Quality Metrics

HEDIS Measures Guide

Complete guide to HEDIS (Healthcare Effectiveness Data and Information Set) quality measures. Learn database design, measure calculation, numerator/denominator logic, and Star Ratings.

What are HEDIS Measures?

HEDIS (Healthcare Effectiveness Data and Information Set) is a comprehensive set of standardized performance measures developed by the National Committee for Quality Assurance (NCQA). Health plans, including Medicare Advantage and commercial insurers, use HEDIS to measure quality of care and service.

HEDIS measures cover prevention, acute and chronic disease management, behavioral health, and patient experience. Results drive quality improvement, bonus payments, and consumer choice through Star Ratings.

Common HEDIS Measures

Diabetes

HbA1c Control (HBD)

Percentage of diabetic patients with HbA1c poor control (>9.0%)

Denominator: Members age 18-75 with diabetes
Numerator: Members with most recent HbA1c >9.0%
Lower is better (fewer patients with poor control)

Prevention

Breast Cancer Screening (BCS)

Percentage of women who had mammogram to screen for breast cancer

Denominator: Women age 50-74
Numerator: Women with mammogram in last 2 years
Higher is better (more women screened)

Prevention

Colorectal Cancer Screening (COL)

Percentage of adults who had appropriate screening for colorectal cancer

Denominator: Adults age 50-75
Numerator: Members with colonoscopy (10 yrs), FIT (1 yr), or Cologuard (3 yrs)
Higher is better

Heart Disease

Controlling High Blood Pressure (CBP)

Percentage of patients with hypertension whose BP is adequately controlled

Denominator: Members age 18-85 with hypertension diagnosis
Numerator: BP <140/90 mmHg
Higher is better

HEDIS Database Schema

-- HEDIS Measure Definitions
CREATE TABLE hedis_measure (
  id                    UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  
  -- Measure Identification
  measure_code          VARCHAR(10) UNIQUE NOT NULL,  -- BCS, HBD, COL, etc.
  measure_name          VARCHAR(255) NOT NULL,
  measure_domain        VARCHAR(100),                  -- Effectiveness, Access, etc.
  
  -- Measure Details
  description           TEXT,
  numerator_definition  TEXT,
  denominator_definition TEXT,
  exclusion_criteria    TEXT,
  
  -- Calculation
  higher_is_better      BOOLEAN,
  
  -- Metadata
  ncqa_year             INTEGER,                       -- HEDIS year
  active                BOOLEAN DEFAULT TRUE,
  
  created_at            TIMESTAMP DEFAULT NOW()
);

-- Patient Measure Attribution
CREATE TABLE patient_measure_attribution (
  id                    UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  
  patient_id            UUID NOT NULL,
  measure_code          VARCHAR(10) REFERENCES hedis_measure(measure_code),
  measurement_year      INTEGER NOT NULL,
  
  -- Eligibility
  in_denominator        BOOLEAN DEFAULT FALSE,
  denominator_date      DATE,
  
  -- Exclusions
  is_excluded           BOOLEAN DEFAULT FALSE,
  exclusion_reason      VARCHAR(255),
  
  -- Numerator Compliance
  in_numerator          BOOLEAN DEFAULT FALSE,
  numerator_date        DATE,
  numerator_value       VARCHAR(100),              -- e.g., HbA1c value
  
  -- Data Sources
  data_source           VARCHAR(50),               -- claims, lab, EHR
  evidence_date         DATE,
  
  -- Status
  status                VARCHAR(50),               -- pending, complete, gap
  last_calculated       TIMESTAMP,
  
  created_at            TIMESTAMP DEFAULT NOW(),
  updated_at            TIMESTAMP DEFAULT NOW(),
  
  UNIQUE(patient_id, measure_code, measurement_year)
);

CREATE INDEX idx_measure_attr_patient ON patient_measure_attribution(patient_id);
CREATE INDEX idx_measure_attr_measure ON patient_measure_attribution(measure_code);
CREATE INDEX idx_measure_attr_year ON patient_measure_attribution(measurement_year);
CREATE INDEX idx_measure_attr_status ON patient_measure_attribution(status);

-- Care Gaps (Patients Not Meeting Measures)
CREATE TABLE care_gap (
  id                    UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  
  patient_id            UUID NOT NULL,
  measure_code          VARCHAR(10) REFERENCES hedis_measure(measure_code),
  gap_year              INTEGER NOT NULL,
  
  -- Gap Details
  gap_type              VARCHAR(50),               -- overdue, never done
  gap_description       TEXT,
  recommended_action    TEXT,
  
  -- Outreach
  outreach_status       VARCHAR(50),               -- pending, contacted, scheduled, closed
  outreach_date         DATE,
  outreach_method       VARCHAR(50),
  
  -- Closure
  gap_closed_date       DATE,
  closure_method        VARCHAR(100),              -- appointment, lab, etc.
  
