The latest data from Apple’s study with Stanford University is encouraging and I think represents a step closer towards a killer use case for wearables as well as a complete disruption for the medical devices industry.
In November 2017, Apple and Stanford ran a study of 400,000 users of the Apple Watch 3 to ascertain just how good the device is at detecting abnormalities that are a sign of a serious health problem.
The study focused on atrial fibrillation which is the irregular beating of the heart that in many cases has no perceptible symptoms.
However, it is an indicator of major problems such as stroke, heart failure and dementia which in many cases are life changing or fatal.
The idea is to use asymptomatic atrial fibrillation to prevent these catastrophic and hugely expensive outcomes.
The data from the study is quite encouraging:
First, false positives:
During the study, 0.5% of the participants received notifications from their devices.
Without knowing the age range of the participants (occurrence increases substantially with age), it is impossible to determine how close this is to reality, but importantly it is in the ball-park.
In the developed world, 2-3%of the population will experience atrial fibrillation but the distribution is large with only 0.14% of the population below 49 being affected, going as high as 13% in people aged over 80.
This is important because false positives could cause a lot of undue stress as well as an unnecessary strain on the health system.
Second, good correlation:
The Apple watch indications of atrial fibrillation were found to be 71% accurate when immediately followed up with an electrocardiogram.
Furthermore, the Apple watch was found to be 84% accurate in detecting atrial fibrillation when the user was being simultaneously monitored with the watch and an ECG.
Third, the intermittent nature:
Of this condition was made very clear as only 34% of those that followed up with an ECG a week later were found to have the condition at that time.
This underlines the importance of continuous monitoring, which historically has been prohibitively expensive and completely impractical.
While the Apple watch is clearly not close to being medical grade, the accuracy of 84% is an excellent first step towards improving the detection of medical conditions are best diagnosed with continual monitoring.
Atrial fibrillation is the easiest to monitor with blood pressure and blood glucose also offering huge potential markets.
This is because high blood pressure is suffered by up to third of a developed market population with only around half of them being aware of it and seeking treatment.
Both high blood pressure and diabetes are conditions best treated with continuous monitoring as well as lifestyle changes.
Medical grade equipment is extremely expensive and bulky meaning that it is not well suited to continuous monitoring.
This is why I have long believed that a killer user-case for wearables is for them to be able to continually monitor these conditions at a medical grade for a fraction of the cost.
There is also a substantial financial incentive as 70% of all medical conditions are thought to be lifestyle related, the prevention of which would substantially reduce the burden being placed on creaky state health systems.
This is a good result for Apple and creates a greater incentive for users to continue paying premium prices for Apple products as there are tangible benefits to be had even at this early stage.
In this space, I am also keeping an eye on Valencell (blood pressure in earbuds), Lehman Medical devices (blood pressure test in a smartphone) and LifePlus (blood glucose in a smartwatch).
There still remains a very long way to go in terms of accuracy but critically I think this will be achieved through software rather than hardware.
The medical device industry is not in danger yet, but the writing is on the wall.