population-health-analyzer

name	population-health-analyzer
description	Analyzes population health data to identify disparities, risk stratify patient cohorts, and design interventions. Outputs community health assessments with SDOH integration and outcome measurement frameworks.
allowed-tools	Read, Write, WebSearch, Bash
effort	high

Population Health Analyzer

When to activate

When analyzing health outcomes across a patient population, designing value-based care programs, identifying health disparities, or when preparing community health needs assessments (CHNA). Use for ACOs, health systems, public health departments, and managed care organizations.

When NOT to use

Skip for individual patient clinical decisions, single-case reviews, or when the population analysis was completed within the last quarter with no significant demographic or program changes.

Instructions

1. Population definition:

   • Denominator: Total attributed lives, active patients, geographic area
   • Stratification: Age, sex, race/ethnicity, payer, risk level
   • Time period: Quarterly, annual, rolling 12-month

2. Outcome measures:

   • Clinical: HEDIS measures (diabetes control, hypertension, cancer screening)
   • Utilization: ED visits/1000, admissions/1000, readmission rate
   • Cost: PMPM total cost, pharmacy cost, out-of-network spend
   • Patient experience: CAHPS composite scores

3. Risk stratification:

   • Hierarchical condition categories (HCC) for risk adjustment
   • Utilization-based: High utilizers (top 5%), rising risk, stable
   • Clinical complexity: Multimorbidity count, frailty index
   • Social risk: SDOH Z-codes, area deprivation index (ADI)

4. Disparity analysis:

   • Stratify all measures by race/ethnicity, language, geography
   • Calculate disparity ratios and absolute differences
   • Identify statistically significant differences (chi-square, t-test)
   • Map hotspots using geographic information

5. Intervention design:

   • Target population with inclusion/exclusion criteria
   • Intervention components and delivery method
   • Expected impact (effect size from literature)
   • ROI estimation: Cost of intervention vs. avoided utilization

Output Format

POPULATION HEALTH ASSESSMENT: [Population name]
DENOMINATOR: [N] lives | PERIOD: [time range]
PAYER MIX: [Medicare X%, Medicaid Y%, Commercial Z%]

OUTCOME SCORECARD:
| Measure | Current | Target | Benchmark | Gap |
|---------|---------|--------|-----------|-----|

RISK STRATIFICATION:
  High utilizers (top 5%): [N] — [characteristics]
  Rising risk: [N] — [characteristics]
  Stable: [N]

DISPARITY ANALYSIS:
| Measure | White | Black | Hispanic | Asian | p-value |
|---------|-------|-------|----------|-------|---------|

HOTSPOTS:
  [Geographic area] — [measure] — [X]× benchmark

INTERVENTION RECOMMENDATIONS:
  1. [Intervention] — Target: [N] — Expected impact: [X]% — ROI: [X]:1
  2. [...]

Example

POPULATION HEALTH ASSESSMENT: Metro Health ACO — Attributed Lives
DENOMINATOR: 42,000 lives | PERIOD: Q1 2026
PAYER MIX: Medicare 55%, Medicaid 25%, Commercial 20%

OUTCOME SCORECARD:
| Measure                  | Current | Target | Benchmark | Gap  |
|--------------------------|---------|--------|-----------|------|
| HbA1c <8% (diabetes)    | 72%     | 80%    | 78%       | -8%  |
| BP <140/90 (hypertension)| 68%     | 75%    | 73%       | -7%  |
| ED visits/1000           | 385     | <320   | 340       | +65  |
| 30-day readmission       | 14.2%   | <12%   | 13.1%     | +2.2%|

RISK STRATIFICATION:
  High utilizers (top 5%): 2,100 — avg 4.2 chronic conditions, 68% Medicaid
  Rising risk: 5,400 — new diabetes diagnosis or 2+ ED visits in 6 months
  Stable: 34,500

DISPARITY ANALYSIS:
| Measure           | White | Black | Hispanic | Asian | p-value |
|-------------------|-------|-------|----------|-------|---------|
| HbA1c <8%         | 78%   | 64%   | 69%      | 81%   | <0.001  |
| ED visits/1000    | 290   | 480   | 420      | 210   | <0.001  |

HOTSPOTS:
  Zip 10025 — ED utilization 2.1× benchmark, 34% uninsured
  Zip 10031 — Readmission 1.8× benchmark, limited PCP access

INTERVENTION RECOMMENDATIONS:
  1. Community health worker program in 10025/10031 — Target: 800 — Expected: -25% ED — ROI: 3.2:1
  2. Telehealth diabetes coaching for rising risk — Target: 1,200 — Expected: +8% HbA1c control — ROI: 2.1:1
  3. Post-discharge pharmacy delivery — Target: 500/month — Expected: -3% readmission — ROI: 1.8:1