The Art of the History: Questions That Change the Diagnosis

If you are a student or new graduate, you have probably spent hours practising orthopaedic special tests and far less time practising how to ask a question. That ratio is backwards.

The classic study on this is now fifty years old. Hampton and colleagues followed medical outpatients through a full workup and found that the history alone produced the correct diagnosis in about three quarters of cases. The physical exam and investigations mostly confirmed what the clinician already suspected.¹ Later work replicated the finding.² Musculoskeletal practice is no different. By the time a patient with knee pain stands up from the chair, a good clinician usually has two or three working hypotheses and a plan for which tests will separate them.

This post is about the questions that most often change the picture, and about the habits that stop clinicians from hearing the answers.

The habit that matters most: let them finish

In a frequently cited study of primary care visits, physicians interrupted the patient's opening statement after an average of about 18 seconds. Once interrupted, patients rarely returned to complete their account.³ A follow-up study fifteen years later found the number had barely moved.⁴

The opening statement is where the patient tells you what they are actually worried about, in the order they are worried about it. Cutting it short to ask "where exactly does it hurt" trades their agenda for yours before you know what theirs is.

The fix is simple to describe and hard to do: ask an open question, then stay quiet until they stop. Then ask "what else?" and stay quiet again. Most people are done inside two minutes, and those are the most valuable minutes in the visit.

Questions that change the diagnosis

"What changed in the two or three weeks before this started?" This is the single most useful question in a sports and rehab setting. Most non-traumatic musculoskeletal pain is a load problem: a new job, a new shoe, a return to running, a different position on the team, a stretch of poor sleep. Patients rarely volunteer this because they do not see the connection. If you do not ask, you will build a treatment plan around a tissue and miss the reason it got overloaded. We touched on this logic in our post on training load and injury.

"What does it stop you doing?" Pain intensity ratings tell you less than you think. Function tells you what matters to the person and gives you a discharge criterion in the same breath. It also tells you about the problem itself: pain that stops someone sitting for twenty minutes is a different problem from pain that stops them lifting a toddler.

"Is it there at night, and if so, does changing position help?" Night pain alarms students because it appears on every red-flag list. The useful follow-up is whether it eases with a change of position. Mechanical pain usually does; pain from serious pathology often does not. Red-flag questions are weak individually and much stronger in combination, so the skill is building the pattern rather than reacting to a single answer.⁵

"Where does it go?" Not "does it radiate" (patients will say no when it does) but where does it go, with a hand gesture if needed. The difference between pain that stops at the buttock, pain that goes to the knee, and pain that reaches the foot is the difference between three quite different problems.

"What do you think is going on?" This tells you about the patient's beliefs, and beliefs drive behaviour. Someone who thinks their disc has "slipped out" will move very differently from someone who thinks they strained a muscle. You will need to address the belief whether or not you agree with it, and you cannot address what you have not heard.

"What have you already tried, and what happened?" This saves you repeating what has failed, and the pattern of responses is diagnostic in itself. A problem that improves with movement and worsens with rest is telling you something. So is one that has not responded to three months of passive treatment.

"Is there anything else you were hoping to talk about today?" Ask this before the exam, not at the door. The "by the way" at the end of a visit is often where the important thing was hiding.

Listening for the things that do not fit

A good history is a running comparison between the story you are hearing and the story you expect for each hypothesis. When something does not fit (a "simple" ankle sprain still swollen at six weeks, back pain with no trigger in someone with a history of cancer, shoulder pain in a patient who mentions they have been short of breath), that is the moment to slow down rather than smooth it over. The things that do not fit are what the exam is for.

It helps to say your reasoning out loud, briefly. "From what you've described, I'm thinking about two possibilities. The exam will help me tell them apart." This keeps the patient involved and forces you to commit to a hypothesis you can test, rather than examining everything and hoping something lights up.

For students and new graduates

If you are early in your career, write down your leading diagnosis after the history and before the exam, then see how often the exam changes it. For most clinicians the answer is: not often. That is not a reason to skip the exam. It is a reason to put more effort into getting the history right.

In our clinic, students on rotation observe histories before they observe treatment, for exactly this reason. Our services page describes the assessment process we teach and use, and our post on developing clinical reasoning picks up where this one leaves off.

The special tests will come. The habit of listening is harder to build later.

References

1. Hampton JR, Harrison MJ, Mitchell JR, Prichard JS, Seymour C. Relative contributions of history-taking, physical examination, and laboratory investigation to diagnosis and management of medical outpatients. Br Med J. 1975;2(5969):486-489.
2. Peterson MC, Holbrook JH, Von Hales D, Smith NL, Staker LV. Contributions of the history, physical examination, and laboratory investigation in making medical diagnoses. West J Med. 1992;156(2):163-165.
3. Beckman HB, Frankel RM. The effect of physician behavior on the collection of data. Ann Intern Med. 1984;101(5):692-696.
4. Marvel MK, Epstein RM, Flowers K, Beckman HB. Soliciting the patient's agenda: have we improved? JAMA. 1999;281(3):283-287.
5. Finucane LM, Downie A, Mercer C, et al. International framework for red flags for potential serious spinal pathologies. J Orthop Sports Phys Ther. 2020;50(7):350-372.