IFU (Instructions for Use): Medical Device Labeling Requirements

IFU (instructions for use) means the device-specific instructions that explain a medical device’s intended purpose, safe use and relevant limitations. This is ordinary medical-device industry usage, not a quoted statutory definition. An IFU helps its intended reader understand when and how to use the device, what to avoid and what to do when a relevant problem occurs.

The following comparison describes typical document functions, not mutually exclusive legal categories. A document can serve more than one function; its title alone does not determine its regulatory treatment.

The useful operational boundary is the intended reader and task, not document length or filename. Where relevant, patients may need self-use steps, caregivers assistance instructions, clinicians preparation and treatment information, and service personnel maintenance procedures. Not every device has all four audiences. Plan content around actual responsibilities rather than giving every reader the same technical detail.

Likewise, a technical file organizes supporting device documentation; it is not a substitute for usable instructions. Keep the user’s task distinct from the manufacturer’s need to retain supporting evidence.

This article starts with US regulatory requirements, including the FDA QMSR, then addresses suggested controls for instruction content, language and delivery variants, packaging use and revision history. Binding requirements are distinguished from practical document-control recommendations.

Under 21 CFR 801.5, adequate directions for use are directions under which a layperson can use a device safely and for its intended purposes. The standard concerns what the user can do with the directions—not simply whether a document titled “IFU” exists.

The prescription-device exemption in the introductory paragraph of 21 CFR 801.109 is conditional: exemption from section 502(f)(1) depends on meeting all the provision’s conditions. Calling a device “prescription-only” does not establish eligibility. Assess the complete provision before relying on the exemption; the content review below is not an exemption determination.

Separate placement from content

21 CFR 801.109(c) places information for use in labeling on or within the package from which the device is dispensed. That location requirement is a separate question from whether the information covers the relevant content categories. The narrow dispensing-package omission question is addressed in the FAQ.

The categories below come from paragraph (c); the review questions are practical prompts, not prescribed headings or a universal template. Interpret them against the device’s actual use rather than inventing drug-like dosing instructions for equipment.

Review distributed use information too

21 CFR 801.109(d) reaches labeling distributed by or on behalf of the manufacturer, packer or distributor that furnishes—or purports to furnish—information for use, whether or not it accompanies the dispensing package. Such labeling must contain adequate information for that use. A practical review therefore examines what each artifact communicates, not just its title or page count.

Worked case: clinical monitor and quick-start card

Assume a US prescription clinical monitor has a guide inside its dispensing package and a manufacturer-distributed quick-start card containing setup and monitoring instructions. These assumptions do not establish that the device satisfies every exemption condition.

• Package guide: map its location and content to paragraph (c). Also assess it under paragraph (d), because it is manufacturer-distributed labeling furnishing use information.
• Quick-start card: assess its use information under paragraph (d), even if distributed separately. If included in the package, consider it in the paragraph (c) package-content review as well.
• Comparison: check whether shortened setup steps omit a relevant precaution or contradict the guide. Record the affected instruction and the review decision rather than accepting brevity as justification.

As a practical documentation method, retain this artifact-to-provision map with the supporting technical documentation, identifying the versions reviewed and unresolved content questions.

Start with product classification, destination market and intended user—not the document title. Use this applicability table before selecting IFU content or delivery controls.

21 CFR 820.1(a) places methods, facilities and controls for design, manufacture, packaging, labeling, storage, installation and servicing of finished devices intended for human use within QMSR scope. See the FDA QMSR overview for context.

For UK planning, use the MHRA overview as orientation, then check current device-specific requirements. For medicines, supplements, food and chemicals, begin with product classification rather than transferring this device-specific map.

• Binding regulatory provisions: requirements tied to their stated product and jurisdiction.
• Official guidance: identified separately from binding provisions; not the basis assigned to this article’s operational suggestions.
• Suggested industry practice: practical controls and review methods, not additional legal duties.
• Software support: implementation assistance, not a legal obligation or compliance certification; humans retain approval and release decisions.

Develop the IFU from device behavior and user tasks, not from a generic document template. A suggested operating method is to trace each safety-relevant instruction through the following chain. This is a content-development method, not a statement of a particular standard’s requirements.

1. Intended user and environment: who performs the task, where, and with what training or constraints?
2. Use task: what action or decision must the user complete?
3. Foreseeable error or hazard: what could go wrong during that task?
4. Necessary instruction: what action, sequence, limit or precaution addresses that concern?
5. Supporting evidence: which test, analysis or observed-use result supports the instruction?
6. Approved content location: which IFU section, figure and revision contain the reviewed wording?

Select topics according to the device: setup, compatible accessories, operating limits, cleaning, maintenance and troubleshooting are examples, not universal mandatory elements. Require an evidence basis before including an instruction. For example, a cleaning method needs support for the specified agent and procedure; familiar wording from another product is insufficient. Link the supporting record to the relevant technical documentation.

