Questions: Analytical Method Validation: Core Performance Parameters

Question 1

A new HPLC method measures a reference standard at 118% recovery consistently across three independent analysts on different days. What validation parameter failure does this indicate, and what does it tell you about the method?

• A. Poor precision — the 18% deviation means different analysts are getting different results
• B. Poor accuracy (systematic positive bias) — the method consistently overestimates by 18%, regardless of analyst or day
• C. Inadequate linearity — 118% recovery means the calibration curve is non-linear at this concentration
• D. Insufficient range — 118% recovery indicates the method is operating above its validated upper limit

Answer: (B)

Consistent 118% recovery across three independent analysts is the signature of systematic positive bias — the method reliably measures 18% more than the true value. This is an accuracy (trueness) failure, not a precision failure. If all three analysts get approximately 118%, the method is actually precise (reproducible across analysts) but inaccurate (biased). Accuracy and precision are independent dimensions: a method can be precisely wrong. This distinction matters because the corrective strategies are completely different — bias requires method investigation (sample preparation, calibration, matrix effects), not repeatability optimization.

Question 2

During robustness testing, an analyst changes mobile phase pH from 3.0 to 3.2 (within the ±0.2 unit range being evaluated). Peak resolution drops from 4.0 to 1.2, falling below the specification of ≥2.0. What does this finding indicate?

• A. The method has failed linearity testing and requires a new calibration model
• B. The method lacks robustness — it is fragile to small, realistic perturbations that will occur in normal laboratory variation
• C. The method has poor intermediate precision and needs to be repeated by a different analyst
• D. The validated range is too narrow; wider concentration range testing will resolve the sensitivity to pH

Answer: (B)

Robustness testing evaluates whether the method tolerates small, intentional perturbations to critical parameters — the same variations that occur unintentionally in routine use (slightly different pH meters, different reagent batches, minor temperature fluctuations). A method that fails specification when pH changes by only 0.2 units will produce out-of-specification results in normal laboratory operation. This is a robustness failure, identified during validation so that the critical parameter can be tightened in the method specification before the method is deployed. It is unrelated to linearity, precision, or range.

Question 3

A method that demonstrates high precision — tight, reproducible results across multiple analysts and days — can be approved for quantitative analysis even if accuracy has not been formally characterized.

• T. True
• F. False

Answer: False

Precision and accuracy are independent dimensions of method performance. A highly precise method consistently produces reproducible results, but those results may be consistently wrong — systematically biased by 20% due to matrix interference, extraction inefficiency, or calibration error. ICH Q2(R2) requires both accuracy and precision to be characterized and to meet acceptance criteria. Precision demonstrates that the method measures something reproducibly; accuracy demonstrates that it measures the right thing. Both are required before a method can be considered validated.

Question 4

The validated range of an analytical method is the concentration interval over which accuracy, precision, and linearity all simultaneously meet their acceptance criteria.

• T. True
• F. False

Answer: True

The validated range is not simply the calibration range or the interval over which the detector produces a response. It is the proven operating space — the concentration interval where the complete set of key performance requirements (accuracy within specification, precision within specification, and linear response) holds simultaneously. A calibration curve might be linear across a wide range while accuracy degrades at very low concentrations (below LOQ) or very high concentrations (approaching saturation). The validated range is bounded by where all requirements are met together, not just where the curve looks good.

Question 5

A researcher says: 'My method has r² = 0.9998 for the calibration curve, so it's ready for use.' What critical dimensions of method performance does this statement overlook?

Model answer: A high r² confirms linear proportionality between analyte concentration and detector response — but this addresses only one of many validation parameters. It says nothing about accuracy (whether the calibration is systematically biased — high r² with 20% positive bias is still r² = 0.9998), precision (repeatability and reproducibility across analysts, days, and laboratories), specificity (whether co-eluting degradation products or matrix components interfere with the analyte peak), robustness (whether performance holds when method parameters like pH or temperature vary slightly), or the validated range boundaries (whether accuracy and precision remain acceptable at the lowest and highest concentrations). A method can have near-perfect linearity and still be consistently inaccurate, irreproducible between analysts, or sensitive to minor mobile phase changes.

r² is the metric students most readily grasp from statistics, making it a tempting proxy for 'the method is validated.' This question tests whether students understand that linearity is one dimension among many, and that each validation parameter answers a distinct, non-redundant question about method fitness for intended use.