Plot Results from Multiply Imputed Resilience Analysis

Description

Plot method for visualizing imputation diagnostics

Usage

## S3 method for class 'prepost_resilience_mi_list'
plot(x, type = "coefficients", ...)

Arguments

Identifying Predictors of Resilience to Stressors in Single-Arm Studies of Pre-Post Change

Description

Studies of resilience in older adults are typically conducted with a single-arm where everyone experiences the stressor. The simplistic approach of regressing change versus baseline yields biased estimates due to mathematical coupling and regression-to-the-mean. This function provides a method to correct the bias.

Usage

prepost(
  formula,
  data,
  change = TRUE,
  k = c(1, 1.5, 2),
  m = 1,
  nboot = 1000,
  ci.level = 0.95,
  boot.method = c("perc", "norm", "basic", "bca"),
  ncores = 2
)

Arguments

Value

A list with the following components:

naive.beta

Unadjusted, naive estimates of regression coefficients

corrected.beta

The corrected coefficients of the variables. A matrix with one column of parameter values for each value of sensitivity parameter k

CI

A list of length equal to the number of sensitivity values. Each component of the list is a matrix with two columns of lower and upper confidence interval for each parameter

Missing Data

This function uses complete-case analysis only. If your data contains missing values, the function will issue a warning and exclude cases with any missing values. For more robust inference with missing data, consider using multiple imputation followed by prepost_resilience_mi_list(), which pools results across multiply imputed datasets using Rubin's rules.

Parallel Processing

The function uses the parallel and foreach packages to perform parallel computations of bootstrap confidence intervals for different values of the sensitivity parameter, k.

Author(s)

Ravi Varadhan

References

Varadhan, R., Zhu, J., and Bandeen-Roche, K (2024). Identifying Predictors of Resilience to Stressors in Single-Arm Studies of Pre-Post Change. Biostatistics. 25(4): 1094-1111.

See Also

prepost_resilience_mi_list for analysis with multiply imputed data.

Examples

## Not run: 
data(tkr)

names(tkr.dat)
dim(tkr.dat)

# pre-post change regression
ans1 <- prepost(post.Y ~ pre.Y + I(age-mean(age)) + I((age - mean(age))^2) + 
  bmi + gender + as.factor(smoker), data=tkr.dat, k=c(1.2, 1.5), nboot=200)
print(ans1)

# Post regression
ans2 <- prepost(post.Y ~ pre.Y + I(age-mean(age)) + I((age - mean(age))^2) + 
  bmi + gender + as.factor(smoker), data=tkr.dat, 
  k=c(1.2, 1.5), change=FALSE, nboot=200, boot.method="norm")
print(ans2)

# without any covariates
ans3 <- prepost(post.Y ~ pre.Y, data=tkr.dat, k=c(1.2, 2.0), nboot=200)
print(ans3)

# Bootstrapping using "bca" - relatively slow
# Not run
# ans4 <- prepost(post.Y ~ pre.Y, data=tkr.dat, k=c(1.2, 2.0), change=FALSE,
#   boot.method = "bca")

## End(Not run)

Prep-Resilience Analysis with Pre-Imputed Datasets

Description

This function takes a list of already imputed datasets and runs resilience analysis on each, then pools the results using Rubin's rules. Users are responsible for the imputation process, allowing maximum flexibility in imputation methods.

Usage

prepost_resilience_mi_list(
  data_list,
  formula,
  k = 1,
  nboot = 500,
  pool_method = "detailed",
  coef_labels = NULL,
  verbose = TRUE,
  ...
)

Arguments

Value

A list containing: - pooled_results: The pooled estimates - results_table: Formatted results table - individual_results: List of results from each imputation - model_info: Information about the model specification - imputation_info: Summary of imputation characteristics

See Also

prepost for complete-case analysis without missing data mice for multiple imputation

Examples

## Not run: 
# Example 1: List of data frames from any imputation method
library(mice)
data <- data.frame(
  post_score = rnorm(100),
  pre_score = rnorm(100),
  age = rnorm(100, 50, 10),
  bmi = rnorm(100, 25, 5)
)

# Create some missing data
data$age[1:10] <- NA
data$bmi[5:15] <- NA

# User does their own imputation (could be from any package/method)
# Method 1: Using mice
imp_mice <- mice(data, m = 5, printFlag = FALSE)
imp_list <- list()
for (i in 1:5) {
  imp_list[[i]] <- complete(imp_mice, i)
}

# Method 2: Using Amelia
if (requireNamespace("Amelia", quietly = TRUE)) {
  imp_amelia <- Amelia::amelia(data, m = 5)
  imp_list <- imp_amelia$imputations
}

# Method 3: Using aregImpute from Hmisc
if (requireNamespace("Hmisc", quietly = TRUE)) {
  set.seed(123)
  imp_areg <- Hmisc::aregImpute(~ post_score + pre_score + age + bmi, data = data, n.impute = 5)
  imp_list <- list()
  for (i in 1:5) {
    imp_data <- data
    imp_data$age[is.na(imp_data$age)] <- imp_areg$imputed$age[, i]
    imp_data$bmi[is.na(imp_data$bmi)] <- imp_areg$imputed$bmi[, i]
    imp_list[[i]] <- imp_data
  }
}

# Run resilience analysis on the pre-imputed list
result <- prepost_resilience_mi_list(
  data_list = imp_list,
  formula = post_score ~ pre_score + age + bmi,
  k = 1.2,
  nboot = 200
)

# Example 2: Directly using a mice mids object
result2 <- prepost_resilience_mi_list(
  data_list = imp_mice,  # mids object
  formula = post_score ~ pre_score + age + bmi
)

## End(Not run)

Print Results from Multiply Imputed Resilience Analysis

Description

S3 print method for prepost_resilience_mi_list

Usage

## S3 method for class 'prepost_resilience_mi_list'
print(x, ...)

Arguments

Summary Results from Multiply Imputed Resilience Analysis

Description

S3 summary method for prepost_resilience_mi_list

Usage

## S3 method for class 'prepost_resilience_mi_list'
summary(object, ...)

Arguments

Example dataset for pre-post resilience analysis

Description

A simulated dataset from a single-arm study with pre/post measurements.

Usage

tkr.dat

Format

A data frame with columns:

post.Y

Post-stressor outcome

pre.Y

Baseline outcome

age

Age in years

bmi

Body mass index

gender

Gender (M/F)

smoker

Smoking status

Source

Simulated for demonstration.