climatehealth: Statistical Tools for Modelling Climate-Health Impacts

Description

Overview

This package provides a suite of analysis functions for measuring the relationship between various climate factors (indicators) and health outcomes.

Included Indicators

• Mortality attributable to high and low outdoor temperatures

• Mortality attributable to wildfire-related PM2.5

• Suicides attributable to extreme heat

• Mortality attributable to short-term exposure to outdoor PM2.5 exposure

• Diarrhea cases attributable to extreme temperatures and rainfall

• Malaria cases attributable to extreme temperatures and rainfall

License

MIT

The full range of topics include

• Temperature-related health effects

• Health effects of wildfires

• Mental Health

• Health effects of air pollution

• Water-borne diseases

• Vector-borne diseases

Author(s)

Maintainer: Kenechi Omeke climate.health@ons.gov.uk

Authors:

• Charlie Browning

• Etse Yawo Dzakpa

• Gladin Jose

• Matt Pearce

• Ellie Watkins

• Claire Hunt

• Beatrice Byukusenge

• Cassien Habyarimana

• Venuste Nyagahakwa

• Felix Scarbrough

• Treesa Shaji

• Bonnie Lewis

• Maquines Odhiambo Sewe

• Vijendra Ingole

Other contributors:

• Sean Lovell [contributor]

• Antony Brown [contributor]

• Euan Soutter [contributor]

• Gillian Flower [contributor]

• David Furley [contributor]

• Joe Panes [contributor]

• Charlotte Romaniuk [contributor]

• Milly Powell [contributor]

• Wellcome [funder]

• Office for National Statistics (SOSCHI Project) [copyright holder]

See Also

Useful links:

• https://climate-health.officialstatistics.org

English day of week names

Description

Provides consistent English day names regardless of system locale

Usage

.english_dow_names(day_numbers = NULL, short = FALSE)

Arguments

Value

Character vector of day names

English month names

Description

Provides consistent English month names regardless of system locale

Usage

.english_month_names(month_numbers = NULL, short = FALSE)

Arguments

Value

Character vector of month names

Temporarily set English locale for date operations

Description

Temporarily sets the locale to English for date parsing and formatting

Usage

.with_english_locale(expr)

Arguments

Value

Result of the expression

Raise a typed error with structured metadata

Description

Creates a classed condition that can be caught and inspected by the API layer. This is the base helper - prefer using specific helpers like abort_column_not_found() or abort_validation() when applicable.

Usage

abort_climate(message, type = "generic_error", ..., call = rlang::caller_env())

Arguments

Value

Never returns; always raises an error.

Examples

# Basic usage
err <- tryCatch(
  abort_climate("Something went wrong", "generic_error"),
  error = identity
)
inherits(err, "climate_error")

# With metadata
err <- tryCatch(
  abort_climate(
    "Invalid lag value",
    "validation_error",
    param = "nlag",
    value = -1,
    expected = "non-negative integer"
  ),
  error = identity
)
err$type

Raise a column-not-found error with available columns

Description

Use this when a required column is missing from a dataset. Includes fuzzy matching to suggest the closest available column name.

Usage

abort_column_not_found(
  column,
  available,
  dataset_name = "dataset",
  call = rlang::caller_env()
)

Arguments

Value

Never returns; always raises an error.

Examples

data <- data.frame(temp = 1)
if (!("tmean" %in% colnames(data))) {
  err <- tryCatch(
    abort_column_not_found("tmean", colnames(data)),
    error = identity
  )
  err$suggestion
}

Raise a model error (statistical/computational failures)

Description

Use this when statistical models fail to converge, produce singular matrices, or encounter other computational issues that aren't due to obvious user error.

Usage

abort_model_error(
  message,
  model_type = "unknown",
  ...,
  call = rlang::caller_env()
)

Arguments

Value

Never returns; always raises an error.

Examples

tryCatch({
  stop("convergence failed")
}, error = function(e) {
  err <- tryCatch(
    abort_model_error(
    "Model failed to converge",
    model_type = "dlnm",
    original_error = conditionMessage(e)
    ),
    error = identity
  )
  inherits(err, "model_error")
})

Raise a validation error (data/parameter issues)

Description

Use this for general validation failures where the user's input or data doesn't meet requirements. For missing columns specifically, use abort_column_not_found().

Usage

abort_validation(message, ..., call = rlang::caller_env())

Arguments

Value

Never returns; always raises an error.

Examples

# Parameter validation
nlag <- -1
if (nlag < 0) {
  err <- tryCatch(
    abort_validation(
      "nlag must be >= 0",
      param = "nlag",
      value = nlag,
      expected = "non-negative integer"
    ),
    error = identity
  )
  inherits(err, "validation_error")
}

Aggregate air pollution results by month

Description

Aggregates daily analysis results to monthly summaries

Usage

aggregate_air_pollution_by_month(
  analysis_results,
  max_lag = 14L,
  include_national = TRUE
)

Arguments

Value

Dataframe with monthly aggregates

Aggregate air pollution results by region

Description

Aggregates daily analysis results to regional summaries

Usage

aggregate_air_pollution_by_region(analysis_results, max_lag = 14L)

Arguments

Value

Dataframe with regional aggregates

Aggregate air pollution results by year

Description

Aggregates daily analysis results to annual summaries

Usage

aggregate_air_pollution_by_year(
  analysis_results,
  max_lag = 14L,
  include_national = TRUE
)

Arguments

Value

Dataframe with annual aggregates

Split dataframe into multiple dataframes, based on a columns value.

Description

Split dataframe into multiple dataframes, based on a columns value.

Usage

aggregate_by_column(df, column_name)

Arguments

Value

A list of dataframes, split up based on the value of column_name.

Descriptive statistics

Description

Generates summary statistics for climate, environmental and health data

Usage

air_pollution_descriptive_stats(
  data,
  env_labels = c(pm25 = "PM2.5 (µg/m³)", tmax = "Max Temperature (°C)", precipitation
    = "Precipitation (mm)", humidity = "Humidity (%)", wind_speed = "Wind Speed (m/s)"),
  save_outputs = FALSE,
  output_dir = NULL,
  moving_average_window = 3L,
  plot_corr_matrix = FALSE,
  correlation_method = "pearson",
  plot_dist = FALSE,
  plot_na_counts = FALSE,
  plot_scatter = FALSE,
  plot_box = FALSE,
  plot_seasonal = FALSE,
  plot_regional = FALSE,
  plot_total = FALSE,
  detect_outliers = FALSE,
  calculate_rate = FALSE
)

Arguments

Value

Invisibly returns the national data with moving averages

Comprehensive Air Pollution Analysis Pipeline

Description

Master function that runs the complete air pollution analysis including data loading, preprocessing (including lags), modeling, plotting, attribution calculations vs reference standards, power analysis and descriptive statistics

Usage

air_pollution_do_analysis(
  data_path,
  date_col = "date",
  region_col = "region",
  pm25_col = "pm25",
  deaths_col = "deaths",
  population_col = "population",
  humidity_col = "humidity",
  precipitation_col = "precipitation",
  tmax_col = "tmax",
  wind_speed_col = "wind_speed",
  categorical_others = NULL,
  continuous_others = NULL,
  Categorical_Others = NULL,
  Continuous_Others = NULL,
  max_lag = 14L,
  df_seasonal = 6,
  family = "quasipoisson",
  reference_standards = list(list(value = 15, name = "WHO")),
  output_dir = "air_pollution_results",
  save_outputs = TRUE,
  run_descriptive = TRUE,
  run_power = TRUE,
  moving_average_window = 3L,
  include_national = TRUE,
  years_filter = NULL,
  regions_filter = NULL,
  attr_thr = 95,
  plot_corr_matrix = TRUE,
  correlation_method = "pearson",
  plot_dist = TRUE,
  plot_na_counts = TRUE,
  plot_scatter = TRUE,
  plot_box = TRUE,
  plot_seasonal = TRUE,
  plot_regional = TRUE,
  plot_total = TRUE,
  detect_outliers = TRUE,
  calculate_rate = FALSE
)

Arguments

Value

List containing:

data

Processed data with lag variables

meta_analysis

Meta-analysis results with AF/AN calculations

lag_analysis

Lag-specific analysis results

distributed_lag_analysis

Distributed lag model results (if requested)

plots

List of generated plots (forest, lags, distributed lags)

power_list

A list containing power information by area

exposure_response_plots

Exposure-response plots for each reference standard (if requested)

reference_specific_af_an

AF/AN calculations specific to each reference standard (if requested)

descriptive_stats

Summary statistics of key variables

Examples

example_data <- data.frame(
  date = seq.Date(as.Date("2020-01-01"), by = "day", length.out = 180),
  province = "Example Province",
  pm25 = stats::runif(180, 8, 35),
  deaths = stats::rpois(180, lambda = 5),
  population = 500000,
  humidity = stats::runif(180, 40, 90),
  precipitation = stats::runif(180, 0, 20),
  tmax = stats::runif(180, 18, 35),
  wind_speed = stats::runif(180, 1, 8)
)
example_path <- tempfile(fileext = ".csv")
utils::write.csv(example_data, example_path, row.names = FALSE)

results <- air_pollution_do_analysis(
  data_path = example_path,
  date_col = "date",
  region_col = "province",
  pm25_col = "pm25",
  deaths_col = "deaths",
  population_col = "population",
  humidity_col = "humidity",
  precipitation_col = "precipitation",
  tmax_col = "tmax",
  wind_speed_col = "wind_speed",
  continuous_others = NULL,
  max_lag = 7L,
  df_seasonal = 4,
  family = "quasipoisson",
  reference_standards = list(list(value = 15, name = "WHO")),
  years_filter = NULL,
  regions_filter = NULL,
  include_national = FALSE,
  output_dir = tempdir(),
  save_outputs = FALSE,
  run_descriptive = FALSE,
  run_power = FALSE,
  moving_average_window = 3L,
  attr_thr = 95,
  plot_corr_matrix = FALSE,
  correlation_method = "pearson",
  plot_dist = FALSE,
  plot_na_counts = FALSE,
  plot_scatter = FALSE,
  plot_box = FALSE,
  plot_seasonal = FALSE,
  plot_regional = FALSE,
  plot_total = FALSE,
  detect_outliers = FALSE,
  calculate_rate = FALSE
)

