Mental shortcuts in clinical judgment

Transcript

RyanA hospital consultation psychologist has had a long week. Three patients in a row came to her confused, and all three turned out to have delirium. Then a fourth patient arrives, mildly confused.

SoheeAnd she's thinking, delirium again.

RyanShe's so sure that she skips the review of other causes her service normally requires. Nothing about this patient told her it was delirium. Her last three patients did.

SoheeWelcome to EPPP Prep, the podcast. I'm Sohee.

RyanI'm Ryan. Today we're talking about the mental shortcuts that can quietly steer a clinical judgment. The knowledge statement here is about factors that influence evidence-based interpretation of data, including base rates, group differences, cultural biases and heuristics.

SoheeLet's start with what a heuristic is, because it isn't an insult. The OpenStax psychology textbook calls it a general problem-solving framework. You can think of heuristics as mental shortcuts. A rule of thumb is an example.

RyanAnd the textbook gives them their due. A shortcut saves time and energy when making a decision. The catch is that, despite those time-saving qualities, it's not always the best method for making a rational decision.

SoheeHere's the part I find most useful for clinicians. OpenStax lists five conditions that make us reach for a heuristic. When we're faced with too much information. When time to decide is limited. When the decision seems unimportant. When there's very little information to go on. And when an appropriate heuristic happens to come to mind at that moment.

RyanRead that list again and picture a busy intake day. Too much information, too little time. That's practically a recipe.

SoheeSo the point isn't that good clinicians never use shortcuts. It's knowing which ones tend to mislead, and building habits that catch them.

RyanLet's name them. Our opening story is the availability heuristic. OpenStax defines it as making a decision based on an example, information, or recent experience that is readily available to you, even though it may not be the best example to inform your decision.

SoheeThe tell is a recent or vivid case that then shapes judgments about a different person. Three delirium patients in a row. A dramatic case in the news. A gripping workshop.

RyanThere's a forensic version in our practice set. An evaluator recently testified in a widely publicized case in which a defendant convincingly faked psychotic symptoms. Over the next month, she begins to suspect faking in most defendants who report hallucinations, even though her test results for them don't suggest it.

SoheeAnd there's the giveaway: her own test results point the other way. One vivid, recent case is outvoting the data in front of her.

RyanNext, confirmation bias, the tendency to focus on information that confirms your existing beliefs. The textbook's example is a student who thinks a professor isn't very nice, and so notices every instance of rude behavior while ignoring the countless pleasant interactions.

SoheeThe tell for confirmation bias is a belief held before the evidence arrives, followed by one-sided attention. Fitting details get collected. Contrary details get brushed aside.

RyanAnchoring bias is focusing on one piece of information when making a decision or solving a problem. OpenStax uses a realtor who shows four friends only run-down houses at their budget, then a lovely house above it. They're so focused on the dollar figure they don't see what's actually available.

SoheeIn clinical work, an anchor can be as small as the first line of a referral note. If it names a diagnosis, every later detail can get read through that one label.

SoheeHindsight bias leads you to believe that an event you experienced was predictable, even though it really wasn't. You knew it all along, after the fact.

RyanPicture someone reviewing a chart after a crisis and saying the signs were all there. Some of them may have been. Hindsight bias makes them look much clearer than they were to the person in the room at the time.

RyanAnd the one I want to linger on, because it connects to cultural bias. Representative bias is making a judgment based on how similar a person seems to your existing stereotypes and prior beliefs.

SoheeThe textbook's example is striking. An urgent care nurse may be more likely to efficiently detect a heart attack in a male patient than in a female patient, because of an unintentional stereotype that heart attacks happen more often to males.

RyanNow think about assessment. If a clinician's picture of who has a given disorder is built on one group, people who don't fit that picture can get missed, or mislabeled. That's a judgment problem, not a test problem, and a more precise test won't fix it by itself.

SoheeOpenStax also offers a line that ties all these biases together. They tend to preserve that which is already established, to maintain our preexisting knowledge, beliefs, attitudes and hypotheses.

RyanWhich is exactly what a first impression in an intake interview can become, if we let it.

SoheeOkay, the exam angle. Practice questions on these biases usually offer you all of them as options, so you need a clean way to tell them apart.

RyanHere's ours. Is something recent or memorable shaping a judgment about someone new? Availability. Is there a belief the person already held, and they're defending it with one-sided evidence? Confirmation. Is the decision hanging on a single detail of the current case? Anchoring. Are they looking back after an outcome and calling it obvious? Hindsight.

SoheeLet's pause and think. Right after attending a gripping workshop on childhood bipolar disorder, a school psychologist starts to think of that diagnosis first for most students referred for irritability. She rarely considered it before, and the referrals themselves haven't changed. What's going on?

RyanAvailability. The recent, memorable workshop makes one diagnosis come to mind first. The referrals didn't change. Her recall did.

SoheeAnd why isn't it confirmation bias? Because there was no long-held belief she was protecting. She didn't think this way before the workshop.

RyanSecond pause and think. A psychologist has long believed that most adults with anxiety have hidden childhood trauma. With a new anxious client, he asks many questions about early adversity, writes down every vague hint of it, and treats the client's detailed descriptions of a stable, supportive childhood as defensiveness.

SoheeConfirmation bias. A long-held belief, then attention to everything that supports it, and contrary evidence explained away.

RyanAnd anchoring is the tempting wrong answer there. But anchoring is fixing on one piece of information. He's seeing lots of information, and keeping only the pieces that fit.

SoheeSo what protects us? A few habits come straight out of the sources we've used in this series. First, clear criteria. OpenStax notes that observers can unconsciously skew what they record to fit their expectations, and recommends clear criteria for what to record and how to classify it.

RyanSecond, more than one set of eyes. Researchers compare observations of the same event by multiple observers to test inter-rater reliability. If two independent people see the same thing, it's less likely to be one person's expectation.

SoheeThird, structure. In a structured interview, everyone gets the same prepared questions, and each answer is rated on a standardized system. OpenStax notes that this lets the interviewer accurately compare candidates. The same idea protects a clinical intake from wandering wherever a first impression leads.

RyanAnd fourth, the base rate, from our last episode. Before deciding a new patient has the same thing as the last three, ask how common that condition really is in the people you see.

SoheeIn our opening story, the safeguard already existed. Her service required a review of other causes. The shortcut was skipping it.

RyanLet's recap. Heuristics are mental shortcuts. They save time and energy, but they aren't always the best route to a rational decision, and we lean on them most when we're rushed or overloaded.

SoheeAvailability: a recent or vivid example drives the judgment. Confirmation: defending a prior belief with one-sided evidence. Anchoring: fixing on one detail. Hindsight: it seemed predictable after the fact. Representative: judging by similarity to a stereotype.

RyanAnd the protections are clear criteria, independent observers, structured procedures, and base rates.

SoheeHere's a gentle thought to end on. If you've had a hard exam result, it's tempting to fall into a kind of hindsight bias about yourself, telling yourself you should have seen it coming. That story usually isn't fair, or accurate.

RyanWhat helps is what helps clinicians: look at the actual evidence. You're studying, you're showing up, and you're learning how judgment works. Take that into your next practice set. See you next time on EPPP Prep.

Loading the read-along timings…