Skip to Main Content

Resources for Visiting Students

Presentation Recommendations

The Chief Concern

Chief Concerns (CCs) are like newspaper ledes: we use them to spark interest, frame presentations, and highlight the patient’s core problem. Effective CCs share essential facts while staying concise.

“Chief Concern” is a Yale-ism. Several years ago, we stopped using "Chief Complaint" so patients wouldn’t think we were calling them complainers. The purpose is the same and, miraculously, the initials are the same too.

CCs should be expressed in full sentences that incorporate the patient's name, age, gender (man/woman, not male/female), relevant social, ethnic, and medical background, and a main problem. For example:

  • Ms. Washington is a 34-year-old homeless woman with a history of Type I diabetes, presenting with three days of polyuria and polydipsia after running out of insulin.
  • Mr. Adams is a 67-year-old man with a history of poorly-controlled hypertension and hyperlipidemia, presenting with two hours of severe substernal chest pain.
  • Ms. Jefferson is a 29-year-old marathon runner, presenting to the ER after losing consciousness at the New Haven Road Race.

I think you’ll agree these CCs grab your interest, highlight the patient’s core issue, provide context, and frame the presentation to come.

CCs fall off the tracks when they include too little or too much information. Some of you probably learned to limit CCs to one word, like “cough,” “headache,” or “dysuria,” but this approach lacks context. Consider these two patient with cough:

  • Mr. Madison is a 20-year-old man with a history of asthma, presenting with three days of non-productive cough, wheeze, and shortness of breath.
  • Mr. Hancock is a 74-year-old man with severe COPD, presenting with three months of cough, productive of bloody sputum.

Don’t play coy with your CCs. Tell your audience what they need to know.

Too much information is also a problem, weighing down the CC with gunk:

  • Ms. Livingston is a 53-year-old woman with a history of CAD, s/p PCI 10 years ago with DES to the LAD, left breast cancer s/p lumpectomy and XRT, sickle cell trait, and IBS, presenting with a bull’s-eye rash following a tick bite.

If you’re like me, you lost focus after the PCI. The CAD, breast cancer, sickle cell trait, and IBS are irrelevant. Just say “Ms. Livingston is 53-year-old woman presenting with a bull’s-eye rash following a tick bite.” File the other details in the “Past Medical History.”

Finally, some special considerations.

The MICU uses a version of the CC called the “Critical Care Chief Concern“ (CCCC). This is another Yale-ism, dating back to my days as MICU Director. The purpose is to highlight why the patient was admitted to the ICU. Consider these examples:

  1. “Mr. Franklin is a 59-year-old man presenting with fever and cough.”
  2. “Mr. Franklin is a 59-year old man presenting with fever, cough, and hypoxemic respiratory failure.”

Version one suggests Mr. Franklin could be seen in clinic; version two signals that Mr. Franklin is critically ill.

Some of you were taught to use race in the CC. Doing so is rarely appropriate, unless your point is to convey a social determinant of health (e.g., a Latino meat packer exposed to COVID-19). On the other hand, ethnicity may be important if it implies genetic heritage as a disease risk factor (e.g., a Swedish man with hypercalcemia and hilar adenopathy).

Some of you were taught to use the patient’s words in the CC. Unfortunately, it may not be feasible or helpful to use patients’ words. Few patients can report lab abnormalities (e.g., a patient with end stage renal disease presenting with a potassium of 8.3.). Similarly, the patient’s concern may not be the concern demanding your attention. For example, a patient may come to the ER for a twisted ankle but require admission for alcohol withdrawal. Of course you will attend to the ankle, but your focus should be on the DTs. Finally, some patients can’t express a concern, for example if they’re unconscious.

It’s amazing how much attention we can devote to one sentence, but the effort is worth it. Your listeners and readers need you to convey Chief Concerns that are concise, targeted, memorable, and, most importantly, useful.


With the CC and HPI behind us, let’s tackle “the lists” today: the Medications, Past Medical History (PMH), Social History (SHx), Family History (FHx), and Review of Systems (ROS). Some considerations:

Imports: EMRs can automatically import lists into your note, which is both a convenience and a hazard. Automatic imports can save you key strokes, but they can also litter your notes with inaccurate, incomplete, and outdated information. If your patient’s “COPD” turned out to be sarcoidosis, you need to update the diagnosis. If your patient no longer takes aspirin, you need to get rid of it. If your patient’s smoking history was left out, you need to put it in. It’s your job to identify what’s true, what’s false, and what’s missing- and make the changes.

Placement: It’s up to you to decide what goes in the HPI and what goes in the lists. In general, information related to the acute problem goes in the HPI. For example, if you admit a patient with acute asthma, you should describe their multiple intubations and chronic steroid treatment in the HPI. In contrast, if the same patient presents with DKA, you should place their asthma in the PMH.