Perform meta analysis with multiple lag structures

Description

Implements distributed lag model. Individual lag coefficients and cumulative effects are extracted and perform meta analysis

Usage

air_pollution_meta_analysis(
  data_with_lags,
  max_lag = 14L,
  df_seasonal = 6L,
  family = "quasipoisson"
)

Arguments

Value

Dataframe with lag-specific results including for regional and national

Air Pollution Power Calculation using Meta Results

Description

Produce a power statistic by region for PM2.5 attributable mortality using meta-analysis results

Usage

air_pollution_power_list(
  meta_results,
  data_with_lags,
  ref_pm25 = 15,
  attr_thr = 95,
  include_national = TRUE
)

Arguments

Value

A list containing power information by region

FUNCTION FOR COMPUTING ATTRIBUTABLE MEASURES FROM DLNM

Description

A function to calculate attributable numbers and fractions derived from (c) Antonio Gasparrini 2015-2017. Modifications to produce daily values with confidence intervals.

Usage

an_attrdl(
  x,
  basis,
  cases,
  coef = NULL,
  vcov = NULL,
  model.link = NULL,
  dir = "back",
  tot = TRUE,
  cen,
  range = NULL,
  nsim = 5000
)

Arguments

Value

• Attributable Fraction

• Attributable Fraction lower confidence intervals

• Attributable Fraction upper confidence intervals

• Attributable Numbers

• Attributable Numbers lower confidence intervals

• Attributable Numbers upper confidence intervals

• Simulation matrix of attributable numbers

Calculate daily RR/AF/AN/AR for region-specific/national distributed lag effects for a chosen PM2.5 reference.

Description

Calculate daily RR/AF/AN/AR for region-specific/national distributed lag effects for a chosen PM2.5 reference.

Usage

analyze_air_pollution_daily(
  data_with_lags,
  meta_results,
  ref_pm25 = 15,
  ref_name = "WHO",
  max_lag = 14L
)

Arguments

Value

List with region-specific/national results for daily RR/AF/AN/AR

FUNCTION FOR COMPUTING ATTRIBUTABLE MEASURES FROM DLNM

Description

A function to calculate attributable numbers and fractions derived from (c) Antonio Gasparrini 2015-2017.

Usage

attrdl(
  x,
  basis,
  cases,
  model = NULL,
  coef = NULL,
  vcov = NULL,
  model.link = NULL,
  type = "af",
  dir = "back",
  tot = TRUE,
  cen,
  range = NULL,
  sim = FALSE,
  nsim = 5000
)

Arguments

Value

Attributable Numbers and Fractions

Calculate Attributable Metrics for Climate-Health Associations.

Description

Computes the attributable number, fraction, and rate of cases associated with specific exposure variables (e.g., temperature or rainfall) using fitted INLA models. The function estimates these metrics at the desired spatial aggregation level (country, region, or district) and optionally disaggregates by month or year.

Usage

attribution_calculation(
  data,
  param_term,
  model,
  level,
  param_threshold = 1,
  max_lag,
  nk,
  filter_year = NULL,
  group_by_year = FALSE,
  case_type,
  output_dir = NULL,
  save_csv = FALSE
)

Arguments

Value

A tibble containing the following columns:

• Grouping variables depending on the level and group_by_year settings.

• MRT: Minimum risk temperature (or equivalent reference exposure).

• AR_Number, AR_Number_LCI, AR_Number_UCI: Estimated, lower, and upper bounds of the attributable number of cases.

• AR_Fraction, AR_Fraction_LCI, AR_Fraction_UCI: Estimated, lower, and upper bounds of the attributable fraction (%).

• AR_per_100k, AR_per_100k_LCI, AR_per_100k_UCI: Estimated, lower, and upper bounds of the attributable rate per 100,000 population.

Generate a grid size for a certain number of plots.

Description

Generate a grid size for a certain number of plots.

Usage

calculate_air_pollution_grid_dims(n_plots)

Arguments

Value

A list containing ncol and nrow values for the grid.

Calculate attributable numbers and fraction of a given health outcome.

Description

Takes a calculated RR and upper and lower CIs, and applies these to the input data to calculate attributable fraction and attributable number, along with upper and lower CIs, for each day in the input data. Uses Lag 1 RR and lower/upper CIs.

Usage

calculate_daily_AF_AN(data, rr_data)

Arguments

Value

A dataframe containing a daily timseries of AF and AN, including upper and lower confidence intervals.

QAIC calculation

Description

Computes the Quasi-Akaike Information Criterion (QAIC) for models, enabling model comparison

Usage

calculate_qaic(
  data,
  save_csv = FALSE,
  output_folder_path = NULL,
  print_results = FALSE
)

Arguments

Value

Dataframe containing QAIC results for each lag.

Passes data to casecrossover_quasipoisson to calculate RR.

Description

Splits data by region if relative_risk_by_region==TRUE. If TRUE, data for each individual region is passed to casecrossover_quasipoisson to calculate RR by region. If FALSE, RR is calculated for the entire dataset.

Usage

calculate_wildfire_rr_by_region(
  data,
  scale_factor_wildfire_pm,
  calc_relative_risk_by_region = FALSE,
  save_fig = FALSE,
  output_folder_path = NULL,
  print_model_summaries = FALSE
)

Arguments

Value

Dataframe of relative risk and confidence intervals for each lag of wildfire-related PM2.5. Split by region if calc_relative_risk_by_region set to TRUE.

Fit quasipoisson regression models for different lags using a time-stratified case-crossover approach.

Description

Fits quasipoisson regression models using gnm

Usage

casecrossover_quasipoisson(
  data,
  scale_factor_wildfire_pm = 10,
  wildfire_lag,
  save_fig = TRUE,
  output_folder_path = NULL,
  print_model_summaries = TRUE
)

Arguments

Value

Dataframe of relative risk and confidence intervals for each lag of wildfire-related PM2.5

Check multicollinearity using VIF and write the results to file

Description

This function runs check_diseases_vif(), reshapes the result into a tabular data frame, and optionally writes the table to VIF_results.csv.

Usage

check_and_write_vif(data, param_term, inla_param, case_type, output_dir = NULL)

Arguments

Value

A data frame with columns variable, VIF, and interpretation.

Check multicollinearity using VIF on model variables

Description

This function checks multicollinearity across the disease outcome, the exposure term(s) of interest, and the additional INLA covariates using a correlation-matrix-based variance inflation factor calculation.

Usage

check_diseases_vif(data, param_term, inla_param, case_type)

Arguments

Value

A list with:

variables

Character vector of variables used in the VIF calculation.

vif

Numeric vector of VIF values aligned to variables.

vif_interpretation

Character vector of qualitative VIF interpretations ("Low", "Moderate", "High", or "Not computed").

Check if a dataframe is empty.

Description

Checks if a dataframe is empty, and raises an error if it is.

Usage

check_empty_dataframe(df)

Arguments

Value

NULL. No return if the dataframe is not empty.

Check that a file exists at a passed path.

Description

Checks the files on disk to assert that the passed file is present.

Usage

check_file_exists(fpath, raise = TRUE)

Arguments

Value

'exists'. Whether or not the file exists on disk.

Check that a file extension on a given path matches the expected.

Description

This function takes an expected file extension, and validates it against a user-inputted file path.

Usage

check_file_extension(fpath, expected_ext, param_nm = "fpath", raise = TRUE)

Arguments

Value

Whether or not the passed file has a valid file extension.

Check for Rtools Installation on Windows

Description

Verifies whether Rtools is installed and properly configured on a Windows system.

Usage

check_has_rtools()

Details

The function uses pkgbuild::check_build_tools(debug = TRUE) to test for the presence of Rtools and its integration with R. If Rtools is missing or misconfigured, the function throws an error with installation instructions.

Value

Returns TRUE invisibly if Rtools is detected and functional. Otherwise, throws an error.

See Also

check_build_tools, https://cran.r-project.org/bin/windows/Rtools/

Check variance inflation factors of predictor variables using a linear model

Description

Checks variance inflation factors of predictor variables using a linear model of the predictor variables on the health outcome. Prints stats if print_vif==TRUE. Raises a warning if VIF for a variables is > 2.

Usage

check_wildfire_vif(
  data,
  predictors,
  save_csv = FALSE,
  output_folder_path = NULL,
  print_vif = FALSE
)

Arguments

Value

Variance inflation factor statistics for each predictor variable.

Read in and combine climate and health data

Description

Read and combine climate and health data prepared for the spatiotemporal and DLNM analysis.

Usage

combine_health_climate_data(
  health_data_path,
  climate_data_path,
  map_path,
  region_col,
  district_col,
  date_col,
  year_col,
  month_col,
  case_col,
  case_type,
  tot_pop_col,
  tmin_col,
  tmean_col,
  tmax_col,
  rainfall_col,
  r_humidity_col,
  geometry_col,
  runoff_col = NULL,
  ndvi_col = NULL,
  spi_col = NULL,
  max_lag,
  output_dir = NULL
)

Arguments

Value

A list of dataframes containing the map, nb.map, data, grid_data, summary

Deprecated alias for run_descriptive_stats().

