot about these questions over the past few weeks as I start my Fall semester at Harvard’s TH Chan School of Public Health. And no that wasn’t a name drop. It was a point of reference drop, to help place the problem I am about to describe, into context.

For the past 3 weeks I’ve been taking a pretty stand out course on quality improvement¬† taught by the incomparable Maureen Bisognano and her enviable squad of colleagues and friends. For a bit of background, Maureen is the President Emerita of the Institute for Healthcare Improvement, basically a healthcare innovation incubator based in Boston, and her friends include people like Don Berwick, Atul Gawande, Brent James, Ellen Goodman, and Derek Feeley, to name a few of the powerful speakers who inspired us. For 3 weeks, we reviewed international examples of ways to radically redesign the US healthcare system.

And then, I had fall registration.

Excited to push the boundaries of existing systems, I was eager to take classes across the University, and like Maureen showed us, find examples from other industries, sectors, and disciplines that could be applied to the great challenges facing medicine and build solutions that translate across public systems and private incentive structures. But to my surprise, that inter-sectional, open learning, not only wasn’t a requirement, but was incredibly difficult to find. And if it is hard to find at places branded on leadership, I worry it may be symptomatic of a larger issue.

Which got me thinking, just how are we going to train this next level of leaders when the educational pipeline through medicine looks remarkably similar to what was suggested in the landmark Flexnor Report in 1910, that established a system of medical education, that with few exceptions, continues today.

Courses continue to be divided by discipline, learners are isolated to their field of study, and for those who select to learn outside that box, they do so singularly, as a function of personal interest or a non-degree program rather than systematic integration of agents and ideas. Then is it any wonder why our systems remained silo’ed with divergent (and sometimes antagonistic) incentive structures despite overlapping public interests?

Don Berwick is often quoted as saying, “Every system is perfectly designed to get the results it gets.” In medicine, as we move upstream in our interventions for disease, to target root causes rather than downstream effects, we must do the same thing for the education pipeline. We have to challenge ourselves to teach future leaders how to think, reason, and execute rather than prescriptively guiding them, formally and informally, to reproduce the same system, with the same challenges we already have.

It has been more than 100 years since the last massive shift in professional medical education, perhaps parting with apprenticed dysfunction and practicing principled disruption is a good start.




Teaching Structural Competency and the Perils of Pretending


Earlier this year, I started teaching a course to first year pediatric residents at Stanford. In it, I challenge the trainees to identify the structural contexts in which patients and families make choices that may impact their health and well-being. Termed structural competency, the goal is to enable young physicians to understand and confront stigma and inequality as key determinants of health. We talk about educational attainment and health literacy, socioeconomic status and health access, social norms and health practices, institutional discrimination and health disparities, and the built environment and health behaviors. Together, we examine the fragile balance between resources and health, recognizing that local forces that manipulate resources effectively legislate health, by structuring choice and opportunity.

To illustrate these fundamental connections, it is often necessary to convert what otherwise exist as invisible forces in society into accessible, clinically-relevant language. This allows us to conceptualize the structural framework in which patients’ live, work, and play, within a medical model. It shrinks what seems like a diffuse and disconnected system of local policies and institutions into tangible drivers of health and disease that require socially-informed, clinical interventions. It transforms inequality, a sociopolitical phenomenon, into a silent but active participant in the clinical encounter. This makes addressing local infrastructure a central component of any community-centered, health promotion strategy.

But as we expand our purview beyond the exam room and encourage young physicians to adopt a global approach to clinical medicine, we must be very careful not to succumb to, what I will call, “the perils of pretending.” Here, there are 3 common pitfalls that warrant discussion.

1. The Poverty Simulator. In any educational endeavor, experience is perhaps, the greatest teacher. Without experiencing poverty first-hand, it may be difficult for residents to understand the challenges families living in poverty face when seeking medical care or selecting medical treatments.

