@pipeworx/who-gho

Connect: https://gateway.pipeworx.io/who-gho/mcp · Install: one-click buttons

Tools: 3

The World Health Organization’s GHO database. ~3,000 health-related indicators across 194 member states: mortality, disease prevalence, healthcare workforce, immunization, environmental health, NCDs, communicable diseases, demographics. The canonical international health-data source. Free, no auth.

Why this matters for AI agents

For international comparisons of health indicators or country-level public-health snapshots, WHO GHO is the source. Where CDC is US-focused, WHO is global. Pair with World Bank (development indicators) and IMF (macro) for full country-level analysis.

Common flows:

• Country indicator. “Life expectancy in Brazil?” → indicator + country query.
• Cross-country comparison. Same indicator across multiple countries.
• Time series. Indicator over years for trend analysis.
• Indicator browse. “What does WHO publish on diabetes?” → search the indicator catalog.

Auth

None. WHO GHO is fully public, free.

Indicator categories

Major categories (each with dozens to hundreds of indicators):

• Mortality and life expectancy
• Communicable diseases (HIV, TB, malaria, COVID-19, vaccine-preventable)
• Non-communicable diseases (cardiovascular, cancer, diabetes, mental health)
• Maternal and child health
• Health workforce (physicians, nurses, beds per 1000)
• Environmental health (air pollution, water/sanitation access)
• Health systems financing
• Risk factors (tobacco, alcohol, BMI, blood pressure)

Common pitfalls

• Country reporting quality varies. Wealthy countries report comprehensively; lower-income countries have data gaps and longer lags. Some indicators are WHO-modeled estimates filling country reporting gaps.
• Disaggregation availability. “Indicator X for country Y” may not break down by sex, age, or urban/rural. Check whether the disaggregation you want exists before building queries that depend on it.
• Definition shifts. WHO occasionally revises indicator methodology (e.g., changing definition of “stunting” cut-points). Long time series across methodology changes need annotation.
• Lag. Most WHO data lags 1-3 years. Recent year may have only modeled estimates. For real-time outbreak data, use WHO’s separate disease-surveillance feeds.
• Country naming. WHO uses ISO 3-letter codes. Some politically-disputed entities (Taiwan, Palestine, Kosovo) have inconsistent treatment in headline data; check coverage explicitly.
• Population denominator. Per-capita rates are usually computed against UN population estimates. Different sources (UN vs. national stats) can differ slightly, especially for fast-growing populations.
• WHO regions. WHO groups countries into 6 regions (Africa, Americas, Eastern Mediterranean, Europe, Southeast Asia, Western Pacific). These don’t match World Bank or other regional groupings.

Tools

• get_indicators — Search or list WHO Global Health Observatory indicators. Returns indicator codes and names. Use the indicator code with get_data to retrieve actual values. Example: get_indicators(“life expectancy”) o
• get_data — Get World Health Organization health statistics for a country and year — life expectancy, immunization coverage, mortality, disease burden, risk factors and health system measures from the Global Heal
• list_countries — Return all WHO-recognized countries with their 3-letter ISO code, display name, and parent region code. Use the ISO codes returned here as the country parameter in get_data (e.g., “USA”, “GBR”, “JPN”)