Healthcare Data Sensitivity Classification Levels
Description
Eight data elements, four tiers, and several deliberate traps. Payer contract terms are not patient data at all and still sit at Confidential, because sensitivity is not the same thing as protected health information. A provider's NPI is a matter of public record and still classifies as Internal, because the classification follows the use rather than the field. Maria Chen's diagnosis code outranks her home address, because an identified patient joined to a clinical fact is the combination that can cost someone a job or their safety. Every correct placement reveals that tier's access rule; every incorrect one returns the card and offers the question that decides the tier. Placing all eight assembles the four rules into one summary panel.
Lesson Plan
Audience: undergraduate students and healthcare data practitioners. Allow 12-18 minutes. Prerequisites: protected health information, and the idea of role-based access.
Learning objective: Given a sample healthcare data element, the learner can classify it into the correct sensitivity tier (Public, Internal, Confidential, Restricted) and identify the access control consequence of that classification.
- Explore: Place the two easy cards first — clinic hours and the marketing brochure — and read the Public access rule that appears.
- Explain: Place the payer contract terms and the de-identified research dataset. Explain what these two have in common that puts them at the same tier as a patient's home address, given that neither identifies a patient.
- Transfer: Once all eight are placed, read the summary panel. Then classify two data elements of your own choosing and name the question you used to decide each.
Assessment: use the Explain prompt as an exit ticket. A complete response separates sensitivity from identifiability, and gives a defensible reason for placing a non-patient data element above a patient one.
Embed This MicroSim
Copy this iframe to your website:
1 | |
Quality Checks
Validated September 8, 2026. Completeness rubric: 100/100, scored with the project's calculate-quality-score.py against the standardization checklist. Checked automatically in a browser at 400, 800, and 1200 px: the page loads with no JavaScript errors, no interactive control is clipped outside the viewport, and the document does not scroll horizontally. Controls, selections, and the reset path were exercised by hand during development at desktop width. The page reports its own height to the parent document, so the declared iframe height above serves only as a fallback.
Specification
The full specification below is extracted from Chapter 28: Data Quality, Stewardship, and Compliance.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 | |
Related Resources
References
- Source chapter — supplied the learning objective and the worked example.
- p5.js reference — canvas, drawing, and input handling (accessed September 8, 2026).
- p5.js web editor — paste the JavaScript source to experiment; no hosted sketch has been published.
- Wikipedia: Data classification (data management) — the tiered scheme this exercise applies.