# What Everyone Wants From a Doctor

*Belief, truth, company, and the person who came in wanting something else*

Kanwar Partap Singh Gill, MD · November 5, 2026 · The Human Pulse

The title of this essay makes a promise I cannot keep, and it seems only honest to say so before going any further. Nobody knows what everyone wants from a doctor. People arrive wanting different conversations. Some want every detail and, if possible, a diagram. Some want the answer to the question they came with and nothing else, thank you. Some enjoy a minute of talk about the drive or the weather; others regard it as a tax on their afternoon, levied by a man who is already running late.

There is evidence for the disagreement. In 2004, Sara Swenson and colleagues reported in the *Journal of General Internal Medicine* on [a randomized study](https://readingroom.law.gsu.edu/faculty_pub/2031/) in which 250 adult patients at the clinics of an academic medical center were asked which of two styles of communication they preferred. Sixty-nine percent chose the patient-centered style, the one more interested in the patient's perspective and in the person. The rest, roughly three in ten, preferred the more conventional, doctor-directed, biomedical style. Younger patients were more likely to prefer the patient-centered approach.

I find that result oddly reassuring, which may not have been the authors' intention. It means that a substantial minority of the people in any waiting room may want me to be briefer, more direct and less curious about their weekend than I am inclined to be, and that being warm at someone who came for efficiency is its own way of not listening. Somewhere a reader wanted this paragraph to get to the point. He has my sympathy, and the next sentence.

With that admitted, three wishes still seem to me worth holding together, as aspirations and not as a finding about all patients: to be believed, to be told the truth, and not to be left alone with it. Each needs care in the defining, because each can be honoured badly. And beside them sits a fourth commitment, which is less a wish than a discipline: to keep discovering that this particular person, today, wants something I did not expect. The question this essay tries to answer is how a doctor can hold a few convictions firmly enough to be trustworthy and loosely enough to be corrected.

## A Preference Is Not a Personality

The Swenson study divided patients into two camps, and any study must simplify in order to measure. But a conversation in a consulting room has more dimensions than a choice between two styles, and they vary independently of one another.

There is the amount of detail a person wants: the headline, the mechanism, the numbers, the literature. There is directness: whether hard news should arrive in the first sentence or be approached in stages. There is the person's preferred role in decisions: to decide, to share the deciding, or to be told what the doctor recommends. There is accompaniment: who should be in the room, who should hear the news first, whether a daughter should be present for the explanation or only for the drive home. And there is timing: whether everything should be discussed now, or whether the person needs the main point today and the rest next week, after sleeping on it.

These dimensions do not travel together. One person wants every number and has no wish to make the decision; she wants to understand exactly why the recommendation is what it is, and then to be told it. Another wants very little detail and insists on deciding; he would like the options in a sentence each and the right to choose among them without a lecture. A third wants the hard news in the first sentence and the explanation never. None of these is a type. Each is a set of answers to several questions, and the answers may change when the decision changes.

That is the trouble with inferring preferences from anything other than the person. Swenson's team found that age was associated with preference, and an association is a fact about a population, not an instruction about the eighty-year-old who wants to read the trial data herself. The same is true of assumptions about culture. There is a widespread belief that families from some traditions want serious diagnoses told to relatives before the patient, and in some families that is true, and in others it is a stereotype that would deprive a competent adult of news about her own body. Earlier visits are a better guide than demographics, but they are a guide to an earlier decision. The man who wanted the short version for his blood pressure may want the long version for the lump.

What predicts the conversation a person wants is mostly the person, today, facing this. The only reliable way to learn it is to ask, and the asking can be brief. *How much detail would be useful to you?* *Would you like my recommendation, or would you rather hear the options first?* *Is there someone you would like with you for this?* Questions of that kind cost seconds and save the far longer conversation that follows a wrong guess.

Preferences can shift even within a single visit. A person who began by asking for everything may, ten minutes into a frightening explanation, have heard as much as she can hold for one day; her questions get shorter, or stop. A person who wanted only the headline may, once it has been delivered and turns out to be bearable, want to understand it after all. The conversation a person wanted at the start is evidence about the conversation she wants now, not a contract that binds her to it.

