Skip to contents

Per-area pollution burden

Source: R/pollution.R
pollution_burden_by_area.Rd

Applies pollution_burden to every row of a table with one row per area (forward sortation area, census tract, neighbourhood, ...) and returns the rows sorted by attributable cases, worst first. Use it to see where in a city or region the pollution burden falls, for example to rank neighbourhoods for an intervention.

Usage

pollution_burden_by_area(
  area_table,
  area_col = "fsa",
  exposure_col = "exposure",
  population_col = "population",
  baseline_rate_col = "baseline_rate",
  pollutant = "NO2",
  outcome = "all_cause_mortality",
  rare_outcome = FALSE,
  high_rate = FALSE
)

Arguments

area_table

A data frame with one row per area: a label, the exposure in micrograms per cubic metre, the population in persons and the baseline rate per person-year (0.008 is 800 per 100,000).

area_col, exposure_col, population_col, baseline_rate_col

Column names (single strings) of the label, exposure, population and rate; defaults "fsa", "exposure", "population" and "baseline_rate".

pollutant, outcome, rare_outcome, high_rate

Passed to pollution_burden for every area: the pollutant (default "NO2"), the outcome (default all-cause mortality), and the switches that admit a rate below 1e-5 or above 0.1 per person-year.

Value

A data frame with one row per area, sorted by attributable_cases, largest first, and row names reset: the four input columns (exposure, population and rate as numbers) plus rr (relative risk), paf (attributable fraction), attributable_cases and baseline_cases (per year).

Details

Each area is treated as wholly exposed at its own concentration (exposure_prevalence = 1), with the pollutant's default reference concentration. Every area is checked as pollution_burden checks a single population, and the first area that fails stops the call with that area's error. Refusals: a column name that is not a single string or not in the table, an empty table, and a non-numeric exposure, population or rate column. The area labels can be of any type.

See also

exposure_concentration_index for how the burden is spread across income.

Examples

t <- data.frame(fsa = c("M6H", "M5V"), exposure = c(28, 18),
                population = c(40000, 60000), baseline_rate = 0.008)
pollution_burden_by_area(t)
#>   fsa exposure population baseline_rate       rr        paf attributable_cases
#> 1 M6H       28      40000         0.008 1.036288 0.03501694          11.205420
#> 2 M5V       18      60000         0.008 1.015968 0.01571728           7.544292
#>   baseline_cases
#> 1            320
#> 2            480

# PM2.5 and different column names.
u <- data.frame(tract = c("A", "B", "C"), pm = c(7, 11, 9),
                pop = c(5000, 8000, 6500), rate = 0.0075)
pollution_burden_by_area(u, area_col = "tract", exposure_col = "pm",
                         population_col = "pop", baseline_rate_col = "rate",
                         pollutant = "PM2.5")
#>   tract pm  pop   rate       rr        paf attributable_cases baseline_cases
#> 1     B 11 8000 0.0075 1.040831 0.03922953          2.3537717          60.00
#> 2     C  9 6500 0.0075 1.024933 0.02432675          1.1859290          48.75
#> 3     A  7 5000 0.0075 1.009278 0.00919281          0.3447304          37.50

# A missing column is named in the error.
try(pollution_burden_by_area(t[, c("fsa", "exposure")]))
#> Error : area_table missing columns: population, baseline_rate