Gadgets and gizmos as a tool to screen for AF 5 reasons for caution

Smart watches, gadgets in the back of phones etc……

I’m not yet persuaded. Why? – Answer in 5 parts –

Accept I’m hardline on this.

1). Optimise treatment in the current diagnosed population first. With robust patient centred care and shared decision making

One of the reasons why the proposition to screen failed NSC test at last ask.

DONT get me started on warfarin v DOAC – my blood pressure will rise.

NICE shouldn’t have said yes in the TAs of DOACs in my view, but it didn’t have much choice.

Sort out treatment optimisation first. The Bradford AF story is here. I kept meaning to write it up properly but failed to find the will to live so I did some pretty slides instead.

Message of chunk 1 is sort out anticoagulant in prevalent pop first.




2). Alive Cor is tool on lots of folk’s phones. Apple (commercial reasons), AHSN (its innovative isn’t it), NHSE (its innovative and makes us look good), and everyone else (its screening, screening saves lives so we should screen) is promoting its use.

I think (may stand corrected) they’ve done the work testing diagnostic accuracy of tool against gold standard (pulse and 12 lead ECG) and stacks up well

So as a tool on back of GP phone to diagnose symptomatic AF it might stack up. Slight caveat, a Bradford GP once diagnosed me with AF with the gadget – I don’t have AF – it’s not perfect.

The Apple Watch is now in the Marketplace. Data has been submitted to the FD it’s worth noting, that simple sums done by @GIDmk tells us in a very useful thread that according to this, the false positive rate for the Apple Watch in detecting atrial fibrillation is 0.04% (99.6% correct). This means that, on average, Apple Watches will be wrong more than 80% of the time

I’m sure other gadgets in the same, or similar space are available.

This is rapidly becoming the largest screening programme anywhere in the world.

Have a look at this thread by John Malondra

And read this blog – Overdiagnosis Only a Matter of Time With ECG Watches. I’d encourage you to read & inwardly digest every word. Remind yourself of the Screening 101 lesson plan also

Many have expressed concerns eg …. “This is arguably my most famous tweet, which has shown up at professional society meetings. Actual pulse is 147 bpm.”

see this tweet on questions for cardiologists before recommending and this one on the scale of mass screening that is not evidence based

However – having it on the back of a phone might lead to a wish for screening. But my fear is it leads to path of “we’ve got a shiny tool, we should screen for AF” . Important to distinguish symptomatic with asymptomatic. And also important to distinguish test vs screening programme. Screening is not a test, it’s a programme.

Message 2 – a) be really really sure the technology stacks up, are you really really sure the kit is good for the job.




b) distinguish symptomatic and asymptomatic.


c) Screening test vs programme



3) the 2014 NSC position on it. And other bodies

a) NSC. NSC position doesn’t recommend screening

I was asked a year ago by a professor of cardiology what it would take to get the NSC to change their view. What I wrote to him I turned into a blog – here

b) NICE.

Check the carefully worded NICE recommendation. There is no hint of a recommendation for screening there.

a year ago NIHR published a cost effectiveness report commissioned by NSC I think on screening for AF

c). AHRQ

it’s worth a read of the AHRQ report on screening

d). The USPTF view is also worth noting, also here

Message 3 = NSC position on screening stands, till they change position.

furthermore, see here for a study that seems to show another reason detection of asymptomatic AF may not make a difference in reducing stroke —> Yes, AF is a single risk factor for stroke, but there are many OTHER causes of stroke. The link to the study is here

e). Screening for AF comes with many snags

see this good thread summarising one of the more important AF screening papers of the last 5 yrs by John Mandrola.

Screening low-risk adults for AF would require 10K people screened to prevent 1 stroke, but 800 of those would get false +ve result. There’s no good clinical evidence that treatment of asymptomatic AF w/ anticoagulnts improves outcomes for those patients

If we are using Apple watches to do this then you’ve got to buy ALOT of Apple watches to prevent a stroke (10,000). And that’s before factor in the cost of DOAC & harms from false +ve







4). However the world is full of forceful patient groups (mostly backed by pharma cash who want to sell us DOACs) who want us to save lives by screening.

They often play the Ahhhhh but it’s not screening, it’s “case finding” card.

There isn’t any difference. IMO. The end

Here’s the blog version of paper I wrote for Ann Mackie on the topic

I once asked Prof Mant if he was able to distinguish the concepts. He couldn’t. See this thread



5). Message 5 = it’s screening, unless it’s testing in symptomatic patient, thus think about NSC position.

6) even then. Operational stuff

Even when all what is addressed, see the conclusion to this editorial

Before we can introduce a national screening programme, we need to know from randomised trials whether or not AF screening is effective at reducing cardiovascular morbidity and mortality. There are ongoing trials exploring this question in Sweden and the Netherlands, and more are planned. If screening is effective, there are several operational issues that need to be addressed before a national screening programme can be introduced: What is the optimal target population? What is the optimal screening technology? How best to confirm that screen-positive people do have AF? How to ensure fully informed patient choice, both with regard to acceptance of screening, and subsequent treatment if found to have AF?

There is potential for public health gain in screening for AF, but, as yet, the evidence base is unlikely sufficient for a national programme. there is far bigger potential in other spaces.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s