Description

Generic wrapper function to compute descriptive statistics and EDA outputs.

Usage

common_descriptive_stats(
  df_list,
  output_path,
  aggregation_column = NULL,
  population_col = NULL,
  plot_corr_matrix = FALSE,
  correlation_method = "pearson",
  plot_dist = FALSE,
  plot_ma = FALSE,
  ma_days = 100,
  ma_sides = 1,
  timeseries_col = NULL,
  dependent_col,
  independent_cols,
  units = NULL,
  plot_na_counts = FALSE,
  plot_scatter = FALSE,
  plot_box = FALSE,
  plot_seasonal = FALSE,
  plot_regional = FALSE,
  plot_total = FALSE,
  detect_outliers = FALSE,
  calculate_rate = FALSE
)

Arguments

Value

Character vector. Backward-compatible output path format.

Deprecated. Use run_descriptive_stats() instead.

Deprecated alias for run_descriptive_stats_api().

Description

Deprecated alias for run_descriptive_stats_api().

Usage

common_descriptive_stats_api(
  data,
  aggregation_column = NULL,
  population_col = NULL,
  dependent_col,
  independent_cols,
  units = NULL,
  plot_correlation = FALSE,
  plot_dist_hists = FALSE,
  plot_ma = FALSE,
  plot_na_counts = FALSE,
  plot_scatter = FALSE,
  plot_box = FALSE,
  plot_seasonal = FALSE,
  plot_regional = FALSE,
  plot_total = FALSE,
  correlation_method = "pearson",
  ma_days = 100,
  ma_sides = 1,
  timeseries_col = NULL,
  detect_outliers = FALSE,
  calculate_rate = FALSE,
  output_path
)

Arguments

Value

Character vector. Backward-compatible output path format.

Deprecated. Use run_descriptive_stats_api() instead.

Deprecated alias for descriptive_stats_core().

Description

Deprecated. Use descriptive_stats_core() instead.

Usage

common_descriptive_stats_core(
  df,
  output_path,
  title,
  aggregation_column = NULL,
  population_col = NULL,
  plot_corr_matrix = FALSE,
  correlation_method = "pearson",
  plot_dist = FALSE,
  dependent_col,
  independent_cols = c(),
  units = NULL,
  plot_na_counts = FALSE,
  plot_scatter = FALSE,
  plot_box = FALSE,
  plot_seasonal = FALSE,
  plot_regional = FALSE,
  plot_total = FALSE,
  timeseries_col = "date",
  detect_outliers = FALSE,
  calculate_rate = FALSE
)

Create and plot the exposure-lag-response relationship (contour plot) at country, region or district level for each disease cases type (diarrhea and malaria).

Description

: Generates a contour plot showing the exposure-lag-response relationship of the exposure tmax and rainfall and the diseases case type.

Usage

contour_plot(
  data,
  param_term,
  model,
  level,
  max_lag,
  nk,
  case_type,
  filter_year = NULL,
  save_fig = FALSE,
  output_dir = NULL
)

Arguments

Value

contour plot at country, Region and District level

Create lagged values for PM2.5 variable and average lag column.

Description

Creates new variables in a dataframe for lags and means over lag periods.

Usage

create_air_pollution_lags(data, max_lag = 14L)

Arguments

Value

Dataframe with added columns for lagged PM2.5 concentration.

Create statistical summaries of columns in a dataframe.

Description

Create statistical summaries of columns in a dataframe.

Usage

create_column_summaries(df, independent_cols = NULL)

Arguments

Value

Dataframe. Column summaries

Create a correlation matrix for columns in a dataframe.

Description

Create a correlation matrix for columns in a dataframe.

Usage

create_correlation_matrix(
  df,
  independent_cols = NULL,
  correlation_method = "pearson"
)

Arguments

Value

Matrix. Correlation matrix for selected columns in the input dataset.

Generate a grid size for a certain number of plots.

Description

This function calculates the minimum grid size required to plot X amount of plots on a a figure. For example, 6 plots would require a 3x2, where as 7 would require a 3x3, and so on.

Usage

create_grid(plot_count)

Arguments

Value

A numeric vector: c(x, y), where x and y define the grid dimensions.

Create indices for INLA models

Description

: For the INLA model, there is a need to set-up regions index, district index, and year index. This function create these indices using the dataset, ndistrict and nregion.

Usage

create_inla_indices(data, case_type)

Arguments

Value

The modified data with the created indices.

Generate lagged values for predictor (temperature) variables

Description

Generates new variables in a dataframe for lags and means over lag periods.

Usage

create_lagged_variables(data, wildfire_lag, temperature_lag)

Arguments

Value

Dataframe with added columns for lagged temperature and wildfire-related PM2.5 concentration

Create a summary of all NA values in a dataset.

Description

Create a summary of all NA values in a dataset.

Usage

create_na_summary(df, independent_cols = NULL)

Arguments

Value

Dataframe. A summary of NA values in the dataset.

Generate splines for temperature variable

Description

Generates temperature splines for each region

Usage

create_temperature_splines(data, nlag = 0, degrees_freedom = 6)

Arguments

Value

Dataframe with additional column for temperature spline.

Emit a consistent deprecation warning for descriptive stats wrappers.

Description

Emit a consistent deprecation warning for descriptive stats wrappers.

Usage

deprecate_descriptive_stats(old_fn, new_fn)

Arguments

Value

None. Emits a warning.

Save descriptive statistics

Description

Generates summary statistics for climate and health data and saves them to the specified file path.

Usage

descriptive_stats(data, variables, bin_width = 5, output_dir = ".")

Arguments

Value

Prints summary statistics and a histogram of the the outcome variable

Core Functionality for Producing Descriptive Statistics

Description

Core Functionality for Producing Descriptive Statistics

Usage

descriptive_stats_core(
  df,
  output_path,
  title,
  aggregation_column = NULL,
  population_col = NULL,
  plot_corr_matrix = FALSE,
  correlation_method = "pearson",
  plot_dist = FALSE,
  dependent_col,
  independent_cols = c(),
  units = NULL,
  plot_na_counts = FALSE,
  plot_scatter = FALSE,
  plot_box = FALSE,
  plot_seasonal = FALSE,
  plot_regional = FALSE,
  plot_total = FALSE,
  timeseries_col = "date",
  write_outlier_table = FALSE,
  calculate_rate = FALSE
)

Arguments

Value

None. Outputs are written to files.

Detect Outliers Using the IQR Method

Description

Detect Outliers Using the IQR Method

Usage

detect_outliers(df, independent_cols = NULL)

Arguments

Value

Dataframe. Column summaries

Code for calculating Diarrhea disease cases attributable to extreme precipitation and extreme temperature Run Full diarrhea-Climate Analysis Pipeline

Description

The diarrhea_do_analysis function runs the complete analysis workflow by combining multiple functions to analyze the association between diarrhea cases and climate variables. It processes health, climate, and spatial data, fits models, generates plots, and calculates attributable risk.

Usage

diarrhea_do_analysis(
  health_data_path,
  climate_data_path,
  map_path,
  region_col,
  district_col,
  date_col = NULL,
  year_col,
  month_col,
  case_col,
  tot_pop_col,
  tmin_col,
  tmean_col,
  tmax_col,
  rainfall_col,
  r_humidity_col,
  runoff_col,
  geometry_col,
  spi_col = NULL,
  ndvi_col = NULL,
  max_lag = 2,
  nk = 2,
  basis_matrices_choices,
  inla_param,
  param_term,
  level,
  param_threshold = 1,
  filter_year = NULL,
  family = "nbinomial",
  group_by_year = FALSE,
  config = TRUE,
  save_csv = FALSE,
  save_model = TRUE,
  save_fig = FALSE,
  cumulative = FALSE,
  output_dir = NULL
)

Arguments

Value

A list containing:

• Model output from INLA

• Monthly random effects plot

• Yearly random effects plot

• Contour plot

• Relative risk map

• Relative risk plot

• Attributable fraction and number summary

Meta-analysis and BLUPs

Description

Run meta-analysis using temperature average and range as meta predictors. Then create the best linear unbiased predictions (BLUPs).

Usage

dlnm_meta_analysis(
  df_list,
  coef_,
  vcov_,
  save_csv = FALSE,
  output_folder_path = NULL
)

Arguments

Value

• mm A model object. A multivariate meta-analysis model.

• blup A list. BLUP (best linear unbiased predictions) from the meta-analysis model for each region.

• meta_test_res A dataframe of results from statistical tests on the meta model.

Define minimum mortality percentiles and temperatures

Description

Calculate the temperature at which there is minimum mortality risk using the product of the basis matrix and BLUPs.

Usage

dlnm_min_mortality_temp(
  df_list,
  var_fun = "bs",
  var_per = c(10, 75, 90),
  var_degree = 2,
  blup = NULL,
  coef_,
  meta_analysis = FALSE,
  outcome_type = c("temperature", "suicide")
)

Arguments

Value

Percentiles and corresponding temperatures for each geography.

Create population totals

Description

Creates a list of population totals by year and region for use in the attributable rate calculations.

Usage

dlnm_pop_totals(df_list, country = "National", meta_analysis = FALSE)

Arguments

Value

List of population totals by year and region

Power calculation

Description

Produce a power statistic by area for the attributable threshold and above as a reference.

Usage

dlnm_power_list(
  df_list,
  pred_list,
  minperc,
  attr_thr_high = 97.5,
  attr_thr_low = 2.5,
  compute_low = TRUE
)

Arguments

Value

A list containing power information by area.