## Believed

The first wish is usually put as *to be believed*, and belief is the right word for part of it and the wrong word for another part.

Punjabi has a sixteenth-century line that names the need underneath. Shah Hussain, the Lahore Sufi poet, begins one of his best-known kafis with a cry that runs, in transliteration, *Māī nī main kinūṅ ākhāṅ, dard vichhoṛe dā hāl*: roughly, in my rendering, *Mother, to whom shall I tell the state of this pain of separation?* The poem is about longing for the divine beloved, and I do not want to flatten it into a clinical text. But its opening question is one that a great many people carry into a doctor's office, whatever their pain is made of. Before they want a diagnosis, before they want a plan, they want someone to tell. *To whom shall I tell it* is the oldest form of the wish to be believed.

What a person reports about their own experience, that it hurts, that they are exhausted, that something has changed, is evidence, and the honest default is to treat it as true. What a person believes is *causing* it is something slightly different. It is a hypothesis, often a good one, sometimes the best one in the room. Taking someone seriously means hearing both. It does not mean adopting the explanation before examining it, and it does not mean dismissing it because the doctor arrived with a different one. *I believe you that this is real* and *I agree that this is your gallbladder* are separate sentences. A person can be offered the first without the second, and is owed the first regardless.

The test of being believed, as I understand it now, is whether the account changes what happens next. Was the person examined where they said it hurt? Was their own description written down in their own words, and not translated at once into a tidier term? When the explanation turned out to differ from the one they brought, were they told why, or only told no? A theory that is considered and set aside, with reasons given, has been taken seriously. A theory that was never really heard has not, however courteous the conversation around it.

The philosopher Miranda Fricker, in *Epistemic Injustice* (2007), gave a name to one way this test is failed. She called it testimonial injustice: the wrong done to a person when a listener gives their word less credibility than it deserves because of a prejudice about the kind of person speaking. The prejudice need not be conscious, and the listener may be entirely courteous. In medicine the deficit can attach to age, to sex, to race, to a psychiatric diagnosis already in the chart, to a history of frequent visits; there is a substantial literature suggesting that the pain of some groups of patients is more often underestimated than that of others. What makes Fricker's idea useful in a consulting room is that it locates the failure in the hearer. The patient told her story accurately. It was discounted on arrival, before anyone had examined it.

Francis Peabody, the Boston physician whose lecture to Harvard medical students was published in 1927 as "The Care of the Patient," warned against a particular way of failing this test. He observed that a person admitted to hospital tends to lose his personal identity, and becomes, in the speech of the ward, a case of some disease in some bed. The danger is not limited to hospitals. A person can become *the chest pain in room four* in any building that sees enough chest pain. Once that has happened, the account the person gives is heard as an instance of the category, and anything in it that does not fit the category is quietly discarded as noise. Belief, in the sense that matters, is the refusal to let the category hear for you.

## Truth the Person Can Use

Truth has its own difficulty, which is that a perfectly accurate sentence can be useless.

Medicine's vocabulary contains traps its speakers have stopped noticing. A *negative* result is usually good news, and a *positive* one often is not, which inverts the ordinary meaning of both words. *Chronic* means long-lasting, not severe, though it is easily heard as a sentence of doom. *Acute* means sudden, not necessarily serious. None of these words is wrong. Each needs a plain sentence beside it, said before anyone has to ask.

Why do experts misjudge this so consistently? The best explanation I know comes from a small experiment. In 1990 Elizabeth Newton, a graduate student in psychology at Stanford, asked one group of people to tap out the rhythms of well-known songs on a table and another group to name the songs from the tapping. The tappers predicted that their listeners would recognize about half of them. The listeners identified roughly one in forty. The tappers could hear the melody in their heads as they tapped and could not stop hearing it, and so they could not imagine how bare the knocking sounded to someone without the tune. Economists had already named the general phenomenon the curse of knowledge. A physician explaining a diagnosis is a tapper. The melody of anatomy and physiology plays under every sentence he speaks, and he cannot easily hear how much of what he says is, to the person across the desk, only knocking.

