Skip to contents

Visualizations for Bayesian Spatial Disease Mapping Results

Source: R/sus_mod_plot_spatial_bayes.R
sus_mod_plot_spatial_bayes.Rd

Produces three types of visualization for a climasus_spatial_bayes object returned by sus_mod_spatial_bayes(): a relative risk choropleth map, a credible interval width (uncertainty) choropleth, a forest plot of fixed effects, or a combined panel of both maps.

Usage

sus_mod_plot_spatial_bayes(
  x,
  municipalities = NULL,
  type = c("rr", "uncertainty", "coef", "both"),
  title = NULL,
  base_size = 12,
  lang = "pt",
  ...
)

Arguments

x

A climasus_spatial_bayes object from sus_mod_spatial_bayes().

municipalities

An sf object containing municipality polygons with a code_muni column (7-digit IBGE code). Required for type = "rr", "uncertainty", and "both". Typically obtained via geobr::read_municipality().

type

Character. One of "rr" (default), "uncertainty", "coef", or "both". See Plot types.

title

Character or NULL. Custom plot title. When NULL (default), a multilingual default is used.

base_size

Numeric. Base font size passed to ggplot2 themes. Default 12.

lang

Character. Language for axis labels and titles: "pt" (default), "en", or "es".

...

Currently unused; reserved for future arguments.

Value

  • For type = "rr", "uncertainty", "coef": a ggplot object.

  • For type = "both": a patchwork object when patchwork is installed, otherwise a named list with elements $rr and $uncertainty.

Plot types (type)

"rr"

Choropleth map of the smoothed posterior mean relative risk (rr_mean). Uses a diverging colour scale centred at 1.0 (navy = below, white = 1.0, red = above). Municipalities where the entire 95\ interval lies above 1 (rr_lower95 > 1) receive a black border to flag significantly elevated risk.

"uncertainty"

Choropleth of the 95\ (rr_upper95 - rr_lower95), rendered with the plasma palette from ggplot2::scale_fill_viridis_c(). Highlights areas with high posterior uncertainty, often due to small populations or sparse data.

"coef"

Horizontal forest plot of the fixed-effect posterior summaries stored in x$fixed. A dashed vertical reference line marks zero. Points show the posterior mean; horizontal bars show the 95\ credible interval.

"both"

Side-by-side panel combining the "rr" and "uncertainty" maps. Uses patchwork when available; falls back to a named list with $rr and $uncertainty otherwise.

See also

sus_mod_spatial_bayes(), sus_mod_spatial_weights()

Examples

if (FALSE) { # \dontrun{
library(climasus4r)
library(geobr)

shp <- geobr::read_municipality(code_muni = "all", year = 2022)
W   <- sus_mod_spatial_weights(shp, return_matrix = TRUE)

result <- sus_mod_spatial_bayes(
  df         = agg,
  outcome    = "deaths",
  W          = W,
  covariates = c("temp_mean", "prec_total"),
  offset     = "expected_deaths"
)

# Relative risk map
sus_mod_plot_spatial_bayes(result, municipalities = shp, type = "rr")

# Uncertainty map
sus_mod_plot_spatial_bayes(result, municipalities = shp, type = "uncertainty")

# Forest plot of fixed effects
sus_mod_plot_spatial_bayes(result, type = "coef", lang = "en")

# Combined panel (requires patchwork)
sus_mod_plot_spatial_bayes(result, municipalities = shp, type = "both")
} # }