treatment.effect {bipd}R Documentation

Calculate patient-specific treatment effect

Description

Function for calculating the patient-specific treatment effect. Patient-specific treatment effect includes the main effect of treatment and treatment-covariate interaction effect (i.e. effect modification). Reports odds ratio for the binary outcome.

Usage

treatment.effect(
  ipd = NULL,
  samples = NULL,
  newpatient = NULL,
  scale_mean = NULL,
  scale_sd = NULL,
  reference = NULL,
  quantiles = c(0.025, 0.5, 0.975)
)

Arguments

ipd

IPD object created from running ipdma.model type function

samples

MCMC samples found from running ipd.run function

newpatient

covariate values of patients that you want to predict treatment effect on. Must have length equal to total number of covariates.

scale_mean

option to specify different overall mean compared to what was calculated in IPD object. can be useful when using multiple imputation.

scale_sd

option to specify different overall standard deviation compared to what was calculated in IPD object.

reference

reference group used for finding patient-specific treatment effect. This is only used for deft approach

quantiles

quantiles for credible interval of the patient-specific treatment effect

Value

patient-specific treatment effect with credible interval at specified quantiles

References

Seo M, White IR, Furukawa TA, et al. Comparing methods for estimating patient-specific treatment effects in individual patient data meta-analysis. Stat Med. 2021;40(6):1553-1573. doi: 10.1002/sim.8859

Riley RD, Debray TP, Fisher D, et al. Individual participant data meta-analysis to examine interactions between treatment effect and participant-level covariates: Statistical recommendations for conduct and planning. Stat Med. 2020:39(15):2115-2137. doi: 10.1002/sim.8516

Examples

ds <- generate_ipdma_example(type = "continuous")
ipd <- with(ds, ipdma.model.onestage(y = y, study = studyid, treat = treat, X = cbind(z1, z2), 
response = "normal", shrinkage = "none"))

samples <- ipd.run(ipd, pars.save = c("beta", "gamma", "delta"), n.chains = 3, n.burnin = 500, 
n.iter = 5000)
treatment.effect(ipd, samples, newpatient = c(1,0.5))


[Package bipd version 0.3 Index]