Run national predictions from meta analysis

Description

Use the meta analysis to create national level predictions

Usage

dlnm_predict_nat(
  df_list,
  var_fun = "bs",
  var_per = c(25, 50, 75),
  var_degree = 2,
  minpercreg,
  mmpredall,
  pred_list,
  country = "National"
)

Arguments

Value

A list containing predictions by region.

Reduce to overall cumulative

Description

Reduce model to the overall cumulative association

Usage

dlnm_reduce_cumulative(
  df_list,
  var_per = c(25, 50, 75),
  var_degree = 2,
  cenper = NULL,
  cb_list,
  model_list
)

Arguments

Value

• coef_ A matrix of coefficients for the reduced model.

• vcov_ A list. Covariance matrices for each region for the reduced model.

Produce variance inflation factor

Description

Produces variance inflation factor for the independent variables.

Usage

dlnm_vif(df_list, independent_cols = NULL)

Arguments

Value

A list. Variance inflation factors for each independent variables by region.

Enforce a file extension on a given path

Description

Ensures that the provided file path ends with the desired file extension.

Usage

enforce_file_extension(path, file_extension)

Arguments

Value

Character. The path with the expected file extension.

Extract metadata from a climate_error

Description

Extracts the structured metadata from a typed climate error for use in API responses or logging.

Usage

extract_error_metadata(error)

Arguments

Value

A list containing the error metadata (type, column, available, etc.)

Extract mean wildfire PM2.5 values for shapefile regions from NetCDF file

Description

Takes a NetCDF file of gridded wildfire data and shapefile for geographical regions and extracts mean values for each shapefile region.

Information on NetCDF files: https://climatedataguide.ucar.edu/climate-tools/NetCDF#:~:text=An%20nc4%20files%20is%20a,readily%20handle%20netCDF%2D4%20files.

We use a daily time series of gridded wildfire-related PM2.5 concentration from the Finnish Meteorological Institute's SILAM-CTM model. This is available open-source: https://doi.org/10.57707/fmi-b2share.d1cac971b3224d438d5304e945e9f16c.

Usage

extract_means_for_geography(
  ncdf_path,
  shp_path,
  region_col = "region",
  output_value_col = "mean_PM"
)

Arguments

Value

Dataframe containing a daily time series with mean wildfire-related PM2.5 values for each region

Fit GAM model

Description

Fit a generalized additive model (mgcv::gam) including pm25 and its lagged variables (pm25_lag1, ..., pm25_lagN)

Usage

fit_air_pollution_gam(
  data_with_lags,
  max_lag = 14L,
  df_seasonal = 6L,
  family = "quasipoisson"
)

Arguments

Value

A list with components:

• model: the fitted mgcv::gam object (or NULL if fit failed)

• coef_table: data.frame with columns: lag (0 for pm25, 1..N for pm25_lag#, and "0-N" for cumulative), pm25_variable, coef, se, ci.lb, ci.ub

• vcov_used_for_cumulative: logical; TRUE if vcov() was used to compute cumulative SE

Calculate p-values for Wald test

Description

Calculate p-values for an explanatory variable.

Usage

fwald(mm, var)

Arguments

Value

A number. The p-value of the explanatory variable.

Generate a run id for descriptive statistics output folders.

Description

Generate a run id for descriptive statistics output folders.

Usage

generate_descriptive_stats_run_id()

Value

Character. Run id in the format YYYYmmdd_HHMMSS_NNNN.

Generate Relative Risk Estimates by Region

Description

Computes relative risk estimates for wildfire-specific PM2.5 exposure across regions as PM values changes.

Usage

generate_rr_pm_by_region(
  data,
  relative_risk_overall,
  scale_factor_wildfire_pm,
  wildfire_lag = 0,
  pm_vals = NULL
)

Arguments

Value

A data frame with relative risk estimates for each region and PM value.

Relative risk estimates across PM2.5 concentrations for a specified lag.

Description

Computes relative risk and confidence intervals across a range of PM2.5 concentrations for a specified wildfire-related lag, using log-linear extrapolation from a reference estimate.

Usage

generate_rr_pm_overall(
  data,
  relative_risk_overall,
  scale_factor_wildfire_pm,
  wildfire_lag = 0,
  pm_vals = NULL
)

Arguments

Value

A data frame with columns: 'pm_levels', 'relative_risk', 'ci_lower', and 'ci_upper', representing estimated relative risk and 95% confidence intervals across the specified PM2.5 levels.

Generate and RGB colour value with alpha from a hex value.

Description

Generate and RGB colour value with alpha from a hex value.

Usage

get_alpha_colour(hex, alpha)

Arguments

Value

The converted RGB colour.

Create lagged columns and provide the mean value.

Description

Creates new columns containing lagged values over n rows and determine the mean of the lagged column.

Usage

get_lags_and_means(data, lagcol, nlags)

Arguments

Value

Dataframe with added columns for lagged values and mean(s) of those lags.

A function to predict relative risk at country, region, and district level

Description

Produces cumulative relative risk at country, region and district level from analysis.

Usage

get_predictions(data, param_term, max_lag, nk, model, level, case_type)

Arguments

Value

A dataframe containing cumulative relative risk at the chosen level.

Process data for national analysis

Description

Aggregate to national data and run crossbasis

Usage

hc_add_national_data(
  df_list,
  pop_list,
  var_fun = "bs",
  var_per = c(10, 75, 90),
  var_degree = 2,
  lagn = 21,
  lagnk = 3,
  country = "National",
  cb_list,
  mm,
  minpercgeog_
)

Arguments

Value

• df_list List. A list of data frames for each geography and national level.

• cb_list List. A list of cross-basis matrices by geography and national level.

• minpercgeog_ Vector. Percentile of minimum mortality temperature for each geography and national level.

• mmpredall List. A list of national coefficients and covariance matrices.

Run ADF test and produce PACF plots for each model combination

Description

Run augmented Dickey-Fuller test for stationarity of dependent variable and produce a partial autocorrelation function plot of residuals for each model combination.

Usage

hc_adf(df_list)

Arguments

Value

'adf_list'. List of ADF results for each geography.

Estimate attributable numbers

Description

Estimate attributable numbers and confidence intervals for each geography using Monte Carlo simulations.

Usage

hc_attr(
  df_list,
  cb_list,
  pred_list,
  minpercgeog_,
  attr_thr_high = 97.5,
  attr_thr_low = 2.5
)

Arguments

Value

'attr_list'. A list containing attributable numbers per geography.

Create attributable estimates tables

Description

Aggregate tables of attributable numbers, rates and fractions for total, yearly and monthly by geography and national level.

Usage

hc_attr_tables(attr_list, country = "National", meta_analysis = FALSE)

Arguments

Value

• res_attr_tot Dataframe. Total attributable fractions, numbers and rates for each geography over the whole time series.

• attr_yr_list List. Dataframes containing yearly estimates of attributable fractions, numbers and rates by geography.

• attr_mth_list List. Dataframes containing total attributable fractions, numbers and rates by calendar month and geography.

Create cross-basis matrix

Description

Creates a cross-basis matrix of the lag-response and exposure-response functions, for each geography.

Usage

hc_create_crossbasis(
  df_list,
  var_fun = "bs",
  var_degree = 2,
  var_per = c(10, 75, 90),
  lagn = 21,
  lagnk = 3,
  dfseas = 8
)

Arguments

Value

'cb_list'. A list of cross-basis matrices by geography.

Produce check results of model combinations

Description

Runs every combination of model based on user selected additional independent variables and returns model diagnostic checks for each.

Usage

hc_model_combo_res(df_list, cb_list, independent_cols = NULL, dfseas = 8)

Arguments

Value

• qaic_results A dataframe of QAIC and dispersion metrics for each model combination.

• residuals_list List. Residuals for each model combination.

Model Validation Assessment

Description

Produces results on QAIC for each model combination, variance inflation factor for each independent variable, ADF test for stationarity, and plots for residuals to assess the models.

Usage

hc_model_validation(
  df_list,
  cb_list,
  independent_cols = NULL,
  dfseas = 8,
  save_fig = FALSE,
  save_csv = FALSE,
  output_folder_path = NULL,
  seed = NULL
)

Arguments

Value

• qaic_results A dataframe of QAIC and dispersion metrics for each model combination and geography.

• qaic_summary A dataframe with the mean QAIC and dispersion metrics for each model combination.

• vif_results A dataframe of variance inflation factors for each independent variables by geography.

• vif_summary A dataframe with the mean variance inflation factors for each independent variable.

• adf_results A dataframe of ADF test results for each geography.

Plot attributable fractions by calendar month - low temperatures

Description

Plot attributable fractions grouped over the whole time series by calendar month to explore seasonality.

Usage

hc_plot_af_cold_monthly(
  attr_mth_list,
  df_list,
  country = "National",
  attr_thr_low = 2.5,
  save_fig = FALSE,
  output_folder_path = NULL
)

Arguments

Value

Plots of attributable fractions by calendar month per geography.

Plot attributable fractions for cold by year

Description

Plot attributable fractions by year and geography with confidence intervals.

Usage

hc_plot_af_cold_yearly(
  attr_yr_list,
  save_fig = FALSE,
  output_folder_path = NULL,
  country = "National"
)

Arguments

Value

Plots of yearly attributable fractions per geography.

Plot attributable fractions by calendar month - high temperatures

Description

Plot attributable fractions grouped over the whole time series by calendar month to explore seasonality.

Usage

hc_plot_af_heat_monthly(
  attr_mth_list,
  df_list,
  country = "National",
  attr_thr_high = 97.5,
  save_fig = FALSE,
  output_folder_path = NULL
)

Arguments

Value

Plots of attributable fractions by calendar month per geography.

Plot attributable fractions for heat by year

Description

Plot attributable fractions by year and geography with confidence intervals.