Order matters as much as vocabulary. The point should usually come first and the details after. A person who has been waiting to learn whether something is serious will absorb very little else until that question has been answered, and a careful explanation delivered ahead of the answer is mostly heard as delay. There are exceptions, and they belong to the person: some people want to be walked toward hard news, and will say so if asked. But the default should favour the answer, because the cost of a wrong guess in that direction is a few seconds of abruptness, while the cost of a wrong guess the other way is several minutes of dread disguised as context.

Then there is quantity. Some people want everything. Others want the headline and an open door to ask for more later. The same person may want different amounts on different days. Communication courses often teach a sequence called *ask–tell–ask*: find out what the person already understands and wants to know, tell them, then ask what they have taken from it. Written down, it sounds mechanical, and no sequence guarantees understanding. In practice it is a way of letting the person who has to live with the information set its depth.

One more distinction belongs here, and it is easy to lose. Satisfaction is not understanding. A person can leave a visit grateful, reassured and entirely mistaken about what was said; a nod in an examination room is a social act, not a report of comprehension. Communication that aims at satisfaction will tend toward warmth and fluency. Communication that aims at understanding will sometimes be slower and a little awkward, because it has to check. Both matter. Only the second is truth the person can use, and it is the one that tends to be sacrificed when the clock runs short, because its absence does not show until later.

## The Wish Not to Know

The second wish, to be told the truth, has an exception that complicates the title more than any other. Some people do not want to be told. Not everything, or not yet, or not by this person, or not in this room. A woman awaiting a biopsy may want to know whether the result means more treatment, and may not want to hear the statistics that come with it. A man may want his son to hear the details and pass on what he thinks his father needs. Someone may want to know the plan and nothing about the odds.

European law has recognized this explicitly. The Council of Europe's Convention on Human Rights and Biomedicine, opened for signature at Oviedo in 1997, states that everyone is entitled to know any information collected about his or her health, and then, in the next sentence, that the wishes of individuals not to be so informed shall be observed. A UNESCO declaration on the human genome adopted the same year speaks in similar terms of a right to decide whether or not to be informed of the results of genetic examination. The pairing is instructive. A right to know that cannot be declined becomes a duty to listen.

The difficulty is to honour the wish not to know without sliding back into an older habit, in which doctors decided on their own authority what patients could bear and withheld serious diagnoses as a matter of course, a habit whose history I have traced [elsewhere](../what-must-not-change/index.html). The two can look alike from outside. What separates them is who decides. The wish not to know must be the patient's own, expressed by the patient, and not inferred by the doctor from her age or her background, and not requested on her behalf by a family that wishes to protect her. It is also best learned before the news exists. *When the results come back, some people like to hear every detail, some prefer the main point, and some would rather I talked it through with someone they choose. What would you like?* Asked in advance, that question is a courtesy. Asked after the result is known, it is itself a piece of news.

There are limits. A decision about treatment cannot responsibly be made by someone who refuses to know what the treatment is for, and some results carry obligations toward other people that a patient's preference cannot cancel. But within those limits, the right to be told the truth includes the right to set its pace, and occasionally to decline it. A doctor committed to truthfulness is committed to not deceiving, which is a different thing from a commitment to say everything, to everyone, at once.

## Help with Deciding

The decision is where the disagreement among patients becomes sharpest, and where medicine has changed its mind most visibly within living memory.

In 1992 the bioethicists Ezekiel and Linda Emanuel published a paper in *JAMA* describing four models of the relationship between physician and patient. In the paternalistic model, the doctor decides what is best and secures the patient's agreement. In the informative model, the doctor supplies the facts and the patient, as a kind of consumer, chooses. In the interpretive model, the doctor helps the patient work out what she actually values and which option fits it. In the deliberative model, the doctor goes further, engaging the patient in a conversation about which values are worth pursuing, as a teacher or a friend might. The Emanuels argued for the last as the ideal in most circumstances. Their scheme has proved durable because almost everyone who has been a patient recognizes at least two of the doctors in it.

