Create era covariate settings
createEraCovariateSettings(
includeEraIds,
excludeEraIds = NULL,
label = "Covariates",
stratifyById = FALSE,
start = 0,
startAnchor = "era start",
end = 0,
endAnchor = "era end",
firstOccurrenceOnly = FALSE,
allowRegularization = FALSE,
profileLikelihood = FALSE,
exposureOfInterest = FALSE
)
One or more IDs of variables in the SccsData object that should be used to construct this covariate. If set to NULL, all variables will be used.
One or more IDs of variables in the [SccsData] object that should not be used to construct this covariate.
A label used to identify the covariates created using these settings.
Should a single covariate be created for every ID in the SccsData object, or should a single covariate be constructed? For example, if the IDs identify exposures to different drugs, should a covariate be constructed for every drug, or a single covariate for exposure to any of these drugs. Note that overlap will be considered a single exposure.
The start of the risk window (in days) relative to the startAnchor
.
The anchor point for the start of the risk window. Can be "era start"
or "era end"
.
The end of the risk window (in days) relative to the endAnchor
.
The anchor point for the end of the risk window. Can be "era start"
or "era end"
.
Should only the first occurrence of the exposure be used?
When fitting the model, should the covariates defined here be allowed to be regularized?
When fitting the model, should the likelihood profile be computed for the covariate defined here? The likelihood profile can be used to avoid making normal approximations on the likelihood and can be used in methods specifically designed to make use of the profile, but may take a while to compute.
If TRUE, the fitted coefficient for this variable will be reported when
using runSccsAnalyses()
. Requires includeEraIds
to be a exposure
reference ID as defined in createExposure()
.
An object of type EraCovariateSettings
.
Create an object specifying how to create a (set of) era-based covariates.