A practical guide for manufacturers on grouping multiple models of active medical devices into a single registration unit for overseas markets, covering regulatory logic, documentation, common pitfalls, and preparation checklists.
Key Summary
When registering active medical devices overseas, the core principle for grouping multiple models is to first determine whether the product falls within the medical device regulatory scope of the target country, and then to group models reasonably based on risk classification, intended use, technical structure, software version, and shared attributes. Companies should avoid forcibly placing models with significant functional differences or differing safety characteristics into the same registration unit, while also avoiding the cost and redundant testing associated with submitting each model individually.
The recommended approach is to first assess the reusability of existing technical documentation from NMPA, CE, FDA, ISO 13485, or MDSAP, and then confirm the target country's acceptance of model coverage. Markets such as GHWP member states, Southeast Asia, the Middle East, and Latin America commonly accept the “representative model plus difference list” approach. Before submission, manufacturers must prepare technical files, performance validation, risk management, clinical evaluation or clinical evidence, labels and instructions, local authorization, and post-market maintenance plans. Common risks include family model classification not meeting local regulations, insufficient comparative evidence, unclear authorization chains, and incomplete change control.
Manufacturers should clarify the relationship between the primary model and derived models at an early stage, establish a difference matrix, properly arrange test samples and biocompatibility data, and confirm the corrective action requirements and fee structure with the registration agent to reduce the risk of repeated deficiencies and delays in multi-country registration.
Applicable Scenarios and Core Issues
Active medical devices include patient monitors, ultrasound diagnostic equipment, therapeutic devices, imaging systems, and rehabilitation equipment. Manufacturers often face a practical challenge when registering overseas: a product series may contain multiple models with different configurations, screen sizes, software versions, probe counts, or operating voltages. Is it necessary to submit each model separately? This question directly impacts registration timelines, testing costs, and market launch schedules.
Grouped registration is not permitted in all countries, nor is it acceptable to arbitrarily place all models under a single certificate. The key determining factor is the target country's regulatory definition of a “registration unit” and whether the technical documentation can demonstrate that the safety and performance of all models are substantially equivalent. Companies must first confirm whether the product falls within the country's medical device regulatory scope, then determine the risk classification and registration pathway, and finally establish a grouping strategy.
Typical scenarios include: a series already registered with NMPA in China and intended for GHWP member states or other overseas markets; products with CE or FDA certification that need to expand to Southeast Asia, the Middle East, or Latin America; or products in development planned for simultaneous submission to multiple countries using one set of core technical files. These scenarios all require clear judgment on model grouping.
Registration Judgment Logic
The decision logic for grouped registration should follow a two-step approach: first, determine which category of medical device the product belongs to; second, determine which models can be grouped into the same registration unit. While each country's regulations differ, they generally consider intended use, technical structure, key performance parameters, core software algorithms, and intended environment.
In practice, manufacturers should first establish the product's risk classification. For example, Class II active devices usually follow a declaration of conformity or registration approval pathway, while Class III devices require more rigorous clinical evaluation and quality management system review. Risk classification determines the required tests, depth of review, and whether local clinical trials or clinical literature support are needed.
Within the same risk classification, assess whether all models are consistent in structural composition, working principle, key components, software version, safety characteristics, and performance specifications. If the primary model can cover the main performance and safety indicators of the derived models, and the differences do not affect safety or effectiveness, they can usually be submitted as the same registration unit.
For multi-country registration, consider documentation reuse and localization. Companies should inventory existing CE, FDA, NMPA, ISO 13485, and MDSAP documentation to identify which reports can be directly translated, which require retesting, and which need additional testing for local standard differences. For example, some GHWP member states require their own EMC or electrical safety standard versions; even if a CE report is valid, supplementary difference testing may be necessary.
Documentation and Evidence
For grouped submissions, the following documents are typically required:
- Product classification determination document
- Registration unit division explanation
- Difference matrix between primary and derived models
- Technical file
- Risk management report
- Performance validation report
- Electrical safety and EMC test reports
- Software description documentation
- Clinical evaluation or clinical evidence
- Labels and instructions for use
- Local agent authorization documents
- Post-market surveillance plan
The difference matrix is the core evidence for grouped submission. It should list, item by item, the similarities and differences among models in hardware configuration, software functionality, electrical parameters, mechanical structure, intended use, contraindications, and optional accessories. The matrix must not only prove that models are “similar” but also explain that differences have no adverse effect on safety or performance.
For active devices, selecting the representative model is critical. Regulatory authorities typically require the most complex, fully functional, and highest-risk model to be selected as the representative model for full testing, while derived models may only require difference testing. Companies must ensure that the representative model can represent the entire series in all target countries.
Labels and instructions must cover all submitted models, not just one model. List all model names, specifications, software versions, and accessory lists, and localize them into the target country's language. Additionally, the authorized representative or local agent's authorization document must explicitly cover all models and be consistent with the registration certificate.
If the product involves software functions, such as blood pressure algorithms, heart rate algorithms, or image processing algorithms, the software version control document is also a critical submission item. Multiple models may use the same software platform but enable different functions; the software version and functional differences must be explained.
For clinical evaluation or clinical evidence, if the target country requires clinical data, provide the latest systematic clinical literature search report, clinical experience data, or clinical trial report. If using NMPA or CE clinical evidence, assess its applicability to the target country's population and clinical practice.
Common Mistakes
- Forcibly grouping models with different intended uses, such as diagnostic and therapeutic devices, into the same registration unit, leading to conflicting review requirements.
- Testing only one model while the labels list all models without adequate coverage in the technical file, resulting in deficiencies.
