Generate perfomance metrics for the OHDSI Methods Benchmark

computeOhdsiBenchmarkMetrics(
  exportFolder,
  mdrr = 1.25,
  stratum = "All",
  trueEffectSize = "Overall",
  calibrated = FALSE,
  comparative = FALSE
)

Arguments

exportFolder: The folder containing the CSV files created using the packageOhdsiBenchmarkResults function. This folder can contain results from various methods, analyses, and databases.
mdrr: The minimum detectable relative risk (MDRR). Only controls with this MDRR will be used to compute the performance metrics. Set to "All" to include all controls.
stratum: The stratum for which to compute the metrics, e.g. 'Acute Pancreatitis'. Set to 'All' to use all controls.
trueEffectSize: Should the analysis be limited to a specific true effect size? Set to "Overall" to include all.
calibrated: Should confidence intervals and p-values be empirically calibrated before computing the metrics?
comparative: Should the methods be evaluated on the task of comprative effect estimation? If FALSE, they will be evaluated on the task of effect estimation.

Value

A data frame with the various metrics per method - analysisId - database combination.