,
Craig Wang [aut] ,
Dominic Magirr [aut]| Title: | Binary Endpoint Estimation with Covariate Adjustment |
|---|---|
| Description: | Performs estimation of marginal treatment effects for binary outcomes when using logistic regression working models with covariate adjustment (see discussions in Magirr et al (2024) <https://osf.io/9mp58/>). Implements the variance estimators of Ge et al (2011) <doi:10.1177/009286151104500409> and Ye et al (2023) <doi:10.1080/24754269.2023.2205802>. |
| Authors: | Alex Przybylski [cre, aut],
Mark Baillie [aut] ,
Craig Wang [aut] ,
Dominic Magirr [aut] |
| Maintainer: | Alex Przybylski <[email protected]> |
| License: | LGPL (>= 3) |
| Version: | 0.2.0 |
| Built: | 2024-11-12 13:25:24 UTC |
| Source: | https://github.com/openpharma/beeca |
Calculates the marginal estimate of treatment effect and its corresponding standard error based on a fitted GLM object using specified contrast (summary measure) methods
apply_contrast( object, contrast = c("diff", "rr", "or", "logrr", "logor"), reference )apply_contrast( object, contrast = c("diff", "rr", "or", "logrr", "logor"), reference )
object |
a fitted |
contrast |
a string specifying the type of contrast to apply.
Accepted values are "diff" (risk difference), "rr" (risk ratio),
"or" (odds ratio), "logrr" (log risk ratio), "logor" (log odds ratio).
Note: log-transformed ratios (logrr and logor) work better compared to rr
and or when computing confidence intervals using normal approximation.
The choice of contrast affects how treatment effects are calculated and
interpreted. Default is |
reference |
a string or list of strings indicating which treatment
group(s) to use as reference level for pairwise comparisons. Accepted values
must be a subset of the levels in the treatment variable. Default to the
first n-1 treatment levels used in the This parameter influences the calculation of treatment effects relative to the chosen reference group. |
The apply_constrast() functions computes the summary measure between two arms
based on the estimated marginal effect and its variance-covariance matrix using
the Delta method.
Note: Ensure that the glm object has been adequately prepared with
average_predictions() and estimate_varcov()
before applying apply_contrast(). Failure to do so may result in
errors indicating missing components.
An updated glm object with two additional components
appended: marginal_est (marginal estimate of the treatment effect)
and marginal_se (standard error of the marginal estimate).
These appended component provide crucial information for interpreting
the treatment effect using the specified contrast method.
get_marginal_effect() for estimating marginal effects directly
from an original glm object
trial01$trtp <- factor(trial01$trtp) fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) |> predict_counterfactuals(trt = "trtp") |> average_predictions() |> estimate_varcov(method = "Ye") |> apply_contrast("diff", reference = "0") # Assuming `trial01` is a dataset with treatment (`trtp`) # and baseline covariate (`bl_cov`) trial01$trtp <- factor(trial01$trtp) fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) # Preprocess fit1 as required by apply_contrast fit2 <- fit1 |> predict_counterfactuals(trt = "trtp") |> average_predictions() |> estimate_varcov(method = "Ye") # Apply contrast to calculate marginal estimates fit3 <- apply_contrast(fit2, contrast = "diff", reference = "0") fit3$marginal_est fit3$marginal_setrial01$trtp <- factor(trial01$trtp) fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) |> predict_counterfactuals(trt = "trtp") |> average_predictions() |> estimate_varcov(method = "Ye") |> apply_contrast("diff", reference = "0") # Assuming `trial01` is a dataset with treatment (`trtp`) # and baseline covariate (`bl_cov`) trial01$trtp <- factor(trial01$trtp) fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) # Preprocess fit1 as required by apply_contrast fit2 <- fit1 |> predict_counterfactuals(trt = "trtp") |> average_predictions() |> estimate_varcov(method = "Ye") # Apply contrast to calculate marginal estimates fit3 <- apply_contrast(fit2, contrast = "diff", reference = "0") fit3$marginal_est fit3$marginal_se
average_predictions() averages counterfactual predictions stored within
a glm object. This is pivotal for estimating treatment contrasts and
associated variance estimates using g-computation. The function assumes
predictions are generated via predict_counterfactuals().
average_predictions(object)average_predictions(object)
object |
a fitted |
The average_predictions() function calculates the average over the
counterfactual predictions which can then be used to estimate a treatment
contrast and associated variance estimate.
The function appends a glm object with the
averaged counterfactual predictions.
Note: Ensure that the glm object has been adequately prepared with
predict_counterfactuals() before applying average_predictions().
