ate_assessment: Assess Analytical Performance Against Allowable Total Error

In valytics: Statistical Methods for Analytical Method Comparison and Validation

Assess Analytical Performance Against Allowable Total Error

Description

Evaluates observed analytical performance (bias and imprecision) against allowable total error specifications. Provides pass/fail assessment for individual components and overall method acceptability, along with the sigma metric.

Usage

ate_assessment(
  bias,
  cv,
  tea,
  allowable_bias = NULL,
  allowable_cv = NULL,
  k = 1.65
)

Arguments

bias	Numeric. Observed bias (systematic error), expressed as a percentage.
cv	Numeric. Observed coefficient of variation (imprecision), expressed as a percentage.
tea	Numeric. Total allowable error specification. Can be provided directly or will be calculated if allowable_bias and allowable_cv are provided with k.
allowable_bias	Numeric. Allowable bias specification (optional). If provided, enables individual bias assessment.
allowable_cv	Numeric. Allowable imprecision specification (optional). If provided, enables individual CV assessment.
k	Numeric. Coverage factor for TEa calculation when using component specifications (default: 1.65).

Details

The assessment evaluates method performance at multiple levels:

Component Assessment (if specifications provided):

• Bias: Pass if |observed bias| <= allowable bias

• CV: Pass if observed CV <= allowable CV

Total Error Assessment:

• Observed TE = k * CV + |Bias| (linear model)

• Pass if observed TE <= TEa

Sigma Metric:

• Sigma = (TEa - |Bias|) / CV

• Provides quality rating from "World Class" to "Unacceptable"

Value

An object of class c("ate_assessment", "valytics_ate", "valytics_result"), which is a list containing:

assessment

List with pass/fail results:

• bias_acceptable: Logical; TRUE if |bias| <= allowable_bias

• cv_acceptable: Logical; TRUE if cv <= allowable_cv

• tea_acceptable: Logical; TRUE if observed TE <= TEa

• overall: Logical; TRUE if method meets specifications

observed

List with observed performance:

• bias: Observed bias

• cv: Observed CV

• te: Observed total error (k * CV + |Bias|)

specifications

List with allowable specifications:

• allowable_bias: Allowable bias (or NULL)

• allowable_cv: Allowable CV (or NULL)

• tea: Total allowable error

sigma

List with sigma metric results:

• value: Sigma metric value

• category: Performance category

settings

List with settings:

• k: Coverage factor used

Overall Assessment

The overall assessment is determined as follows:

• If only TEa is provided: based on total error assessment

• If component specs provided: all components must pass

• Sigma >= 3 is generally considered minimum acceptable

References

Westgard JO (2008). Basic Method Validation (3rd ed.). Westgard QC, Inc.

Fraser CG (2001). Biological Variation: From Principles to Practice. AACC Press.

See Also

ate_from_bv() for calculating specifications from biological variation, sigma_metric() for sigma calculation details

Examples

# Basic assessment with TEa only
assess <- ate_assessment(bias = 1.5, cv = 2.5, tea = 10)
assess

# Assessment with all component specifications
assess_full <- ate_assessment(
  bias = 1.5,
  cv = 2.5,
  tea = 10,
  allowable_bias = 3.0,
  allowable_cv = 4.0
)
assess_full

# Using specifications from ate_from_bv()
specs <- ate_from_bv(cvi = 5.6, cvg = 7.5)
assess <- ate_assessment(
  bias = 1.5,
  cv = 2.5,
  tea = specs$specifications$tea,
  allowable_bias = specs$specifications$allowable_bias,
  allowable_cv = specs$specifications$allowable_cv
)
summary(assess)

# Check if method passes
assess$assessment$overall

valytics documentation built on Feb. 19, 2026, 5:06 p.m.