Blog by Sumana Harihareswara, Changeset founder

19 Dec 2023, 11:05 a.m.

Nailing Down An Answer, Or At Least A Question

Someone in a group chat mentioned a particular annoyance in dealing with doctors, nurses, and similar medical experts. Sometimes you take pains to ask them a specific, deliberately-worded question, and they don't really seem to listen. They answer a question different from the one you asked, maybe a more common one. They recommend xyz, not noticing that you said that won't work for you. And so on. It's maddening! How can we avoid it or make it less maddening?

I've dealt with a lot of medical folks recently. Here (edited) is what I said about why this problem happens and how I deal with it, mostly in the US. Some of this also applies to other kinds of experts. Hope it helps.

Why they don't seem to listen

As the questioner mentioned, analogous situations happen in politics: journalists ask politicians questions, and politicians duck, dissemble, and keep "on message" at the expense of answering that particular question. Customer service representatives do stuff like this too.... it's common! In public-facing jobs where professionals get a lot of (often repetitive) questions from the public, we can notice some patterns in how those professionals respond to that stimulus.

Low knowledge: Sometimes the respondent is just ignorant/unskilled enough that they genuinely cannot discern the difference between the question asked and the question they are used to!

Low attention: Some of us are basically only paying partial attention to the specific words of the question and pattern-match on the general topic, because we are distracted, mentally multitasking, kind of phoning it in, or something.

I have come to believe that a big proportion of people in everyday life are doing what we perceive to be routine things while paying way less attention than they would need to reliably notice that stimuli are meaningfully different than we expect. And that this is normal, probably kind of inherent in human cognition or something. A less charitable way of putting this: a lot of people are zoning out a lot of the time, and that's unlikely to change soon.

Medical providers these days especially have a hard time paying full attention to conversation with a patient, because they're often simultaneously fighting/using a terrible, mandatory electronic medical records system that takes twenty clicks to do the most common things. And they're rushed by administrators and billing incentives that want them to churn through patient visits quickly.

So, sometimes, I need to flag for someone else "hey, this isn't what you expect, please pay full attention" (pre-emptively, or when I perceive that they may be doing the cognitive-miser pattern-matching thing). Later in this post, I'll describe a few ways I do that.

Optimizing for the common case: Also! Medical professionals try to have ready-to-mind short speeches that educate patients about topics most of their patients don't know and need to know. And, in my opinion, that's a genuinely good thing! It's great that they have a desire to explain the basics clearly and reliably, and furthermore, to ensure the median patient doesn't get confused with a level of complexity they can't yet comprehend.

Sure, they could always deliver this sort of speech along with a lot of diagnostic "do you already know what X is?" checks. But: having to say "no I don't" is humiliating for a lot of people, so they will feel bad, and that bad feeling will get in the way of them listening or learning. And many patients will lie and say "yes" to avoid the humiliation, and then they won't learn it at all.

So that's an unfortunate structural reason a busy medical provider has an understandable incentive to launch into the basics speech without those diagnostic checks. They're optimizing for the common case, and thus aren't being as careful to avoid annoying edge cases like me. And this pattern happens in combination with the "only paying partial attention" pattern, the "too ignorant to notice the differences between questions" pattern, etc.

Status concerns: Sometimes we are averse to the humiliation of saying "I don't know," or the vulnerability of visibly pausing and thinking and demonstrating that our answer was not already ready. (A lot of this got clearer for me when I read this list of behaviors that raise or lower one's relative status, from a wiki for improv performers.) And so we kind of deploy a pre-prepared speech on an adjacent topic.

Or, we are averse to letting other people know what we think about the specific question, and we have an agenda to push that we prioritize above having a respectful and collaborative interaction with this questioner. (I think this one is usually subconscious in, like, healthcare and customer service people, but I bet they explicitly teach it to future politicians in those fancy British boarding schools.)

But even if a questioner doesn't humiliate an expert or derail an agenda with a specific question, the expert's status sensitivity might still get in the way of general rapport:

Along the way: acting deferential enough

Speaking as a patient and a patient advocate: while I'm dealing with providers not listening or responding to my specific question, I usually don't point that out directly. And a big reason is because they are often averse to me acting like an equal in the conversation at all.

Quite a lot of medical providers are averse to insufficiently deferential patient behavior. For some of them, this is just pure insecurity that gets in the way of you and your medical experts being a team together. In some cases providers are leery of patients who think they're informed because they've done a single surface-level web search (the dismissive term "Dr. Google" comes up when they talk about that bad pattern); dealing with those patients involves picking through an exhausting nest of misinformation, outrage, and basic misunderstandings of anatomy, the pharmaceutical industry, and how to read clinical research.

So, as I interact with practitioners who might be annoyed by my initiative, or who don't know whether I'm actually good at doing my own research, I mitigate those bad impressions a few ways.