Usage

hc_plot_af_heat_yearly(
  attr_yr_list,
  save_fig = FALSE,
  output_folder_path = NULL,
  country = "National"
)

Arguments

Value

Plots of yearly attributable fractions per geography.

Plot attributable rates by calendar month - low temperatures

Description

Plot attributable rates grouped over the whole time series by calendar month to explore seasonality.

Usage

hc_plot_ar_cold_monthly(
  attr_mth_list,
  df_list,
  country = "National",
  attr_thr_low = 2.5,
  save_fig = FALSE,
  output_folder_path = NULL
)

Arguments

Value

Plots of attributable rates by calendar month per geography.

Plot attributable rates by year - low temperatures

Description

Plot attributable rates by year and geography with confidence intervals.

Usage

hc_plot_ar_cold_yearly(
  attr_yr_list,
  save_fig = FALSE,
  output_folder_path = NULL,
  country = "National"
)

Arguments

Value

Plots of yearly attributable rates per geography.

Plot attributable rates by calendar month - high temperatures

Description

Plot attributable rates grouped over the whole time series by calendar month to explore seasonality.

Usage

hc_plot_ar_heat_monthly(
  attr_mth_list,
  df_list,
  country = "National",
  attr_thr_high = 97.5,
  save_fig = FALSE,
  output_folder_path = NULL
)

Arguments

Value

Plots of attributable rates by calendar month per geography.

Plot attributable rates by year - high temperatures

Description

Plot attributable rates by year and geography with confidence intervals.

Usage

hc_plot_ar_heat_yearly(
  attr_yr_list,
  save_fig = FALSE,
  output_folder_path = NULL,
  country = "National"
)

Arguments

Value

Plots of yearly attributable rates per geography.

Plot total attributable fractions and rates - low temperatures

Description

Plot total attributable fractions and rates over the whole time series by geography.

Usage

hc_plot_attr_cold_totals(
  df_list,
  res_attr_tot,
  save_fig = FALSE,
  output_folder_path = NULL,
  country = "National"
)

Arguments

Value

Plots of total attributable fractions and rates by geography

Plot total attributable fractions and rates - high temperatures

Description

Plot total attributable fractions and rates over the whole time series by geography.

Usage

hc_plot_attr_heat_totals(
  df_list,
  res_attr_tot,
  save_fig = FALSE,
  output_folder_path = NULL,
  country = "National"
)

Arguments

Value

Plots of total attributable fractions and rates by geography.

Plot statistical power for temperature mortality analysis

Description

Plots the power statistic for each reference temperature at and above the attributable risk threshold for each geography.

Usage

hc_plot_power(
  power_list_high,
  power_list_low,
  save_fig = FALSE,
  output_folder_path = NULL,
  country = "National"
)

Arguments

Value

Plots of power by temperature for the attributable thresholds and above for each geography.

Plot results of relative risk analysis

Description

Plots cumulative lag exposure-response function with histogram of temperature distribution for each geography.

Usage

hc_plot_rr(
  df_list,
  pred_list,
  attr_thr_high = 97.5,
  attr_thr_low = 2.5,
  minpercgeog_,
  country = "National",
  save_fig = FALSE,
  output_folder_path = NULL
)

Arguments

Value

Plots of cumulative lag exposure-response function with histogram of temperature distribution for each geography.

Run predictions from model

Description

Use model to run predictions. Predictions can be produced for a single input geography, or multiple disaggregated geographies.

Usage

hc_predict_subnat(
  df_list,
  var_fun = "bs",
  var_per = c(10, 75, 90),
  var_degree = 2,
  mintempgeog_,
  blup,
  coef_,
  vcov_,
  meta_analysis = FALSE
)

Arguments

Value

'pred_list'. A list containing predictions by geography.

Define and run quasi-Poisson regression with DLNM

Description

Fits a quasi-Poisson case-crossover with a distributed lag non-linear model.

Usage

hc_quasipoisson_dlnm(df_list, control_cols = NULL, cb_list, dfseas = 8)

Arguments

Value

'model_list'. List containing models by geography.

Read temperature-related mortality indicator data

Description

Reads in data and geography names for analysis from a CSV file.

Usage

hc_read_data(
  input_csv_path,
  dependent_col,
  date_col,
  region_col,
  temperature_col,
  population_col
)

Arguments

Value

'df_list'. A list of dataframes for each geography with formatted and renamed columns.

Produce cumulative relative risk results of analysis

Description

Produces cumulative relative risk and confidence intervals from analysis.

Usage

hc_rr_results(
  pred_list,
  df_list,
  minpercgeog_,
  attr_thr_high = 97.5,
  attr_thr_low = 2.5
)

Arguments

Value

'rr_results'. Dataframe containing cumulative relative risk and confidence intervals from analysis.

Save results of analysis

Description

Saves a CSV file of cumulative relative risk and confidence intervals.

Usage

hc_save_results(
  rr_results,
  res_attr_tot,
  attr_yr_list,
  attr_mth_list,
  power_list_high,
  power_list_low,
  output_folder_path = NULL
)

Arguments

Install the INLA Package from Its Official Repository

Description

This function installs the INLA package from its official repository at https://inla.r-inla-download.org/R/stable/. On Windows, it checks whether Rtools is available and installs the official binary package directly.

Usage

install_INLA(os = .Platform$OS.type)

Arguments

Details

On Windows systems, the function verifies that Rtools is installed using pkgbuild::has_build_tools(). If Rtools is missing, it displays a warning and aborts the installation. The function then installs the matching Windows binary package from the official INLA repository.

On non-Windows systems, the package is installed normally from the repository.

Value

Invisibly returns NULL. The function is called for its side effect.

Examples

## Not run: 
install_INLA()

## End(Not run)

Install the terra Package from the CRAN Archive

Description

This function installs the terra package at version 1.8-60 from the CRAN archive.

Usage

install_terra(os = .Platform$OS.type)

Arguments

Details

On Windows systems, the function verifies that Rtools is installed using pkgbuild::has_build_tools(). If Rtools is missing, it displays a warning and aborts the installation. The function then forces installation from source.

Value

Invisibly returns NULL. The function is called for its side effect.

Examples

## Not run: 
install_terra()

## End(Not run)

Check if an error is a climate_error

Description

Utility function to check if a caught condition is a typed climate error.

Usage

is_climate_error(error)

Arguments

Value

TRUE if the error inherits from "climate_error", FALSE otherwise.

Examples

tryCatch({
  stop("example error")
}, error = function(e) {
  if (is_climate_error(e)) {
    # Handle structured error
  } else {
    # Handle untyped error
  }
})

Join monthly PM2.5 estimates with attributable risk data by region and time

Description

Aggregates PM2.5 data to monthly averages by region and joins it with attributable risk data using year, month, and region as keys.

Usage

join_ar_and_pm_monthly(pm_data, an_ar_data)

Arguments

Value

A data frame with monthly average PM2.5 values joined to attributable risk data.

Join health and climate data

Description

Joins a daily time series of wildfire PM2.5 data with a daily time series of health data.

Usage

join_health_and_climate_data(
  climate_data,
  health_data,
  region_col = "region",
  date_col = "date",
  exposure_col = "mean_PM"
)

Arguments

Value

Dataframe containing a daily time series of the joined climate and health data.

Append the units to a column label.

Description

Append the units to a column label.

Usage

label_with_unit(col, units)

Arguments

Value

The new column label containing units (if col in units).

Read in climate, environmental and health data and rename columns

Description

Reads in a CSV file for a daily time series of climate, environmental and health data and renames them to standardised names. This function creates year, month, day, and day of week columns derived from the date.

Usage

load_air_pollution_data(
  data_path,
  date_col = "date",
  region_col = "region",
  pm25_col = "pm25",
  deaths_col = "deaths",
  population_col = "population",
  humidity_col = "humidity",
  precipitation_col = "precipitation",
  tmax_col = "tmax",
  wind_speed_col = "wind_speed",
  categorical_others = NULL,
  continuous_others = NULL
)

Arguments

Value

Dataframe with formatted and renamed with standardized column names.

Read in and format climate data

Description

Read in a monthly time series of climate data, rename columns and create lag variable for spatiotemporal and DLNM analysis. The climate data should start a year before a start year in the health data to allow the lag variables calculation.

Usage

load_and_process_climatedata(
  climate_data_path,
  district_col,
  year_col,
  month_col,
  tmin_col,
  tmean_col,
  tmax_col,
  rainfall_col,
  r_humidity_col,
  runoff_col = NULL,
  ndvi_col = NULL,
  spi_col = NULL,
  max_lag
)

Arguments

Value

climate dataframe with formatted and renamed columns, and the lag variables

Read in and format health data - diseases cases type

Description

Read in a csv file containing a monthly time series of health outcomes and population data. Renames columns and creates time variables for spatiotemporal analysis.

Usage

load_and_process_data(
  health_data_path,
  region_col,
  district_col,
  date_col = NULL,
  year_col = NULL,
  month_col = NULL,
  case_col,
  case_type,
  tot_pop_col
)

Arguments

Value

A dataframe with formatted and renamed columns.

Read in and format country map data

Description

: Read in a shape file, rename columns and create the adjacency matrix for spatiotemporal analysis.

Usage

load_and_process_map(
  map_path,
  region_col,
  district_col,
  geometry_col,
  output_dir = NULL
)

Arguments

Value

• 'map' The processed map

• 'nb.map'

• 'graph_file'

Load wildfire and health data

Description

Loads a dataframe containing a daily time series of health and climate data, which may be disaggregated by region.

