Generating Focused, Accurate, and Relevant Differential Diagnoses
If you’ve been following the diagnostic reasoning blog, you are familiar with the concept of diagnostic reasoning, pre-test probability, diagnostic accuracy (sensitivity and specificity, negative and positive predictive value), and general principles around creating a differential diagnosis for the patient in front of you. Some presentations do not require an exhaustive differential list – for example, identifying a plantar wart – but other cases do require us to sit down and think things through. Whether we’re working in community health, urgent care, or the hospital, it’s important to organize our differential in a coherent way. This fosters communication between team members (e.g. when handing off to another provider, or communicating via documentation in a patient’s chart), when presenting a case to our preceptor as a student, or for writing up a referral to a specialist. Let’s review a common way of organizing our differentials.
Case: You’re seeing a 60-year-old female with a 3-month history of cough. Before seeing her, you have a general list of things you want to rule in and rule out. You think of your common and less common causes of chronic cough (defined as a cough lasting more than 8 weeks) (1).
An easy way to distill things as you’re going through your evaluation of a patient is separating your thoughts into 3 categories: what is the most likely diagnosis? What is the least likely diagnosis? And what needs to be ruled out (i.e. diagnoses we don’t want to miss).
Back to the case: she has a history of allergic rhinitis, which flares up during change of season and she ran out of her allergy medications (nasal spray and antihistamine). The cough is dry and seems to be worse at night and first thing in the morning. She does not have a history of asthma or COPD however she has been smoking ½ pack per day of cigarettes for the last 30 years. She denies coughing up blood, drenching night sweats, unintentional weight loss, fevers, reflux, indigestion, nor does she have any leg swelling or abdominal pain. She has not travelled anywhere recently and lives alone. She has been taking perindopril for years for high blood pressure. Her vitals are stable, she has no weight loss, and remainder of exam is non-contributory. Let’s take a look at some differential diagnoses:
Presentation time: let’s say you’re a student presenting to your preceptor. A succinct way to summarise would be to go through all lists, and end with what you think the most likely differential(s) are, and what your thoughts are for management and follow up.
1. Start with the most likely: “Most likely she has upper airway cough syndrome, because of her history of allergic rhinitis and the fact that the cough is worse when she lies down. We could start her back on nasal spray and/or antihistamines to see if the cough resolves. Because of her smoking history, I would want to rule out COPD and send her for spirometry testing. A third thing to consider would be ACE inhibitor use, and she could be switched to another agent if the cough doesn’t resolve.”
2. Less likely: “It’s less likely asthma without a previous history of asthma. It’s less likely congestive heart failure because she doesn’t have any other suggestive signs or symptoms. Non-asthmatic eosinophilic bronchitis is less common but could be considered down the road. Her vaccines are up to date so pertussis is unlikely.”
3. Lastly, must-not-miss: “I’ve ruled out the must-not-miss diagnoses like TB and cancer – she doesn’t have night sweats, weight loss, fevers, hemoptysis, or sick contacts. I’m still concerned with her smoking history, so lung cancer should be considered down the road if the cough doesn’t resolve despite initial investigations and management.”
Key Take Home Point: sorting your differential diagnosis into most likely, less likely, and must not miss allows you to deliver relevant information in a succinct manner. Not all details of the history and physical exam need to be reviewed – just your pertinent positives and negatives for the differentials you are trying to rule in and out!
References/Readings
British Medical Journal [Internet]. Assessment of chronic cough [October 2023], available from https://bestpractice.bmj.com/topics/en-gb/69
A helpful video from Anna Pickens on presenting your differential diagnosis: https://www.youtube.com/watch?v=8J8VN1acM3o