One such simulator offers “players” a chance to live on a low-income budget. Other programs ask residents to navigate public transit to various appointments. At Stanford, I ask the residents to live on the average food stamp budget for a week. These self-reflective exercises are meant to influence learner attitudes about inequality and build empathy among providers as they realize what it takes to survive under certain conditions.

The problem with poverty simulators is that the process of pretending to be poor unfairly and inaccurately reduces the daily struggle of living in poverty to a series of poor choices, no pun intended. The “game” motif insinuates that some choices are superior to others while completely obscuring the larger network of policies and institutions that concentrate disadvantage and manipulate choice in low-income communities.

For example, if you live in a food desert, the choice to eat fresh produce is constrained by the proximity of those resources to your home. This “trade-off” requires bargaining between necessities and results in a loss either way. Buying cheaper food in your neighborhood may have adverse health consequences and expending the time and money to obtain healthy food on a fixed income makes other necessities unaffordable. This zero-sum reality profoundly limits choice.

To avoid this pitfall, it is important to be clear about the purpose of the exercise, which is to acknowledge that resource limitations have health consequences. The lesson is that poverty is not a deficiency of ingenuity or the manifestation of good or bad choices. There are no “right” choices when selecting between food and medicine. So if poverty is the result of eroding urban infrastructure and imbalanced resource allocation and is associated with poor health outcomes, then building infrastructure is a health intervention.

2. The Absence of Clinical Models. While the associations between social determinants of health and poor health outcomes are well-documented, we lack comprehensive, evidenced-based clinical models for addressing complex trauma and chronic stress, physiologically significant exposures that are the downstream sequelae of poverty and inequality. Short of co-locating same-day necessities in medical clinics, like food pantries or legal assistance, there are few models to describe how physicians in particular and the medical system at large, should engage the sociopolitical drivers of health through clinical work.

In the absence of these models, some physicians pretend there is nothing that can be done, or worst yet, that these issues are not “medical.”

The problem is that we are complacent in our current clinical practice. Stagnated by the dearth of evidence and overwhelmed by the magnitude of the issue, we simply avoid it. We fail to universally screen patients for social determinants of health because we don’t know where to refer them. We refuse to inquire about these issues because we essentially lack confidence in our ability or aptitude to address them.

The solution here is to do it any way. Just as all politics are local, so too will be the formation and dissemination of novel clinical models that address these issues. So we must encourage our trainees to identify the most pressing needs in their communities and trial socially-savvy interventions in their continuity clinics. This is quality improvement at its best.

3. The Conflation of Race and Risk. When seeking to address the “cultural” influences in a clinical encounter, it can be easy to minimize “culture,” to the readily identifiable traits in the visit. Here, “culture” becomes a monolithic, static archetype we project onto patients based on our unconscious bias about their physical attributes, like ethnicity, nationality, or language.

When we do this, we are pretending that socially-assigned attributes, like race, are a proxy for risk. We track patterns of disease prevalence by these attributes and over time, come to associate the attribute with the disease. This logical fallacy then informs clinical practice and leads clinicians make inaccurate assumptions about certain patient populations, their relationship with disease, and the efficacy of certain medications to address their complaints (remember BiDil?).

The solution is to replace cultural competence with structural competence and educate young providers to interrogate the local context in which patients live, the resources at their disposal, and the networks they rely on to make medical decisions. We must of course, when doing this, not turn a blind eye to the ways in which local policies and historical discrimination produce predictable patterns of disease in certain communities. These patterns may make it seem as if the risk factor is easily recognized in the exam room (race, nationality) as opposed to the real risk factor that lives in our communities – structural inequality.

As medicine advances to address inequality as an important driver of health, we must be thoughtful about the way we educate our trainees to tackle this new frontier in primary care. While there will be pitfalls along way, if we tread carefully and together, we can transform the future of medicine in powerful and meaningful ways.

* Update: To learn more about my course, please check out the syllabus and reference guide I have shared below.

Rhea’s Stanford Course Details

Rhea’s Stanford Course Resources