Slashdot Mirror
Hopkins Study Finds Popular Blood Pressure App Wildly Inaccurate (jamanetwork.com)
An anonymous reader writes to point out a software review of the kind you can't generally find in an app store. A group from Johns Hopkins checked the accuracy of the Instant Blood Pressure app, which has sold more than 148,000 copies, and purports to measure blood pressure with just an iPhone -- no cuff required -- and found it wanting, to put it mildly. In the researcher's study, published in JAMA Internal Medicine, the app missed elevated blood pressures four out of five times.
If there's no cuff or anything, how does it work? Magic?
And this surprises anyone (other than those 148,000 customers, of course) because...
No one should trust a digital health app. Unless it's been validated in a clinical study, you shouldn't trust anything that's put out to run a smartphone.
Especially one like this. It uses the phone's accelerometer to measure your heart rate by putting it up against your chest and then measures the coloration of your fingertip by putting your finger over the camera lens. iPhone camera lenses and accelerometers were not designed for this use case, and this type of measurement has never been clinically studied or validated, and this is the common trend in nearly every single software health app out there.
I'm sorry but phones weren't designed for this kind of use and the software isn't either. People making these apps are wishful thinking at best and dangerous at worst.
$100? I've been using a $50 Walgreen wrist bp monitor for 3 years now that's been reliable and accurate. I just went to the doctor last Tuesday and before I went I checked my bp with my monitor and it was 127/84. 45 minutes later at the doctors office the nurse checked it and my bp was 126/79. Unless you've got circulatory problems in your arms, the monitors are decently accurate. When buying one, take it to the doctor's office and compare with the nurse's readings.
I have taken my Omron to the doctor's office to compare to his measurement several times and it's been very close every time.
Even if you measure your BP with the standard cuff, it's difficult to measure. Patient beware. Your BP can be thrown off by a number of factors such as recent caffeine intake, urinating, not urinating (because you're anticipating the need for a sample), nerves known as "white coat hypertension", and probably some other things I don't know because I'm not a doctor.
If you have a high reading, I advise you to freak out like you're GOING TO DIE RIGHT NOW. Just kidding of course; but that's exactly the kind of dynamic that could get you placed on medication when you don't really need it.
Likewise, don't ignore consistently high readings. Keyword--consistent. Obviously, don't listen to me. Listen to a lot of other people, especially doctors; but what I'm telling you is pretty much in line with consensus to the best of my knowledge.
Bonus: that pinkish hibiscus tea? It's a natural diuretic. If you drink it and get tested, it could mask symptoms. Alternatively, it could act in tandem with your medication and bring your BP lower than you want it, causing you to get dizzy or even pass out.
Anyway, IMHO it's not necessarily bad to have home testing; but you have to understand what it is you're measuring, how hard it is to measure, and what the measurements mean. I'd say the device might be useful if you can get consistent results, and then at some point you start getting a change for no good reason. Then, see a doctor; but you should probably be getting regular checkups anyway...
so a badly coded app estimated blood pressure by superficial analysis of inaccurate measurements based on pseudo science.
and blood pressure in everyday conditions( as opposed to condition of a patient in controlled situation (eg hooked up in a bed) with certain other symptoms), even when measured correctly, tells us almost nothing by itself.
what a fake world!
Is as accurate as TornadoGuard?
Wow, who could have possibly predicted that an app for your phone, using no extra sensory hardware whatsoever, could possibly be wildly inaccurate! After all it must have been written with the Best Intentions of the programmer, who fills his life with unicorns and puppies and fluffy kittens, sugar, spice, and everything nice, what could have possibly have gone wrong??? Obviously the silly testers and end-users must not be using it correctly. Seriously, those people who had heart attacks and strokes and died? Must be their fault!
Are YOU using the TOOL, or is the TOOL using YOU? Think about it!
What happens when your doctor's office doesn't use an old fashion cuff and instead use this one? Every time I go in I have to position my arm in a different way. Sometimes it's stretched out, sometimes it's across by chest, once it was straight up. It's not really a surprise that my blood pressure always seems to be different (both significantly lower or higher) than what a traditional reading normally is for me.
Suppose I have an unseasoned nurse. What seasoning would you recommend? Thyme and Oregano? Maybe a little Cilantro?
Guns don't kill people; Physics kills people! - John Lithgow as Dick Solomon on Third Rock From The Sun
Oh, who do you think you are kidding? If he is just average looking, but pays you any attention at all, that alone would get your blood pumping and you know it.
Guns don't kill people; Physics kills people! - John Lithgow as Dick Solomon on Third Rock From The Sun
All this time I have been waiting for the damn cuff to fall into the correct position of its own volition!
Guns don't kill people; Physics kills people! - John Lithgow as Dick Solomon on Third Rock From The Sun
I remember when I wanted to take a more proactive approach to monitoring my BP, so I went out and bought one of those machines. Paid almost $100 for it, so wasn't exactly the cracker-jack model.
Plugged it in and performed several tests on both arms. Measurements swung so wildly that I ended up returning the damn thing the next day. Utterly pointless.
You want to measure your blood pressure accurately and consistently? Do it the old-fashioned way. Find a seasoned nurse with a good ear.
I have an older Omron and the readings it gives seem to be both consistent and broadly in line with the readings at my doctor's office. Sometimes getting different readings on different arms can be a sign of something wrong.
Every time I go in I have to position my arm in a different way. Sometimes it's stretched out, sometimes it's across by chest, once it was straight up. It's not really a surprise that my blood pressure always seems to be different
This is not the fault of the device, it is a problem in training and experience. If your health care professionals are as incompetent at the use of a standard blood pressure cuff as they are with the wrist cuff from Amazon, you'll get "significantly higher or lower" readings using the "traditional method", too.
"Measuring device at heart level" is the rule. Which medical professional do you go to that thinks that your wrist held straight up above your head is "at heart level"?
A blood pressure monitor is a class II medical device, according to the FDA: http://www.fda.gov/MedicalDevi... Seems like the app creators could be in some pretty hot water if their device doesn't work and they didn't get certification from the FDA to distribute it...
Just because I can hook a shark from a boat, I do no offer to wrestle it in the water.
That they haven't had a visit from the FDA. Even with all their weasel-words, I don't see how they can get out of being considered a medical device by the FDA.
Considering the amount of crap my company goes through to comply with the relevant regs, I get really annoyed by people flaunting them and getting away with it.
A thousand pounds of wood moving at 300 feet per minute. Don't get in the way.
Just like blood oxygen saturation. It's 99% +/- 1% for 99% of all users.
The thinge probably measures the time it takes for the arterial pressure spike from a heartbeat to propagate to the tip of your finger, and then uses some wild guessing to translate that into a blood pressure value. The technique kind of works, but only for measuring changes in blood pressure. You have to start with a regular (oscillometric or RR) blood pressure measurement and then track changes using the arterial pressure propagation delay.
Or make sure you actually know how to use one of those machines. You need to position the cuff very deliberately, make sure you are siting correctly, make sure the arm is at the proper level. There's more to it then it seems. I'm not say you specifically didn't use it correctly but in general people don't.
I'm not sure I've ever heard a more polite way of accusing me that I didn't RTFM, but thank you.
Oh, and by the way, yes, I did RTFM.