Trend Reporting and Serious Incident Classification Under MDR

Two separate tracks — and why confusing them is costly

MDR contains two distinct reporting obligations that operate in parallel and address different scenarios. Serious incident reporting (Article 87) covers individual events that meet a specific severity threshold. Trend reporting (Article 88) covers patterns of events that, individually, do not meet that threshold but collectively represent a significant change that needs to be investigated. Most manufacturers are reasonably aware of the first. Many are underequipped for the second.

Getting this wrong in either direction creates problems. Failing to classify an event as a serious incident when it meets the criteria is a non-conformity and can expose patients to ongoing risk. But over-reporting — treating every adverse event as a serious incident — puts unnecessary strain on the competent authority relationship and on your internal investigation capacity. The classification decision matters, and it needs to be made systematically.

What makes something a serious incident

MDR Article 2(64) defines a serious incident as any malfunction or deterioration in the characteristics or performance of a device made available on the market that directly or indirectly led to, or could have led to, the death of a patient, user, or other person, a serious injury, or a serious public health threat. The "could have led to" language is important — you don't need a confirmed harm. If a failure could plausibly have caused death or serious injury under different circumstances, that qualifies.

Three things trip teams up here. First, the near-miss problem: an event where patient harm was avoided by chance or by intervention still qualifies if the failure itself met the criteria. Second, the definition of "serious injury" under MDR is broader than many teams assume — it includes any life-threatening illness, permanent impairment of a body function, permanent damage to a body structure, or any condition requiring medical or surgical intervention. Third, device malfunction that does not result in any patient contact but creates a risk of the above — for example, contamination of a sterile product line — can still be reportable.

The decision tree for classification should be in your PMS procedures, not in the minds of individuals. What question do you ask first? What criteria determine the answer? What happens when the answer is ambiguous? Ambiguity in individual cases is normal — the procedure should handle it explicitly.

Trend reporting: below-threshold events that accumulate

Article 88 trend reporting is triggered when statistically significant increases in the frequency of incidents that, while not serious incidents under Article 87, represent a significant change in the expected rate. This is deliberately abstract. What "statistically significant" means for your device is something you need to define in your PMS plan, not improvise at the time of review.

In practice, trend reporting is underused. The most common pattern is that manufacturers track complaint rates but don't have a documented baseline that would let them identify when an increase is statistically meaningful. Without that baseline, "trend reporting" becomes a theoretical obligation rather than a practical one.

Setting up trend reporting correctly requires three things. First, a documented baseline rate for the incident category you are monitoring — established from historical data or, for new products, from literature on comparable devices. Second, a statistical method for identifying a significant deviation from that baseline — this doesn't need to be complex, but it needs to be defined. Third, a procedure for what happens when the threshold is crossed: who is notified, what investigation is launched, and whether the competent authority needs to be informed.

Note that trend reporting does not automatically generate the same external reporting obligation as a serious incident report. But if the investigation reveals that the trend is due to a systematic device or process problem, a field safety corrective action (FSCA) may follow — and if it does, an FSN goes with it.

MIR vs FSN: what each one is and when it's required

A Medical Incident Report (MIR) — referred to in MDR as a serious incident report — is the initial notification to the competent authority that a serious incident has occurred or may have occurred. It must be filed within the timeframes set by Article 87(5): immediately and no later than 15 days for serious public health threats, 10 days for death or unexpected serious deterioration in state of health, and 30 days for other reportable incidents. The report is submitted to the competent authority in the member state where the incident occurred.

A Field Safety Notice (FSN) is a communication to users, operators, or patients about a field safety corrective action (FSCA). An FSCA is any corrective action taken by the manufacturer to reduce the risk of death or serious injury associated with a device already placed on the market. An FSN is required whenever an FSCA is taken, and it must be communicated to the appropriate parties — in practice, this means customers, distributors, and sometimes directly to patients depending on the device type and the nature of the risk.

The MIR and FSN are related but not identical in their triggers. A serious incident always triggers a MIR. An MIR may or may not lead to an FSCA — it depends on the investigation outcome. An FSCA always requires an FSN. A field safety action can sometimes be initiated before the serious incident investigation is complete, if the risk is immediate. In that case, the MIR and FSN may be issued around the same time but driven by different processes.

How to handle the grey zone

Most difficult classification decisions involve incidents that might be serious under one interpretation and non-serious under another. The most useful principle here is to default toward reporting when the outcome is uncertain. Competent authorities do not penalise manufacturers for over-reporting in good faith. They do follow up on failures to report that later become visible.

For grey-zone cases, document your classification decision and the rationale explicitly — not just the outcome. If you decided an event was not a serious incident, note what evidence supported that decision. If the same type of event appears again and you have to reconsider, that documentation tells you what you thought at the time and whether the new information changes the picture.

Where to go from here

The vigilance reporting resource in this category covers the full reporting timeline and member state notification pathways in more detail. The PMS data sources and thresholds resource explains how to set baseline rates for trend monitoring. If you are uncertain about how a specific event should be classified, the competent authority's guidance documents and the MDCG 2023-3 guidance on serious incidents and FSCAs are the primary references.