Usage

load_wildfire_data(
  health_path,
  ncdf_path,
  shp_path,
  join_wildfire_data = TRUE,
  date_col,
  region_col,
  shape_region_col = NULL,
  mean_temperature_col,
  health_outcome_col,
  population_col = NULL,
  rh_col = NULL,
  wind_speed_col = NULL,
  pm_2_5_col = NULL
)

Arguments

Value

Dataframe containing a daily time series of climate and health data.

Code for calculating Malaria disease cases attributable to extreme rainfall and extreme temperature Run Full Malaria-Climate Analysis Pipeline

Description

The Malaria_do_analysis() function executes the complete workflow for analyzing the association between malaria cases and climate variables. It integrates health, climate, and spatial data; fits spatio-temporal models using INLA; and generates a suite of diagnostic and inferential outputs, including plots and attributable risk estimates.

Usage

malaria_do_analysis(
  health_data_path,
  climate_data_path,
  map_path,
  region_col,
  district_col,
  date_col = NULL,
  year_col,
  month_col,
  case_col,
  tot_pop_col,
  tmin_col,
  tmean_col,
  tmax_col,
  rainfall_col,
  r_humidity_col,
  runoff_col,
  geometry_col,
  spi_col = NULL,
  ndvi_col = NULL,
  max_lag = 2,
  nk = 2,
  basis_matrices_choices,
  inla_param,
  param_term,
  level,
  param_threshold = 1,
  filter_year = NULL,
  family = "nbinomial",
  group_by_year = FALSE,
  cumulative = FALSE,
  config = FALSE,
  save_csv = FALSE,
  save_model = FALSE,
  save_fig = FALSE,
  output_dir = NULL
)

Arguments

Value

A named list containing:

• inla_result - Fitted INLA model object and summaries.

• plot_malaria, plot_tmax, plot_rainfall - Exploratory time-series plots.

• reff_plot_monthly - Monthly random effects plot.

• reff_plot_yearly - Yearly spatial random effects plot.

• contour_plot - Exposure-response contour plot.

• rr_map_plot - Spatial relative risk map.

• rr_plot, rr_df - Relative risk plot and associated data.

• attr_frac_num - Attributable risk summary table.

• plot_AR_num, plot_AR_frac, plot_AR_per_100k - Plots of attributable number, fraction, and rate.

Process data for national analysis

Description

Aggregate to national data and run crossbasis

Usage

mh_add_national_data(
  df_list,
  pop_list,
  var_fun = "bs",
  var_per = c(25, 50, 75),
  var_degree = 2,
  lag_fun = "strata",
  lag_breaks = 1,
  lag_days = 2,
  country = "National",
  cb_list,
  mm,
  minpercreg
)

Arguments

Value

• df_list List. A list of data frames for each region and nation.

• cb_list List. A list of cross-basis matrices by region and nation.

• minpercreg Vector. Percentile of minimum suicide temperature for each region and nation.

• mmpredall List. A list of national coefficients and covariance matrices.

Estimate attributable numbers

Description

Estimate attributable numbers for each region and confidence intervals using Monte Carlo simulations.

Usage

mh_attr(df_list, cb_list, pred_list, minpercreg, attr_thr = 97.5)

Arguments

Value

A list containing attributable numbers per region

Create attributable estimates tables

Description

Aggregate tables of attributable numbers, rates and fractions for total, yearly and monthly by region and nation

Usage

mh_attr_tables(attr_list, country = "National", meta_analysis = FALSE)

Arguments

Value

• res_attr_tot Dataframe. Total attributable fractions, numbers and rates for each area over the whole time series.

• attr_yr_list List. Dataframes containing yearly estimates of attributable fractions, numbers and rates by area.

• attr_mth_list List. Dataframes containing total attributable fractions, numbers and rates by calendar month and area.

Quasi-Poisson Case-Crossover model with DLNM

Description

Fits a quasi-Poisson case-crossover with a distributed lag non-linear model

Usage

mh_casecrossover_dlnm(df_list, control_cols = NULL, cb_list)

Arguments

Value

List containing models by region

Create cross-basis matrix

Description

Creates a cross-basis matrix for each region

Usage

mh_create_crossbasis(
  df_list,
  var_fun = "bs",
  var_degree = 2,
  var_per = c(25, 50, 75),
  lag_fun = "strata",
  lag_breaks = 1,
  lag_days = 2
)

Arguments

Value

A list of cross-basis matrices by region

Produce check results of model combinations

Description

Runs every combination of model based on user selected additional independent variables and returns model diagnostic checks for each.

Usage

mh_model_combo_res(df_list, cb_list, independent_cols = NULL)

Arguments

Value

• qaic_results A dataframe of QAIC and dispersion metrics for each model combination.

• residuals_list A list. Residuals for each model combination.

Model Validation Assessment

Description

Produces results on QAIC for each model combination, variance inflation factor for each independent variable, and plots for residuals to assess the models

Usage

mh_model_validation(
  df_list,
  cb_list,
  independent_cols = NULL,
  save_fig = FALSE,
  save_csv = FALSE,
  output_folder_path = NULL,
  seed = NULL
)

Arguments

Value

• qaic_results A dataframe of QAIC and dispersion metrics for each model combination and geography.

• qaic_summary A dataframe with the mean QAIC and dispersion metrics for each model combination.

• vif_results A dataframe. Variance inflation factors for each independent variables by region.

• vif_summary A dataframe with the mean variance inflation factors for each independent variable.

Plot attributable fractions by calendar month

Description

Plot attributable fractions grouped over the whole time series by calendar month to explore seasonality.

Usage

mh_plot_af_monthly(
  attr_mth_list,
  df_list,
  country = "National",
  attr_thr = 97.5,
  save_fig = FALSE,
  output_folder_path = NULL
)

Arguments

Value

Plots of attributable fractions by calendar month per area

Plot attributable fractions by year

Description

Plot attributable fractions by year and area with confidence intervals

Usage

mh_plot_af_yearly(
  attr_yr_list,
  save_fig = FALSE,
  output_folder_path = NULL,
  country = "National"
)

Arguments

Value

Plots of yearly attributable fractions per area

Plot attributable rates by calendar month

Description

Plot attributable rates grouped over the whole time series by calendar month to explore seasonality.

Usage

mh_plot_ar_monthly(
  attr_mth_list,
  df_list,
  country = "National",
  attr_thr = 97.5,
  save_fig = FALSE,
  output_folder_path = NULL
)

Arguments

Value

Plots of attributable rates by calendar month per area

Plot attributable rates by year

Description

Plot attributable rates by year and area with confidence intervals

Usage

mh_plot_ar_yearly(
  attr_yr_list,
  save_fig = FALSE,
  output_folder_path = NULL,
  country = "National"
)

Arguments

Value

Plots of yearly attributable rates per area

Plot total attributable fractions and rates

Description

Plot total attributable fractions and rates over the whole time series by area.

Usage

mh_plot_attr_totals(
  df_list,
  res_attr_tot,
  save_fig = FALSE,
  output_folder_path = NULL,
  country = "National"
)

Arguments

Value

Plots of total attributable fractions and rates by area

Plot power

Description

Plots the power statistic for each reference temperature at and above the attributable risk threshold for each area.

Usage

mh_plot_power(
  power_list,
  save_fig = FALSE,
  output_folder_path = NULL,
  country = "National"
)

Arguments

Value

Plots of power by temperature for the attributable threshold and above for each area.

Plot results of relative risk analysis - Mental Health

Description

Plots cumulative lag exposure-response function with histogram of temperature distribution for each region

Usage

mh_plot_rr(
  df_list,
  pred_list,
  attr_thr = 97.5,
  minpercreg,
  country = "National",
  save_fig = FALSE,
  output_folder_path = NULL
)

Arguments

Value

Plots of cumulative lag exposure-response function with histogram of temperature distribution for each region

Run regional predictions from model

Description

Use model to run regional predictions

Usage

mh_predict_reg(
  df_list,
  var_fun = "bs",
  var_per = c(25, 50, 75),
  var_degree = 2,
  minpercreg,
  blup,
  coef_,
  vcov_,
  meta_analysis = FALSE
)

Arguments

Value

A list containing predictions by region

Read in and format data - Mental Health

Description

Reads in a CSV file for a daily time series of health and climate data, renames columns and creates stratum for case-crossover analysis.

Usage

mh_read_and_format_data(
  data_path,
  date_col,
  region_col = NULL,
  temperature_col,
  health_outcome_col,
  population_col
)

Arguments

Value

A list of dataframes with formatted and renamed columns.

Produce cumulative relative risk results of analysis

Description

Produces cumulative relative risk and confidence intervals from analysis.

Usage

mh_rr_results(pred_list, df_list, attr_thr = 97.5, minpercreg)

Arguments

Value

Dataframe containing cumulative relative risk and confidence intervals from analysis.

Save results of analysis - Mental Health

Description

Saves a CSV file of cumulative relative risk and confidence intervals.

Usage

mh_save_results(
  rr_results,
  res_attr_tot,
  attr_yr_list,
  attr_mth_list,
  power_list,
  output_folder_path = NULL
)

Arguments

Plot relative risk results by region (if available).

Description

Plots relative risk and confidence intervals for each lag value of wildfire-related PM2.5

Usage

plot_RR(
  rr_data,
  wildfire_lag,
  by_region = FALSE,
  save_fig = FALSE,
  output_folder_path = NULL
)

Arguments

Value

Plot of relative risk and confidence intervals for each lag of wildfire-related PM2.5

Core functionality for plotting results of relative risk analysis.

Description

Plots relative risk and confidence intervals for each lag value of wildfire-related PM2.5.