Atul Gawande, in *Being Mortal* (2014), gave the first three of these types their comic names: Dr. Knows-Best, Dr. Informative and Dr. Interpretive. His point, made with some self-criticism, was that medicine had largely abandoned the first for the second, and that the second, for all its respect, could leave people alone at the moment they most needed help. A doctor who lays out the options, with their probabilities, and then steps back has respected the patient's right to choose. He may also have handed a frightened person a problem she has no means of solving, and called it autonomy.

The patient who asks *What would you do?* deserves a fair hearing, and in some circles she does not get one. Her question is sometimes treated as an abdication, a failure to engage, a thing to be gently deflected back to her. I think that is a mistake. A recommendation is information, and frequently the most useful thing a frightened person can be given, because it compresses a great deal of experience into a form a person can act on. Giving it does not remove her choice. It adds a voice to it. What turns a recommendation into the paternalism of the first model is not its existence but its manner: whether the reasons are given, whether they are tied to what the person has said matters to her, and whether she remains genuinely free to decline it without paying for the refusal in the doctor's coolness.

There is a subtlety in answering her, though. *What would you do?* can be heard as a question about the doctor's own values, and those may be a poor guide to hers. A physician's tolerance of risk, his feelings about hospitals, his circumstances and his fears are his own, and it is a much-discussed observation among doctors themselves that physicians facing serious illness sometimes choose differently for themselves from what they routinely offer others. The more honest answer is usually a translation. *If I were advising someone who had told me what you have told me, that staying in your own home matters more to you than anything, this is what I would recommend, and this is why.* That answers the question she asked in the form she needs: a recommendation fitted to her, and not a confession about me.

Respectful participation, in other words, need not mean forcing everyone to decide in the same way. Some people want to weigh every option; they should be given the means to weigh them. Some want a recommendation with reasons; they should be given one. Some want to decide together with a spouse, a son, a sister, and should be given time for that. Some want the decision made by someone they trust, and that, too, is a choice, provided it is theirs and was not made for them by default. The shared-decision ideal is a good ideal. Applied as a compulsory style, it becomes its own kind of paternalism: the doctor deciding how the patient must decide.

## Not Alone with It

The third wish is the easiest to promise badly.

To my ear, the Punjabi ਭਰੋਸਾ, *bharosā*, names what is at stake more exactly than the English *trust*. It means reliance, the confidence one can lean on: to have *bharosā* in someone is to be able to put weight on them and expect them to hold. It also travels in the other direction. To *give bharosā*, ਭਰੋਸਾ ਦੇਣਾ, is to assure someone, to offer them something to lean on. A patient's *bharosā* in a doctor and the doctor's *bharosā* to the patient are the same word moving across the desk, and each is worth exactly as much as the promises behind it.

The old physicians knew that this reliance was itself a kind of medicine. In the Hippocratic collection there is a short, late text called *Precepts*, almost certainly not by Hippocrates himself, which in W. H. S. Jones's translation of 1923 observes that "some patients, though conscious that their condition is perilous, recover their health simply through their contentment with the goodness of the physician." One can quarrel with the word *simply*. But the observation underneath it is sound: a person who feels held by a reliable relationship bears illness differently from one who does not. And the same chapter, characteristically, is largely about fees. It urges the physician not to be unkind about fees, to consider the patient's means, and sometimes to give his services for nothing, and it is in that practical context that it says "where there is love of man, there is also love of the art." The ancient writer did not separate trust from the arrangements that make it possible. Neither should we.

Which is why *bharosā* must be given carefully. A doctor cannot offer unlimited presence. There are other patients, nights and weekends, and systems that send messages to whoever happens to be covering. Promising more than the actual team can deliver does not create reliance; it arranges a disappointment for later, and the disappointment will fall precisely when the person leans. What can be offered is more modest and more honest: that a difficult conversation does not end on the difficult sentence. That there is a next step, and the person knows what it is. That they know when the next conversation will happen and how to reach someone before then. That the limits are stated plainly, *I won't have that result until next week*, *if this changes over the weekend, this is who will answer*, so that no one mistakes silence for neglect.