Assign distinct review questions. Engineering can confirm technical correctness, such as accessory compatibility and operating limits. Clinical or application specialists can assess whether the sequence fits actual use and the intended audience. Regulatory reviewers can check consistency with intended use. Quality reviewers can check that evidence, wording and unresolved comments agree before approval. These are suggested responsibilities; organizational assignments vary.

Assess critical-task clarity through human factors work with representative users performing realistic tasks. Observe missed steps, incorrect connections, hesitation and recovery rather than asking only whether the text seems clear. Keep this assessment separate from proofreading. If users repeatedly misconnect an accessory, investigate the interface and task design as well as the wording; adding a warning does not necessarily resolve the underlying risk.

Manage multilingual IFUs through a variant matrix under document control. As a practical control, connect device model, configuration, market, language, source revision, translated artifact identifier and approval state. Determine language coverage through the market-specific assessment described above; do not equate one market with one language. Record each applicable combination so missing variants remain visible.

Treat source-content review and translation review as linked but distinct decisions: the first establishes whether the instructions are technically correct; the second checks whether each language version preserves that meaning and remains usable. Source approval alone should not close translation review. Useful review controls include:

• Assign a terminology owner to resolve device names, user-interface terms and specialist vocabulary consistently.
• Check units and decimal conventions against the intended values; local formatting must not change quantities.
• Compare warning meaning, including severity, prohibited actions and consequences—not merely sentence structure.
• Review diagrams, translated callouts and cross-references against the matching model and configuration.
• Inspect the final layout for clipped text, misplaced warnings, broken references and legibility; review the deliverable, not just the translation worksheet.

Worked case: counting controlled variants

Assume two device models, three total languages including the source language, and one separate PDF for each model-language pair. The controlled set contains 2 × 3 = 6 files. If four files have completed review and approval, 6 − 4 = 2 variants remain unapproved. This calculation describes artifact status, not permission to distribute any device.

Treat electronic IFUs as three separate operational decisions, not one move away from paper:

• Internal electronic master: the instruction source maintained under document control, whether users ultimately receive paper or electronic instructions.
• Externally published electronic copy: a user-facing version available online, potentially alongside printed instructions.
• Electronic-only delivery: a strategy that replaces paper instructions for a defined device and market.

Check current jurisdiction- and device-specific rules before replacing paper instructions. Do not treat a website, QR code, PDF or professional user population as sufficient evidence that electronic-only delivery is permitted.

For electronic publication, use practical access controls rather than assuming that uploading a file completes the task:

• Provide stable access routes and clear selection by device model, configuration, market and language.
• Check file readability on expected devices, including legibility of diagrams, and review accessibility for intended users.
• Keep version-specific archives with explicit applicability information.
• Assign an owner for availability monitoring and recovery.

Displaying the current revision is not the same as retrieving the revision relevant to an older device configuration. Label both routes clearly so that “latest” does not silently substitute for “applicable.”

For manufacturers subject to the US QMSR, 21 CFR 820.35(c) requires recording the UDI for each medical device or batch of medical devices. This recording requirement should not be read as specifying where an identifier must appear in the IFU.

As an information-management practice, keep device identity, instruction revision and language variant as separate linked records. Resolve instruction applicability using the model, relevant software version, destination market and applicability dates—not merely the newest document revision. Production context tagging can help organize those selection criteria.

Suggested IFU evidence model:

Device configuration → applicable instruction revision → language-specific artifact → human approval decision → actual packaging-use record.

In this schema, the approval decision identifies the exact artifact reviewed, its approver and decision date. The packaging-use record links that artifact to the device or lot actually packed, with a reference to the corresponding device history record. Applicability establishes what should accompany the configuration; packaging use records what actually accompanied it.

Preserve approved artifacts and historical relationships rather than overwriting them when instructions change. This lets a reviewer reconstruct an earlier shipment without relying on today's active IFU. Assess identifier-change and database-update decisions against current applicable rules; do not infer either solely from an instruction revision number.

For device manufacturers working under the FDA QMSR, an approved IFU master is only the starting point for packaging-line control. The operational question is whether the correct instruction document reaches the correct device package.

Binding requirements: 21 CFR 820.45 requires documented, maintained procedures describing activities to ensure labeling and packaging integrity, inspection, storage and operations. Under 21 CFR 820.45(a), manufacturers must ensure labeling and packaging are examined for accuracy before release or storage, where applicable.

21 CFR 820.45(c) requires established and maintained operations that prevent mixups, including inspection before use to assure correct labeling and packaging as specified in the medical device file. Results of that labeling inspection must be documented. These are required outcomes and activities; the operating examples below illustrate implementation rather than a prescribed workflow.

