DATASUS - Mortality Data (SIM)

The load_mortality function provides access to the System of Mortality Information (SIM) datasets, which contain detailed information about deaths in Brazil. Each original SIM data file includes rows corresponding to a declaration of death (DO) and columns with several characteristics of the person, the place of death, and the cause of death.

The load_mortality function offers the following parameters:

1. dataset: Specifies the SIM dataset to download:

   • SIM Datasets:
     • "general" – Main Declarations of Death. (National dataset available — states = "all") Contains records of all non-fetal Death Certificates (DO) in Brazil, including socio-demographic data, location, and causes of death (ICD-10). It’s the base for general mortality analysis. (since 1979 to present)
     • "fetal" – Fetal mortality data. (National dataset not available) Contains records of fetal deaths, with information on the mother, pregnancy, and causes of fetal death. It’s essential for maternal and child health. (since 1979 to present)
     • "external_causes" – Mortality data from external causes. (National dataset not available) Contains a subset of "general" focusing on deaths due to accidents, violence, and other unnatural causes. Used for safety and prevention studies. (since 1979 to present)
     • "infant" – Infant mortality data (children). (National dataset not available) Contains a subset of "general" recording deaths of children under 1 year old, detailing causes and birth-related factors. Crucial for assessing child health. (since 1979 to present)
     • "maternal" – Maternal mortality data. (National dataset not available) Contains a subset of "general" for deaths of women during or shortly after pregnancy/childbirth, detailing obstetric causes. Important for women’s health. (since 1996 to present)

2. time_period: a numeric value or vector indicating the year(s) of the data to be downloaded. For example, 2020 or 2015:2020.

3. states: (valid only for the general dataset) — a string or a vector of strings indicating the Brazilian state(s) for which the data should be downloaded. The default is "all", which downloads data for the entire country. For specific states, use the official abbreviations such as "SP" (São Paulo), "RJ" (Rio de Janeiro), or c("SP", "RJ").

4. raw_data: Logical, default is FALSE.

   • TRUE: If TRUE, returns the raw data exactly as provided by DATASUS.
   • FALSE: If FALSE (default), returns a cleaned and standardized version of the dataset.

5. keep_all: A boolean choosing whether to aggregate the data by municipality, losing individual-level variables (FALSE) or to keep all original variables (TRUE). Only applies when raw_data is FALSE.

6. language: A string indicating the desired language of variable names and labels. Accepts "eng" (default) for English or "pt" for Portuguese (only when raw_data = FALSE).

Examples:

library(datazoom.saude)

# Download raw data for general mortality - State of Rio de Janeiro, 2022.
raw_data_general_rj <- load_mortality(
  dataset = "general",
  time_period = 2022,
  states = "RJ",
  raw_data = TRUE
)

# Download treated data for general mortality - States of Rio and São Paulo, 2022.
trated_data_general_rj <- load_mortality(
  dataset = "general",
  time_period = 2022,
  states = c("RJ", "SP"),
  raw_data = FALSE,
  keep_all = FALSE # Explicitly stating default behavior
)

# Download treated data for Maternal Deaths - Brazil, 2020 to 2022.
# Descriptions in Portuguese.
# Note: `maternal` does not provide separate files by state.
data_maternal_pt <- load_mortality(
  dataset = "maternal",
  time_period = 2020:2022,
  states = "all",
  raw_data = FALSE,
  language = "pt"
)

# Download treated data for Infant Deaths - Brazil, 2017.
# Keeping all individual variables (not aggregated).
data_infant_full <- load_mortality(
  dataset = "infant",
  time_period = 2017,
  states = "all",
  raw_data = FALSE,
  keep_all = TRUE,
  language = "eng"
)
  
# Download treated data for Fetal Deaths - State of Amazonas, 2000.
data_infant_full <- load_mortality(
  dataset = "fetal",
  time_period = 2000,
  states = "AM", 
  raw_data = FALSE,
  language = "eng"
)

# Download treated data for External Causes Deaths - State of Acre, 2022.
data_infant_full <- load_mortality(
  dataset = "fetal",
  time_period = 2022,
  states = "AC", 
  raw_data = FALSE,
  language = "eng"
)