Information leaves the examination room with the person who received it. It goes home to kitchens and cars and the hours after midnight. Accompanying someone, in the modest sense I mean, is making sure they know what they are carrying, and whom to call if it grows heavier. That is a smaller promise than *I will always be there*. It has the great advantage of being one that can be kept.

It also has to be a promise that survives the doctor's absence, because modern care is given by teams and the person who explained the plan will not be the person who answers every call. Reliance placed entirely in one individual is fragile for exactly that reason. Part of giving *bharosā* honestly is making it transferable: telling the person who else is involved, what the nurse who calls on Thursday will know, what the covering physician over the weekend will be able to see, and what they will not. A patient who has been introduced, even in a sentence, to the people behind her doctor can lean on the arrangement without feeling that she has been handed to strangers. A patient who has not been introduced discovers the team only when her doctor is unavailable, which is the worst possible moment to learn that her trust was resting on one pair of shoulders.

## When the Answer Is No

A sceptic could fairly press the argument from the other side. If preferences are so varied and so important, why not simply give people what they ask for? And if some requests should be refused, who is the doctor to decide which?

Some requests do collide with evidence, with professional duty or with the needs of other people. A person may ask for a prescription the evidence does not support for the illness in question, such as an antibiotic for what is almost certainly a viral cold, which will not help and may harm. A person may ask for a scan whose result could not change anything but whose radiation and incidental findings are real. A person may ask for a letter to an employer that would state something the doctor does not believe to be true. A person may ask for a treatment whose risks fall partly on others. In each case the wish is understandable, and in each case agreement would fail the person or someone else. Unquestioning agreement is not respect. It is a way of withdrawing from the relationship while appearing to honour it.

But the opposite response fails too. A doctor who meets a request with a flat refusal, perhaps accompanied by a short lecture, has asserted control and explained nothing, and the person leaves with the same wish and less *bharosā* than before. Between those two failures lies a narrower path, and it seems to me to have four parts. First, find the wish behind the request: the person who asks for antibiotics usually wants to get better quickly, or needs to be back at work on Monday, or was given them last time and got better. Second, say what can be done toward that wish, because something usually can. Third, give the reason for the refusal in terms of the person's own interests, not the doctor's rules. Fourth, leave a route open: a time to come back if things change, or a second opinion if the person remains unconvinced.

Suppose, for illustration, a man with a streaming cold asks for antibiotics because he has an important week at work and was given them for something similar two winters ago. The flat refusal says that antibiotics do not treat viruses, which is true and answers nothing he cares about. The longer version takes perhaps a minute more. It acknowledges the week at work, which is the real subject. It explains that the tablets would not shorten this illness and could bring side effects of their own at exactly the wrong time. It turns to what might actually help him get through the week, and it tells him plainly which changes in how he feels would be a reason to come back. He may still be disappointed. He has, however, been treated as someone with a reason, and not as someone with a demand.

This approach does not always end in agreement, and it should not be judged by whether it does. A person may leave still disagreeing. What matters is that they leave having been told why, by someone who understood what they were after, with their dignity and their options intact. A respectful limit is a form of honesty. It tells the person that the doctor's yes is worth something, because it is not automatic.

## Repair

A conversation can begin with good intentions and still miss the person.

Imagine a clinician giving a careful explanation of a new diagnosis: what it is, what causes it, what the treatment involves, how it will be monitored. The explanation is accurate, well ordered by the standards of a textbook, and delivered with warmth. The patient, meanwhile, is trying to discover one fact: whether the plan will interfere with getting to work tomorrow, at a job where a missed day is a serious matter. Every sentence of the explanation is true. Its order has made it difficult to use, because the one thing the person needed is buried somewhere after the mechanism, or not addressed at all.

The mismatch may show itself in small ways. The patient's eyes go to the clock. A question arrives that seems off the subject: *What time does the pharmacy close?* Or the patient says nothing, and nods, and the visit ends with a satisfied clinician and a person who has learned nothing she can act on before morning.

