Diagnostic Checklists

Home / Lectures / Diagnostic Checklists


Checklists are a great way to approach common problems. In the medical world Atul Gawande’s Checklist Manifesto was received as a checklist idea that improved patient safety. Below you will find an article addressing a differential diagnosis checklist addressing common complaints and the differentials associated with them. Read the beginning of the article below and check out the attached document:

Diagnostic Checklists (Last saved 7/17/2013 8:58 AM)
1. The purpose of these checklists is to prompt physicians to consider a broad differential diagnosis for common symptoms in primary care and to help resist the most common cause of missing a diagnosis: failure to consider it. The lists are not exhaustive, but the goal is feasible because
a. It is possible to cover 99% of diagnostically challenging symptoms with a short list, whereas 100% coverage would require a long list.
b. Similarly, for each symptom it is possible to cover 99% of the diseases that cause it with a short list.
c. Checklists are not needed for complaints that rarely pose diagnostic challenges, either because the list of causes is short (e.g., constipation, breast lump, sore throat, nasal congestion), the diagnosis is obvious (e.g., minor trauma, wart), there is a single cause that accounts for the overwhelming majority of cases (e.g., hypertension, obesity), the complaint prompts a standard diagnostic approach which catches important diagnoses even when not initially considered (e.g, rectal bleeding followed by colonoscopy), or the complaint is idiosyncratic (e.g., “funny feeling” in cheek).
d. The checklists do not include diagnoses that are exceedingly rare in North America or diagnoses in which the symptom is a late manifestation rather than a presenting complaint (e.g., abnormal uterine bleeding caused by end-stage renal disease).
2. The diagnoses are listed in approximate order of prevalence in primary care (most common diseases at the top). This ordering is based on limited published data and the authors’ experience. (Cherry DK, et al. National Ambulatory Medical Care Survey: 2006 Summary. National Health Statistics Reports. Number 3. August 6, 2008.)

3. Unless otherwise specified, the checklists address adults rather than children.
4. These checklists may have five advantages over more traditional differential diagnoses:
a. A shorter list, practical at the point of care
b. Diseases ordered by prevalence in primary care
c. Identification of “must-not-miss” diagnoses, designated by an ace of spades (♠)
d. Identification of diagnoses that are commonly missed, designated by an asterisk (*) and partly based on Schiff GD, et al. Arch Intern Med. 2009;169:1881-7; and on Zwaan L, et al. Arch Intern Med. 2010 ;170:1015-21).
e. Lumping and splitting to meet the needs of clinicians rather than pathologists or billers (e.g., better to split “myocardial infarction” from “heart disease,” but better to lump “abnormal uterine bleeding” rather than splitting into “dysfunctional uterine bleeding” vs. “structural causes” because this distinction may not be apparent initially).
5. The checklists are designed to be printed on 4×6 cards, two-sided, laminated, with a 3/4 inch metal binder ring in the left upper corner. The lists could also be adapted for a handheld computer.
The checklists are modified from Ely JW, Graber ML, Croskerry P. Checklists to reduce diagnostic errors. Acad Med. 2011 Mar;86(3):307-13.

Differential Diagnosis Checklist


  • alvarezzy

    great (extensive) list. keep in mind, this was published in Academic Medicine by an internist, a family medicine doc, and a (canadian) ED doc. ED docs typically see 2-4 patients an hour, Medicine people usually see 0.5 patient an hour.

    from a practical point of view, i’m sure my charge nurse/and the doc i’m signing out to are going to bite my head off if I’m ruling out sarcoidosis on all my patients. also, if you do write all this in the chart and not justify why you don’t think a certain diagnosis is not so, it probably is not going to help you when you do miss the diagnosis.

    It’s definitely a good list to trigger consideration of diseases not part of your heuristics (to avoid cognitive errors), but too extensive for practical ED flow.

    i tend to use the skills chiefing year taught me. the “sick or not sick” dichotomy. if sick, assign, or in my case now, prioritize/treat/admit. if not sick, how do i botch up this case and get me another M&M/friday morning rounds/QI review bounceback.

    Here’s a plug for electronic medical records. Epic, for instance, asks questions in history to trigger consideration of not-to-miss diagnoses. The PMH and FH also has list of not-to-miss diagnoses based on chief complaint. It’s a subtle yet arguably effective way of avoiding cognitive biases.

    definitely not a perfect system, but curious what everyone else does.

    i’m sure siuf would disagree, and would discharge everyone =). I know dr. schechter has an interesting way of addressing chief complaints. side-by-side, i’m sure everyone agrees dr. schechter (extensive differential dx list driven) and conroy (heuristics driven) are equally good physicians, just have different styles/speed.

Leave a Reply