Pertinent Negatives in History Taking, With Three Worked Cases

You present on rounds and list "no fever, no weight loss, no night sweats" by reflex. The senior asks which of those changed anything.

This post's about choosing negatives on purpose, tied to your differential. Building the list lives in turning symptoms into a differential. Here, we practice the history side with three cases where one plain "no" separates lookalikes.

What is a pertinent negative?

A pertinent negative is a symptom or history feature the patient does not have, that you asked about because its presence would have pushed a specific diagnosis on your differential up the list. Its absence lowers that diagnosis's probability; it does not remove it.

A routine negative is any "no" you collect out of habit. In pertinent negatives history taking, if it doesn't affect a named diagnosis you're considering, it's noise. You pick them from your differential, not from a template.

That is the teaching in a Loyola teaching worksheet that has you write the chief complaint, list three diagnostic hypotheses, then add positives and negatives. The hypotheses come first, which is the whole point.

In its chest pain example, the negatives are "No history of trauma to chest", "No history of fever and productive cough", and "No past history of similar symptoms". Each is only pertinent if the paired hypothesis sits on your list for this patient.

Why examiners and seniors care about them

On the wards, seniors listen for whether the negatives you include track a real differential, not a checklist. As one example, UNC's student milestones put this at the top "Aspirational" level, phrased as "Targeted and appropriate history, including pertinent positives and negatives."

Students usually ask the questions, then under-report them. In a 2012 Family Medicine study of 55 third-year students across two standardized-patient cases, what hit the note lagged what happened in the room.

• 87% of both positive and negative items were obtained in the history.
• 75% of positive items made it into the written note.
• 52% of negative items made it into the written note.

Which means the asking is fine, and the gap is noticing which no matters enough to write down and say aloud. One school, 55 students, so take it as a signal, not a law.

How to choose pertinent negatives from your differential

Pertinent negatives come from your differential, not from a template. Use them to separate lookalikes while you're reasoning with an incomplete story.

1. Name your top three diagnoses after the opening complaint, out loud or on paper. Commit early so you can test them.
2. For each, ask what this one would give the patient that the other two would not. That feature is your discriminating question.
3. Ask it directly and specifically. Patients rarely volunteer an absence or a time course unprompted.
4. Keep the no answers that moved a named diagnosis down. Drop the ones that did not change your ranking.
5. Reorder, then repeat with the new top three.

A can't miss diagnosis earns its negatives even if unlikely, because the cost of missing it is high.

Worked case 1: sudden breathlessness with chest pain

Martin Rowe, 55, usually fit, runs a joinery firm. He opens with, "I've suddenly gone short of breath with chest pain over the last two days," and he is worried he is having a heart attack.

Unprompted you get this: sharp pain on the right, worse with a deep breath, and breathlessness that began at the same time, both starting two days ago. No other details until you start asking targeted questions.

Top three on this history are pulmonary embolism, acute coronary syndrome, and community-acquired pneumonia. A simple scaffold helps you keep the questions tight, see structuring a chest pain history.

• Acute coronary syndrome. The question: is it a heavy, crushing pressure in the middle of the chest, going to the arm or jaw, or the exertional tightness he knows as heart pain. The answer: no, when asked. What it does: makes ACS less likely in this case.
• Community-acquired pneumonia. The question: any fever, shivers, or green productive sputum over the two days. The answer: no, when asked. What it does: lowers the probability of pneumonia here.

You still ask the "can't miss" negative, even if PE is leading: any blackout, faint, or feeling about to pass out. He says no. It helps you gauge immediate risk and separate collapse-type presentations on this case's own list.

Then you uncover the positives that move PE up, none volunteered: he coughed blood twice, his left calf has been swollen, tight and tender for a week, he flew long-haul from Australia five days ago, and he had keyhole knee surgery three weeks back.

The case answer is pulmonary embolism. The negatives did not prove it. A negative lowers probability, it never rules a diagnosis out. The negatives made ACS and pneumonia less likely while the positives, once asked for, pushed PE up.

Try pertinent negatives Take a history from an AI patient who offers negatives only when you ask, then check your reasoning. Start a case

Worked case 2: back pain with weak legs

Neil Ashworth, 45, a van driver and fitter for a flooring firm who smokes, opens with, "I hurt my back a few weeks ago and now both my legs have gone weak." The pain began three weeks ago after lifting a roll of carpet at work.

He volunteers that the pain used to shoot down one leg and now shoots down both, with both legs weak and heavy, worse over the last two days.

He thinks he has aggravated a slipped disc, wants stronger painkillers, and to be signed off work. He readily blames his bladder problem on the painkillers.

The case differential runs like this: mechanical back pain, sciatica, spinal cord compression, spinal tuberculosis, peripheral neuropathy, and cauda equina syndrome.