Repair begins when the mismatch becomes visible, which means it begins with noticing. It does not require an elaborate apology that asks the patient to reassure the clinician that no offence was taken; that only adds a second task to the first. It requires a change of direction. Name the question that was missed: *I've been explaining the condition, and I think the thing you need to know is whether you can work tomorrow.* Answer what can be answered. Then ask how much of the remaining explanation would be useful now, and how much can wait. The response shows the patient that her correction altered the encounter, which is the only proof that it was heard.

What makes repair difficult is seldom ignorance of how to do it. It is the clinician's attachment to the explanation already under way: its logic, its completeness, the sense that a good explanation should be allowed to finish. Letting it go mid-sentence can feel like a loss of control. It is better understood as the conversation finally finding its subject. A good conversation is not one in which the clinician never guesses wrong. It is one in which a mistaken guess can be corrected without the patient having to struggle for the right to be heard.

The same repair runs in the other direction, and is needed as often. A clinician who has learned that brevity is a kindness may give a crisp answer to someone who wanted to understand, and watch her leave with a plan she cannot explain to anyone at home. The signs are different: a question that circles back to *why*, a request to have something written down, a pause before agreeing. The remedy is the same. Name what was missed, offer the fuller account, and ask how much of it would help. Neither brevity nor thoroughness is the virtue. The virtue is noticing which one this person needed, and changing course when the first guess was wrong.

## Asking Again

Continuity can help with all of this, but only if memory remains revisable. A note that someone usually wants a concise answer may be helpful at a later visit. It should not become permission to withhold detail when the person is now facing a different decision. Likewise, someone who previously wanted every number may have a day when they need the main point and time to think. I would like the record of a preference to work as an opening question and not as a permanent instruction about a person's character. *Last time you wanted the short version; is that still right for this?* takes a few seconds. It keeps the benefit of remembering without losing the ability to learn.

Peabody ended his lecture with a sentence that has been quoted so often it is in danger of sounding like a greeting card: "One of the essential qualities of the clinician is interest in humanity, for the secret of the care of the patient is in caring for the patient." Read slowly, it is stranger and better than its reputation. It looks like a tautology, and the apparent tautology is the point. *Caring for* has two senses in English: the work of looking after someone, and the feeling of concern for them. Peabody's sentence insists that the first depends on the second, that the tasks of care will not be done well by someone who is not interested in the person receiving them. Elsewhere in the lecture he put the same thought more bluntly: the treatment of a disease may be entirely impersonal, but the care of a patient must be completely personal. He died later in 1927, in his mid-forties, and the lecture has often been read in the light of his own illness.

*Completely personal* is the phrase that answers my title. If the care of a patient must be personal, then what everyone wants from a doctor cannot be specified in advance, because it is different for each person and for each day. What can be specified is the attitude that finds it out: enough conviction to offer belief, truth and accompaniment to anyone who walks in, and enough curiosity to ask what this person, today, came in hoping for. The answer is usually some version of the title. Often enough to matter, it is not, and the visit goes better when the doctor is ready to hear the difference.

