Philippines Medical Device Grouping Requirements: Single, Family, System, Set and IVD Cluster Explained

Philippines Medical Device Grouping Requirements: What Decides Whether You File Once or Twenty Times

Grouping is the rule set that decides how many separate FDA applications your product list becomes. Two or more items may sit on one application only if they share one product owner, one proprietary or generic name and one common intended purpose — and only if they then fit one of the ASEAN grouping categories: single, family, system, set or group, plus the test kit and cluster categories that exist for IVDs. Your fee bill, your launch date and your renewal workload all follow from that single decision.

The reason it carries so much weight in the Philippines is mechanical: official fees and technical review both run per application, not per model. Twenty variants filed as twenty singles are charged twenty times, reviewed twenty times, renewed twenty times and change-controlled twenty times. The same twenty filed as one properly justified family are charged and reviewed once. Nothing else in the registration process moves cost and calendar by that kind of multiple, and nothing else is settled so early — grouping is decided before the first line of the dossier is written.

Before the categories, there are four gates every grouping has to pass. The first three are the ASEAN foundation rules; the fourth is the one that catches most first-time filers:

• One product owner. The legal manufacturer or brand owner named on the quality-system certificate and the free-sale certificate has to be the same for everything in the group. Two manufacturers means two applications, however identical the products look. A system or set can sometimes carry a compatible accessory from another owner, but that is an exception you have to evidence, not the default.
• One proprietary or generic name. The grouped items must be presented under a single product name. A second brand name is a second application — which is where a marketing decision quietly becomes a regulatory one, because launching a variant under its own brand splits a family you had already paid for.
• One common intended purpose. Intended purpose is what the label, the instructions for use and the dossier say the device is for. If one model in the group is additionally indicated for a second clinical use, the common purpose is gone and so is the grouping.
• One risk class. Devices of different classes generally cannot ride on one application. Where a legitimate set or system does contain mixed classes, the highest class present governs the whole application — one higher-risk component drags an entire tray of low-risk consumables onto the heavier track.

Classification therefore comes first in time: you cannot test "same class" until every item has a class. If the A/B/C/D letters are not settled yet, start with the A/B/C/D risk classes and what each class has to file and come back. Grouping is the step immediately after classification and immediately before the dossier — and it is the last cheap moment to change your mind.

Single, Family, System, Set and Group: What Each Grouping Category Actually Covers

Five categories cover general medical devices, and they are not interchangeable — each one answers a different question about how your products relate to one another. Choose the category that describes the real relationship. Choosing the one that produces the smallest fee and hoping the reviewer agrees is the single most expensive habit in this part of the process.

• Single. One device in one configuration, filed on its own. Permitted variation is limited to things that do not change the device itself — typically pack size or the number of units in a box. If you are filing singles while telling yourself the models are "basically the same", you are in fact describing a family and should be justifying it as one.
• Family. Several models from the same owner sharing the same intended purpose, the same risk class and the same basic design and manufacturing process, differing only in permitted variations such as size, colour, shape, model designation or the number of components. This is the category that saves the most money and the one that gets stretched furthest past its limits; it has its own section below.
• System. Devices from the same owner that the owner intends to be used in combination to achieve one clinical purpose, presented under a system name — a console with its dedicated handpieces and consumables, for example. The test is not that the items can work together; it is that the manufacturer specifies them to be used together and supports that in the labelling and the dossier. Accessories from another owner can sometimes be included where compatibility is demonstrated with evidence.
• Set. Two or more different types of device, from the same owner and under a common intended purpose, supplied and sold together for a defined procedure — a procedure tray or a surgical kit. The mixed-class rule bites hardest here: the highest-risk item in the set determines the class of the whole set, and therefore its route and its dossier depth.
• Group. Devices physically packaged and labelled as one sale unit for one purpose. In regional practice "set" and "group" are used for overlapping situations and the wording differs between ASEAN regulators, so describe the physical and commercial reality of your pack accurately in the application and use the terminology of the FDA issuance in force when you file rather than terminology imported from another market.

