IVDR ties a device's intended purpose to eight named facts

This blog has already compared how US law and EU MDR Annex I test a device's instructions for use — one against a performance standard, the other against a named list of required content. IVDR Annex I runs its own named list for in vitro diagnostics, and one entry on that list doesn't behave like the rest of it. Where most of Annex I Section 20 asks for a single fact stated once — the manufacturer's address, the expiry date, the storage conditions — the entry for a device's intended purpose asks for eight separate facts, each with its own content test, bundled into what a lot of files still draft as one descriptive sentence.

One list entry, eight sub-questions

IVDR Annex I, point 20.4.1, lists the particulars an in vitro diagnostic's instructions for use have to contain, and point (c) on that list is the device's intended purpose. Unlike the labeling particulars sitting next to it, 20.4.1(c) doesn't resolve with a single statement. It breaks the intended purpose into a named set of elements a complete statement has to address: what the device detects or measures, the function it performs — screening, monitoring, diagnosis or aid to diagnosis, prognosis, prediction, or companion diagnostic use — whether the result is qualitative, semi-quantitative, or quantitative, whether the test runs automated or manual, and the specimen type the device requires. A statement that names the analyte and stops there has answered one of several questions the same sentence is supposed to carry.

A companion diagnostic owes a name the others don't

Where the device's function is companion diagnostic use, the intended purpose statement owes one more fact none of the other function categories require: the International Non-proprietary Name of the specific medicinal product the test is linked to. That's not a labeling nicety — it's the detail that ties the IVD's own clinical claim to a specific drug's own approval, and it only attaches once the function element above has already been answered as “companion diagnostic.” A statement that calls a device a companion diagnostic in its marketing material but never names the linked product in the instructions for use has left the one fact that makes the claim checkable out of the document that's supposed to carry it.

The statement answers to a definition, not just a list

Article 2(12) defines intended purpose as the use for which a device is intended according to the data the manufacturer supplies — on the label, in the instructions for use, in promotional or sales material, and in the performance evaluation. That last source is the one the old IVD Directive never named, and it changes where a drafting team has to look before finalizing the statement at 20.4.1(c): the performance evaluation report has to support every element the instructions for use claim, not the other way around. A statement broadened after the performance evaluation was already locked, to cover a use the study never tested, is a labeling claim without the evaluation behind it.

Self-testing devices carry a ninth duty on top

Section 20.4.2 adds a requirement specific to self-testing devices: the intended purpose has to give a lay user enough information to understand the medical context and interpret the device's own result correctly. That's not a plainer rewrite of the eight-element statement a professional-use IFU carries — it's an added test asking whether the same facts, once stated, actually land with a reader who has no clinical training to fall back on. A self-test IFU that lifts its intended purpose statement directly from a professional-use version of the same assay has satisfied 20.4.1(c)'s elements without satisfying 20.4.2's separate audience test.

The notified body checks this before it checks anything else

A technical documentation assessment doesn't start by reviewing performance data against the intended purpose on faith. In practice, a notified body validates the intended purpose statement against 20.4.1(c)'s own elements first, because the rest of the file — the classification under Annex VIII, the performance evaluation plan, the risk analysis — is built to match whatever that statement says. A statement that's vague on result type or specimen at the start of review doesn't just risk a comment back from the assessor; it risks the classification and performance-evaluation sections built against it having to be revised once the statement itself gets corrected.

Where this meets the file

An IVD's labeling file needs the intended purpose statement broken into its own named fields — analyte, function, result type, automation, specimen, and the companion-diagnostic INN where it applies — checked against the performance evaluation report line by line, rather than carried as one paragraph nobody revisits once it's drafted. An intended-purpose worksheet built around those named fields, alongside the launch catalog's other IVDR submission templates, is previewed there now. If your program drafts this statement differently, the shelf takes that correction directly.