  • avoiding saying things they might consider confrontational, such as explicitly pointing out "I think you didn't understand my question"
  • asking "am I getting that right?" along the way during my question
  • mentioning a specific reliable patient-oriented source I learned basics from, such as the official hospital pamphlet, or a American Society Of [This Specialty] handout
  • if I'm the patient, having a friend ask the question as my patient advocate - they're acting in a socially permissible role to protect me, so they don't seem as uppity as I would if I asked it myself
  • saying that I am asking this because [OTHER PERSON]/[MASS MEDIA THING] brought it up - this way I have not committed the sin of independent initiative! I was acted upon! I am not exercising agency, don't worry! It's far more comfortable for them if I ask a question because a cousin or a subway ad suggested it, rather than because I exercised any speck of curiosity about the body I live in.
  • performing self-deprecating femininity in gesture, word choice, vocal tone, sprinkling apologies everywhere, etc.

I don't let these get in the way of urgent concerns, but I figure they at least help make sure I don't get written off as too annoying to listen to.

Also: I have lucked out sometimes with providers who are also researchers or who often treat academics as patients; they seem used to actually noticing when patients are intellectual peers and conversing accordingly.

A specific script

In these situations, when it's one-on-one (you and the provider), what do you do when this sort of thing is happening? Do you try to redirect to your specific question? Do you wait till they're done speaking to do so, or do you interrupt as soon as it’s clear they are being too basic? What kinds of words/gestures/facial expressions do you deploy to do so?

Here's how I flag "you have misheard/not paid attention to this specific question, and please pay attention to the nuance here" during in-person medical care. I kind of counter the "might seem argumentative" bit here by being very friendly and even deferential elsewhere in the conversation, as I mentioned previously.

Me: [question]
Them: [adjacent answer]
Me: [interrupting politely, like with a hand wave, when they take a breath, etc.] Hey, so, actually I think maybe I wasn't clear.

Notice here how I'm offering them the option to save face by pretending that the error was mine. I probably don't full-on smile here, but I also don't scowl, roll my eyes, or shout.

Me [continued]: I actually already [MENTION A SPECIFIC THING I READ OR WATCHED TO LEARN THE BASICS] and so now I have a kind of more advanced question. Is that OK?

Here I try to genuinely have a collaborative tone, not an annoyed one. Notice that I'm explicitly checking in to see whether they're up for this; maybe they are busy, or want to refer me to a colleague. If not, now their attention is reset and primed for "advanced question, not routine". And now I repeat my question but with more interactivity, structuring it with checks along the way about my assumptions.

Me [continued]: So I think we have 3 options here, and I want to double check my assumptions about 2 of them. The options are x, y, and z, right? [WAIT FOR CONFIRMATION] OK, so first, about x....

During that interaction, or during their original adjacent-question answer, I may perceive that they are too unskilled to perceive and respond to the nuance. If that's the case, I find a face-saving way to end the chat and then go do my own research. I might try to ask another practitioner as well. It would be great if I could ask the provider for a referral to someone who can answer better, but in my experience, if I cannot redirect them and prompt them successfully to notice the difference between my question and the one they know how to answer, they also are unlikely to think usefully about who, or what reliable sources, could help me further.

I have had pretty good success with this approach in the US and with experts from Europe and Australia. My social skills in India are not up to the task, and if you have advice on how to deal with this sort of thing in India, please tell me because I consistently find out after the fact that I've inadvertently irritated medical folks.

Email: I think this approach adapts reasonably well to live textual chat. I sometimes run into this problem in email interactions with experts, and I've had good luck replying with something like:

Thanks for your answer. I do want to better understand some specifics, though. I have three questions:



Each of A, B, and C ought to be a single paragraph ending in a question mark. And it's often worth it to think hard about what specific question I really want them to answer. Like: do I really want to know "medicine x is listed as a contraindication for treatment y, so which should we choose?" or do I want to know "given that medicine x is a contraindication for treatment y, do you think it's worth the risk to do both?"

This gives the respondent a scaffold or template to help them respond to each sub-question, instead of missing one while answering another.

Liberate yourself from attachment

There is of course a meta-problem undergirding the patient's frustration here: simply the expectation that practitioners will pay full attention to the patient's actual words and thoughtfully customize a response, and that to neglect that responsibility is aberrant behavior.

I think I partly avoid this frustration by being less attached to that expectation.

That is a fraught thing to say and touches on neurodiversity, fairness, justice, anxiety, etc. but it is true for me.

And, as my tongue-in-cheek reference to the third of Buddha's Four Noble Truths implies, this is somewhat more widely applicable as well.


19 Dec 2023, 12:04 p.m.

In any public facing job where you are interacting with a wide range of people on a narrow set of topics, you get the same questions multiple times, and I think it is very hard (or at least, I found it very hard) to remember this is coming from a different person every time, who may have different reasons for asking the question, rather than one generic and very stupid person.