Failure to do so may result in errors indicating missing components.
an updated glm object appended with an additional component
counterfactual.means.
predict_counterfactuals() for generating counterfactual
predictions.
estimate_varcov() for estimating the variance-covariate matrix
of mariginal effects
get_marginal_effect() for estimating marginal effects directly
from an original glm object
# Use the trial01 dataset data(trial01) # ensure the treatment indicator is a factor trial01$trtp <- factor(trial01$trtp) # fit glm model for trial data fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) # Preprocess fit1 as required by average_predictions fit2 <- fit1 |> predict_counterfactuals(trt = "trtp") # average over the counterfactual predictions fit3 <- average_predictions(fit2) # display the average predictions fit3$counterfactual.means# Use the trial01 dataset data(trial01) # ensure the treatment indicator is a factor trial01$trtp <- factor(trial01$trtp) # fit glm model for trial data fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) # Preprocess fit1 as required by average_predictions fit2 <- fit1 |> predict_counterfactuals(trt = "trtp") # average over the counterfactual predictions fit3 <- average_predictions(fit2) # display the average predictions fit3$counterfactual.means
Main variance estimation function. Estimates the variance-covariance matrix of a marginal estimand for a generalized linear model (GLM) object using specified methods. This function supports both Ge's and Ye's methods for variance estimation, accommodating different estimand specifications.
estimate_varcov( object, strata = NULL, method = c("Ge", "Ye"), type = c("HC0", "model-based", "HC3", "HC", "HC1", "HC2", "HC4", "HC4m", "HC5"), mod = FALSE )estimate_varcov( object, strata = NULL, method = c("Ge", "Ye"), type = c("HC0", "model-based", "HC3", "HC", "HC1", "HC2", "HC4", "HC4m", "HC5"), mod = FALSE )
object |
a fitted |
strata |
an optional string or vector of strings specifying the names of stratification variables. Relevant only for Ye's method and used to adjust the variance-covariance estimation for stratification. If provided, each specified variable must be present in the model. |
method |
a string indicating the chosen method for variance estimation.
Supported methods are |
type |
a string indicating the type of variance estimator to use (only applicable for Ge's method). Supported types include HC0 (default), model-based, HC3, HC, HC1, HC2, HC4, HC4m, and HC5. See vcovHC for heteroscedasticity-consistent estimators. This parameter allows for flexibility in handling heteroscedasticity and model specification errors. |
mod |
For Ye's method, the implementation of open-source RobinCar package
has an additional variance decomposition step when estimating the robust variance,
which then utilizes different counterfactual outcomes than the original reference.
Set |
The estimate_varcov function facilitates robust variance estimation
techniques for GLM models, particularly useful in clinical trial analysis
and other fields requiring robust statistical inference. It allows
researchers to account for complex study designs,
including stratification and different treatment contrasts,
by providing a flexible interface for variance-covariance estimation.
Note: Ensure that the glm object has been adequately prepared with
predict_counterfactuals and average_predictions
before applying estimate_varcov(). Failure to do so may result in
errors indicating missing components.
an updated glm object appended with an
additional component robust_varcov, which is the estimated variance-covariance matrix
of the marginal effect. The matrix format and estimation method are
indicated in the matrix attributes.
Ye T. et al. (2023) Robust variance estimation for covariate-adjusted unconditional treatment effect in randomized clinical trials with binary outcomes. Statistical Theory and Related Fields
Ge M. et al. (2011) Covariate-Adjusted Difference in Proportions from Clinical Trials Using Logistic Regression and Weighted Risk Differences. Drug Information Journal.
Bannick, M. S., et al. A General Form of Covariate Adjustment in Randomized Clinical Trials. arXiv preprint arXiv:2306.10213 (2023).
average_predictions() for averaging counterfactual
predictions.
apply_contrast() for computing a summary measure.
