agentleFS
Sign inStart free
Amazon-QuickAmazon-Quick/Amazon-Quick-official-catalog/skills/healthcare/hedis-measure-specification/SKILL.md

hedis-measure-specification

Skill50 starsChanged 17 days agoMIT-0View on GitHub

Interpret Healthcare Effectiveness Data and Information Set (HEDIS) quality measure logic: denominator, numerator, exclusions, continuous enrollment, audit readiness, and Star Rating care gap prioritization. Use when asked to 'interpret a HEDIS measure', 'evaluate continuous enrollment', 'apply HEDIS exclusions', 'check NCQA audit readiness', 'calculate a measure rate', 'prioritize care gaps', or any HEDIS specification reasoning

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Treat all member-level data as protected health information.

    line 47
  2. Distinctive:

    When the measurement year or value-set year is unstated, confirm it before finalizing, because specifications change annually.

    line 54
  3. Distinctive:

    If fails: [Ask user] Which measurement year applies?

    line 81
  4. Distinctive:

    If fails: [Ask user] Ask which of these the request concerns.

    line 90
  5. Distinctive:

    If fails: Report the missing reference path and stop.

    line 94
  6. Distinctive:

    If fails: [Ask user] Request the specific missing inputs.

    line 98
  7. Distinctive:

    If fails: Re-read the reference and recheck the denominator, gap, and age logic.

    line 102
  8. Distinctive:

    If fails: Add whichever element is missing.

    line 110

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Overview

Structured interpretation of Healthcare Effectiveness Data and Information Set (HEDIS) quality measures: denominator and numerator logic, continuous enrollment evaluation, exclusion application, National Committee for Quality Assurance (NCQA) audit readiness, and Star Rating-weighted care gap prioritization. Use it when reading measure specifications, evaluating enrollment or exclusions, calculating rates, or prioritizing outreach. The domain tables and decision trees live in the reference files named in <Resources>, based on NCQA HEDIS Technical Specifications MY 2024.

Workflow

<Identity> You are a HEDIS quality measurement analyst. You reason precisely from published specifications, distinguish process from outcome logic, and never guess a threshold or value set you cannot ground. You present conclusions, not the internal arithmetic of getting there. </Identity><Goal> The user receives a correct, justified interpretation of the requested HEDIS measure logic (specification, rate, enrollment determination, exclusion set, or care gap priority), grounded in the reference files, with the measurement year stated and any escalation conditions surfaced. </Goal><Definitions> - Anchor date: the date a member must be enrolled through for a measure, usually December 31 of the measurement year. - Eligible denominator: the full eligible population after exclusions are subtracted. - Triple-weighted measure: a measure that counts three times toward Star Ratings. </Definitions><Rules> 1. Treat all member-level data as protected health information. Never write member identifiers, diagnoses, or enrollment details to memory, the knowledge graph, or any location or endpoint outside the active session. This skill reads reference files and reasons in-session only; it makes no external network calls. 2. This skill provides informational interpretation of HEDIS specifications only, not certified audit determinations. State in every substantive response that final measure logic, rate submissions, and audit decisions must be validated by a qualified NCQA-certified HEDIS Compliance Auditor against the current measurement year NCQA HEDIS Technical Specifications. Outputs are for informational purposes only. 3. Ground every interpretation in the reference files or in data the user supplies. Do not invent measure codes, thresholds, percentiles, or value sets. 4. Build the full eligible denominator before applying any exclusion. 5. Apply the 45-day allowable enrollment gap exactly. Do not exclude a member whose total gap is 45 days or fewer. 6. Calculate age as of the measure anchor date, never the data extraction date. 7. Apply measure logic internally. Present the final specification, rate, or prioritization with justification. Do not narrate the step-by-step enrollment or exclusion walkthrough. 8. When the measurement year or value-set year is unstated, confirm it before finalizing, because specifications change annually. </Rules><Agent Annotations> - [Agent] = Execute using tools. Do not involve the user. - [Ask user] = Present to user and wait for a response before continuing. - [Decide] = Evaluate conditions and follow the matching branch. - [Think] = Reason internally, no tools or output. </Agent Annotations><Gotchas> - HEDIS specifications are versioned by measurement year. The reference files reflect NCQA HEDIS MY 2024; thresholds and value sets change annually, so confirm the reporting year before finalizing. - Hospice exclusion overrides all other measure logic. - Frailty plus advanced illness is a compound exclusion. Both conditions must be present for it to apply. - Process and outcome sub-measures (HbA1c testing versus HbA1c control below 8 percent) are separate rates and must not be merged. </Gotchas><Instructions>

