A dermatologist's father is diagnosed with glioblastoma in a city eleven hundred miles away.
Within an hour of the phone call, the family has arranged itself around a single assumption: she will handle it. She will read the pathology report. She will judge whether the oncologist is any good. She will find out whether the recommended protocol is standard or dated. She will get the appointment moved up. And in some months' time, she will be the one who says the word hospice.
She is a dermatologist. She has not thought seriously about neuro-oncology since a rotation two decades ago. She is also, and this is the part nobody in the family registers, a daughter whose father is dying, which is not a state compatible with careful clinical reasoning.
She does what every physician in this position does. She starts texting.
A fellowship friend who might know someone. A residency classmate who is now a department chair somewhere. Anyone, anywhere, who can spend twenty minutes explaining the landscape to a colleague who is out of her depth and cannot say so to her family.
If she trained at a large academic center and kept in touch, she will find someone within a day. If she did not, she will read papers at midnight and guess.
This is the oldest reciprocity system in medicine and it has never been given any infrastructure at all.
It happens to essentially everyone
The scale of this is not in doubt, and the foundational study is thirty-five years old because the phenomenon is that stable.
La Puma and colleagues, publishing in the New England Journal of Medicine in 1991, surveyed physicians about requests from family members:
- 99 percent (461 of 465) had received requests for medical advice, diagnosis, or treatment from relatives.
- 83 percent had been asked to prescribe.
- 80 percent had been asked to diagnose.
Ninety-nine percent. This is not a subset of physicians with particular family dynamics. It is the universal condition of holding a medical degree.
And the profession's formal guidance addresses only part of it. Ethical codes counsel against treating family members, for excellent reasons: objectivity is compromised, examination is awkward, records are absent, and the relationship distorts judgment in both directions.
But the guidance answers the treatment question and says essentially nothing about the coordination burden, which is the far more common and more consuming demand. Nobody is asking the dermatologist to prescribe for her father. They are asking her to steer.
The other 1993 study, which is the important one
There is a second New England Journal paper that deserves far more attention than it receives, because it documents the mechanism that actually carries this.
Levy and colleagues, in 1993, studied professional courtesy:
- 96 percent of physicians (2,127 of 2,224) offered free or discounted care to fellow physicians and their families.
- 79 percent said professional courtesy solidified the bonds between physicians.
Read that as an anthropological finding rather than a billing one. Nearly every physician in the country extended a real economic favor to colleagues and their families, and four in five understood it as the thing that held the profession together.
The economics that supported free care have largely been dismantled by insurance structures and compliance rules. The underlying reciprocity did not disappear. It changed currency.
What physicians now extend to each other is not free care. It is access and attention: the appointment moved up, the pathology looked at, the twenty minutes on the phone explaining what the options actually mean, the honest answer to "is this oncologist any good."
That favor economy is running right now, at very large scale, entirely on personal networks. And unlike the 1993 version, nobody has studied it, named it, or built anything for it.
What it costs the physician-caregiver
The burden here is real and increasingly documented, under the term "double-duty caregiver."
Research in The Gerontologist found double-duty caregivers are expected to be involved, by their families and by themselves, and that they "experience presenteeism and make errors" at work. A PLOS One scoping review found healthcare professionals acting as family caregivers face "multifaceted expectations" and report "stress, guilt, and potential burnout," with support needs largely unmet.
The specific pressures are distinctive:
You cannot admit uncertainty. Your family's confidence in the plan rests on your competence. Saying "I genuinely do not know if this is the right treatment" removes the only expert they have.
You know too much and not enough simultaneously. You can read the pathology and cannot contextualize it. You understand the statistics and cannot judge this particular oncologist.
You cannot stop being the doctor. Every family member's question routes to you, including at the funeral.
And there is a documented friction with the treating team. The physician-relative asking detailed questions and requesting expedited access can trigger the dynamic clinicians recognize as VIP syndrome, which is associated with worse rather than better care.
The invisible inequity
Here is the part that makes this a coordination problem rather than a private hardship.