Details: Without going overboard, share details. What caused the CKD? How extensive is the RA? How much does the patient drink?

What You Write and What You Say: These lists can get long, so you will generally say less than you write. You don’t need to discuss your 99-year-old’s family history on rounds, even if you included it in your note.* Just say, “The Family History is non-contributory.” Limit what you say out loud to issues needing attention. If your audience has questions, they can ask.

For the individual sections:

Meds: Confirm what the patient is truly taking. Inaccurate med lists are a major source of medical error, so attend to the details, including doses. Include OTC meds and supplements.

PMH: Don’t ask me why it’s called the “past” medical history. Are there present and future histories too? I guess “past” is meant to distinguish prior illnesses from the “present” one described in the HPI. The PMH is a common source of chart lore, and I’ve personally cured many patients of chronic diseases they never actually had by hitting the delete button.** The PMH often contains duplicate information, like “CHF” and “HFrEF.” In general, keep the more descriptive diagnoses (HFrEF) and delete the generic ones (CHF).

Social History: The SHx is your chance to describe your patient as a full person.*** One day in the MICU, we discovered two famous musicians on our service, because we asked about their jobs. One day on Fitkin, we met a lineman from the Green Bay Packers. Another day, we met a construction worker who helped build Smilow. There’s so much to learn: Where was your patient born? Where have they lived? Where have they traveled? What are their hobbies? Did they serve in the military and, if so, did they see action? You can learn about exposures- at home, at work, at play, from pets, from sexual partners, etc. Historically, we’ve placed tobacco, alcohol, and recreational drug use in the SHx, but it really belongs in the PMH.

Family History: The FHx is especially important when you’re considering inheritable diseases, like early onset CAD and familial cancers. You should always ask about biological parents and siblings. If you’re hunting for rare genetic diseases, extend the pedigree with grandparents, aunt and uncles, cousins, and biological children.

Review of Systems: It bears repeating that the ROS is distinct from pertinent positives and negatives, which go in the HPI. Consider the ROS your chance to uncover problems distinct from the main concern- for example, discovering that your patient with cellulitis has lost weight, or that your patient with Lyme Disease found a lump in her breast. You can use the ROS to ask about vaccinations and age-appropriate cancer screening, like mammograms and Pap smears. Your comprehensive review can save a life.

Last Words: Long, chaotic, inaccurate lists are useless: they overwhelm your audience and bury crucial information. Your mission is to create lists that are streamlined, accurate, and thoughtfully detailed. Your patients and teammates are counting on you.

Physical Exam

Physical exams save lives. As a student, I found a suboccipital lymph node in a patient with an FUO; she had miliary TB. As a resident, I heard a diastolic murmur in a college student with a sore throat; he had aortic insufficiency. In the Yale MICU, I diagnosed pseudomonas bacteremia in a patient with ecthyma gangrenosum, and in clinic, I diagnosed small cell cancer in a patient with a hard supraclavicular lymph node.

My exam skills aren’t extraordinary, but I do my best to pay attention, and I think I’ve helped patients along the way. Here are a dozen suggestions to consider as you document your own exams:

  1. Provide a general description: Is the patient comfortable? In distress? Conversant? Calm? Agitated? This is how we distinguish sick from not sick, and your documentation will prove crucial when an RRT is called or the night shift cross-covers.
  2. Be thorough: H&Ps should include comprehensive exams. Wouldn’t it be tragic to miss a breast mass or melanoma, regardless of the patient’s reason for admission? With time, you’ll surely make surprising and life-changing diagnoses.
  3. Follow a template: Templates foster efficiency and completeness. If you’re not using a template yet, get one, then add and subtract as necessary.
  4. Highlight abnormalities: Use bold font to signify key findings. Readers will thank you.
  5. Elaborate as needed: Share details. If your patient has abdominal pain, tell us about distention, bowel sounds, tenderness (including severity and location), and whether she had guarding or rebound. Thorough exams can distinguish between appendicitis, bowel obstruction, mesenteric ischemia, kidney stones, and a host of other diagnoses.
  6. Report pertinent negatives: Anticipate what your audience will want to know, even if the exam is non-revealing. If your patient has nausea but his abdominal exam was benign, tell us. If your patient has a headache but no neck stiffness, say it. If your patient has a foot ulcer, but the pulses and sensory exam were normal, we need to know. If you don’t document normal findings, your audience will assume you didn’t check.
  7. Take pictures: Especially for rashes, there’s no substitute.
  8. Report the vitals: It’s a common error to report 24-hour ranges in the exam section—"the heart rate was 52 to 152”—but this is confusing. Just tell us what the vitals were when you saw the patient; 24-hour ranges belong in the history, not the exam. Also, provide context: if the patient was on pressors when the blood pressure was checked, or if the patient was on oxygen when the SpO2 was checked, we need to know.
  9. Include devices: Tell us if the patient has lines and tubes- and if the entry sites were clean and dry (or not).
  10. Establish a baseline: Follow up exams will be compared to your baseline. If your patient has CHF, how extensive were the crackles and how severe was the edema? If your patient has asthma, how many words could she say before pausing to inhale? When documenting follow up, refer to prior observations (e.g., “she was breathing more comfortably than the day before”).
  11. Note the timing: Your audience will assume your exam coincides with the rest of the note. If that’s not true, make it clear. It makes no sense to say the patient was intubated in the history and then say they were on high flow nasal cannula on your exam. Similarly, if the patient’s lungs were clear after getting albuterol, tell us.
  12. Never copy and paste: Exams that are copied forward always make me wonder if the patient was actually examined. Copied exams can also cause medical-legal trouble: “Doctor, on the day Mr. Smith perforated his appendix, you wrote that his exam was ‘benign,’ which is exactly what you said the day before and the day before that. Did you actually examine the patient?” Equally important, don’t ever, ever, ever copy someone else’s exam. The only exam you should document is your own- a subspecialty fellow once copied my joint exam on a patient with gout and I’m still shaking my head. Document a fresh exam every day.

