Verified against ChatGPT · 2026-08-11
Build a non-diagnostic health risk screening checklist for a workplace wellness program
Produces a self-report risk screening checklist for a workplace or community wellness program that flags who should talk to a clinician next, without ever scoring or labeling an individual's actual risk level itself.
The prompt
Ready to copy — highlighted parts are example details you can swap.
You are building a self-report health risk screening checklist for a workplace or community wellness program. The checklist's only job is to help someone notice patterns worth raising with a clinician — it must never produce a risk score, label, or verdict about an individual's actual health status.
PROGRAM CONTEXT
An annual voluntary wellness screening event at a mid-size company, self-administered on paper.
RISK AREAS TO COVER
Sleep quality, stress levels, physical activity frequency, and family history awareness.
HOW RESULTS WILL BE USED
Kept entirely by the individual — not collected or reported to HR or the employer in any form.
AUDIENCE
Adult employees across a wide age range, mixed health literacy, completing it voluntarily on their own time.
RULES
Write each checklist item as a plain factual question about a behavior, symptom frequency, or history ("in the past month, how many days did you...") rather than a question that requires the person to already have interpreted their own risk ("do you have high blood pressure risk"). Group items by risk area, and after each group, write a plain instruction for what a person should do with a pattern of "yes" or concerning answers in that group — always "consider discussing this with a doctor," never a score, tier, or label like "moderate risk" or "high risk," since assigning a risk tier from a self-report checklist without clinical context is exactly the kind of judgment this tool should not make. Keep the checklist anonymous-compatible by design — do not require or reference anything that would only make sense if results were being centrally scored or ranked against other participants, since that shifts the tool from self-reflection aid toward something that resembles a clinical assessment.
WHAT NOT TO DO
Do not include a scoring formula, point system, or cutoff threshold anywhere in the checklist, even one you're inclined to add for user-friendliness — an unscored checklist that prompts a conversation is safer and more accurate than a scored one that implies a false level of diagnostic precision.
MANDATORY DISCLAIMER
Open the checklist with a note stating plainly that this is a self-reflection tool, not a medical screening or diagnostic instrument, that it does not calculate a risk score, and that any pattern of concerning answers should be discussed with a qualified clinician, who is the only one positioned to actually assess health risk.
OUTPUT FORMAT
1. The disclaimer.
2. Checklist items grouped by risk area, each as a plain yes/no or frequency question.
3. After each group, the plain instruction line ("if you answered yes to several of these, consider raising it with your doctor").
4. A closing line restating that this checklist doesn't score or diagnose anything.Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The most dangerous thing a lay-audience risk checklist can do is imply diagnostic precision it doesn't have — a point-scored checklist that outputs "you are at moderate risk" sounds authoritative but is built on self-report data with no clinical context, and a person reading a false "low risk" result may reasonably (and wrongly) decide a real symptom doesn't need attention, which is worse than no checklist at all. Explicitly banning any scoring formula or cutoff threshold, and requiring every group to end in "consider discussing this with a doctor" rather than a tier label, keeps the tool doing the one thing it can safely do: prompting a conversation, not replacing clinical judgment with an unearned number. Phrasing items as plain behavioral or frequency questions rather than ones requiring the respondent to have already assessed their own risk ("do you have high blood pressure risk") matters because self-assessment questions produce unreliable answers from people without clinical training — a frequency question ("how many days in the past month") is something anyone can answer accurately regardless of health literacy, which is exactly the design principle behind validated screening instruments used in real preventive care programs. The anonymous-compatible design rule exists because centralizing or ranking individual results turns a private self-reflection tool into something that functions like workplace health surveillance, which raises a different set of concerns entirely and isn't what this checklist is meant to be. The disclaimer stating explicitly that no score is calculated addresses a specific expectation mismatch: many people assume any checklist with yes/no questions must produce some kind of number, so the note has to say plainly that this one deliberately doesn't, and why.
What you get back
Note: this is a self-reflection tool, not a medical screening — it does not calculate a risk score. Sleep: In the past month, how many nights did you get less than 6 hours of sleep? / How many days did you feel tired despite a full night's sleep? If you answered yes to several of these, consider raising your sleep patterns with your doctor rather than assuming it's just normal tiredness.
Verified against
ChatGPT GPT-5.1 · 2026-08-11
Changelog
- 2026-08-11 — 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
