Walking the Mountain Together: Lessons From 30 Years of Shared Decisions

Secretary Kennedy has been talking a lot about “shared decision making.” The phrase sounds democratic, humane, empowering; a rebuke to paternalistic medicine and a promise of respect for patient autonomy. But in today’s clinics and policy debates, “shared decision making” often functions more as a rhetorical talisman than a clinical method. Many people, including some clinicians, would struggle to explain what it looks like in the exam room.

That matters because medicine is not a vending machine of options, nor a priesthood of unquestionable authority. It is a domain of uncertainty, asymmetry of knowledge, vulnerability, conflicting values, and irreversible consequences. How decisions get made inside that terrain determines whether patients are genuinely empowered or quietly abandoned.

Shared decision making does not mean “the doctor gives you a menu and you pick what you want,” “do your own research and tell me what drug you’d like,” or “everyone’s opinion is equally valid.” And it certainly does not mean the physician is neutral and morally off the hook.

It means something much more demanding, more relational—and more difficult to do well.

Shepherds, Sherpas, and Who Owns the Mountain

The traditional metaphor for doctors is that of a shepherd. The patient is the sheep. The doctor knows best. The doctor decides. The doctor leads. The patient follows.

That metaphor is dying, and for good reason. It implies passivity, dependency, and one-way authority. It fits badly with chronic disease, preventive care, end-of-life choices, and the modern reality that patients live with their conditions long after the clinic visit ends.

A better metaphor is the sherpa.

A sherpa does not choose the mountain. The climber does. The climber chooses the goal, the pace, and how much risk they are willing to accept. The sherpa, however, knows the terrain, the weather patterns, the crevasses, the false summits, and the places where people die when they push their luck.

They travel together. The sherpa carries some of the load. The sherpa advises. And sometimes the sherpa says, “This route isn’t safe. You can take it if you insist, but you should know what it really costs.”

That preserves the patient as the protagonist of their health “journey" while acknowledging the physician’s asymmetric knowledge and fiduciary duty. It respects autonomy without pretending expertise is optional.

The Cruel Joke: “Do Your Own Research”

Most medical decisions involve probabilities, trade-offs, delayed consequences, and incomplete data. Even professionals misinterpret risk statistics. Even specialists disagree. Clinical trials are done on populations that rarely resemble the messy individuality of the person in front of you. Guidelines lag behind emerging evidence and are shaped by committees, incentives, and politics.

Telling a layperson, “Here are your options, you decide,” without scaffolding, is not empowerment. It is abdication. Autonomy without literacy is theater. When a clinician responds, “I can’t tell you what to do,” the patient is stranded, not liberated. 

The question is not whether values belong in medical decisions. They always do. The question is who helps translate evidence into choices a real person can live with. If individuals are to be true stewards of their own health, the physician’s role is to make real stewardship possible. That requires something deeper than consent forms and pamphlets. Shared decision making, properly understood, has four interlocking parts.

1. Expert Guide to the Evidence and the Body

While a physician diagnoses, diagnosis is not just labeling. It frames the problem in a way that enables decision-making.

Is this condition self-limited or progressive? Immediately dangerous or slowly corrosive? Does treatment change outcomes meaningfully or mostly shift symptoms around?

Then come the options. Each option carries probabilities, uncertainties, and trade-offs. Some risks are immediate and visible. Others are delayed and abstract. Some benefits are dramatic but rare. Others are modest but reliable. Population-level data must be translated into what it means for you, your body, and life context. For example, a medication that reduces stroke risk by 2% over ten years may mean little to a healthy 40-year-old with a strong aversion to daily pills. However, it will mean a great deal to a 75-year-old who has already had a stroke and fears losing independence more than almost anything else.

This translation is not mechanical work amenable to algorithmic “ best practice.” It is interpretive work, suffused with human empathy and judgment.

2. Builder of Literacy and Context

Shared decision-making only works if the patient can reason meaningfully about the choice.

This often means slowing down to explain basics clinicians take for granted: the difference between relative and absolute risk, what “watchful waiting” actually entails, or why feeling better is not always the same as being safer. A patient who understands these distinctions is far better equipped to make a choice. 

That means teaching. Explaining physiology in plain language. Explaining what the disease does over time. Explaining what risk actually means. Correcting misinformation without humiliation. It means giving just enough information, at the right level, so the patient is neither overwhelmed nor infantilized.

It means making explicit how values shape choices.

Is short-term relief more important than long-term risk? Is preserving independence more important than extending life? Is avoiding side effects more important than maximizing survival odds?

None of these are “medical questions,” they are human ones, forced by circumstances and immediacy into the open.

3. Moral and Relational Fiduciary

Here is where the fantasy of “neutral information delivery” collapses.

The patient is vulnerable. The physician’s expertise creates responsibility, not moral neutrality.

A doctor who says, “Here are your options, it’s entirely up to you,” and withholds their judgment is not respecting autonomy. They are hiding behind it.

A responsible answer contains both guidance and respect:

“If I were you, given what you’ve told me matters most to you, I would lean toward option B. It better aligns with your priorities and avoids risks that you find unacceptable. But you are allowed to choose otherwise, and I will support you.”

Trust also lives here, built on transparency about uncertainty, honesty about conflicts of interest, system pressures, and algorithmic nudges that shape care behind the curtain. Trust is not built by pretending medicine is purer than it is.

4. Co-Agent in Decisions

Shared decision making is not “you decide, I obey,” or vice versa. It is joint agency.

The physician clarifies goals and constraints, matching options to those goals, creating a plan that the patient understands, endorses, and can realistically carry out. And then, inevitably, come adjustments to side effects, second thoughts, new information, and changing life circumstances. Real decisions unfold over time. There is no single moment of consent.

The degree of “sharing” also adapts to the situation. Some patients want strong guidance. Others want maximal control. Capacity, distress, culture, and health literacy all matter.

Rigid ideology has no place here.

RFK Jr.’s Framing

When RFK Jr. invokes shared decision-making, he gestures toward something real: the legitimate frustration people feel about being processed through a medical-industrial machine that often ignores their values, minimizes their concerns, and treats guidelines as holy writ. That anger is not delusional. It is grounded in lived experience.

But when shared decision-making is used as a cudgel against expertise or as a justification for elevating fringe claims to equal footing with evidence, it mutates into something else entirely. There is a difference between respecting autonomy and promulgating epistemic chaos; a difference between empowering patients and burdening them with impossible interpretive labor.

Shared decision-making does not mean every opinion is equally valid. It means evidence is filtered through values, not replaced by them.

A Quiet Radicalism

Proper shared decision-making is slow. It is relational. It resists standardization. It cannot be compressed into seven minutes and a checkbox. That makes it expensive in a system that rewards throughput, procedures, and coding.

Because it 

• Treats patients as moral agents, not consumers, and physicians as fiduciaries, not providers.
• Amits uncertainty without surrendering judgment.
• Builds literacy over time, rather than outsourcing it to Google.
• Accepts that medicine is not about optimizing metrics. It is about helping particular people live with the consequences of particular choices.

It is quietly radical.

If individuals are to be true stewards of their own health, the physician’s role is not to replace their agency. It is to cultivate it. Interpreting complexity, enhancing literacy and agency over time, growing to become a trusted advocate and recommender using judgment and relationship, to help a person choose and live their path.

So, when you hear “shared decision making” invoked as a political slogan, ask what version is being smuggled in.

Is it the vending machine model? The abdication model? The culture-war model?

The Sherpa model does not promise certainty or perfect outcomes. It promises companionship, honesty, and judgment offered in service of another person’s life. In a world of medical complexity, that may be the most humane promise medicine can make.