Premier internists distinguish themselves by performing—and documenting—thoughtful, thorough exams. In doing so, you will pick up subtle findings, make elusive diagnoses, and show patients that you care. Nothing is more important.


  1. Summarize. Start every assessment by highlighting salient findings from the patient’s history, exam, and test results. Be brief: the purpose of the summary is to focus attention. Don’t regurgitate raw data like “the creatinine is 4.7.” Instead, interpret- “the patient has AKI.” Here’s a concise summary: “Mr. Smith is a middle-aged man with multiple cardiac risk factors, presenting with three hours of substernal chest pain, dyspnea, mild hypoxemia, bibasilar crackles, a mildly elevated troponin, non-specific ST-T wave changes, and a chest x-ray with bilateral infiltrates.”
  2. Name the problem(s): Identify a discrete issue or issues that you’ll use to generate a differential diagnosis. It’s generally easiest to choose a single issue like chest pain or hypoxemia because single issues lend themselves to straightforward illness scripts. But beware: choosing single issues can lead to lists that are long, unwieldy, and impractical (how many causes of chest pain can you name?). Sometimes it’s better to combine problems like chest pain and hypoxemia. Doing so narrows your focus, because you’ll just have to entertain diagnoses that explain the combined problem. For example, costochondritis causes chest pain but not hypoxemia, so it’s not a consideration, whereas PE is a consideration, because it can cause both. The downside to composite problems is that they may not lead to obvious illness scripts. For example, if you have a patient with a rash and diarrhea, you may want to tackle each problem separately, rather than combine them prematurely. Once you’ve created two separate differential diagnoses, you can see if they overlap. Finally, if you create a composite problem, be sure the multiple issues stem from a single diagnosis so you don’t miss diagnoses occurring in parallel. For example, a patient could have chest pain from one problem, myocardial ischemia, and hypoxemia from something else, like COPD.
  3. Generate a differential diagnosis. Create a list including all possible diagnoses that merit attention- either because they’re likely or because they’re so dangerous that you don’t want to miss them. Use your summary statement to create this list. In the case of Mr. Smith above, you might consider MI, pneumonia, and pulmonary embolism, while also mentioning a rare, “can’t miss” diagnosis like aortic dissection. Don’t bother with remote possibilities, like Zoster without a rash, or diagnoses that are self-evidently ruled out, like pneumothorax not seen on imaging.
  4. Assess probabilities. Show your clinical reasoning, citing your evidence. Again, consider Mr. Smith: “Given the patient’s cardiac risk factors, the location and quality of his pain, elevated troponin, and EKG abnormalities, I’m most concerned about NSTEMI, complicated by CHF. Pneumonia is less likely without fever or sputum. PE is also less likely given the infiltrates and presence of a more compelling explanation for his symptoms. Although it’s always worth considering aortic dissection, I am reassured by the quality and location of his chest pain, which is neither tearing nor radiating to the back, and the absence of mediastinal widening on CXR.”
  5. Commit yourself. Finish by naming the diagnosis or diagnoses that warrant further attention. If you’re convinced you’ve made the diagnosis, say it: “I believe Mr. Smith has an NSTEMI complicated by CHF.” If you can’t rule out other diagnoses yet, then that’s what you should say: “I believe Mr. Smith has an NSTEMI, but I cannot entirely rule out PE, pending further evaluation.”