This article is more than 1 year old

Surprise! That blood-pressure app doesn't measure blood pressure

Be still my beating heart? The iPhone won't notice

Quantified self types not only fill Twitter feeds with reports from every walking, running, breathing and bonking app around – but the spewed data isn't always particularly accurate.

In a letter published in the Journal of the American Medical Association (JAMA), a bunch of US medical academics decided to test AuraLife's Instant Blood Pressure app, and they aren't pleased with what they found.

The short version is: unless you're treating the app as a novelty, don't bother with it.

The researchers tested the iPhone version of the popular $4.99 app – it's sold roughly 148,000 copies, and it's also available for Android – and found it “highly inaccurate.”

Most seriously, it under-reports hypertensive readings: “The low sensitivity for hypertensive measurements means that approximately four-fifths (77.5 per cent) of individuals with hypertensive BP levels will be falsely reassured that their BP is in the nonhypertensive range,” the researchers conclude.

Of course, with such a slap down in a serious test, AuraLife has responded in two ways: it's added a warning to its website, and it's written a blog post.

The warning looks like this:

Instant Blood Pressure Website with warning

Instantbloodpressure.com's disclaimer

… and it wasn't present on the site when Archive.org's Wayback Machine snapped it in January:

Instant Blood Pressure Website pre-warning

What crisis? Instantbloodpressure.com captured by Wayback in January 2016

In its front-page blog post, AuraLife complains that its app isn't designed to work in the way the JAMA authors expected.

The app only works for blood pressure between 102 and 158 systolic and 65 to 99 diastolic, the company says, and won't work outside this range.

Marketing BP fears

Instantbloodpressure.com marketing in January - but it's not designed to detect high blood pressure, says the company

The biz also complains that the letter's authors didn't take into account changes to its software during the study period.

“Their data set was effectively exposed to a combination of at least five different, unique sets of data variability because our servers-side algorithm was updated at least five different times on our server infrastructure during that time period,” we're told.

The measurements can vary enough in six months to invalidate external testing, apparently, and the blog post's author didn't stop to consider how that might be interpreted by actual normal users, we note.

The post concludes with the warning newly added to the website, that Instant Blood Pressure is “not a medical device.” That's good to know. ®

More about

More about

More about

TIP US OFF

Send us news


Other stories you might like