USP Analytical Instrument Qualification: Group A, B, and C

USP General Chapter <1058> — Analytical Instrument Qualification (AIQ) — is the pharmacopeial framework that tells a lab how much qualification effort a given instrument actually needs. Its central idea is a three-tier classification, Group A, Group B, and Group C, that scales the rigor of Design, Installation, Operational, and Performance Qualification (DQ/IQ/OQ/PQ) to how much an instrument’s measurement result depends on its own internal complexity. A magnetic stirrer and an HPLC system are both ‘lab equipment,’ but <1058> does not ask a lab to qualify them the same way, and understanding why is what keeps a qualification program both defensible and proportionate.

What USP <1058> actually is

<1058> sits in USP–NF’s General Chapters <1000>–<1999> range, which USP designates as informational rather than mandatory: it does not carry independent regulatory force the way a numbered chapter under <1000> (like the balance requirements in USP <41>) can when a monograph invokes it directly. In practice, <1058> functions as the de facto industry standard for analytical instrument qualification across pharmaceutical, biotech, and GMP-adjacent laboratories — FDA investigators and quality auditors routinely expect a lab’s qualification program to be consistent with it, and most commercial AIQ programs (vendor qualification services included) are built directly around its structure. See CASRAI’s guide to IQ/OQ/PQ for the underlying DQ/IQ/OQ/PQ qualification sequence itself; this page covers the classification layer <1058> adds on top of it — which instruments need which stages, and how much testing within each stage. If you’re unsure how qualification itself differs from calibration and validation as terms, CASRAI’s comparison of equipment qualification vs. validation vs. calibration untangles that first.

The three groups, not four

<1058> sorts instruments into three groups by asking one question: can the instrument’s fitness for use be confirmed by observation and basic function checks alone, or does confirming it requires measuring a value against a standard, and if so, how many interacting variables and how much software sit between the measurement and the number a user reads?

Group A — no measurement, or none requiring calibration

Group A covers equipment with no measurement function, or equipment whose measurement/display does not require calibration to be fit for its intended use. There is nothing here for OQ or PQ to test in the traditional sense — qualification is essentially confirming correct installation and that the unit functions as the manufacturer intends.

• Typical instruments: magnetic stirrers, vortex mixers, orbital shakers, general-purpose water baths used only to hold temperature loosely, basic centrifuges with no validated speed readout used for compendial testing, stir/heat plates.
• Qualification effort: IQ (confirm correct receipt, installation, utilities connected per manufacturer specification) plus a documented functional check that the unit powers on and performs its basic function. No formal OQ/PQ protocol, no calibration certificate, no periodic requalification beyond routine preventive maintenance.

Group B — measured values, verifiable by simple standards

Group B covers instruments that provide measured values, but where the correctness of that value can be verified using simple, well-characterized physical or chemical reference standards, or standard test procedures, without needing an automated or computerized data system to interpret the result.

• Typical instruments: analytical balances, pH meters, melting-point apparatus, non-automated titrators, refractometers, muffle furnaces and drying ovens, viscometers, simple conductivity meters.
• Qualification effort: full DQ/IQ/OQ/PQ against defined acceptance criteria, using traceable reference standards or certified weights — e.g., a balance’s OQ checks linearity, repeatability, and eccentricity against certified test weights per USP <41>. Requires a documented calibration/requalification schedule at a defined interval, not a one-time qualification event. See CASRAI’s guide to balance qualification for a worked Group B example end to end.

Group C — complex systems with computerized control

Group C covers instruments and systems whose correct operation depends on many interacting parameters that cannot be tested by simple, isolated checks, and that include computerized data acquisition, processing, or control. Confirming fitness for use here requires testing the instrument’s functional modules individually and as an integrated system, plus validating the software/data-system layer that touches the result.

• Typical instruments: HPLC and UPLC systems, GC systems, mass spectrometers, UV-Vis, IR, and NMR spectrophotometers, dissolution testers with automated sampling, ICP and atomic absorption spectrometers, automated titrators with software-driven endpoint calculation.
• Qualification effort: the most extensive of the three — DQ against documented user requirements, IQ covering both hardware and software installation, OQ that exercises each functional module (pump flow accuracy, detector linearity, autosampler precision, and so on) against specification, and PQ that runs the instrument through its actual intended method under real conditions of use. The computerized-system component is governed by the same GAMP 5 / 21 CFR Part 11 framework covered in CASRAI’s guide to Computer System Validation (CSV) — a Group C instrument is qualified under <1058> and its software is validated under GAMP 5 as two coordinated, not redundant, activities.

Classifying an instrument that doesn’t obviously fit

Two questions resolve most edge cases: does the result depend on a value the instrument measures (if not, it’s Group A), and if it does, can that value be verified with a simple external standard without a computerized data system standing between the sensor and the number (Group B), or does confirming correct operation require testing multiple interacting modules and a validated software layer (Group C)? A borderline example: a basic tabletop centrifuge with no speed display is Group A, the same centrifuge with a calibrated digital speed readout used to document run conditions for a regulated method is Group B, and a refrigerated ultracentrifuge with programmable, software-controlled run profiles feeding into an LIMS record is Group C. The classification is about the instrument’s role and complexity as actually used in your lab, not the equipment category in the abstract — the same model of instrument can sit in different groups at two different sites depending on how each site relies on it.

What doesn’t change across groups

Regardless of group, <1058> expects every instrument to have documented ownership of its qualification status, a defined periodic maintenance and (where applicable) requalification interval, and a change-control trigger — a firmware update, a relocation, or a major repair should prompt a documented review of whether requalification is needed, not an assumption that the original qualification still holds. A lab’s Validation Master Plan is the document that should record each instrument’s group assignment and qualification schedule in one place, so an auditor (or an incoming quality lead) does not have to reconstruct the logic instrument by instrument.

Frequently asked questions

Is USP <1058> mandatory?

Not on its own — it sits in USP–NF’s informational <1000>–<1999> chapter range, which is not independently enforceable the way a compendial requirement invoked by a monograph (like USP <41> for balances) is. In practice it is the de facto expected framework: auditors and FDA investigators assess pharmaceutical and GMP-adjacent qualification programs against it even where it isn’t cited as a hard requirement.

What’s the difference between USP <1058> and IQ/OQ/PQ?

IQ/OQ/PQ (with optional DQ ahead of it) is the qualification sequence — the stages themselves, used across GxP equipment qualification generally. USP <1058> is a classification scheme that determines how much of that sequence, and how much testing within each stage, a given instrument actually needs, based on its Group A/B/C complexity.

Does a Group A instrument still need calibration?

Only if its output is actually used to produce or support a measured result. A stir plate used purely to mix a solution needs no calibration; the same physical device with a temperature display that a method relies on to document run conditions moves it toward Group B.

How does <1058> relate to GAMP 5?

They cover different layers of the same Group C instrument. <1058> governs qualifying the physical instrument’s hardware performance; GAMP 5 governs validating the computerized system (software, data integrity, 21 CFR Part 11 controls) that acquires, processes, or stores the instrument’s data. A Group C instrument typically needs both, run as coordinated activities against a shared risk assessment rather than two separate, disconnected programs.

Who decides an instrument’s group assignment?

The lab does, documented in its qualification SOP or Validation Master Plan — <1058> provides the criteria and illustrative examples, not a definitive master list, because the same instrument model can be used in ways that place it in different groups at different sites.