  -- Priority
  priority              VARCHAR(20),               -- high, medium, low
  
  created_at            TIMESTAMP DEFAULT NOW(),
  updated_at            TIMESTAMP DEFAULT NOW()
);

CREATE INDEX idx_gap_patient ON care_gap(patient_id);
CREATE INDEX idx_gap_measure ON care_gap(measure_code);
CREATE INDEX idx_gap_status ON care_gap(outreach_status);

-- Measure Results (Aggregated)
CREATE TABLE measure_results (
  id                    UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  
  measure_code          VARCHAR(10) REFERENCES hedis_measure(measure_code),
  measurement_year      INTEGER NOT NULL,
  
  -- Results by Segment (optional)
  provider_id           UUID,
  plan_id               UUID,
  region                VARCHAR(100),
  
  -- Calculation
  denominator_count     INTEGER NOT NULL,
  numerator_count       INTEGER NOT NULL,
  exclusion_count       INTEGER,
  
  -- Rate
  measure_rate          DECIMAL(5,2),              -- Percentage
  
  -- Benchmarks
  ncqa_percentile       INTEGER,                   -- 10th, 25th, 50th, 75th, 90th
  
  -- Metadata
  calculation_date      DATE NOT NULL,
  
  created_at            TIMESTAMP DEFAULT NOW(),
  
  UNIQUE(measure_code, measurement_year, provider_id, plan_id)
);

CREATE INDEX idx_results_measure ON measure_results(measure_code);
CREATE INDEX idx_results_year ON measure_results(measurement_year);

Measure Calculation Logic

Step-by-Step Calculation

Identify Denominator

Find all patients who meet age, diagnosis, and enrollment criteria for the measure period

Apply Exclusions

Remove patients with valid exclusion criteria (hospice, bilateral mastectomy, etc.)

Calculate Numerator

Count patients who met the measure criteria (screening done, test result in range)

Compute Rate

Rate = (Numerator / [Denominator - Exclusions]) × 100

SQL Query Examples

Calculate Breast Cancer Screening Rate

WITH eligible_patients AS (
  SELECT patient_id
  FROM patient_measure_attribution
  WHERE measure_code = 'BCS'
    AND measurement_year = 2024
    AND in_denominator = TRUE
    AND is_excluded = FALSE
),
numerator_patients AS (
  SELECT patient_id
  FROM patient_measure_attribution
  WHERE measure_code = 'BCS'
    AND measurement_year = 2024
    AND in_numerator = TRUE
)
SELECT 
  COUNT(DISTINCT ep.patient_id) as denominator,
  COUNT(DISTINCT np.patient_id) as numerator,
  ROUND(100.0 * COUNT(DISTINCT np.patient_id) / COUNT(DISTINCT ep.patient_id), 2) as rate
FROM eligible_patients ep
LEFT JOIN numerator_patients np ON ep.patient_id = np.patient_id;

Find Patients with Open Care Gaps

SELECT 
  p.mrn,
  p.first_name || ' ' || p.last_name as patient_name,
  hm.measure_name,
  cg.gap_description,
  cg.recommended_action,
  cg.priority
FROM care_gap cg
JOIN patient p ON cg.patient_id = p.id
JOIN hedis_measure hm ON cg.measure_code = hm.measure_code
WHERE cg.gap_year = 2024
  AND cg.outreach_status IN ('pending', 'contacted')
  AND cg.gap_closed_date IS NULL
ORDER BY 
  CASE cg.priority 
    WHEN 'high' THEN 1 
    WHEN 'medium' THEN 2 
    ELSE 3 
  END,
  p.last_name;

HEDIS & Star Ratings

How HEDIS Drives Star Ratings

Medicare Advantage Star Ratings (1-5 stars) include ~40 measures, with many derived from HEDIS. Higher stars = quality bonus payments (up to 5% increase) + increased enrollment.

4+ stars: Eligible for quality bonus payments
5 stars: Can enroll members year-round (not just open enrollment)
3 stars or below: Risk of contract termination

Key HEDIS measures in Stars: Diabetes care (HbA1c, eye exam, kidney test), controlling blood pressure, medication adherence, breast cancer screening

Related Healthcare Guides

Healthcare Analytics →Value-Based Care →Care Management →Healthcare Glossary →

mdatool.