computeEase: Compute Expected Absolute Systematic Error (EASE)

In SelfControlledCohort: Self-Controlled Cohort Population-Level Estimation

Compute Expected Absolute Systematic Error (EASE)

Description

Computes the expected absolute systematic error from the null distribution fitted on negative control estimates. EASE summarizes both bias (mean of null) and imprecision (spread of null) into a single metric.

Usage

computeEase(negatives)

Arguments

negatives

Data frame of negative control results with columns 'rr' and 'seLogRr'

Details

EASE is computed by fitting a null distribution to the negative control log rate ratios using EmpiricalCalibration::fitNull(), then calling EmpiricalCalibration::computeExpectedAbsoluteSystematicError().

Lower values indicate less systematic error. A value of 0 means no detectable bias. The default threshold of 0.25 is aligned with SCCS package conventions.

Value

Numeric EASE value, or NA if the null distribution could not be fitted.

References

Schuemie MJ, Hripcsak G, Ryan PB, Madigan D, Suchard MA. Empirical confidence interval calibration for population-level effect estimation studies in observational healthcare data. PNAS. 2018;115(11):2571-2577.

Examples

if (interactive()) {
# Compute EASE from negative control results
negativeControls <- data.frame(
  rr = c(0.95, 1.02, 0.98, 1.05),
  seLogRr = c(0.2, 0.18, 0.22, 0.19)
)
ease <- computeEase(negativeControls)
}

SelfControlledCohort documentation built on June 17, 2026, 5:07 p.m.