Usage

plot_RR_core(
  rr_data,
  save_fig = FALSE,
  wildfire_lag,
  output_folder_path = NULL,
  region_name = "All regions",
  ylims = NULL
)

Arguments

Value

Plot of relative risk and confidence intervals for each lag of wildfire-related PM2.5.

Plots attributable fractions and CI across years by regions

Description

Generates a PDF containing one or more plots of average attributable fractions over time. If by_region is TRUE, the function creates separate plots for each region. All plots are saved to a single PDF file named "aggregated_AF_by_region.pdf" in the specified output_dir.

Usage

plot_aggregated_AF(data, by_region = FALSE, output_dir = ".")

Arguments

Value

No return value. A PDF file is created.

Create a plot of aggregated annual attributable fractions with CI

Description

Aggregates annual average attributable fraction estimates and generates a ggplot showing the central estimate and CI.

Usage

plot_aggregated_AF_core(data, region_name = NULL)

Arguments

Value

A ggplot object showing annual attributable rates with confidence intervals.

Combined AN and AR plots by region

Description

Creates both Attributable Number (AN) and Attributable Rate (AR) bar charts by region in a single function call.

Usage

plot_air_pollution_an_ar_by_region(
  analysis_results,
  max_lag = 14L,
  include_national = TRUE,
  output_dir = NULL,
  save_plot = FALSE
)

Arguments

Value

List with two ggplot objects: an_plot and ar_plot

Plot the AN and AR by year

Description

Creates both Attributable Number (AN) and Attributable Rate (AR) plots aggregated by year in a single function call.

Usage

plot_air_pollution_an_ar_by_year(
  analysis_results,
  max_lag = 14L,
  include_national = TRUE,
  output_dir = NULL,
  save_plot = FALSE
)

Arguments

Value

List with two ggplot objects: an_plot and ar_plot

Combined Monthly Time Series Plots of AN and AR

Description

Creates both Attributable Number (AN) and Attributable Rate (AR) monthly time series plots in a single function call.

Usage

plot_air_pollution_an_ar_monthly(
  analysis_results,
  max_lag = 14L,
  include_national = TRUE,
  output_dir = NULL,
  save_plot = FALSE
)

Arguments

Value

List with two ggplot objects: an_plot and ar_plot

Plot exposure-response relationship with confidence intervals by region

Description

Creates faceted exposure-response plots showing RR with confidence intervals across PM2.5 concentrations for each region

Usage

plot_air_pollution_exposure_response(
  analysis_results,
  max_lag = 14L,
  include_national = TRUE,
  ref_pm25 = 15,
  output_dir = NULL,
  save_plot = FALSE
)

Arguments

Value

ggplot object

Plot Relative Risk (RR) by lag

Description

Plot Relative Risk (RR) by lag

Usage

plot_air_pollution_forest_by_lag(
  analysis_results,
  max_lag = 14L,
  output_dir = NULL,
  save_plot = FALSE
)

Arguments

Value

ggplot object

Plot forest plot for PM2.5 effects by region

Description

Plot forest plot for PM2.5 effects by region

Usage

plot_air_pollution_forest_by_region(
  analysis_results,
  max_lag = 14L,
  include_national = TRUE,
  output_dir = NULL,
  save_plot = FALSE
)

Arguments

Value

ggplot object

Plot histograms for AN and AR by month

Description

Creates histogram plots for Attributable Number (AN) and Attributable Rate (AR) aggregated by month with connecting lines

Usage

plot_air_pollution_monthly_histograms(
  analysis_results,
  max_lag = 14L,
  include_national = TRUE,
  output_dir = NULL,
  save_plot = FALSE
)

Arguments

Value

List with ggplot objects

Plot Power vs PM2.5 Concentration

Description

Plots the power statistic for each reference PM2.5 at and above the attributable risk threshold for each region.

Usage

plot_air_pollution_power(
  power_list,
  output_dir = NULL,
  save_plot = FALSE,
  ref_name = "WHO",
  include_national = TRUE
)

Arguments

Value

Invisible list of plot information

Plot Total Attributable Number by Region

Description

Aggregates wildfire smoke-related PM2.5 attributable numbers by region and creates a bar plot showing the total attributable number of deaths per region.

Usage

plot_an_by_region(data, output_dir = ".")

Arguments

Value

A ggplot object representing the bar plot.

Plot Attributable Risk by Region

Description

Aggregates wildfire-specific PM2.5 attributable risk (deaths per 100k) by region and creates a bar plot showing the mean attributable risk per region.

Usage

plot_ar_by_region(data, output_dir = ".")

Arguments

Value

A ggplot object representing the bar plot.

Plot monthly deaths and PM2.5 concentrations with dual y-axes

Description

Aggregates data by month and creates a dual-axis plot showing average deaths per 100,000 and mean PM2.5 concentrations.

Usage

plot_ar_pm_monthly(data, save_outputs = FALSE, output_dir = NULL)

Arguments

Value

No return value. Generates a plot and optionally saves files.

Plot Attributable Health Metrics Across Spatial and Temporal Levels

Description

Visualizes attributable health metrics (e.g., attributable number, fraction, or rate) derived from attribution_calculation() across different spatial scales and time periods. The function automatically adapts plots to the selected spatial level (country, region, or district) and handles both single- and multi-year visualizations. It supports faceted, grouped, or aggregated visualizations and can optionally save output plots as PDF files.

Usage

plot_attribution_metric(
  attr_data,
  level = c("country", "region", "district"),
  metrics = c("AR_Number", "AR_Fraction", "AR_per_100k"),
  filter_year = NULL,
  param_term,
  case_type,
  save_fig = FALSE,
  output_dir = NULL
)

Arguments

Details

This function produces publication-ready plots of attributable metrics:

• Country level: Time series line plots with 95% confidence ribbons.

• Region/District level (no filter): Horizontal bar plots showing aggregated metrics, grouped by administrative unit.

• Region/District level (multi-year): Grouped bar plots comparing metrics across years.

The function automatically adjusts y-axis limits, formats numeric labels with commas, and includes optional text annotations (e.g., showing both attributable numbers and fractions). When save_fig = TRUE, one PDF file is created per metric and spatial level, and each file may contain multiple pages if many regions or districts are present.

Value

A named list of ggplot or patchwork plot objects, grouped by metric. Each element corresponds to one metric ("AR_Number", "AR_Fraction", "AR_per_100k") and may include one or more plots, depending on the level and year filters.

Plot Average Monthly Attributable Health Metrics with Climate Overlays

Description

Visualizes average monthly attributable health metrics (e.g., attributable number, fraction, or rate) derived from attribution analyses across different spatial scales. The function automatically adapts plots to the selected spatial level (country, region, or district) and summarizes seasonal patterns using monthly aggregation. Optionally, corresponding monthly climate variables can be overlaid on a secondary axis to support joint interpretation of health impacts and climate seasonality.

Usage

plot_avg_monthly(
  attr_data,
  level = c("country", "region", "district"),
  metrics = c("AR_Number", "AR_per_100k", "AR_Fraction"),
  c_data,
  param_term,
  case_type,
  filter_year = NULL,
  save_fig = FALSE,
  output_dir = NULL
)

Arguments

Details

This function produces publication-ready visualizations of average monthly attributable health metrics:

• Country level: A single bar plot summarizing national average monthly attribution patterns.

• Region/District level: One bar plot per administrative unit, showing average monthly attribution, with automatic pagination when many units are present.

• Climate overlay (optional): Monthly climate exposure plotted as a line on a secondary y-axis to facilitate comparison with seasonal health impacts.

Metric-specific aggregation rules (sum or mean) and numeric formatting are applied automatically. Axis limits and breaks are dynamically adjusted to improve readability. When save_fig = TRUE, a single PDF file is created per metric and spatial level, with multiple pages used for region- or district-level outputs when necessary.

Value

A named list of ggplot objects. Each element corresponds to the country or an individual region or district and contains a monthly attribution plot. The list is returned invisibly when plots are saved to file.

Plot a grid of box plots for multiple numeric variables

Description

Plot a grid of box plots for multiple numeric variables

Usage

plot_boxplots(
  df,
  columns = NULL,
  select_numeric = FALSE,
  title = "Boxplots",
  ylabs = NULL,
  save_plot = FALSE,
  output_path = NULL
)

Arguments

Plot a correlation matrix include a heatmap.

Description

Plot a correlation matrix include a heatmap.

Usage

plot_correlation_matrix(matrix_, title, output_path)

Arguments

Plot histograms of column distributions.

Description

Plot histograms of column distributions.

Usage

plot_distributions(
  df,
  columns,
  title,
  xlabs = NULL,
  save_hists = FALSE,
  output_path = NULL
)

Arguments

Plot Time Series of Health and Climate Variables

Description

Generate time series plots for combined health and climate data prepared for spatiotemporal and DLNM analysis. Supports aggregation at the country, region, or district level.

Usage

plot_health_climate_timeseries(
  data,
  param_term,
  level = "country",
  filter_year = NULL,
  case_type,
  save_fig = FALSE,
  output_dir = NULL
)

Arguments

Value

A ggplot object.

Visualise monthly random effects for selected INLA model

Description

Generates and saves a plot of monthly random effects for different regions, visualizing their contribution to Malaria Incidence Rate.

Usage

plot_monthly_random_effects(
  combined_data,
  model,
  save_fig = FALSE,
  output_dir = NULL
)

Arguments

Value

THe monthly random effects plot.

Plot the moving average of a column.

Description

Plot the moving average of a column.

Usage

plot_moving_average(
  df,
  time_col,
  value_col,
  ma_days,
  ma_sides,
  title,
  save_plot = FALSE,
  output_path = "",
  units = NULL
)

Arguments

Plot the rate of a dependent variable per 100,000 population per year.