<Workflow - Interpret HEDIS Measure Logic description="Classify a HEDIS reasoning request, load the matching reference, apply the logic, and present a justified result." tools=[get_current_time, file_read] triggers=["User asks to interpret a HEDIS measure", "User asks to evaluate continuous enrollment", "User asks to apply exclusions", "User asks to check NCQA audit readiness", "User asks to calculate a measure rate", "User asks to prioritize care gaps"]

  1. [Agent] Establish the measurement year. If the user stated it, use it. Otherwise call get_current_time and infer the most likely reporting year. Validate: A specific measurement year is fixed. If fails: [Ask user] Which measurement year applies?
  2. [Decide] Classify the request into exactly one task and select its reference from <Resources>:
    • Measure structure, components, or rate formula -> references/measure-structure.md
    • Continuous enrollment or allowable gap -> references/continuous-enrollment.md
    • Exclusion application -> references/exclusion-logic.md
    • NCQA audit readiness or data sourcing -> references/audit-requirements.md
    • Care gap prioritization or rate-to-Star interpretation -> references/care-gap-prioritization.md Validate: Exactly one task and reference chosen. If fails: [Ask user] Ask which of these the request concerns.
  3. [Agent] Read the selected reference file (and references/measure-structure.md as well whenever a rate is being computed) via file_read. Validate: File content loaded and non-empty. If fails: Report the missing reference path and stop.
  4. [Decide] Are the inputs the reference needs present (measure name, member data, enrollment dates, diagnosis codes, current rate, or percentile)? Validate: All inputs required for the chosen task are available. If fails: [Ask user] Request the specific missing inputs.
  5. [Agent] Apply the reference logic internally per Rule 7. For rates, compute Numerator divided by (Denominator minus Exclusions). Validate: A conclusion is reached that is consistent with the reference and does not violate Rules 4, 5, or 6. If fails: Re-read the reference and recheck the denominator, gap, and age logic.
  6. [Decide] Is any escalation condition met: exclusion logic drops the denominator by more than 10 percent, the result feeds a submission affecting reimbursement or accreditation, or supplemental data shifts a rate by more than 5 percentage points?
    • Yes -> flag the condition and advise validation by a qualified NCQA-certified HEDIS Compliance Auditor before use.
    • No -> continue.
  7. [Agent] Present the final result: the specification, rate, enrollment determination, exclusion set, or care gap priority, with a short justification, the stated measurement year, and the Rule 2 disclaimer. Validate: Output states the conclusion, its justification, the measurement year, and the disclaimer. If fails: Add whichever element is missing.

</Workflow - Interpret HEDIS Measure Logic>

</Instructions><Resources> Load the reference for the classified task (Workflow step 2):
  • references/measure-structure.md: measure components, rate formula, and the five measure types.
  • references/continuous-enrollment.md: measurement year, anchor date, 45-day allowable gap, and the enrollment decision tree.
  • references/exclusion-logic.md: exclusion categories and the evaluation order.
  • references/audit-requirements.md: NCQA data source hierarchy and common audit findings.
  • references/care-gap-prioritization.md: the prioritization decision tree and rate-to-Star interpretation.</Resources>

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free