The deaths are associated but not necessarily caused by the incorrect readings, but as is often the case, medical interventions treat all observed conditions as side effects (this is similar in the case of drug and device trials), and the FDA has typically operated from an abundance of caution (though policies are somewhat erratic under the current administration).
I share the skepticism of the top-level comment by jimrandomh, in I understand that CGMs are used to guide treatment but not determine it, and that the consequence of spurious low blood-glucose readings is not likely to be immediately threatening (that is: the consequence of mistreating based on the mis-reading would be an actual high blood glucose event), though of course over the long term, high blood glucose levels are precisely the mechanism by which long-term and late-stage diabetes symptoms and conditions emerge.
Given the large number of devices (38% of US adults, or ~125 million), and millions of CGMs in use, seven associated deaths seems a relatively low number and correspondingly low risk.
TFA also would seem to misclassify the problem as one principally of software where the actual principle issue would be of potential patient noncompliance with protocols. That itself is complex, and isn't necessarily a matter of blame (the very young, otherwise ill, or cognitively-impaired might well be expected to comply poorly with instructions), but is a concern providers and dispensing pharmacists would have to be exceedingly cognizant of. As well as device manufacturers.
By the same logic, there's no such thing as "cancer", a "cold" (or more accurately: upper respiritory illness), or a broken leg, since each of these have many distinct causes.
All models are wrong, some models are useful. And some are based in at least part on historical accident and sequence of understanding. Diabetes (etymology, Greek diabetes, excessive discharge of urine), is one such of these.
Of the multiple distinct types of diabetes currently recognised (types 1 & 2, which you note, gestational, MODY, 5, and possibly several others), there is a commonality of primary symptoms (unregulated, often high, blood sugar), treatments (most must or may be treated with supplemental insulin), monitoring (of blood glucose levels typically by finger stick or CGM, as well as HgA1C for longer-term status and progression), of healthcare providers specialising in the diseases (generally endocrinologists), and of long-term complications: high blood pressure, heart disease and failure, neuropathy, poor circulation, various infections, and often peripheral limb amputations.
Thus the medical literature notes that diabetes is a group of common endocrine diseases all sharing high blood sugar levels, though of distinct types having distinct causes but largely similar treatments.
In the same sense, treatment for a broken leg largely doesn't distinguish on the cause of the fracture (blunt trauma, falls, osteoperosis, gunshot), treatment of respiratory illnesses is similar despite different infectious agents, and cancers, whilst varying greatly in prognosis and treatment, share the commonality of unregulated growth and metastases, with similar end-stage consequences.
All labels and concepts are human constructs to simplify a complex world. Absolutism over definitions tends not to be especially enlightening. Or useful.
I'm not nitpicking. Using "diabetes" instead of "type 1 diabetes" or "type 2 diabetes" really hurts people with type 1 diabetes. It's a dangerous confusion.
By default, people assume "type 2" when they hear "diabetes." They don't understand that type 1 is a completely different disease - and an absolutely terrifying one. Type 1 and type 2 are as different as day and night. It's like having runny nose vs having no nose.
This confusion harms awareness of type 1 diabetes. It undermines the urgency of finding a cure and shifts attention away from type 1.
When people are diagnosed with type 1 diabetes (or, more often, when their toddlers or children are), they get furious that this confusion exists at all - and that they knew nothing about type 1 diabetes beforehand.
Even more fundamental than data collection is advertising.
Books, and to some extent film, are the only media which aren't absolutely flooded with advertising. Print serial media are (though it's rapidly vanishing), music is (through both broadcast and streaming services, though not via direct media purchases), serial video (television, cable, streaming services, YouTube, and of course social media, are all absolutely saturated with advertising. Which is what the data are feeding, of course.
That Newport fails to make this distinction, and that the goal of TikTok et al are to absolutely engross your attention and time, is a critical failure of this piece.
What you've done above and beyond GTD as I noted in an earlier comment (<https://news.ycombinator.com/item?id=46376558>) is that you've time-blocked when you make that assessment, on the (IMO correct and insightful) notion that triage itself is expensive. That is, determining the importance and length of a task / item of email itself consumes limited cognitive resources.
Adding to what I wrote earlier, another advantage of postal mail is that it comes at fixed intervals, typically once a day (historically possibly more often especially in cities with "morning" and "afternoon" mail, making one-day responses possible, currently with curtailments in service possibly only a few days a week). This automatically batches mail processing.
Early in the corporate adoption of email a firm I worked at only polled periodically for new external email (every 20--30 minutes or so). Whilst internal email was pretty instant, this meant that at most external emails would give cause for interruption only a few times an hour, rather than at any given moment. I've given thought to reimplementing this on my own systems from time to time, perhaps even only 2--3 times a day, say, "morning email" (limited to priority recipients), and afternoon email (the Great Unwashed Masses have their opportunity).
In reality, I've adopted Inbox Black Hole, in which I rarely if ever check personal email. Circumstances make this reasonably viable, though those are decidedly atypical and most professionals would be unable to adopt a similar tactic.
Tariffs, as with taxes, may serve positive, market-favourable functions, particularly in addressing market failures, uneven regulation (e.g., higher pension, safety, environmental, and/or medical-care burdens in the importing country), as well as anti-dumping or anti-interference actions. British-Korean economist Ha-Joon Chang writes of this, particularly in Kicking Away the Ladder:
-- Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, 1776
More significantly, one of the first discussions within the text of precisely what wealth is (on which Smith has several, and occasionally inconsistent, answers, and which he seems to think of as more a flow than a stock.)
One of Smith's principle complaints in his text was what we now call "market failures" and "regulatory capture".
I've realised a few things dealing with time and attention, and devised a few strategies with varying degrees of success:
- Information consumes attention (as has been long observed).
- Corollary: excess information demands fast, cheap, regret-free rejection mechanisms. TFA describes several such approaches. The "DBTC" folder is one, but specifically refusing to use other, unmanageable, message queues (Twitter, FB, Slack, etc.) would be others. If a tool refuses to respect your boundaries, reject that tool.
- Time-blocking for low-urgency, but still significant tasks is useful. You're shifting from interrupt-driven mode to scheduled flow. This also means you can assess how your schedule relates to the incoming message flow, and whether or not that flow still exceeds your (now far more readily quantifiable) time devoted to it.
- There's still the question of how to prioritise items you're responding to. I'd suggest a rough triage method of:
1. Identifying high-priority senders (immediate family, work (management, colleagues, business relations), friends/social, and pretty much all else.
2. Randomly selecting from lower-priority queues is a way of fairly distributing your attention. If you can't do everything, sample a handful of items.
3. Quick "no"s (and learning how to phrase these delicately, if necessary) are useful. In some cases you might point the correspondent in a more useful direction. There's the physics professor's tactic of dealing with crackpot questions by directing them to one another, which preserves both attention and sanity....
My first exposure to the correspondence-limits problem came in one of the SF author Arthur C. Clarke's essay collections published in the 1970s or 1980s, in which he wrote of having had to resort to the tactic of responding to most of his own voluminous postal mail correspondence (and that international postal mail, for the most part, as he lived in Sri Lanka whilst most of his correspondents were elsewhere) with a pre-printed post-card with a set of checkboxes which answered most common inquiries. He'd already considered two further options: "Mr. Clarke regrets", and silence.
reply