• No fever, night sweats or weight loss. These lower the probability of an infective cause such as spinal tuberculosis and other non disc compressive causes, but they are not proof of safety. A negative lowers probability, it never rules a diagnosis out.
• No cancer history, no injected drugs, no spinal injection. These also lower the probability of an infective or other non disc compressive cause, and nothing more. A negative never rules a diagnosis out.

The findings that changed the case were positives he never volunteered. Only when asked did he describe numbness between his legs and around his back passage, two days unable to pass water properly with dribbling, no feeling of bladder filling, no sense of when he needs to open his bowels, and new erection difficulty.

Specific questions find both the positives and the negatives, and stop tidy patient theories from becoming false reassurance. Miss them, and you are back reading about conditions that slip past overnight.

Case answer: cauda equina syndrome from a prolapsed intervertebral disc.

Worked case 3: thunderclap headache and the meningitis question

Andrew, 50, says, "I've suddenly got the worst headache of my life and I've been sick." It hit at the back of his head, reached maximum within seconds while lifting a heavy box, and he has high blood pressure he doesn't always treat, smokes, and has an aunt with a brain aneurysm.

Meningitis sits on the differential because, when asked, his neck feels stiff and sore and bright light hurts his eyes. Here the negatives you call out matter, but fever is something you shouldn't over-trust.

• No fever. He denies fever, which lowers the probability of meningitis, but doesn't exclude it.
• No head injury. When asked, he reports no head injury or trauma and says it came out of the blue, which moves traumatic causes down.
• Never had migraines, few headaches normally. With no migraine history and few headaches, migraine becomes a less comfortable explanation for this event.
• No limb weakness or speech disturbance at present. Helpful to document, but this does not reassure you about a bleed. Absent focal deficits do not exclude one.

Only when pressed he adds he briefly blacked out right as the pain struck, had a similar but milder sudden headache about a week ago, and keeps becoming drowsy and losing the thread. Those positives pull away from meningitis and migraine and toward a vascular cause launched by exertion.

Case answer: subarachnoid hemorrhage.

One caveat. Our case patient is a man, but subarachnoid hemorrhage is more common in women, with incidence about 1.24 times higher in women per a 2007 systematic review, a gap that starts near 55 and widens after. The reasoning about negatives is the same whoever is in front of you.

How do you present pertinent negatives without a laundry list?

Present pertinent negatives by tying each one to a diagnosis on your differential. Say only those that changed your thinking. Put them immediately after the pertinent positives in your summary, grouped by diagnosis rather than by organ system, so your listener follows the argument.

Here is how that sounds in practice with the negatives you say out loud. Stack your positives first, then add the targeted negatives in the same diagnostic groups, which mirrors how handovers flow in the full ward-round presentation.

Name the diagnosis each negative is pushing down, and skip negatives without a target. Negatives that never moved your differential can stay in the notes, not in your spoken summary.

• Before. No fever, no shivers, no green phlegm, no blackout or near-faint, chest pain not central, not crushing, not radiating to the arm or jaw...
• After. ACS less likely, because the pain is not central, crushing, or radiating to the arm or jaw. Pneumonia less likely, because there is no fever, shivers, or green phlegm. No blackout or near-faint, reported because it bears on how unwell he is.

Common mistakes with pertinent negatives

• Negatives you never asked about. Don't write what you didn't ask. Writing "no X" when you never asked is a fabricated history, it erodes trust, and it reads like wishful thinking rather than a record of what actually happened. If you didn't ask, say you didn't, then go back and ask.
• Negatives that change nothing. A "no" that sits against nothing on your list is clutter, it hides the ones that matter in a long bullet dump. Pick your top three working diagnoses first, then choose negatives that meaningfully move those up or down in probability. If you can't tie the "no" to a named option, leave it out for now.
• Treating a negative as proof. A negative lowers probability, it doesn't settle it, so treat it as one data point among many. Case 2's man had no fever, night sweats or weight loss and still had a compressed spinal canal from his disc. Case 3's man had no fever and no limb weakness and still had a bleed.

The study above found students asked the right questions, then left the pertinent negatives out of their notes.

Practice it on a patient who only says no when asked

You build this by asking, not reading. In Diagnosica, you take a history from an AI patient by voice or text, a video avatar on screen, and negatives only appear when you ask the exact question you planned to test.

In the three cases in this post, none of the key negatives are volunteered. You ask, they answer, then you order investigations, commit to a diagnosis and a plan, and you get a scorecard with teaching points after every case, free tier included.

Across the library, 88 of the 136 cases carry at least one key finding marked as a negative, so the habit gets exercised repeatedly, and every case is written and signed off by a doctor.

There's no physical examination practice here, you can't elicit signs, and "It doesn't give advice about real patients, and it's not a diagnostic system. Treat every output as educational."

If you want a feel for the flow, see how an AI patient case runs. Before your next case, write your top three and one discriminating question for each, then ask them.

Educational use only. Not medical advice. AI-generated; verify clinically against primary sources.

Clinical review pending.

Work the differential Run a case free, write your top three first, then ask the questions that split them cleanly. Start a case free