- Difference matrices listing only hardware differences and ignoring software version and operating mode differences, causing post-market regulatory scrutiny.
- Writing the registration unit division basis too vaguely without explaining the relationship between the primary and derived models, prompting reviewers to request additional evidence.
- Failing to confirm the target country's specific requirements for family models. For example, some countries do not accept “family certificates” and require each model to be registered separately, wasting early planning.
- Ignoring the scope of the local agent's authorization. The authorization letter may not include all models, or the agent may change without timely notification to the regulatory body.
- Lack of post-market change control. New models are added without difference assessment or change submission, leading to certificate suspension.
- Using expired CE or FDA reports directly for other country registrations without verifying standard versions and validity dates.
Manufacturer Preparation Checklist
- List all models intended for registration, including hardware versions, software versions, optional accessories, and power supply types.
- Identify target countries and regulatory pathways. It is recommended to list specific regulations for GHWP member states, Southeast Asian countries, the Middle East, and Latin American markets.
- Establish a complete difference matrix comparing safety, performance, and functional differences.
- Select the representative model, prepare a full testing plan, and determine the difference testing items for derived models.
- Inventory existing test reports, risk management documents, software documentation, clinical evidence, and ISO 13485 certificates.
- Confirm requirements for labels, instructions, language, and local agents in each target country.
- Sign authorization agreements with local agents, ensuring clear roles for certificate holder, applicant, and agent.
- Develop a post-market maintenance plan, including adverse event reporting, annual updates, change submissions, and certificate renewals.
AIMEILI Regulatory Interpretation
In the matter of grouping multiple models, the most common misjudgment is that “technical similarity” does not equal “regulatory grouping eligibility.” Some companies assume that because all models are on the same platform, they can be placed on one registration certificate. They later face requests to split the application or provide substantial additional evidence during review. We advise companies to conduct a regulatory gap analysis before project initiation rather than discovering problems after testing or submission.
At the early stage, two actions are essential: first, study the target country's definition of a “registration unit” and common review interpretations; second, establish a unified model naming and version management logic. Many companies have inconsistent version naming across different development batches, making it difficult to build a difference matrix and wasting preparation effort.
Most hardware and software documentation can be reused across countries, such as electrical safety standards, EMC testing, and risk management documentation. However, biocompatibility, environmental adaptability, label language, clinical evaluation conclusions, and country-specific technical specifications must be localized. Companies should not directly copy the NMPA format but should reorganize technical documents according to the target country's CTD format or local template.
Local agents, certificate control, changes, and renewals are the most overlooked risk points in grouped submissions. The registration certificate is usually held by the applicant, and the agreement between the manufacturer and agent must clearly define change notification responsibilities. If the agent changes or the certificate contact person changes, the company must promptly handle change procedures; otherwise, market access for the entire model series can be affected.
For multi-country registration, we recommend a collaboration model of a “core registration team plus local agents in each country,” using a unified core technical document base and localizing difference analysis according to each target country's requirements. This approach significantly reduces the burden of repeating documentation preparation and enables a quick response to different countries' review comments during the deficiency response phase.
Frequently Asked Questions
Can 20 models in the same series all be placed under one registration certificate?
This depends on the target country's regulations and the degree of difference among the 20 models. If all models have the same intended use, similar safety characteristics, similar technical structure, and the representative model can cover all derived models, most countries will accept a single certificate covering all models. However, some countries, such as the Saudi FDA, may require key differing models to be listed separately. It is recommended to first perform a self-assessment using a difference matrix and then confirm with a local agent or regulatory body.
If only the representative model is tested, are test reports required for derived models?
Usually, difference test reports are required. For active devices, the common practice is to conduct full testing on the representative model, while derived models require at least tests related to the differences, such as safety tests for different input power, display sizes, or optional accessory functions. If a derived model only has reduced software functions, additional hardware testing may not be needed, but the software documentation must explain the differences.
After grouped registration, must a newly added model undergo new registration?
This depends on whether the new model falls within the coverage of the existing certificate. If model extension rules were specified at the time of registration and the new model complies with those rules, an add-on change application is generally sufficient. However, if the new model changes the intended use or introduces new risks, a new registration application may be required. Companies should define “change levels” in their post-market surveillance program in advance.
Continue Reading
Previous: Why Do Overseas Registration Timelines for Active Medical Devices Get Extended?
Next: How Should Performance Validation Documentation Be Prepared for Overseas Registration of Active Medical Devices?
Recommended reading:
- How to Select a Local Agent for Overseas Registration of AI Medical Devices?
- Why Do Overseas Registration Timelines for Sterile Consumables Get Extended?
- Why Do Overseas Registration Timelines for Active Medical Devices Get Extended?
- How Should Performance Validation Documentation Be Prepared for Overseas Registration of Active Medical Devices?
- What to Do If Quality System Certificates Are Inconsistent for Sterile Consumables Overseas Registration?
- How to Localize Labels and Instructions for Overseas Registration of Active Medical Devices?
Back to FAQ Center
Content author: AIMEILI Regulatory Editorial Department. Professional review: AIMEILI Medical Device International Registration Project Team. Source principle: priority is given to official regulatory agencies, international organizations, standard organizations, and publicly available regulatory materials; industry media and project experience are only used as auxiliary reference. This article is intended for preliminary understanding, document preparation, and project planning, and does not replace the formal requirements, test conclusions, or legal opinions of the target country's regulatory authorities.
Need a registration pathway assessment?
Send product type, intended use, target countries and existing certificates. AIMEILI can help evaluate registration pathway, documentation gaps and compliance risks.
Contact AIMEILI