Verified against ChatGPT · 2026-08-09
Translate a lab report full of jargon into plain language without inventing a diagnosis
Rewrites a lab or imaging report's dense terminology into plain-language explanations of what each value or line actually measures — strictly stopping short of interpreting what the results mean for the patient's specific situation.
The prompt
Ready to copy — highlighted parts are example details you can swap.
You are translating the technical language on a lab or imaging report into plain English, one line at a time, for someone who wants to understand what they're reading before their follow-up call with the ordering clinician. REPORT TEXT OR VALUES TSH: 6.8 mIU/L (ref 0.4-4.0). Free T4: 1.1 ng/dL (ref 0.8-1.8). Vitamin D, 25-OH: 22 ng/mL (ref 30-100). WHAT THE PERSON ALREADY UNDERSTANDS I know vitamin D is about bone health and sunlight, but I have no idea what TSH or T4 measure. SPECIFIC LINES THEY'RE CONFUSED BY What does TSH actually stand for and measure, and why does the range look different from Free T4's range? HARD RULE — DO NOT CROSS THIS LINE For every line or value in the report, explain only what it measures and, if the report itself states a reference range, where the value falls relative to that range in neutral terms (within range, above range, below range). Do not state or imply what an out-of-range value means for this specific person's health, do not suggest a likely cause, and do not use words like "concerning," "good news," or "nothing to worry about" — those are judgments only the ordering clinician can make with the full clinical picture, which this report alone doesn't contain. If a term appears without an accompanying reference range in the report, say plainly that no range was given rather than supplying a general population range from outside knowledge, since ranges vary by lab and by the person's specific circumstances. HOW TO WRITE EACH EXPLANATION Use one short paragraph per value or finding: what it is, in plain language; what it's generally used to help assess, one sentence, kept general rather than specific to this person; and where this result falls against the report's own stated range, stated neutrally. Skip explanation for anything the person already says they understand, and give the most detail to whatever's listed under confusing terms. OUTPUT FORMAT 1. A short glossary-style list, one entry per value or finding, plain language, neutral, in the order the report lists them. 2. A closing section titled "Questions worth asking the ordering clinician," turning any out-of-range or unclear item into a specific question the person can ask, rather than an answer this document is supplying itself. 3. One final line stating clearly that this is a plain-language reading aid, not an interpretation of what the results mean, and that only the ordering clinician can explain what these results mean for this person specifically.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The explicit ban on words like "concerning" or "nothing to worry about" targets a very specific and common model behavior: when a language model is shown an out-of-range lab value, its default instinct — trained on medical text where out-of-range values are usually discussed alongside their clinical significance — is to append an implied severity judgment even when only asked to define the term, because definition and interpretation are rarely separated in the source material it learned from. Forcing the explanation to stop at "what it measures" and "where it falls against the report's own stated range" breaks that habitual pairing and keeps the tool inside its actual competence, which is language translation, not clinical judgment that depends on symptoms, history, and other results the report alone doesn't contain. Refusing to backfill a missing reference range from general population data closes a subtler risk: lab reference ranges vary meaningfully by lab, by assay method, and by patient population (age, sex, pregnancy status), so a plausible-looking generic range presented as if it were this report's own range could make a genuinely normal result look abnormal or vice versa. Converting every out-of-range or unclear item into a question for the ordering clinician, rather than leaving it as a flat statement, keeps the deliverable oriented toward the actual conversation that needs to happen, and the closing disclaimer is stated in terms of what the document is (a reading aid) rather than generic boilerplate, which keeps it from being skimmed as filler.
What you get back
TSH (thyroid-stimulating hormone): a hormone that signals the thyroid gland to produce thyroid hormone. This report's value is above the stated reference range. Free T4: the active thyroid hormone TSH regulates. This value is within the stated reference range. Questions worth asking: My TSH is above range but my Free T4 is within range — what does that combination usually get checked next for? This is a plain-language reading aid, not an interpretation of these results — only the ordering clinician can explain what they mean for you specifically.
Verified against
ChatGPT GPT-5.1 · 2026-08-09
Changelog
- 2026-08-09 — Initial publish, verified against ChatGPT GPT-5.1.
Need this built into your business?
If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.
EXPLORE WHAT SCULT BUILDS
