Here’s something I don’t like: the idea that a piece of technology I rely on to keep me alive could somehow kill me. I’m speaking not of Toyotas, but of insulin pumps — according to this piece in the Wall Street Journal, “the Food and Drug Administration said Wednesday it has seen an increasing number of hardware and software problems with insulin pumps, tiny devices worn by thousands of diabetics to deliver insulin.” And so on Friday, the FDA brought together an advisory panel of outside medical experts to discuss what actions might be taken to “‘minimize risks associated with the devices in these recall situations.'”
As someone whose pump once suffered a “button error” and began spraying insulin into the air like a fountain (or, less poetically, a peeing dog) I’m glad to hear that pump problems are something people are paying attention to. But I’m also curious as to what kind of issues they’re talking about. According to the Journal,
Manufacturers are required to report problems potentially associated with devices to the FDA. The FDA conducted a review of insulin pump-related adverse-event reports and found nearly 17,000 reports from Oct. 1, 2006, through Sept. 30, 2009. The reports don’t necessarily mean a device caused a problem but serve as a signal for more investigation. Even if a device is functioning properly patients can inadvertently misuse the device. Of the reports, about 12,000 reported a patient injury (such as problems with blood glucose levels) and 310 deaths.
The agency said the information provided by manufacturers involving deaths “was typically incomplete.” The agency said in 225 of the deaths reported the device problem was listed as “unknown,” although in many cases the device was never returned to the manufacturer for additional follow-up.
However, in 41 death reports, a device problem wasn’t identified but the circumstances involving the death involved diabetic coma and problems associated with blood-sugar levels being too high or too low, suggesting the device may not have been working properly.
So, uh, what are we supposed to conclude? Is the problem the devices, or the people using them?
I think it’s very important to keep track of these reports — since this is a situation where a product failure could kill someone, I want the companies who make insulin pumps to feel like there’s a fire under their ass. But at the same time, one of the reasons we don’t yet have a closed loop system (i.e. artificial pancreas) is because companies and the FDA are worried that the devices (or people using them) could make mistakes. Which, granted, they probably would — the question is whether those mistakes would be more severe and dangerous than the mistakes people with diabetes make every single day when we try to gauge interactions between insulin, exercise and meals.
I guess this is what I’m saying: it’s hard to draw conclusions from this article about how often the insulin pumps themselves are the problem (and how concerned we users should be about possible recalls). Looking at it on a more global scale, though, I have two competing desires. First, for companies to pay close attention to product flaws that might kill people. At the same time, I don’t want an obsession with technical perfection — in the case of an artificial pancreas — to get in the way of progress.
Bonus: Reading materials for the FDA conference yesterday