One structural point that regularly surprises people: an accessory is not automatically covered by the device it plugs into. Whether it rides on the parent application depends on whether it is genuinely presented as part of that single, family, system or set. A spare part or consumable sold separately under its own name is generally its own filing, and discovering that after the parent device is approved is a familiar and avoidable delay.

The Family Test: Deciding Whether Two Models Belong on One Application

Two models belong in the same family only when the differences between them are variations of one device rather than two different devices. The working test is five "sames": same product owner, same intended purpose, same risk class, same basic design and same manufacturing process — with every difference confined to attributes that do not affect safety or essential performance.

Generally accepted as permitted variation inside a family:

• Size or dimensional range across models that are otherwise identical.
• Colour and cosmetic finish.
• Shape variations that do not change how the device works or where it contacts the patient.
• Model, catalogue or article numbers.
• Number of units or components per pack.
• Configuration options that do not change function — a bundled carry case, an alternative mounting bracket.

What breaks a family is anything that changes what a reviewer would have to assess:

• A different patient-contact material. Different biocompatibility evidence, therefore a different assessment.
• A different sterilization method, or a change between sterile and non-sterile supply. Some regulators in the region permit sterilization variants inside a family and others do not. Treat this as a question to answer from the Philippine issuance in force, never as an assumption carried over from another market.
• An added or different indication. A second intended use is a second common purpose, so the family fails at rule one.
• A second manufacturing site or a second legal manufacturer. The quality-system certificate and the free-sale certificate will not line up, and the reviewer reads all three documents against each other.
• A different technology or operating principle. A manual version and a powered version of the same instrument are two devices with one marketing story.
• A different brand name, even with identical hardware.
• A different risk class, by definition.

There is one discipline that prevents the expensive version of this mistake, and almost nobody does it: write the family justification before you file — one line per model, naming the attribute that varies and the evidence that the variation does not affect safety or essential performance. If a line is hard to write, that model does not belong in the family. You have then found the problem for free, at a desk, instead of finding it in a deficiency letter months later when the fix is a refile.

Philippines IVD Grouping Requirements: Test Kit, IVD Cluster and Why the Rules Differ From Devices

IVDs run on the same three foundation rules — one owner, one name, one common intended purpose — but they add two grouping categories of their own, the IVD test kit and the IVD cluster, because an IVD's purpose is to produce a result about an analyte rather than to act on the body. That difference in nature is why the general device categories do not map cleanly onto reagents.

• IVD test kit. Reagents, calibrators, controls and components packaged and labelled together as a kit to perform a defined test. The usual boundary is that a kit covers the reagent side of the test: an instrument or analyzer is generally not part of the kit and files separately.
• IVD cluster. A set of reagents from the same owner sharing a common detection methodology and falling inside a recognised cluster category — the mechanism that lets a panel of related assays running on one platform go on one application instead of one application per analyte. Clustering is typically confined to the lower risk classes, and a higher-risk analyte in the group generally pulls the whole application up to the highest class present, which usually destroys the reason you wanted to cluster it.

Three practical differences between device grouping and IVD grouping are worth holding in mind:

• The analyte is part of the intended purpose. Two assays measuring different analytes are two intended purposes, however identical the underlying chemistry. This is why "same platform" on its own never justifies a group — you need the cluster rule, and you need your analytes to fall inside one recognised cluster category.
• Instruments separate from reagents. The analyzer, the reader and the software are their own filings, and their intended-use statements have to stay consistent with the kits that run on them. A mismatch between an instrument's claimed menu and the assays actually registered is a recurring deficiency.
• Sample type and use setting matter. Moving an assay from professional use to self-testing, or adding a sample matrix, changes what has to be proven and will usually break a grouping that otherwise looked clean.

As with devices, classification comes before grouping: you cannot test "same class" until every assay has a class. The classification logic for reagents, along with labelling and the licensing route, is in our IVD certification, classification and import licence guide — map your clusters after that, not before.

