Device failures: lessons from FDA’s hidden trove of incident reports

Chloe Kent 15 July 2019 (Last Updated July 15th, 2019 15:10)

The US Food and Drug Administration has recently published over 20 years’ worth of previously secret data on medical device failures, following a series of international investigations into global safety breaches across the industry. But is the agency doing enough to remedy its past mistakes?

Device failures: lessons from FDA’s hidden trove of incident reports
Alternative summary reporting was designed to reduce paperwork and save time. Credit: Flickr

On 22 June 2019 the US Food and Drug Administration (FDA) released over six million previously hidden medical device ‘incident’ reports filed between 1997 and 2019 by device makers. This ended a controversial program known as alternative summary reporting.

The publication of these files comes mere months after the release of the ‘Implant Files’, an investigation by the International Consortium of Investigative Journalists (ICIJ) into global medical device failings.

Alternative summary reporting was designed to reduce paperwork and save time, according to the FDA. It allowed manufacturers to submit private quarterly spreadsheets summarising reports of common problems with approved devices, rather than filing an individual report on each issue to the publically-available Manufacturer and User Facility Device Experience (MAUDE) database. Since the program’s inception the FDA granted 108 exemptions to individual manufacturers for well-known events associated with specific devices.

This system was supposed to specifically exclude severe or unexpected malfunctions requiring serious attention or remedial action, but industry experts have long been sceptical of this claim.

Speaking to the ICIJ, former FDA data analyst Madris Tomes said: “My concern has always been that these reports contain serious injuries and in reviewing the reports for the last few days, my concerns were well-founded.”

Ineligible deaths

In total, 176 deaths were recorded through FDA alternative summary reporting. While this system did allow the reporting of fatalities related to certain kinds of heart valves implanted five or more years beforehand, these made up only two thirds of deaths in the hidden database.

The ineligible death reports related to two kinds of pacemakers, a breast implant, an intra-aortic balloon pump and a ventilator.

The FDA said: “When an ineligible event was reported through an ASR, the FDA’s standard practice has been to contact the manufacturer to obtain additional information regarding the event. If the event was labelled in error, the manufacturer was asked to submit a supplemental ASR to correct it.”

However, of all 59 ineligible deaths, only eight appeared to have been revised in later reports, leaving 51 ineligible deaths unaccounted for.

Serious incidents

Outside of fatalities, some of the most serious events reported were related to breast implant injuries which had leaked, deflated or migrated. More than half a million unique reports were made, with 6,600 in 2019 alone. The most common issue was rupturing implants.

Blood glucose meters for patients with diabetes were found to have the most unique incidents of any type of device, with 2.4 million reports made over the last 20 years. This is actually relatively unsurprising – about 10% of the US population has diabetes, so the high demand for blood glucose monitors will inevitably lead to a notably high number of failure reports for those devices due to their prevalence.

What next?

The FDA will be keeping its Voluntary Malfunction Summary Reporting (VMSR) program, which still allows for less serious device malfunctions to be reported in a summary form every quarter. However, unlike under alternative summary reporting, this data will all be publicly available through MAUDE. Serious injuries and deaths cannot be reported through VMSR.

The FDA said: “The VMSR Program enables the FDA to efficiently detect potential safety signals and free up resources to better focus on addressing the highest risks, such as deaths and serious injuries, associated with medical devices.”

The regulator will now be using new funding to make the MAUDE database more user-friendly over the next few years, making medical device reporting data easier to find. It also plans to begin developing active surveillance capabilities for its National Evaluation System for health Technology (NEST), which has been developed since 2012 to systematically use real-world data to identify and address safety concerns around devices are on the market through active surveillance.

The agency said: “Active medical device surveillance will better protect patients by, for example, continuously using analytical software algorithms to evaluate large data sets on device performance and patient safety associated with device use in routine clinical practice.”