159,503 events, and more than half the reports arrived late
What happened
On 14 October 2026, The BMJ published a study of how medical device problems get reported to regulators in the United States, and found the system often fails to raise the alarm quickly. Here is the background. When a medical device malfunctions or harms someone, the maker is supposed to tell the US Food and Drug Administration within 30 days. Those reports go into a public database called MAUDE. Doctors, researchers and regulators around the world use it to spot patterns. For years, some manufacturers were allowed to send bulk summaries instead of one report per incident. The idea was to cut paperwork. That scheme, called Alternative Summary Reporting, was closed in 2019 because it could hide safety problems. But similar summary arrangements carried on in other forms. The researchers looked at every adverse event summary report sent to MAUDE between July 2019 and June 2025. That came to 6,308 summary reports, containing 159,503 individual adverse events. Most involved high-risk devices. What they found: - 57 per cent named no device problem or use problem at all. - 38 per cent did not include enough information to understand what had gone wrong. - 11 per cent were wrongly labelled as something other than a summary report. - 7 per cent said "one event" in the number-of-events field, even though they were bulk reports covering many. - 55 per cent were filed after the 30-day deadline. Summary reports, and the number of events inside them, more than doubled between 2020 and 2024. The authors wrote that "immediate improvements are needed to better protect patient safety", and that "making serious adverse event reports fully visible and accessible is critical to patient safety".
Why this matters
A medical device is not just a hip replacement or a pacemaker. It is also your blood glucose meter, your inhaler spacer, a surgical mesh, a feeding tube, an infusion pump, a hearing aid. When something goes wrong with one of these, the only way anyone notices a pattern is if the individual reports get counted and read. One faulty pump is bad luck. Four hundred faulty pumps in six months is a recall. That is why the detail in these reports matters so much. If a bulk report says "one event" when it covers forty, the pattern simply does not show up in the numbers. If 57 per cent name no problem at all, there is nothing for a safety analyst to search for. This is an American study about an American database, so it is fair to ask why it matters in Lincolnshire. Two reasons. The first is that devices are global. A pump sold in Boston is usually the same pump sold in Boston, Lincolnshire. Regulators share signals, and the MHRA and other agencies routinely look at MAUDE data when deciding whether to act. If that data is thin, everyone's early warning is weaker. The second is timing. The MHRA published its own 2030 strategy on the same day, promising faster and more modern regulation. This study is a reminder that speed in approving things and speed in spotting problems afterwards are two different jobs, and the second one is easier to quietly neglect.
What the evidence actually says
This is a study of reporting quality, not a study of harm. That distinction is important and the authors are clear about it. Nobody has shown that patients were injured because these reports were late or vague. What has been shown is that the warning system, as used, often does not carry the information it was designed to carry. The authors list their own limits. Passive reporting, where people choose to send in reports rather than being systematically counted, always undercounts. And there is no publicly available information setting out exactly which rules govern which summary reporting exemption, so the researchers could not always tell what a given manufacturer was permitted to do. A linked editorial by Koji Todaka of Kyushu University Hospital in Japan argued that regulators should require accurate, timely summary reporting, standard ways of identifying devices, and clear links to patient registries. He also suggested researchers should combine spontaneous reports with registries and linked healthcare records, rather than leaning on any one source. "Transparency builds trust, evidence improves safety, and both are essential," he wrote. The FDA is moving to a new Adverse Event Monitoring System which is intended to pull reporting systems together and publish reports in real time. Whether that fixes the problems in this study is not yet known.
Practical advice
You are not going to audit a regulator's database. But there are two useful habits here. Report problems when they happen. In the UK, the Yellow Card scheme covers side effects from medicines and problems with medical devices, including things that nearly went wrong. You can do it yourself at yellowcard.mhra.gov.uk, or your pharmacist or GP can do it with you. Most people assume someone else will have already reported it. Often nobody has. The whole system only works because ordinary people bother. Keep the details. If a device fails, write down the make, the model, the batch or serial number, the date and what happened. Keep the packaging if you can. That is exactly the information that makes a report useful instead of vague, and it is the information this study found missing. Do not stop using a prescribed device because of a news story. Speak to your pharmacist, GP or specialist nurse first. Stopping something like an insulin pump or a feeding device without advice carries its own risk. Check recalls if you are worried about a specific product. The MHRA publishes device safety alerts and field safety notices on GOV.UK, and your pharmacy can look them up for you. If you get a letter about a recall, read it properly. Some recalls ask you to stop using something immediately. Others ask you to do nothing except be aware. They are not the same, and the letter will say which.
Qurexa perspective
Some of what Qurexa delivers is a device, not a medicine. Blood glucose test strips, inhaler spacers, dressings, stoma supplies, continence products. Two things follow from that. We keep batch records for what we deliver. If you ring us about a problem with something we brought you, we can usually tell you the batch, which saves you hunting for packaging you may have already thrown away. And our pharmacy team can submit a Yellow Card report with you rather than leaving you to do it alone. That takes a few minutes on the phone and means the report actually gets filed with the detail that makes it count. If something we delivered has failed, has arrived damaged, or just does not seem right, tell us. We would rather chase a false alarm than miss a real one.
What to know
A study published in The BMJ on 14 October 2026 examined all 6,308 adverse event summary reports sent to the US FDA's MAUDE database between July 2019 and June 2025, covering 159,503 adverse events, mostly involving high-risk devices. It found 57 per cent identified no device or use problem, 38 per cent gave insufficient information, 11 per cent were wrongly labelled, 7 per cent recorded only one event despite being bulk reports, and 55 per cent missed the FDA's 30-day deadline. Summary reports more than doubled between 2020 and 2024. The authors call this a public health concern and say immediate improvements are needed. This is a study of reporting quality, not evidence that specific patients were harmed. In the UK, you can report a problem with a medicine or a medical device through the MHRA's Yellow Card scheme. Keeping the make, model, batch number and date makes a report far more useful. Sources: The BMJ, "Bulk adverse event reporting may delay medical device safety warnings" (DOI 10.1136/bmj-2026-100812), 14 October 2026, reported at https://www.news-medical.net/news/20261014/Bulk-adverse-event-reporting-may-delay-medical-device-safety-warnings.aspx; MHRA, "MHRA 2030 Strategy to modernise patient-centred regulation for the era of AI and personalised medicines", 14 October 2026, https://www.gov.uk/government/organisations/medicines-and-healthcare-products-regulatory-agency; MHRA Yellow Card scheme, https://yellowcard.mhra.gov.uk This article is for general information and does not replace advice from a doctor, pharmacist or other qualified healthcare professional. Do not stop using a prescribed device or medicine on the strength of a news report.
Related articles
47% stayed off, 14% called anyone: what 68,000 surveys say about how we handle flu
On 15 October 2026, the UK Health Security Agency (UKHSA) published new findings about what people actually do when they come down with flu-like symptoms. The…
One dose, not two: what WHO's advisers have just said about the HPV jab
The World Health Organization's expert vaccine panel has changed its advice on the HPV jab. The panel is called SAGE, short for the Strategic Advisory Group of…
3.5 million deaths a year, and no level of it is safe
On 15 October 2026, the World Health Organization (WHO) published an update calling on countries to speed up action against lead poisoning. It came just before…