get_marginal_effect() for estimating marginal effects directly
from an original glm object
# Example usage with a binary outcome GLM model trial01$trtp <- factor(trial01$trtp) fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) #' # Preprocess fit1 as required by estimate_varcov fit2 <- fit1 |> predict_counterfactuals(trt = "trtp") |> average_predictions() # Estimate variance-covariance using Ge's method fit3_ge <- estimate_varcov(fit2, method = "Ge") print(fit3_ge$robust_varcov) # Estimate variance-covariance using Ye's method with stratification fit4 <- glm(aval ~ trtp + bl_cov_c, family = "binomial", data = trial01) |> predict_counterfactuals(trt = "trtp") |> average_predictions() fit4_ye <- estimate_varcov(fit4, method = "Ye", strata = "bl_cov_c") print(fit4_ye$robust_varcov)# Example usage with a binary outcome GLM model trial01$trtp <- factor(trial01$trtp) fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) #' # Preprocess fit1 as required by estimate_varcov fit2 <- fit1 |> predict_counterfactuals(trt = "trtp") |> average_predictions() # Estimate variance-covariance using Ge's method fit3_ge <- estimate_varcov(fit2, method = "Ge") print(fit3_ge$robust_varcov) # Estimate variance-covariance using Ye's method with stratification fit4 <- glm(aval ~ trtp + bl_cov_c, family = "binomial", data = trial01) |> predict_counterfactuals(trt = "trtp") |> average_predictions() fit4_ye <- estimate_varcov(fit4, method = "Ye", strata = "bl_cov_c") print(fit4_ye$robust_varcov)
For purposes of implementation comparisons, these are the result outputs from the SAS macro provided with the Ge et al (2011) publication (https://doi.org/10.1177/009286151104500409), applied to the trial01 dataset included with beeca, adjusting for treatment (trtp) and a single covariate (bl_cov) and targeting a risk difference contrast.
ge_macro_trial01ge_macro_trial01
ge_macro_trial01
A tibble with 1 row and 6 columns:
Marginal risk difference estimate
Standard error of marginal risk difference estimate
Marginal risk in treated
Marginal risk in controls
Lower bound of 95 percent confidence interval of risk difference estimate
Upper bound of 95 percent confidence interval of risk difference estimate
Estimates the marginal treatment effect from a logistic regression working model using a specified choice of variance estimator and contrast.
get_marginal_effect( object, trt, strata = NULL, method = "Ge", type = "HC0", contrast = "diff", reference, mod = FALSE )get_marginal_effect( object, trt, strata = NULL, method = "Ge", type = "HC0", contrast = "diff", reference, mod = FALSE )
object |
a fitted glm object. |
trt |
a string specifying the name of the treatment variable
in the model formula. It must be one of the linear predictor variables used
in fitting the |
strata |
an optional string or vector of strings specifying the names of stratification variables. Relevant only for Ye's method and used to adjust the variance-covariance estimation for stratification. If provided, each specified variable must be present in the model. |
method |
a string indicating the chosen method for variance estimation.
Supported methods are |
type |
a string indicating the type of variance estimator to use (only applicable for Ge's method). Supported types include HC0 (default), model-based, HC3, HC, HC1, HC2, HC4, HC4m, and HC5. See vcovHC for heteroscedasticity-consistent estimators. |
contrast |
a string indicating choice of contrast. Defaults to 'diff' for a risk difference. See apply_contrast. |
reference |
a string or list of strings indicating which treatment
group(s) to use as reference level for pairwise comparisons. Accepted values
must be a subset of the levels in the treatment variable. Default to the
first n-1 treatment levels used in the |
mod |
for Ye's method, the implementation of open-source RobinCar package
has an additional variance decomposition step when estimating the robust variance,
which then utilizes different counterfactual outcomes than the original reference.
Set |
The get_marginal_effect function is a wrapper that facilitates
advanced variance estimation techniques for GLM models with covariate adjustment
targeting a population average treatment effect. It is particularly useful in clinical trial analysis
and other fields requiring robust statistical inference.
It allows researchers to account for complex study designs,
including stratification and treatment contrasts, by providing a flexible
interface for variance-covariance estimation.
an updated glm object appended with marginal estimate components:
counterfactual.predictions (see predict_counterfactuals),
counterfactual.means (see average_predictions),
robust_varcov (see estimate_varcov),
marginal_est, marginal_se (see apply_contrast) and marginal_results. A summary is shown below
| counterfactual.predictions | Counterfactual predictions based on the working model. For each subject in the input glm data, the potential outcomes are obtained by assigning subjects to each of the possible treatment variable levels. Each prediction is associated with a descriptive label explaining the counterfactual scenario. |
| counterfactual.means | Average of the counterfactual predictions for each level of the treatment variable. |
| robust_varcov | Variance-covariance matrix of the marginal effect estimate for each level of treatment variable, with estimation method indicated in the attributes. |
| marginal_est | Marginal treatment effect estimate for a given contrast. |
| marginal_se | Standard error estimate of the marginal treatment effect estimate. |
| marginal_results | Analysis results data (ARD) containing a summary of the analysis for subsequent reporting. |
trial01$trtp <- factor(trial01$trtp) fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) |> get_marginal_effect(trt = "trtp", method = "Ye", contrast = "diff", reference = "0") fit1$marginal_resultstrial01$trtp <- factor(trial01$trtp) fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) |> get_marginal_effect(trt = "trtp", method = "Ye", contrast = "diff", reference = "0") fit1$marginal_results
For purposes of implementation comparisons, these are the result outputs from the SAS Margins macro (https://support.sas.com/kb/63/038.html), applied to the trial01 dataset included with beeca, adjusting for treatment (trtp) and a single covariate (bl_cov) and targeting a risk difference contrast.
