Report Templates
- A report template is a pre-built skeleton — headings, default-normal lines, and blanks — that you dictate into instead of building a report from scratch every time.
- Templates make reports consistent and complete: the same exam always covers the same checklist, so things get described in the same place and don't get forgotten.
- Two flavors: simple "boilerplate" text macros, and true structured templates with labeled, machine-readable fields.
- The famous risk-stratification systems (BI-RADS, Lung-RADS, LI-RADS, and friends) are templates with teeth — they force a category and a recommendation.
- The danger isn't the template; it's trusting the template — leaving a default-normal line in when the thing is, in fact, abnormal.
Imagine if every time you wrote an email you started from a blank screen, re-typed your own greeting, re-invented where the subject line goes, and tried to remember whether you'd mentioned the attachment. Exhausting, and you'd forget the attachment about a third of the time. A report template is the email client that already has the fields laid out — you just fill in what's different today. That's the whole idea, and it's quietly one of the most powerful tools in the reading room.
What a template actually is
At its simplest, a template is a saved chunk of text you drop into a report with a keystroke or a spoken trigger phrase. Radiologists call these macros (or "normals," or "canned text"). Say "normal chest" into the microphone and a full, pre-written paragraph of "the lungs are clear, the heart is normal in size, no effusion" appears, ready for you to edit only the parts that aren't normal.
A template is just that idea, scaled up: a full scaffold for an entire exam type. It usually has three ingredients baked in.
- Fixed headings — TECHNIQUE, COMPARISON, FINDINGS, IMPRESSION — so the report's architecture never wobbles.
- Default-normal lines under each finding heading, pre-filled with the most common (normal) result.
- Blanks or fields you must actively fill — the measurement, the laterality, the impression.
The art is that the radiologist edits deltas only. Everything normal is already written; you change the lines that aren't. Build the report by subtraction, not addition.
A "structured report" and a "report template" aren't quite the same thing. A template is the starting scaffold; structure is whether the report is organized into discrete, labeled, ideally machine-readable fields. Most structured reports are built from templates, but you can have a beautifully templated report that's still just free-flowing prose. For the why-it-matters, see structured reporting foundations.
The two big families
Think of it as the difference between a fill-in-the-blank greeting card and an actual spreadsheet.
| Text macro / boilerplate | Structured template | |
|---|---|---|
| What it is | Pre-written prose you drop in and edit | Labeled fields with defined slots |
| Machine-readable? | Not really — it's still a paragraph | Yes — fields can be pulled into databases |
| Best for | Speed, consistent phrasing | Data, registries, decision support |
| Familiar example | "Normal chest" macro | A standardized cancer-staging worksheet |
The structured kind is what lets a computer later count how many "category 4" thyroid nodules your department reported last year — because that category lived in a defined field, not buried in a sentence. The prose kind just makes you faster and your wording cleaner. Most reading rooms run a healthy mix of both.
Templates with teeth: the -RADS systems
The celebrity templates are the risk-stratification systems — BI-RADS for breast, Lung-RADS for lung screening, LI-RADS for liver lesions in at-risk patients, and a growing alphabet of cousins (TI-RADS, PI-RADS, O-RADS). These are templates that don't just organize your findings; they make you commit. You describe the features, the system assigns a category, and each category carries a management recommendation — short-interval follow-up, biopsy, routine screening.
That's the magic: a number that means the same thing to every radiologist and every referring clinician on the planet. "BI-RADS 4" isn't an opinion about the vibe of a mammogram; it's a defined bucket with a defined next step. The template enforces a shared language so nobody has to guess what "probably fine, but watch it" means.
Many of these systems also require an impression that explicitly states the category and the recommended action. The template won't let you trail off with "clinical correlation recommended" and call it a day — and your referrers will quietly love you for it.
The trap hiding in the convenience
Here's the thing about a report that fills itself in: it'll happily lie to you with a straight face. Every default-normal line is a small bet that today's patient is normal there. Most of the time you win that bet. The day you don't — the day you dictate around a finding but forget to change "no acute findings" at the bottom — the template has now confidently contradicted you in writing.
The classic template error is the leftover default. You describe a fracture in the FINDINGS section but never touch the pre-filled "no acute osseous abnormality" line, and the report contradicts itself. Worse: a stray field from the previous patient (laterality, gender, a measurement) survives into this report. Always read the finished report as if you'd never seen the template — the template's job is to help you, not to sign for you.
The fix isn't to abandon templates — it's to treat the default text as a prompt to verify, not a fact already verified. Speed is the gift; complacency is the bill.
Where this connects
Templates are the delivery vehicle for two skills covered elsewhere: clean, consistent description of a finding, and reliable handoff of urgent results through critical-results and closed-loop communication. A great template makes the first nearly automatic and reminds you to do the second.
If you remember one thing: a template is a checklist that types for you. It guarantees you'll cover everything — but it never guarantees you'll check everything. That part is still your job.