What Correct Grouping Saves, and Where the Saving Is an Illusion

The saving from grouping is a multiplier rather than a discount: it is the difference between paying and waiting once and paying and waiting N times. We publish no figures here — official fees apply per the FDA's current schedule and are revised by issuance — but the arithmetic does not depend on knowing the numbers, only on knowing that they are charged per application.

Where a correct grouping pays:

• Official fees. Charged per application at the product layer, so a dozen models on one family application are billed once instead of a dozen times.
• Review load and elapsed time. One dossier, one review, one set of deficiency letters. A dozen separate files means a dozen queue positions and a dozen separate chances to be asked for the same missing document.
• Renewal. The multiplier repeats every validity cycle for the commercial life of the product. This is the part that is systematically under-counted, because in year one it is invisible.
• Change control. Adding a size to an approved family is usually handled as a change against an existing authorization. Adding a size that does not fit the family is a new application from zero.
• Everything downstream. One authorization number to print on artwork, one number quoted on the customs entry, one line in a hospital tender document — instead of a dozen of each, all of which have to be kept current.

Where the saving is not real: a family that is too wide has negative value. It is not "cheaper but riskier". If the reviewer splits it you pay for the singles anyway, plus the months you spent waiting to be told. And an over-wide family that survives review is worse still, because the gap between what was approved and what you actually ship stops being a filing problem and becomes a post-market and enforcement problem, surfacing during an inspection or a complaint rather than during a review you could have argued.

Grouping is a place to be accurate, not aggressive. The two-layer structure this sits inside — establishment licence first, product authorization second — is set out in our FDA product registration guide, and it matters here because the product application is filed under the licence holder's account and in the licence holder's name.

Filing the Wrong Grouping: Why It Is a Refile, Not an Amendment

Grouping errors hurt out of all proportion to their size because they are normally fixed by filing again rather than by correcting the file you already have. The grouping decision is the shape of the application itself — you cannot edit one family into three singles the way you correct a typo in an address.

What "start over" actually costs, roughly in the order it lands on you:

1. The time already spent is gone. The months in review bought you a deficiency letter, not an authorization.
2. You return to the back of the queue — three times over, if the family split into three.
3. Fees paid on the rejected application are generally not portable to the replacement filings. Treat them as sunk unless the FDA issuance in force says otherwise.
4. The dossier is rebuilt, not copied. Each replacement application needs its own coherent device description, classification rationale and evidence set. Reviewers cross-read, and a dossier that still reads like a slice of an abandoned family invites a second rejection.
5. Artwork and labels are re-worked. The authorization number appears on the label, so one number becoming three means new artwork, new print runs, and stock already printed against a number that no longer exists.
6. Cargo stops. A shipment arriving against an authorization that does not actually cover the model in the carton is a customs problem, not a paperwork problem — see how Philippine customs handles regulated goods at entry for what happens to consignments that arrive without valid cover.
7. Commercial deadlines are missed. Hospital tenders, distributor agreements and launch commitments do not pause for a refile, and the registration is usually the item sitting on the critical path.
8. Your regulatory history carries it. The refiling sits in the same account and the same establishment record that every later application of yours is read against.

Notice that none of this is a scientific objection to your device. It is the cost of a structural decision taken in an afternoon and then not revisited. That decision costs an argument before filing and costs a quarter of a year after it.

One rejected grouping and the whole catalogue goes back to the end of the FDA queue? → FDA licensing and product registration

Eight Grouping Misjudgments That Get Files Sent Back

Almost every rejected grouping is one of eight recurring errors, and all eight are visible at a desk before filing. Each one below is followed by the fix.