margins_trial01margins_trial01
margins_trial01
A tibble with 1 row and 11 columns:
Marginal risk difference estimate
Wald Chi-Square statistic
Row number
Standard error of marginal risk difference estimate
Lower bound of 95 percent confidence interval of estimate
Upper bound of 95 percent confidence interval of estimate
Descriptive label for contrast
Degrees of freedom
p-value
Significance level alpha
Label for contrast
This function calculates counterfactual predictions for each level of a specified treatment variable in a generalized linear model (GLM). It is designed to aid in the assessment of treatment effects by predicting outcomes under different treatments under causal inference framework.
predict_counterfactuals(object, trt)predict_counterfactuals(object, trt)
object |
a fitted |
trt |
a string specifying the name of the treatment variable
in the model formula. It must be one of the linear predictor variables used
in fitting the |
The function works by creating new datasets from the original data used to fit the GLM model. In these datasets, the treatment variable for all records (e.g., patients) is set to each possible treatment level.
Predictions are then made for each dataset based on the fitted GLM model, simulating the response variable under each treatment condition.
The results are stored in a tidy format and appended to the original model object for further analysis or inspection.
For averaging counterfactual outcomes, apply average_predictions().
an updated glm object appended with an
additional component counterfactual.predictions.
This component contains a tibble with columns representing counterfactual
predictions for each level of the treatment variable. A descriptive label
attribute explains the counterfactual scenario associated with each column.
average_predictions() for averaging counterfactual
predictions.
get_marginal_effect() for estimating marginal effects directly
from an original glm object
# Preparing data and fitting a GLM model trial01$trtp <- factor(trial01$trtp) fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) # Generating counterfactual predictions fit2 <- predict_counterfactuals(fit1, "trtp") # Accessing the counterfactual predictions fit2$counterfactual.predictions attributes(fit2$counterfactual.predictions)# Preparing data and fitting a GLM model trial01$trtp <- factor(trial01$trtp) fit1 <- glm(aval ~ trtp + bl_cov, family = "binomial", data = trial01) # Generating counterfactual predictions fit2 <- predict_counterfactuals(fit1, "trtp") # Accessing the counterfactual predictions fit2$counterfactual.predictions attributes(fit2$counterfactual.predictions)
A simplified example of a simulated trial dataset, with missing data.
trial01trial01
trial01
A data frame with 268 rows and 9 columns:
Unique subject identifier
Primary outcome variable (1 = yes/0 = no)
Planned treatment
Baseline covariate (numeric)
Dichotomized version of bl_cov (category of 1 or 0)
Indicators for region (1 = yes/0 = no)
This dataset is a simplified, binary outcome version of a sample Phase 2 clinical trial dataset formatted according to the Analysis Data Model (ADaM) standards set by the Clinical Data Interchange Standards Consortium (CDISC). It is designed for training and educational purposes, showcasing how clinical trial data can be structured for statistical analysis.
trial02_cdisctrial02_cdisc
A data frame with 254 rows and 13 columns, representing trial participants and key variables:
Unique subject identifier (alphanumeric code). A code unique to the clinical trial
Parameter name indicating the specific measurement or outcome assessed.
Age of the participant at study enrollment, in years.
Categorical representation of age groups.
Numeric code representing age groups, used for statistical modeling.
Self-identified race of the participant
Numeric representation of race categories, used for statistical modeling.
Participant's sex at birth.
Planned treatment assignment, indicating the specific intervention or control condition.
Numeric code for the planned treatment, simplifying data analysis procedures.
Analysis value, representing the primary outcome measure for each participant.
Character representation of the analysis value, used in descriptive summaries.
Full analysis set flag, indicating if the participant's data is included in the full analysis set.
This dataset serves as an illustrative example for those learning about the ADaM standard in clinical trials. It includes common variables like demographic information, treatment assignments, and outcome measures.
Data privacy and ethical considerations have been addressed through the anonymization of subject identifiers and other sensitive information. The dataset is intended for educational and training purposes only.
The numeric codes for categorical variables such as RACEN and TRTPN
are arbitrary and should be interpreted within the context of this dataset.
For example, refer to the categorical representations for additional context.
This dataset has been reformatted for educational use from the
safetyData package, specifically adam_adtte. For the original data and
more detailed information, please refer to the
safetyData documentation.