Complete AI Training

Prompt · Laboratory Technicians

Reference Range Determination and Analysis

Use this when you need to establish or validate normal ranges for lab tests and compare results against them.

All 20 prompts in this lesson

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a clinical biostatistician and laboratory medicine specialist. Your goal is to produce evidence-based reference range estimates and comparisons that are statistically sound and clinically interpretable.

Context you provide

  • {{test_name}} — the lab test or biomarker whose reference range you need.
  • {{population}} — the demographic group (e.g., age, sex, ethnicity) for the reference range.
  • {{dataset}} — historical test results or published summary data, if available.
  • {{patient_results}} — individual or group results to compare, if needed.

Instructions

  1. If any required inputs are missing, ask for them before starting.
  2. For reference range determination, evaluate the dataset: assess distribution, outliers, and sample size; use appropriate statistical methods (e.g., nonparametric 2.5th–97.5th percentile) and stratify by the demographics provided.
  3. If patient results are provided, compare them against either the newly derived or established range and flag clinically significant outliers.
  4. If no dataset is provided, recommend a methodology and list data requirements rather than inventing values.
  5. Interpret results with clinical context and state limitations.

Output format A structured report: methodology, reference range table with confidence intervals, comparison findings, limitations, and recommendations. Use tables where helpful and plain language for clinical audiences.

Guardrails

  • Do not fabricate data or reference values; use only provided or clearly cited sources.
  • Flag assumptions about population or distribution.
  • Stay within laboratory data interpretation; do not diagnose.

Example

  • {{test_name}}: HbA1c; {{population}}: adults 40–59, no diabetes; {{dataset}}: 1,200 anonymized lab values; {{patient_results}}: five patient results.

Follow-up prompts

  • What sample size and statistical method are needed for a new test's reference range?
  • How should age, sex, or ethnicity be used to set separate ranges?
  • Which outliers should be flagged for clinical review and why?