Hacker News Reader

Comment by FireBeyond | Hacker News Reader

Life Alert, horrendous as their ads may be, would probably be surprised to hear that it's only now that you're finding a "device which can contact emergency services in the event of an accident" so life changing. As might OnStar, etc.

View on HN

As well as what the sibling commentors have pointed out, Life Alert also never bothered to partner with a company that would put their system into a piece of consumer-electronics that the average healthy (maybe even young) adult would buy for themselves. Nobody wants to wear a LifeAlert bracelet and a watch/fitness-tracker; they want their watch/fitness-tracker to just have that functionality. And so they make a choice, and end up dying in a ditch somewhere.

This aspect, I absolutely agree with - the differing markets for devices. I hope I would not need to buy such a device for 20 or 30 more years. But I do own a Fitbit. And may own an Apple Watch. Very different sets, with differing benefits.

To be fair, almost all in home medical alert systems use a fixed base station. So even if they wore both, the Life Alert bracelet would only save lives in a ditch very near the home.

This is the key thing.

There are many good ideas that exist out there that are "unknown" because the product they are in misses some aspect of how consumers would wish to consume them. My favorite was that the Tesla Roadster wasn't the first electric car, but it was the first that I could see myself buying at some point.

The nice thing about this being a feature of the watch is that the watch brings a bunch of great stuff all on its own. That it can also call emergency services or diagnose Afib is like bonus time.

> or diagnose Afib is like bonus time

Careful with that as an assumption:

> In people younger than 55, Apple Watch’s positive predictive value is just 19.6 percent. That means in this group — which constitutes more than 90 percent of users of wearable devices like the Apple Watch — the app incorrectly diagnoses atrial fibrillation 79.4 percent of the time. (You can try the calculation yourself using this Bayesian calculator: enter 0.001 for prevalence, 0.98 for sensitivity, and 0.996 for specificity).

> The electrocardiogram app becomes more reliable in older individuals: The positive predictive value is 76 percent among users between the ages of 60 and 64, 91 percent among those aged 70 to 74, and 96 percent for those older than 85.

(Source: https://www.statnews.com/2019/01/08/apple-watch-iffy-atrial-...)

That being said, I will not at all dispute the ECG tools as very valuable.

https://www.ajmc.com/conferences/acc-2019/giant-study-sugges... - discusses the findings in the context of the Apple Health "Study" announced at one of their events.

In this case, it's better to have a false positive diagnosis than a false negative.

Why is that? IANAPhysician, but this seems like a fairly rare malady in this age group. Do we really want all of these people rushing to the hospital? If no one trusts the Apple Watch's diagnosis, how useful could it be?

I think the rate is low enough that most people are not getting multiple spurious notifications to rush to the hospital, so trust is probably not the issue. The only problem I see is hospitals being overwhelmed, but again, I'm not sure if we're really overloading them.

The fact that the baseline rate is so low is why this sort of widespread testing isn't really helpful, and can do more harm than good. The wider the net you cast, and the more rare the target condition, the lest useful the test becomes.

Sure, most people are not going to hospital at the suggestion of their Apple watch. However, of those who do go to hospital, nearly all are false positives. This will lead hospital staff to distrust the Apple watch. It may not have that effect on Apple watch owners...

You're unlikely to get rushed to the hospital due to your watch asking you if you're OK while you're consciously engaged in strenuous activity. It's a sign, not a diagnosis.

Yes I would really like to get multi-thousand dollar hospital bills because my watch made a mistake.

Not necessarily, no... There are costs to false positives (money, time, stress, etc).

As with everything, there is some give and take, but one point in favor of something like a watch doing fall detection is a potentially better anti-false-positive detection. Set your life alert pendant down on the dresser a little hard and it may think you've fallen, and maybe you don't hear the callback for verification, so you get a visit from the police in your shower after they've forced the front door open. This happened to my grandfather ;-). The Apple Watch should be able to correctly avoid a false fall detection in that situation because you will have taken it off your wrist (though not strictly necessary since it is waterproof) and it knows that it's not on your wrist.

All of those services rely on manual intervention from staff in a call center.

If any of those services have reached the point that the end device can detect an emergency and contact local emergency services by itself, then they've done a terrible job of advertising that functionality.

I do have a horse in the race. Not the one you might think, but as a paramedic/firefighter, I'm not sure you can argue that "attempts to verify (human or otherwise) that there is a medical incident before dispatching EMS" is automatically worse than "calls 911 from the device automatically".

> If any of those services have reached the point that the end device can detect an emergency and contact local emergency services by itself, then they've done a terrible job of advertising that functionality.

They all have (see my sibling comment about Automatic Fall Detection), and I agree.

Apple watches have audible and haptic alerts and a delay before calling emergency services.

LifeAlert makes you press a button, it is designed for people who might fall off the toilet, not someone who might fall off a mountain bike in the wilderness and be unconscious for hours.

OnStar is part of an entire car.

This is a watch with greater capability than both put together. Let it be awesome.

LifeAlert definitely has accelerometer based fall detection, not just a button press. It doesn't have a cell radio though, so it isn't useful when it's away from its base station.

Life alert & OnStar are fairly singular function devices. Life alert is marketed pretty much exclusively to the elderly and does not have the ability to automatically call 911 in this way. OnStar is available in a single manufacturer's vehicles only, and you can't take it with you when you step onto the sidewalk. A device that monitors vitals as well as other potential situations and automatically gets an ambulance brought to your locations is quite a bit different than either of these.

I'm not claiming Apple has done something revolutionary. It's an evolution that builds on and combines various capabilities in a novel way. It's the sort of thing I could imagine seeing growing up watching "future" tv shows (Beyond 2000 ftw) and I love when technology produces something like that.

If they're surprised by this then that explains a lot.

Ah, but they don't sell chiche phones, laptops, and watches :)

This, but unironically

Irony was not my intention.

Not really. All those other services require the person to hit a button to trigger it, which they can't do while unconscious. The Apple watch called automatically on its own.

No. All those services absolutely offer Automatic Fall Detection: https://www.theseniorlist.com/medical-alert-systems/best/fal...

And OnStar has it easier, with existing collision detection.

Interesting, how much does the Apple Watch fall detection cost monthly?

$10/month if you factor in the need for a cellular plan for the watch?

$15/month for a fall detection device vs $399 + $10/month buys a lot of months. Yes, I know the Apple Watch does more, but I wasn't the one who tried to make the comparison.

Does the Apple Watch stay on the phone with you until EMS arrives? Is it capable of giving medical information to dispatch?

Another "Apple was the first to do this!"... "No it wasn't"... "Well, the cost is different!"... circumlocution.

I could care less, either way. Whichever device gives me less false positives (but not false negatives) works best for me as a first responder.

So long as the non-cellular version of the watch has an active bluetooth link with your phone, it will make the same callout.

What if you own the cellular version but didn't pay for a plan? Many phones will still allow 911 calls even without as paid plan.

AFAIK all phones allow 911 calls if they have cell signal even without any SIM inserted.

I had no idea some services could be so expensive. Life Alert starts at $50/month. Depending on what you get from it, the Apple Watch would be cheaper in less than a year. Interesting!

"I've fallen and I can't get up!"

edit: its a reference to the ads, you bozos.

Reply on news.ycombinator.com