Description

Plot the rate of a dependent variable per 100,000 population per year.

Usage

plot_rate_overall(
  df,
  dependent_col,
  population_col,
  date_col,
  title,
  save_rate = FALSE,
  output_path = NULL
)

Arguments

Plot regional trends of a climate and healthoutcome.

Description

Plot regional trends of a climate and healthoutcome.

Usage

plot_regional_trends(
  df,
  region_col,
  outcome_cols,
  title = "Regional Averages",
  ylabs = NULL,
  save_plot = FALSE,
  output_path = ""
)

Arguments

Read in Relative Risk plot at country, Region, and District level

Description

Plots the relative risk of Malaria cases by the maximum temperature and cumulative rainfall at country, Region and District level

Usage

plot_relative_risk(
  data,
  model,
  param_term,
  max_lag,
  nk,
  level,
  case_type,
  filter_year = NULL,
  output_dir = NULL,
  save_csv = FALSE,
  save_fig = FALSE
)

Arguments

Value

Relative risk plot at country, region, and district levels.

Plot relative risk by PM2.5 levels for all regions and individually

Description

Generates one or more plots showing relative risk estimates across PM2.5 levels. If multiple regions are present, plots are created per region and for all regions combined. Optionally saves the output as a PDF.

Usage

plot_rr_by_pm(data, save_fig = FALSE, output_dir = NULL)

Arguments

Value

No return value. Generates one or more plots and optionally saves them to disk.

Create a relative risk plot across PM2.5 levels for a single region

Description

Generates a ggplot showing relative risk estimates and confidence intervals across PM2.5 levels for a given region.

Usage

plot_rr_by_pm_core(data, region_name = "All Regions", ylims = c(-2, 2))

Arguments

Value

A ggplot object showing relative risk and CI.

Plot Relative Risk Map at sub-national Level

Description

Generates a map of the relative risk of the diseases cases associated with climate hazards, including extreme temperature and cumulative rainfall, at a specified geographical level (district or region).

Usage

plot_rr_map(
  combined_data,
  model,
  param_term = "tmax",
  max_lag,
  nk,
  level = "district",
  case_type,
  filter_year = NULL,
  output_dir = NULL,
  save_fig = FALSE,
  save_csv = FALSE,
  cumulative = FALSE
)

Arguments

Value

Relative risk map at the chosen level.

Plot a grid of scatter graphs comparing one column to various others.

Description

Plot a grid of scatter graphs comparing one column to various others.

Usage

plot_scatter_grid(
  df,
  main_col,
  comparison_cols,
  title,
  save_scatters = FALSE,
  output_path = "",
  units = NULL
)

Arguments

Plot seasonal trends of a health outcome and climate by month.

Description

Plot seasonal trends of a health outcome and climate by month.

Usage

plot_seasonal_trends(
  df,
  date_col,
  outcome_cols,
  title = "Seasonal Averages",
  ylabs = NULL,
  save_plot = FALSE,
  output_path = ""
)

Arguments

Plot the total of selected variables per year.

Description

Plot the total of selected variables per year.

Usage

plot_total_variables_by_year(
  df,
  date_col,
  variables,
  title,
  save_total = FALSE,
  output_path = ""
)

Arguments

Value

Plots are printed to the console or saved as PDF files.

Visualize yearly spatial random effect of the Diseases Incidence Rate (DIR).

Description

Generates and saves plots of yearly spatial random effect of the diseases incidence rate at district level.

Usage

plot_yearly_spatial_random_effect(
  combined_data,
  model,
  case_type,
  save_fig = FALSE,
  output_dir = NULL
)

Arguments

Value

The yearly space random effect for the disease incidence rate plot.

Normalise descriptive stats data input to combined and regional dataframes.

Description

Normalise descriptive stats data input to combined and regional dataframes.

Usage

prepare_descriptive_input(
  data,
  aggregation_column = NULL,
  timeseries_col = NULL
)

Arguments

Value

A list with combined_df and region_df_list.

Validate and prepare base output directory for descriptive stats.

Description

Validate and prepare base output directory for descriptive stats.

Usage

prepare_descriptive_output_dir(output_path, create_base_dir = FALSE)

Arguments

Value

Character. Validated output path.

Raise an Error if a Parameter's Value is NULL

Description

Raise an Error if a Parameter's Value is NULL

Usage

raise_if_null(param_nm, value)

Arguments

Value

None. Stops execution if value is NULL.

Code for producing analysis for health effects of extreme weather events - wildfires Read in and format health data

Description

Reads in a CSV file for a daily time series of health and climate data, renames columns to standard names. Creates columns for day of week, month, and year columns derived from the date.

Usage

read_and_format_data(
  health_path,
  date_col,
  mean_temperature_col,
  health_outcome_col,
  population_col = NULL,
  region_col = NULL,
  rh_col = NULL,
  wind_speed_col = NULL
)

Arguments

Value

Dataframe with formatted and renamed columns

Read a csv file into memory as a data frame.

Description

Read a csv file into memory as a data frame.

Usage

read_input_data(input_csv_path)

Arguments

Value

A dataframe containing the data from the csv.

Examples

input_csv_path <- "directory/file_name.csv"

Reformat a dataframe using various different cleaning techniques.

Description

Take a dataframe, and apply various different cleaning methods to it in order to prepare the data for use with a climate indicator.

Usage

reformat_data(df, reformat_date = TRUE, fill_na = c(), year_from_date = TRUE)

Arguments

Value

The cleaned/reformatted data frame.

Run generic descriptive statistics and EDA outputs for indicator datasets.

Description

Run generic descriptive statistics and EDA outputs for indicator datasets.

Usage

run_descriptive_stats(
  data,
  output_path,
  aggregation_column = NULL,
  population_col = NULL,
  plot_corr_matrix = FALSE,
  correlation_method = "pearson",
  plot_dist = FALSE,
  plot_ma = FALSE,
  ma_days = 100,
  ma_sides = 1,
  timeseries_col = NULL,
  dependent_col,
  independent_cols,
  units = NULL,
  plot_na_counts = FALSE,
  plot_scatter = FALSE,
  plot_box = FALSE,
  plot_seasonal = FALSE,
  plot_regional = FALSE,
  plot_total = FALSE,
  detect_outliers = FALSE,
  calculate_rate = FALSE,
  run_id = NULL,
  create_base_dir = FALSE
)

Arguments

Value

A list with base_output_path, run_id, run_output_path, and region_output_paths.

Examples

df <- data.frame(
  date = as.Date("2024-01-01") + 0:29,
  region = rep(c("A", "B"), each = 15),
  outcome = sample(1:20, 30, replace = TRUE),
  temp = rnorm(30, 25, 3)
)

run_descriptive_stats(
  data = df,
  output_path = tempdir(),
  aggregation_column = "region",
  dependent_col = "outcome",
  independent_cols = c("temp"),
  timeseries_col = "date",
  run_id = NULL
)

Create descriptive statistics via API-friendly inputs.

Description

Create descriptive statistics via API-friendly inputs.

Usage

run_descriptive_stats_api(
  data,
  output_path,
  aggregation_column = NULL,
  population_col = NULL,
  dependent_col,
  independent_cols,
  units = NULL,
  plot_corr_matrix = FALSE,
  plot_dist = FALSE,
  plot_ma = FALSE,
  plot_na_counts = FALSE,
  plot_scatter = FALSE,
  plot_box = FALSE,
  plot_seasonal = FALSE,
  plot_regional = FALSE,
  plot_total = FALSE,
  correlation_method = "pearson",
  ma_days = 100,
  ma_sides = 1,
  timeseries_col = NULL,
  detect_outliers = FALSE,
  calculate_rate = FALSE,
  run_id = NULL,
  create_base_dir = TRUE
)

Arguments

Value

A list with base_output_path, run_id, run_output_path, and region_output_paths.

Examples

run_descriptive_stats_api(
  data = list(
    date = as.character(as.Date("2024-01-01") + 0:29),
    region = rep(c("A", "B"), each = 15),
    outcome = sample(1:20, 30, replace = TRUE),
    temp = rnorm(30, 25, 3)
  ),
  output_path = tempdir(),
  aggregation_column = "region",
  dependent_col = "outcome",
  independent_cols = c("temp"),
  timeseries_col = "date",
  plot_corr_matrix = TRUE
)

Run models of increasing complexity in INLA: Fit a baseline model including spatiotemporal random effects.

Description

: Create and run multiple INLA (Integrated Nested Laplace Approximation) models to the dataset, evaluates them using DIC (Deviance Information Criterion), and identifies the best-fitting model.

Usage

run_inla_models(
  combined_data,
  basis_matrices_choices,
  inla_param,
  max_lag,
  nk,
  case_type,
  output_dir = NULL,
  save_model = FALSE,
  family = "nbinomial",
  config = FALSE
)

Arguments

Value

A list containing the model, baseline_model, and the dic_table.

Save air pollution plot with standardized dimensions

Description

Save air pollution plot with standardized dimensions

Usage

save_air_pollution_plot(plot_object, output_dir, filename)

Arguments

Value

Invisibly returns the output path

Save results of wildfire related analysis

Description

Saves a CSV file of relative risk and confidence intervals for each lag value of wildfire-related PM2.5. Also optionally save results of attributable numbers/fractions.

Usage

save_wildfire_results(
  rr_results,
  an_ar_results = NULL,
  annual_af_an_results = NULL,
  output_folder_path
)

Arguments

Create a cross-basis matrix set for DLNM analysis

Description

Generates cross-basis matrices for lagged climate variables in a dataset, for use in Distributed Lag Nonlinear Models (DLNM).

Usage

set_cross_basis(data, max_lag = 2, nk = 2)

Arguments