Overview
The ADI Master Dashboard monitors climate-health decision intelligence across 194 WHO member states using 25 composite KPI indicators computed from 7881 raw variables. Each KPI aggregates multiple raw data sources into a single 0-100 score, enabling cross-country comparison and resource prioritization.
KPIs are organized along two dimensions: category (who uses the indicator) and functional role (what aspect of risk it measures). Six analytical frameworks layer on top of the KPIs to provide strategic decision support.
Category × Role Matrix
| Hazard | Vulnerability | Readiness | Burden | Macro | Total | |
|---|---|---|---|---|---|---|
| Universal | 1 | 1 | 3 | 1 | — | 6 |
| Donor | — | 2 | 2 | — | 2 | 6 |
| Government | — | — | 4 | 2 | — | 6 |
| INGO | — | 2 | — | 2 | — | 4 |
| Private Sector | 1 | — | — | 1 | 1 | 3 |
KPI Reference
UniversalTracked across all country contexts6 indicators
DonorSupporting evidence-based resource allocation6 indicators
GovernmentSupporting national health system strengthening6 indicators
INGOInforming crisis response and preparedness4 indicators
Private SectorInforming commercial and investment decisions3 indicators
Analytical Frameworks
Data Sources
The 25 KPIs draw from 26 unique data sources, aggregated below by the number of KPIs each source feeds into.
World Bank Development Indicators
17 KPIsClimate-Health Vulnerability Score, Economic Resilience Index, Child Vulnerability Index, Pandemic Preparedness Index, Aid Effectiveness Index, Equity Gap Score, Fragility Index, SDG 3 Progress Tracker, Resource Mobilization Score, Healthcare Workforce Density, NCD Risk Index, Primary Healthcare Strength, Humanitarian Need Severity, Access to Care Barriers, Protection Risk Score, Supply Chain Vulnerability, Market Opportunity Score
WHO Global Health Observatory
12 KPIsHuman Capital Risk Index, Climate-Health Vulnerability Score, Universal Health Coverage Index, Pandemic Preparedness Index, Aid Effectiveness Index, Financial Risk Protection Index, Equity Gap Score, SDG 3 Progress Tracker, Essential Medicines Availability, NCD Risk Index, Access to Care Barriers, Market Opportunity Score
IHME Global Burden of Disease
3 KPIsHuman Capital Risk Index, NCD Risk Index, Workforce Health Index
UNICEF Data
3 KPIsChild Vulnerability Index, Equity Gap Score, SDG 3 Progress Tracker
WHO Global Health Expenditure Database
2 KPIsEconomic Resilience Index, Resource Mobilization Score
World Bank Human Capital Project
1 KPIHuman Capital Risk Index
ERA5 Climate Reanalysis (ECMWF)
1 KPIClimate-Health Vulnerability Score
WHO/World Bank UHC Monitoring Report
1 KPIUniversal Health Coverage Index
UNICEF/DHS/MICS Survey Data
1 KPIUniversal Health Coverage Index
WHO IHR SPAR (State Party Reports)
1 KPIPandemic Preparedness Index
World Bank Living Standards Measurement Studies
1 KPIFinancial Risk Protection Index
World Bank Worldwide Governance Indicators
1 KPIFragility Index
UN SDG Global Database
1 KPISDG 3 Progress Tracker
WHO National Health Workforce Accounts
1 KPIHealthcare Workforce Density
IHME Global Burden of Disease Study
1 KPIDisease Burden Index
WHO Global Health Estimates
1 KPIDisease Burden Index
WHO/UNICEF Joint Monitoring Programme
1 KPIWASH Coverage Index
World Bank WASH Data
1 KPIWASH Coverage Index
WHO PHCPI Vital Signs Profiles
1 KPIPrimary Healthcare Strength
WFP Vulnerability Analysis and Mapping
1 KPINutrition Crisis Index
WHO/UNICEF Nutrition Data
1 KPINutrition Crisis Index
UNICEF Data Warehouse
1 KPINutrition Crisis Index
ILO Occupational Safety & Health Data
1 KPIWorkforce Health Index
World Bank Labor Statistics
1 KPIWorkforce Health Index
World Bank Logistics Performance Index
1 KPISupply Chain Vulnerability
ERA5 Climate Data
1 KPISupply Chain Vulnerability
Data Governance
Normalization
Raw variables are normalized to a 0-100 scale before aggregation. Three approaches are used depending on indicator type:
- •Min-Max Normalization: Used for continuous variables with defined bounds (e.g., mortality rates, coverage percentages). x_norm = (x - min) / (max - min) × 100.
- •Z-Score / PCA: Used when component weights should reflect variance structure. Principal Component Analysis derives data-driven weights from the first principal component.
- •Geometric Mean: Used for multiplicative composites where zero values in any dimension should dominate (e.g., Human Development Index-style construction).
Directionality Alignment
Before aggregation, all variables are aligned so that higher values represent higher risk (or lower performance). Protective indicators are inverted:
- •Higher-is-worse indicators (e.g., mortality rate, pollution): used as-is after normalization.
- •Higher-is-better indicators (e.g., UHC coverage, life expectancy): inverted via x_risk = 100 - x_norm before aggregation.
Missing Data Handling
When raw variables are unavailable for a country, the following imputation strategies are applied in order of preference:
- •Regional median imputation: Use the median value from the same WHO region.
- •Income-group median fallback: If regional data is also sparse, use the median from the same World Bank income group.
- •Exclusion with transparency: If imputation is not defensible, the country is excluded from that KPI and flagged as partial data.
Weighting Methodology
Component weights vary by KPI design:
- •PCA-derived weights: Data-driven, reflecting variance structure (used by Human Capital Risk Index, Climate-Health Vulnerability Score).
- •Equal weights: When no theoretical basis exists for differential weighting (used by simpler composites).
- •Expert/MCDA weights: Multi-Criteria Decision Analysis with stakeholder input (used by Priority Engine dimension weights).
Update Cadence
KPI update frequency depends on underlying data source availability:
- •Monthly: Climate and environmental indicators (ERA5 reanalysis, OpenAQ).
- •Annual: Most health and economic indicators (WHO, World Bank, IHME).
- •Every 2-5 years: Survey-based indicators (DHS, MICS) and census data.
- •Real-time recomputation: Composite scores, priority rankings, and matrix positions are recomputed on each dashboard load from the latest available data.