impact_fraction {graphPAF}

General calculations of impact fractions

Description

General calculations of impact fractions

Usage

impact_fraction(
  model,
  data,
  new_data,
  calculation_method = "B",
  prev = NULL,
  ci = FALSE,
  boot_rep = 50,
  t_vector = NULL,
  ci_level = 0.95,
  ci_type = c("norm"),
  weight_vec = NULL,
  verbose = TRUE
)

Arguments

model

Either a clogit, glm or coxph fitted model object. Non-linear effects should be specified via ns(x, df=y), where ns is the natural spline function from the splines library.

data

A dataframe containing variables used for fitting the model

new_data

A dataframe (of the same variables and size as data) representing an alternative distribution of risk factors

calculation_method

A character either 'B' (Bruzzi) or 'D' (Direct method). For case control data, the method described in Bruzzi 1985 is recommended. Bruzzi's method estimates PAF from relative risks and prevalence of exposure to the risk factor. The Direct method estimates PAF by summing estimated probabilities of disease in the absence of exposure on the individual level

prev

estimated prevalence of disease. This only needs to be specified if the data source is from a case control study, and the direct method is used

ci

Logical. If TRUE, a bootstrap confidence interval is computed along with point estimate (default FALSE)

boot_rep

Integer. Number of bootstrap replications (Only necessary to specify if ci=TRUE)

t_vector

Numeric. A vector of times at which to calculate PAF (only specified if model is coxph)

ci_level

Numeric. Default 0.95. A number between 0 and 1 specifying the confidence level

ci_type

Character. Default norm. A vector specifying the types of confidence interval desired. "norm", "basic", "perc" and "bca" are the available methods

weight_vec

An optional vector of inverse sampling weights for survey data (note that variance will not be calculated correctly if sampling isn't independent). Note that this vector will be ignored if prev is specified, and the weights will be calibrated so that the weighted sample prevalence of disease equals prev.

verbose

A logical indicator for whether extended output is produced when ci=TRUE, default TRUE

Value

A numeric estimated impact fraction if ci=FALSE, or for survival data a vector of estimated impact corresponding to event times in the data. If ci=TRUE, estimated impact fractions and other information are bundled into an object of class IF_summary.

References

Bruzzi, P., Green, S.B., Byar, D.P., Brinton, L.A. and Schairer, C., 1985. Estimating the population attributable risk for multiple risk factors using case-control data. American journal of epidemiology, 122(5), pp.904-914

Examples

library(splines)
library(survival)
new_data <- stroke_reduced
N <- nrow(new_data)
inactive_patients <- (1:N)[stroke_reduced$exercise==1]
N_inactive <- sum(stroke_reduced$exercise)
newly_active_patients <- inactive_patients[sample(1:N_inactive,0.2*N_inactive)]
new_data$exercise[newly_active_patients] <- 0
model_exercise <- clogit(formula = case ~ age + education +exercise +
ns(diet, df = 3) + smoking + alcohol + stress + ns(lipids,df = 3) +
ns(waist_hip_ratio, df = 3) + high_blood_pressure +strata(strata),
data=stroke_reduced)
impact_fraction(model=model_exercise,stroke_reduced,new_data,
calculation_method = "B")