1. "Same catalogue page, therefore same family." A product line as marketing defines it is not a family as the regulator defines it. Fix: run the five "sames" over every model individually and let the line split where it splits.
2. "Same platform, therefore one IVD group." Shared instrumentation does not create a common intended purpose when the analytes differ. Fix: use the cluster rule and confirm your analytes sit inside one recognised cluster category.
3. Mixed risk classes on one application. Bundling a higher-class item with low-class consumables either fails outright or pulls everything onto the heavier track. Fix: classify first, group second, and price the set at the highest class in it.
4. A second brand name inside a family. Identical hardware sold under two names is two applications. Fix: settle brand architecture before filing, not after the distributor asks for a private label.
5. Sterile and non-sterile versions bundled without checking. This is jurisdiction-specific and cannot be assumed. Fix: confirm against the current Philippine issuance before you build the model list.
6. A second manufacturing site inside a family. The ISO evidence and the free-sale certificate stop matching the application. Fix: map site coverage against every model before grouping, and re-check it whenever production moves.
7. Accessories assumed to be covered by the parent device. Separately sold accessories and consumables commonly need their own authorization. Fix: list every separately saleable item and decide explicitly whether it is inside the group or outside it.
8. Grouping copied from another market's approval. A Singapore, Malaysian, EU or US grouping is a useful starting hypothesis and nothing more; adopting it wholesale is one of the most common reasons a first Philippine filing comes back. Fix: re-derive the grouping under the Philippine rules and keep the foreign approval as supporting evidence, not as the answer.

The pattern across all eight is the same one that runs through the rest of Philippine FDA work: these are consistency failures, not technical ones — which is exactly why they are cheap to prevent and expensive to discover.

How to Build a Grouping Map Before You Draft the Dossier

Do the grouping on paper first, in a fixed order, and finish it before anyone starts writing the technical file. The whole exercise is a day or two of work and it is the highest-leverage day or two in the project.

1. Freeze the model list. Every catalogue number you intend to sell in the Philippines in the next validity cycle, including sizes, presentations, accessories and consumables sold separately. Models added later are new applications or change filings, so pull them forward now if you can.
2. Classify each item into its risk class under the ASEAN framework the Philippines has adopted, and write down the reasoning. Do not classify the "product line"; classify each item.
3. Group by intended purpose first, not by product family. Sort the list by the intended-use statement you will actually print. Items that share an intended-use sentence are grouping candidates; items that do not, are not.
4. Apply the four gates — one owner, one name, one intended purpose, one class — and drop out everything that fails.
5. Write one justification line per model naming the varying attribute and the evidence that it does not affect safety or essential performance. A line you cannot write is a model you cannot group.
6. Price and time the map. Count the applications it produces and multiply that by the current official fee schedule and by the current processing period. If the count is uncomfortable, the answer is to revisit the model list, not to widen the families.
7. Have the licence holder review it. The application is filed under the establishment's account and issued in the establishment's name, so the party who will carry the consequences should sign off on the map. If you do not yet have a Philippine entity, the same review has to happen with whoever will act for you — see how an importer of record arrangement works for the structure and its limits.

Keep the map as a living document. When a model is added, a material changes, a site moves or an indication is extended, the first question is always the same one: does this still fit the group it was approved in? Answering that early is change control. Answering it late is a refile.

Summary and Disclaimer

In order: classify every item → apply the four gates (one owner, one name, one intended purpose, one class) → choose the honest category among single, family, system, set and group, or test kit and cluster for IVDs → write a justification line per model → count the applications and price them → then draft the dossier. Grouping is decided in an afternoon, priced for the life of the product, and corrected only by filing again.

This article is general information only and is not legal, regulatory or medical advice. The grouping categories, their definitions, the permitted variations inside a family, the cluster categories for IVDs, official fees and processing periods are set by Philippine FDA administrative orders and circulars and are revised over time, and every judgement here is product-specific — what governs your case is the FDA issuance in force on your filing date together with professional advice on your actual products. We deliberately quote no fee amounts, no day counts and no issuance numbers for that reason.

If you want the map checked before you spend anything on a dossier, bring your model list, your intended-use statements and your quality documents and start with a classification and grouping review from Yixing. It is the cheapest hour in the entire process, and it is the only one that can still change the shape of the filing.