Access to this favor economy is distributed by training network.
The physician who trained at a large academic center, in a well-connected specialty, and who maintained those relationships, can reach a world expert in almost anything within a day. That is not an exaggeration; it is the ordinary experience of a well-connected physician with a sick parent.
The physician who trained at a community program, or who is an international medical graduate whose classmates are scattered across other countries, or who has been in solo practice for fifteen years and lost touch, has essentially none of it.
The referral literature confirms the structure. Physicians route to co-trainees at meaningfully higher rates than to comparable strangers, with the effect driven by residency and fellowship co-training rather than medical school. The favor economy runs on the co-training graph, which means it delivers superb care to the families of well-networked physicians and nothing to everyone else.
And a further consequence worth sitting with: the patients of physicians outside these networks get the same care as everyone else, while the patients of well-connected physicians get expedited, expert-reviewed, quietly better care. That is the system working exactly as its structure dictates, and nobody designed it.
Why this is harder than it looks
It is worth being honest about why nobody has built anything here, because the obstacles are real.
The demand is episodic. A physician needs this a handful of times in a career. That is a difficult basis for any service, because nobody signs up in advance for something they hope not to need.
It is emotionally loaded. People in this situation are not evaluating options rationally. They are frightened and moving fast.
The ask is ill-defined. It is not a consultation, not a second opinion, and not a referral. It is closest to orientation: help me understand the landscape well enough to make decisions with my family. Medicine has no name for that transaction and therefore no structure for it.
And the guardrails are genuinely important. Anyone building here must handle: no patient records shared without the patient's own consent, the peer orienting rather than treating or documenting, clinical responsibility remaining entirely with the treating team, and the physician-caregiver being told explicitly that this is not a second opinion. Without those boundaries stated up front, the whole thing becomes an informal second-opinion service, which is a different and more regulated activity.
What would work
The realistic version is not a navigation product. It is making the existing favor economy visible, routable, and reciprocal.
A willingness signal. Physicians declaring, privately, "I will take colleague-caregiver calls about my area." Many physicians would say yes to this and are never asked, because there is no mechanism to ask.
Matching on specialty and situation, not on who you know. The dermatologist needs a neuro-oncologist willing to spend twenty minutes with a colleague. That is a small, specific, findable set of people if anyone had built the index.
A named transaction. Call it orientation, or a colleague call, or anything with defined boundaries: twenty minutes, explanation and landscape only, no records, no documentation, no treatment relationship. Naming it makes it askable, and the absence of a name is a substantial part of why physicians hesitate.
A reciprocity record. The reason the favor economy strains is that givers are asked repeatedly and receive nothing. Counting the give changes the sustainability of the whole arrangement, which is precisely what the 1993 study's 79 percent understood about professional courtesy.
And the obvious supply. Retired and semi-retired physicians in the relevant specialty are ideal for this: deep experience, time available, no competing clinic, no institutional conflict, and frequently a strong desire to remain useful. They are also, once again, invisible to every existing channel.
What you can do now
If you are the doctor in the family
Say out loud what you do not know. To your family, in plain terms: "I am a dermatologist. I can read this report and I cannot tell you whether this is the right treatment. I am going to find someone who can." That sentence relieves an enormous burden and is almost never said.
Ask for orientation, not for a second opinion. When you contact a colleague, be explicit: "I do not want you to review the case. I want twenty minutes to understand the landscape so I can help my family think." That framing gets a yes far more often, because it is a much smaller and safer ask.
Ask your own cohort first, and ask them to ask. Your residency classmates may not know neuro-oncology and one of them almost certainly knows someone who does. Two hops is usually enough and most people stop at one.
Do not read the portal at 2 a.m. Every physician-caregiver does this and it is corrosive. Set a time to look at results and do not look outside it.
Respect the treating team. The VIP dynamic is real and harmful. Ask questions directly, disclose that you are a physician, and be explicit that you are not trying to direct care.