## Evidence and Editorial Note

**Documented findings.** The essay relies on the author's public biography as a family physician practising in California since 2010, whose first language is Punjabi. Swenson SL, Buell S, Zettler P, White M, Ruston DC, Lo B, "Patient-centered communication: do patients really prefer it?" *Journal of General Internal Medicine* 2004;19(11):1069–1079 (randomized study; 250 English-speaking adult patients in urgent care and ambulatory clinics of an academic medical center; 69 percent preferring the patient-centered style; younger age among the factors associated with that preference): abstract on the Georgia State University College of Law digital repository (linked). The proportion preferring the biomedical style is inferred by subtraction. Emanuel EJ and Emanuel LL, "Four models of the physician-patient relationship," *JAMA* 1992;267(16):2221–2226 (paternalistic, informative, interpretive and deliberative models; the deliberative favoured as the ideal in most circumstances): standard reference knowledge. Atul Gawande, *Being Mortal* (2014): the doctor types Dr. Knows-Best, Dr. Informative and Dr. Interpretive and the argument that the informative model can leave patients alone with decisions, described without quotation from standard reference knowledge of the book. Francis W. Peabody, "The Care of the Patient," *JAMA* 1927;88:877–882, delivered to Harvard medical students and published in March 1927; Peabody's death later that year in his mid-forties, and the reading of the lecture in the light of his illness: standard reference knowledge; the closing sentence is quoted, and the passages on the hospital patient losing his personal identity and on the impersonal treatment of disease and personal care of the patient are paraphrased. Hippocratic *Precepts*, chapter VI, in W. H. S. Jones's Loeb translation (1923): "some patients, though conscious that their condition is perilous, recover their health simply through their contentment with the goodness of the physician," "where there is love of man, there is also love of the art," and the chapter's advice on fees, quoted and paraphrased from the public-domain translation (*Hippocrates*, volume I, Loeb Classical Library, 1923); its late date: Jones's introduction to the treatise in the same volume; its non-Hippocratic authorship: standard reference knowledge. Shah Hussain's kafi beginning *Māī nī main kinūṅ ākhāṅ, dard vichhoṛe dā hāl*: opening line given in transliteration with the author's own rendering; Shah Hussain as a sixteenth-century Sufi poet of Lahore: standard reference knowledge. Miranda Fricker, *Epistemic Injustice: Power and the Ethics of Knowing* (2007), and her concept of testimonial injustice, described without quotation; the literature on the underestimation of pain in some groups of patients, stated generally: standard reference knowledge. Elizabeth Newton's 1990 Stanford doctoral study of "tappers" and "listeners" (tappers predicting about half of the songs would be recognized; listeners identifying about one in forty), and the term "curse of knowledge" from Camerer, Loewenstein and Weber (1989): standard reference knowledge. Council of Europe, Convention on Human Rights and Biomedicine (Oviedo, opened for signature 1997), Article 10(2), on the right to know and the wish not to be informed; UNESCO, Universal Declaration on the Human Genome and Human Rights (1997), Article 5(c), on the right to decide whether or not to be informed of the results of genetic examination: standard reference knowledge, paraphrased. That physicians facing serious illness sometimes choose differently for themselves from what they offer others: a much-discussed observation, stated generally. The meanings of ਭਰੋਸਾ (*bharosā*) and ਭਰੋਸਾ ਦੇਣਾ: standard Punjabi usage. The lay misreading of *negative*, *positive*, *chronic* and *acute*; the *ask–tell–ask* sequence as a commonly taught communication technique; that antibiotics do not treat viral colds: standard reference knowledge of clinical practice and medical education. **Vendor representations.** None.

**Recollection and tradition.** No real patient, encounter or conversation is described, and no remembered scene is claimed. The clinician explaining a new diagnosis to a patient worried about work, and the man with a cold who asks for antibiotics, are hypothetical and illustrative, as are the example requests and the example sentences, which illustrate the author's present-day views and are not quotations from any visit. The varied preferences described are general possibilities, not accounts of particular patients.

**Analysis and forecast.** The five dimensions of conversational preference, the distinction between believing an experience and accepting an explanation of it, the remarks on vocabulary, the curse of knowledge, order, quantity and the difference between satisfaction and understanding, the treatment of the wish not to know, the reading of the Emanuels and Gawande, the reading of *bharosā* and the *Precepts*, the four-part approach to a respectful refusal, the account of repair and the reading of Peabody are the author's interpretation as of November 6, 2026. This essay is not a claim that all patients share the same priorities, and it makes no promise about any physician's availability. **Independence.** This essay is not sponsored by any institution, company, health system or technology developer. It is personal reflection, not medical advice; readers seeking care should contact their own physician's office.



## Related reading

- [What the Chart Cannot See](https://personal.kpsgill.com/essays/what-the-chart-cannot-see/)
- [Hope, Measured Carefully](https://personal.kpsgill.com/essays/hope-measured-carefully/)
- [The Older I Get, the More I Listen](https://personal.kpsgill.com/essays/the-older-i-get-the-more-i-listen/)