Sample Dictations

Auto Texts

Hall FM. Language of the Radiology Report: Primer for Residents and Wayward Radiologists. AJR 2000;175:1239-1242

Based primarily on teachings of Drs. Pierre Sasson and Alex Heffess, as well as the other radiologists at Mount Auburn:

Our product is our report

Many doctors (and patients!) know radiologists only by their reports, and judge their quality based on the quality of their reports. Referring physicians lose confidence if are factual or typographical errors. Referring physicians often often only read impression. Imagine the patient reading report, and the report read being out loud in court.

Reports should be clear, concise, accurate, and precise. Brevity is often valued, although reports can be too long or too short. It is a balance between addressing all of the important points without extraneous words. Answer the clinical question and describe pertinent positive and negative findings; include relevant points to avoid phone calls.

Interventional procedural reports are different than typical diagnostic reports, often with a longer more detailed technique section. Mammography reports must use the specialized BIRADS terminology.


Not rule out

Need symptom, sign, or diagnosis

Key history to know:


Key for billing


Eg CT of abdomen and pelvis with intravenous contrast,mention angiogram, etc, post processing and 3d reconstructions

Sometimes combine studies if pertinent and same day


List pertinent studies you actually looked at, not every study the patient ever had


The findings section is an objective description of what is visible on the study images, including the pertinent positive and negative aspects. Many reports are moving towards a structured findings section (for instance, Liver: [], Pancreas: [], Spleen: []), although for some radiologists and in some instances a more conversational structure is used.


Always start the impression by answering the referring physician's question. Next order the important results by importance.

Interpretation, differential diagnosis, management

Don't just repeat findings, although sometimes brief reiteration of most pertinent findings useful in case where there is a differential



"should be considered"

"similar appearance"



Caution with word recommend, can box in referring physician

Use ACR appropriateness criteria

And other evidence based

Check that patient hasn't already had the exam

Can vary by who talking to - eg NP may need more guidance than specialized surgeon

Stay within realm of expertise - imaging or which sub specialty consultation

Can include citation - indication/increase quality of care, if referring MD wants to know can look it up

To our knowledge, the best managemnt at this time is


Discussed with, include date time

For unexpected finding or with immediate clinical relevance - protects you if they don't act on it

Record changes between significant changes from preliminary reports


Chatty versus structured versus personal narrative

Separate topics by paragraphs

Present tense for diagnostic studies

Past tense for procedures

Edit carefully

Attend to details

Speech recognition can insert similar words - ascending descending, prosthetic prostatic, mm cm, degenerative degenerate, leave out 'no', hyper hypo, atypical

If you can leave a word out, do

Be clear and brief

Inconsistent serial measurements

Measure largest dimension, then 90 degrees

If disagree with prior re measure and describe

Check measure carefully, try stick with one type length either cm or mm

Terminology to use with Caution

Avoid jargons, eponyms, abbreviations (unless common and universal such as AP/PA)

Caution with: is noted, identified, evidence of, visualized, demonstrates, exhibits, appreciated,

Extra caution for

clinical correlation. Buzz term gets clinicians upset, considered insulting, shifting responsibility. Instead be specific eg consider eval blood counts, focal tenderness at this site.

Cannot exclude - instead indicate suspected likelihood

Retrospectively - don't throw prior reader under bus, esp in mammo

Normal vs. unremarkable

Compatible, consistent with

Normal cxr vs. no acute cp da

Pet peeves to avoid

Lung fields

Infiltrate - opacity,

Caution with

Progression, regression - see resist criteria


Poor inspiration

Stable vs unchanged



Pleural based

Cardiac silhouette - unless think pericardial effusion

Slow move to standard nomenclature, structured reporting

More Pending

want to sound strong or lame?

Include links from JACR - don't use cannot exclude, etc

but also include what words TO use