Notice what this is doing to your work. The presenteeism and error findings are about you. Tell someone at work what is happening.
If a colleague asks you
Say yes to the twenty minutes. You will not remember it in a month. They will remember it for the rest of their life. This is the single highest-value use of twenty minutes available in the profession.
Orient rather than opine. What the standard options are, what questions to ask, what would make you worried, who else to talk to. That is enormously useful and carries none of the hazards of case review.
Be honest about the local team. Physicians ask each other about the quality of a named clinician precisely because it is the one thing they cannot look up, and the honest answer, delivered privately to a colleague, is what the whole favor economy exists to move.
If you are senior or retired
Tell people you will take these calls. In your specialty, for colleagues with a family diagnosis. It costs a few hours a year, it is the kind of contribution retired physicians consistently say they want, and there is no other route by which anyone will find you.
Frequently asked questions
How common is it for physicians to be asked for medical help by family? Essentially universal. A 1991 New England Journal of Medicine study found 99 percent of physicians had received requests from family members for advice, diagnosis, or treatment, with 83 percent asked to prescribe and 80 percent asked to diagnose.
Should physicians treat their own family members? Professional ethical guidance generally counsels against it, citing compromised objectivity, awkward examination, absent records, and distorted judgment. That guidance addresses treatment and says comparatively little about the far more common coordination and interpretation burden that families actually place on physician relatives.
What is professional courtesy in medicine? Historically, the practice of providing free or discounted care to fellow physicians and their families. A 1993 New England Journal study found 96 percent of physicians extended it and 79 percent believed it solidified professional bonds. The economics have largely ended, while the underlying reciprocity persists in the form of access, attention, and informal advice.
What is a double-duty caregiver? A healthcare professional who also acts as a family caregiver. Research finds they are expected to be involved by families and by themselves, experience presenteeism and workplace errors, and report stress, guilt, and burnout with largely unmet support needs.
How do physicians get expert opinions for their own families? Almost entirely through personal networks built during training. Physicians route to former co-trainees at measurably higher rates, which means access to informal expert input is distributed by where someone trained and whether they maintained those relationships, rather than by need.
What should a physician ask a colleague for when a relative is diagnosed? Orientation rather than case review: twenty minutes to understand the standard options, the questions worth asking, what would be concerning, and who else to speak to. This framing is a smaller and safer request than a second opinion, requires no records, creates no treatment relationship, and is what most physician-caregivers actually need.
The bottom line
Ninety-nine percent of physicians are asked by their families for medical help. Ninety-six percent, in the era when it was possible, extended free care to colleagues and their families, and four in five understood that reciprocity as the thing that held the profession together.
The free care ended. The reciprocity did not. It simply changed into something harder to see: the expedited appointment, the pathology looked at as a favor, the twenty-minute phone call that lets a frightened colleague understand what is happening to their father.
That system is running at enormous scale right now, entirely on personal networks, with no routing, no capacity signal, no record, and no way in for anyone who did not train in the right place.
So a dermatologist eleven hundred miles from her dying father either finds someone within a day or reads papers alone at midnight, and which of those happens depends on decisions she made about where to do residency twenty years ago.
The profession's oldest mutual aid arrangement is also its least engineered. It works beautifully for the well-connected and not at all for everyone else, and nobody has ever tried to fix that, because it has never occurred to anyone that it is a system at all.
Part of a series on the missing professional infrastructure of healthcare. Previously: The Solo Generalist's Backup
Evidence note: prevalence of family requests comes from La Puma et al., New England Journal of Medicine (1991). Professional courtesy figures come from Levy et al., New England Journal of Medicine (1993). Double-duty caregiver findings come from The Gerontologist (2025) and a PLOS One scoping review (2025). Co-training referral effects come from Pany and McWilliams in Health Services Research (2021). Ethical guidance on requests from relatives is discussed in Eastwood, Journal of General Internal Medicine (2009), and in professional codes of medical ethics. Nothing in this article is clinical or ethical guidance for any specific situation.
