Ir para o conteúdo

climasus4py — English Documentation

climasus4py is a Python package for analysing Brazilian SUS (public health system) microdata, with built-in support for INMET climate data enrichment. It uses DuckDB as the query engine, keeping the entire pipeline lazy until final materialisation — allowing you to process full national datasets with low memory usage.

  • :books: API Reference — Auto-generated documentation for all public functions.
  • :house: Home

Quick example: respiratory mortality in São Paulo (2020–2023)

import climasus4py as cs

# Full pipeline — auto-download + cache + filter + aggregate
result = cs.sus_pipeline(
    system="SIM-DO",
    uf="SP",
    year=[2020, 2021, 2022, 2023],
    groups="respiratory",      # ICD-10 group: J00-J99
    age_min=18,                # adults only
    time="month",              # monthly aggregation
    geo="state",               # state level
    lang="en",                 # English column names
)

# The relation is lazy — materialises only here:
df = result.df()
print(df.head(10))

Available disease groups

Group Description
respiratory Respiratory diseases (J00–J99)
cardiovascular Cardiovascular diseases
dengue Dengue fever (A90, A91)
covid19 COVID-19 (U07.1, U07.2)
diabetes Diabetes mellitus (E10–E14)
neoplasms Neoplasms (C00–D48)
external_causes External causes (V01–Y98)
maternal_causes Maternal causes (O00–O99)
malaria Malaria (B50–B54)
tuberculosis_respiratory Pulmonary tuberculosis
zika_chikungunya Zika and Chikungunya
climate_sensitive Climate-sensitive disease bundle

Supported DATASUS systems

Code System
SIM-DO Death Certificates (mortality)
SINASC Live Births
SINAN-DENGUE Dengue notifications
SIH Hospital Admissions
SIA Outpatient Visits