Log inSubmit server
SUS Brasil MCP logo

SUS Brasil MCP

by SIH

Query Brazil's SUS hospital admissions by cause, geography, demographics, ICSAP, trends, and population-adjusted rates.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install SUS Brasil MCP

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "sih-sus-brasil-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "sih-br-mcp"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Install not yet sandbox-tested— the server itself is confirmed live (we called its tools/list endpoint). The stdio install command is a separate, ongoing check.

Overview

SIH/SUS Brasil MCP connects AI assistants to aggregated DATASUS SIH/SUS hospital admission data for Brazil from 1992 through 2025. Its 12 tools cover causes, monthly and annual trends, regional comparisons, ICSAP classification and indicators, rankings, and hospitalization rates. Queries can be filtered or grouped by year, month, UF, municipality, CID, sex, age, and race when the underlying year supports those fields. Reach for it when you need auditable aggregate analysis without downloading DBC files, decoding microdata, or writing SQL.

Use cases

  • Count hospitalizations by diagnosis, age, sex, or geography
  • Compare hospitalization rates between Brazilian states or regions
  • Track monthly or annual admission trends
  • Rank conditions sensitive to primary care by admissions or cost
  • Classify CID-10 codes and calculate ICSAP indicators

Key features

  • DATASUS SIH/SUS aggregates from 1992–2025
  • CID-9 and CID-10 historical diagnosis support
  • ICSAP classification, rankings, and indicators
  • Monthly, annual, regional, and comparative analyses
  • Crude, age-specific, and age-standardized rates
  • Cube caching with provenance and freshness metadata

Tools12

What SUS Brasil MCP lets your agent do. Expand a tool for its parameters.

Verified live~3.6k tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 1d ago

Recent health check succeeded.

9/9 checks healthy over the last 45d

What it can do

From the 12 tools this server returned when we ran it; 12 declare their own safety hints.

Makes network requests · 1Read-only tools · 12
Network at startup3 connections
Files written at startupNone
Install scriptsprepare
npm provenanceSigned build
Package age51 days
Maintainers1

Observed 2026-10-30 in an isolated sandbox. A description of reach, not a security audit.

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 14d ago via OSV.dev · sih-br-mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars1Stargazers on the source repository.
npm downloads1.8kPackage downloads in the last 30 days.
Last commit3d agoMost recent push to the default branch.
Tools exposed12Callable tools this server registers over MCP.

How SUS Brasil MCP works

What SIH/SUS Brasil MCP does

SIH/SUS Brasil MCP exposes aggregated Brazilian SUS hospital admission data through the Model Context Protocol. The dataset covers 1992–2025 and includes admissions, length of stay, SUS-paid value, and deaths across supported dimensions such as year, month, UF, municipality, diagnosis classification, sex, age, race or color, and ICSAP group.

The server is designed for questions such as counting pneumonia admissions by age and state, comparing hospitalization rates between regions, finding the leading avoidable-condition groups in a municipality, or examining a monthly dengue series. Responses include data provenance and source citations, helping users inspect the origin and revision status of returned figures.

How it works

The server consumes annual SIH/SUS cubes published through the healthbr-data project. The source pipeline starts with DATASUS files, converts them to Parquet, builds aggregated cubes, and publishes those cubes with manifests and provenance sidecars. The MCP server downloads only the years needed by a request and keeps them in a local cache. SHA-256 checks are used to verify downloaded cubes against the published manifest.

Some supported queries can use derived summaries instead of full annual cubes. Requests that require dimensions unavailable in a summary are routed to the relevant cube. Responses can indicate whether a result is provisional, whether a cube is behind the upstream mirror, and how the data was retrieved.

Setup and configuration

Run the package with:

Terminal
npx -y sih-br-mcp

The server uses standard MCP client configuration and does not require a database or a local SQL system. Downloaded cubes are cached under ~/.cache/sih-br-mcp/cubos/ by default.

Optional configuration includes:

  • SIH_CACHE_DIR changes the local cache directory.
  • SIH_CUBES_BASE_URL selects another cube source.
  • SIH_CUBES_CACHE=off disables cube caching, which is useful for smoke or golden tests.

Tools and capabilities

The server provides 12 tools:

  • List the 19 Brazilian CSAP groups and their CID-10 mappings.
  • List CID-10 chapters and diagnosis ranges.
  • Inspect available years, population coverage, data eras, and cube freshness.
  • Query hospitalizations with multidimensional filters and aggregations.
  • Produce monthly or annual hospitalization trends.
  • Compare admissions across Brazilian states or regions.
  • Query ICSAP admissions by group and demographic or geographic filters.
  • Calculate ICSAP percentages and related indicators.
  • Rank CSAP groups by admissions, days, or paid value.
  • Classify CID-10 codes as CSAP or non-CSAP.
  • Calculate crude, age-specific, or age-standardized hospitalization rates.
  • Compare ICSAP trends across states or CSAP groups.

Limitations and notes

This service returns aggregates rather than individual AIH records. It is not intended for analyses requiring procedure details, establishment CNES, attendance character, secondary diagnoses, or data from other DATASUS systems such as SIM, SINASC, SIA, or SINAN.

Historical fields require careful interpretation. From 1992–1997, diagnoses use CID-9 data and the UF represents the file or establishment location rather than residence; 1997 contains both CID-9 and CID-10 records. The early ICSAP classification is derived and non-official, with some groups not comparable to later years. Race or color is available from 2008 onward. Population-based age filters before 2000 are constrained by five-year population bands. Recent years may remain provisional while delayed admissions are incorporated, and DATASUS can revise past files.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about SUS Brasil MCP

SIH/SUS Brasil MCP is an MCP server that connects AI assistants to aggregated DATASUS SIH/SUS hospital admission data from Brazil. Its main tools query hospitalizations, trends, regional comparisons, ICSAP groups and indicators, CID classifications, and population-based hospitalization rates for data spanning 1992–2025.

Run `npx -y sih-br-mcp` from an MCP client configuration. The server downloads requested annual cubes on demand and stores them in `~/.cache/sih-br-mcp/cubos/` unless `SIH_CACHE_DIR` or `SIH_CUBES_CACHE` changes that behavior.

Yes, it calculates crude, age-specific, and age-standardized rates using population data supplied through the same data channel. Rates can use bases of 1,000, 10,000, or 100,000, with 100,000 as the default.

No, it provides aggregated hospital admission data rather than individual AIH records. Use a microdata workflow when you need procedure details, CNES, attendance character, secondary diagnoses, or variables outside the published cubes.

Interpret pre-1998 results as a separate historical era: diagnoses use CID-9, 1992–1997 UF refers to the establishment file rather than residence, and ICSAP uses a derived CID-9 list. Race data is unavailable for those years, and 1997 includes both diagnosis revisions.

Related MCP servers

All in Biology & Bioinformatics All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewSUS Brasil MCP AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/sih-sus-brasil-mcp?style=directory)](https://allmcps.com/mcp/sih-sus-brasil-mcp)
HTML Embed
<a href="https://allmcps.com/mcp/sih-sus-brasil-mcp"><img src="https://allmcps.com/api/badge/sih-sus-brasil-mcp?style=directory" alt="SUS Brasil MCP on AllMCPs" /></a>