# ICH Analytical Validation Framework

> ICH analytical validation in Pharmacy: Q2(R2) and Q14, which parameters each test type needs, forced degradation, verification of pharmacopoeial methods.

- Canonical URL: https://prepelephant.com/topics/allied/pharmacy/validation-analytical-ich
- Exam / course: Allied Health · Subject: Pharmacy
- Publisher: PrepElephant (https://prepelephant.com) — Prepared and reviewed by the PrepElephant Academic Review Team
- First published: 2026-10-02
- Last updated: 2026-10-02
- How to cite: "ICH Analytical Validation Framework", PrepElephant, https://prepelephant.com/topics/allied/pharmacy/validation-analytical-ich

## Direct answer

ICH Q2(R2), adopted on 1 November 2023 together with Q14 on analytical procedure development, is the framework that decides what must be proven before an analytical procedure may release product: specificity or selectivity, linearity, range, accuracy, precision (repeatability, intermediate precision, reproducibility), detection limit, quantitation limit and robustness — applied selectively according to the test type, since an identification test needs specificity alone while an assay or impurity method carries the full load. The 2023 revision extends validation to multivariate procedures such as NIR and Raman and ties validation into the analytical target profile and lifecycle thinking of Q14. Two boundaries shape it: pharmacopoeial methods are verified, not revalidated, under actual laboratory conditions, and any in-house method must first be shown stability-indicating by forced degradation. Detection and quantitation limits are anchored to the formulas DL = 3.3σ/S and QL = 10σ/S, standard deviations over slope.

## What you must remember

- **Guideline identity:** Q2(R2) for validation, Q14 for development, both adopted November 2023, replacing Q2(R1) of 1994-96 vintage — the update examiners have begun to probe.
- **Parameter list:** specificity, linearity, range, accuracy, precision, detection limit, quantitation limit, robustness — the eight-item core that Q2 tables against test type.
- **Test-type mapping:** identification needs specificity only; assay needs linearity, accuracy, precision, range, robustness; impurity quantitation adds DL and QL; limit tests need specificity and DL.
- **Formulas:** DL = 3.3σ/S and QL = 10σ/S; σ may come from calibration residuals or the response of blank/low-level samples.
- **Verification, not validation:** compendial methods are verified in the receiving lab by confirming suitability under actual conditions (USP <1226> logic); only in-house or modified methods are fully validated.
- **Stability-indicating requirement:** a method controlling stability samples must resolve drug from all relevant degradants, demonstrated by forced degradation (acid, base, oxidant, heat, light).
- **Multivariate scope:** Q2(R2) newly covers models with latent variables — NIR, Raman — validating calibration, cross-validation and model maintenance.
- **Revalidation triggers:** changes in method, equipment, component sources or synthesis route reopen selected validation experiments, decided through risk assessment.

## Building a stability-indicating HPLC assay

A new fixed-dose tablet needs a release method. Development (the Q14 half) begins with the analytical target profile: assay both actives against 95-105 per cent of claim and quantify each impurity well below specification. Forced degradation generates the challenge set — acid, base, peroxide, thermal and photolytic stress targeting 5-20 per cent degradation. The chromatographer iterates pH, organic fraction and gradient until drug peaks emerge pure by photodiode-array analysis with degradants resolved — that demonstration is what "stability-indicating" means, and no validation package survives inspection without it.

Validation then matches experiments to claims: accuracy by recovery at 80, 100 and 120 per cent; precision as replicate RSD within and between days, analysts and instruments; linearity across five levels; robustness by wobbling flow, pH, temperature and wavelength; DL and QL for the impurity side only. A later transfer to a contract laboratory runs either re-validation of key elements or a transfer study with pre-agreed criteria, and a faster UPLC revision scopes revalidation by risk rather than rewriting it. The framework's essence: experiments answer specific claims, and every claim is documented against data.

## Where students slip

Validation is confused with system suitability at every level: suitability (replicate RSD, tailing, resolution) runs with each sequence and proves today's system, while validation is the one-time, change-triggered demonstration that the method works — answers merging them lose the conceptual mark. Second, candidates validate everything on every method; Q2's table exists to prevent this, and assigning DL/QL to a release assay reveals rote learning. Third, verification versus validation: an IP monograph method still must be verified on the receiving lab's column, instrument and matrix. Fourth, the revision blind spot: answers describing Q2(R1) as current miss that Q2(R2) with Q14 (November 2023) added multivariate validation and lifecycle development. Finally, specificity, per Q2(R2), is the ability to assess the analyte unequivocally in the presence of expected components — define it in those terms.

## Frequently asked questions

### Which validation parameters does ICH Q2 require for an assay method?

Specificity, linearity, range, accuracy, precision and robustness — detection and quantitation limits belong to impurity and limit-test methods, not to release assays.

### What changed from Q2(R1) to Q2(R2)?

The 2023 revision adds validation of multivariate procedures such as NIR and Raman, aligns with Q14 development and lifecycle concepts, and restructures evaluation of precision and accuracy across technologies.

### How does verification differ from validation?

Pharmacopoeial methods need only verification — demonstrating under the receiving lab's conditions that suitability criteria are met — whereas in-house or modified methods require full validation.

### Why is forced degradation performed before validation?

It generates realistic degradants to prove the method is stability-indicating — able to separate intact drug from all relevant degradation products — without which stability data cannot be trusted.

### How are detection and quantitation limits calculated?

DL = 3.3σ/S and QL = 10σ/S, where σ is the standard deviation of response or residuals and S the calibration slope; signal-to-noise ratios of about 3 and 10 are acceptable alternatives.