getSMR: Calculate the standardized mortality/morbidity ratios

In diseasemapping: Modelling Spatial Variation in Disease Risk for Areal Data

Calculate the standardized mortality/morbidity ratios

Description

Calculates the rate of observe value over expected value. It will also merge back the observed value, expected value and the ratio back to the population data set.

Usage

## S4 method for signature 'SpatVector,ANY,ANY,ANY,ANY'
getSMR(
popdata, model, casedata, regionCode , regionCodeCases , 
area.scale = 1,  sex=c('m','f'),...
)
## S4 method for signature 'list,ANY,ANY,ANY,ANY'
getSMR(
popdata, model, casedata, regionCode , regionCodeCases , 
area.scale=1, sex=c('m','f'), ...
)
## S4 method for signature 'data.frame,ANY,missing,missing,missing'
getSMR(
popdata, model, casedata, regionCode , regionCodeCases , 
area.scale = 1,  sex=c('m','f'),...
)

## S4 method for signature 'data.frame,ANY,data.frame,missing,missing'
getSMR(
popdata, model, casedata, regionCode , regionCodeCases , 
area.scale = 1,  sex=c('m','f'),...
)
## S4 method for signature 'data.frame,ANY,data.frame,character,missing'
getSMR(
popdata, model, casedata, regionCode , regionCodeCases , 
area.scale = 1,  sex=c('m','f'),...
)
## S4 method for signature 'data.frame,ANY,missing,character,missing'
getSMR(
popdata, model, casedata, regionCode , regionCodeCases , 
area.scale = 1,  sex=c('m','f'),...
)

## S4 method for signature 'data.frame,ANY,data.frame,character,character'
getSMR(
popdata, model, casedata, regionCode , regionCodeCases , 
area.scale = 1,  sex=c('m','f'),...
)

Arguments

popdata	the name of population data set
model	rates, either fitted model (usually a glm object), or a vector of rates.
casedata	the name of case data set
regionCode	the name of district area column in population data set
regionCodeCases	the name of district area column in case data set
area.scale	scale the unit of area. e.g $10^6$: if your spatial coordinates are metres and you want intensity in cases per km2
sex	possible subsetting of cases and population, set sex='f' for females only.
...	additional arguments. When popdata is a list, arguments can be personYears (logical, convert rates to person years), years (a vector with the year of each dataset), or year.range (two dimensional vector with first and last year)

Details

If model is numeric, it's assumed to be a vector of rates, with the names of the elements corresponding to columns of the population data set. Names do not need to match exactly (can have M in one set of names, male in another for instance).

Otherwise, model is passed to the predict function.

Value

Returns a new population data set contains expected number of cases, observed number of cases and SMR. It has the same format as the population data set which put into the function.

Examples

data(kentucky)
kentucky = terra::unwrap(kentucky)

kentucky2 = getSMR(kentucky, larynxRates, larynx, 
		regionCode="County")

terra::values(kentucky2)[1:10,grep("^F|^M", names(kentucky2), invert=TRUE)]

theBreaks = signif(seq(0, max(kentucky2$SMR, na.rm=TRUE), len=9),1)
theCol = heat.colors(length(theBreaks)-1)
terra::plot(kentucky2, col=theCol, breaks = theBreaks)
legend('left', fill=theCol, legend = paste(theBreaks[-length(theBreaks)], ' - ', theBreaks[-1]))

diseasemapping documentation built on Sept. 22, 2023, 1:07 a.m.