Operating examples: from specification to inspection evidence

• Approved packaging specifications: identify the IFU document number, revision and language variant permitted for the device configuration. Give operators an unambiguous reference rather than relying on leaflet appearance.
• Segregated printed stock: separate visually similar IFUs and distinguish approved stock from held material. Use identified storage locations and controlled issue quantities to reduce opportunities for substitution.
• Job-specific variant checks: compare the issued IFU against the packaging job’s device model, destination and specified revision before use; a matching product family name alone is a weak check.
• Line clearance: remove prior-job leaflets, labels and packaging from feeders, benches and staging areas, then record the clearance check before introducing the next job’s materials.
• Inspection evidence: record the job, IFU identifier and revision, inspection reference, result, inspector and time. Preserve enough detail to reconstruct what was checked, not merely that a checkbox was selected.
• Discrepancy escalation: stop affected packaging, identify and hold potentially affected units, and route the mismatch to designated reviewers. Resume only after authorized disposition and confirmation that the line contains the intended materials.

For prescription-device labeling, 21 CFR 801.109(e) requires labeling bearing information for use to carry its issuance date or latest-revision date, except labels and cartons. An IFU revision identifier alone does not communicate that required date.

As a practical change-control convention, distinguish three fields: document revision identifies the content version; approval timestamp records when an authorized person approved it; operational effective date establishes when it becomes applicable to specified operations. Approval need not mean immediate production use. Keep historical versions retrievable without leaving them selectable for unintended new work.

Before effectiveness, define affected device configurations, markets, work orders, printed stock and published copies. Assign stock disposition explicitly: continued use for defined configurations, segregation, authorized correction or destruction. Scope technical-change reviews to changed instructions and supporting evidence; field-feedback reviews to the reported use problem and affected populations; translation-only reviews to meaning, terminology and layout. Do not assume unchanged source-language text makes a translation change inconsequential.

A proposed integration architecture passes explicit artifact and revision identifiers from document control to product configuration, ERP, print services, electronic publication and manufacturing records. Include language, applicability and effective date rather than passing only a filename or “latest” link. For MES–ERP integration, compare the requested revision with the acknowledged revision and the revision actually used. Reconcile failed or delayed transfers before retrying affected jobs; check for duplicate print jobs, stale downloads and orders retaining superseded instructions.

Worked case 3: mixed revisions in packaged stock

Assume a packaging order contains 1,200 kits and its approved specification calls for revision C. Inspection finds 780 kits with C and 420 with B: 1,200 − 780 = 420 kits require disposition. After authorized correction and reinspection, 400 of those pass and 20 remain unresolved. The instruction check is satisfactory for 780 + 400 = 1,180 kits; 20 remain held.

Record the original findings, correction authorization, replacement revision and reinspection results against the affected kits. Passing this check is not final product release. Document approval and product disposition remain authorized human decisions; a successful system transfer neither releases product nor establishes compliance.

These illustrative failure mechanisms are diagnostic examples, not a ranking of inspection findings. Separate immediate containment from investigation of the cause and assessment of whether the broader control failed.

Under 21 CFR 820.35(a), manufacturers must maintain records of complaint review, evaluation and investigation when complaints involve possible failure of a device, labeling or packaging to meet specifications. Treat a possible IFU specification failure within this FDA QMSR complaint-record framework.

Before selecting corrective action, reconstruct the actual artifact, affected device configurations, packaging events and distribution scope. Use production context tagging to connect those facts; record uncertainties rather than assuming the current master was shipped. Test whether the error was isolated or whether approval, transfer or retention controls also failed elsewhere. Apply the earlier stock-disposition case to affected inventory without repeating its calculation. Route recall, reporting and identifier questions for device-specific regulatory assessment rather than deciding them from the instruction correction alone.

Use this suggested review exercise on one device configuration and one completed packaging event. Record the model, relevant software version, destination market, language, lot or serial range, and packaging date. Test whether a reviewer can answer the following with retrievable records, not verbal assurances.

1. Identify applicability: Which IFU revision and language variant applied to this configuration at that event? What record supports that selection?
2. Retrieve the artifact: Can the reviewer obtain the exact approved file through document control, rather than a similarly named working copy?
3. Trace approval: Who approved that artifact, when, and for what scope? Can the decision be connected unambiguously to the revision used?
4. Reconcile use: Do the packaging instruction, issued IFU stock or publication reference, inspection results, and batch record agree? Account for substitutions and discrepancies.
5. Assign exceptions: For each unresolved mismatch, identify an accountable owner, affected stock or publication, interim control, decision deadline, and escalation recipient.

Then test responsibility boundaries: Who owns technical content, market applicability, translation acceptance, publication, production implementation, and field-feedback assessment? Name both the decision-maker and escalation route for each function. Where responsibilities overlap, specify who resolves disagreement and who authorizes the next action; do not assume one department owns every decision. Retain open actions with the sampled review.