Article

What Your Apple Watch AFib Burden Percentage Actually Means

Published 5 August 2026 · 7 minute read · by Bryan, developer of My AFib Companion

It arrives on a Monday. Your watch buzzes, and there is a percentage attached to your heart.

Maybe it says 4%. Maybe 22%. Maybe “2% or less.” Whatever it says, it arrives with almost no explanation, and if you were diagnosed recently, a bare number about your heart is an unpleasant thing to get over breakfast.

Nobody explained it to you. That isn’t your fault — Apple documents it across a support page and a legal PDF. Here is the plain version.

What the number actually measures

AFib burden is the percentage of time your heart showed signs of AFib relative to the time you were wearing the watch, over the previous seven days. Apple’s Instructions for Use puts it precisely: “the proportion of time your heart was beating in atrial fibrillation relative to Watch wear over the previous week.”

The denominator is the part nobody mentions. It is not the week. It is your wrist.

Apple delivers the estimate every Monday, covering the seven days before. Monday’s number describes last week — not today, and not right now.

Why the denominator changes the answer

Say you wear the watch all day and take it off overnight to charge, and your episodes tend to come at night — an extremely common pattern; there’s a whole genre of forum post titled some version of “afraid to sleep, most of mine happen at night.”

Then the watch is on the nightstand for exactly the hours you most want measured. Those hours land in neither half of the fraction. They’re absent.

Worth knowing: Apple states the estimate never displays 0%. The lowest it shows is “2% or less.” A very low reading is a floor, not a zero — and a floor calculated only from the hours the watch was on you.

And even while worn, it isn’t watching constantly: Apple says the watch “only checks for signs of atrial fibrillation periodically” and “may not find every instance.”

None of that means the watch is broken. It means it’s answering a narrower question than the one you’re asking.

Why some weeks are blank

Apple’s stated requirement is at least 12 hours a day, 5 days a week to consistently receive estimates. Heart Rate and Wrist Detection have to be on, and Low Power Mode switches off background heart measurements entirely. Fall short and you get no estimate — just a notification saying there was no data that week.

If your history has holes, that’s almost always a charging schedule, not a malfunction. Charging while you shower and while you eat dinner usually clears the bar without giving up overnight wear. Two weeks with different wear time also aren’t comparable to each other.

How precise is it?

Apple published this and almost nobody reads it. In the validation study, 413 people wore an Apple Watch and a reference ECG patch together for up to 13 days; 280 contributed to the main accuracy analysis. Against the patch:

• the average difference from the ECG was 0.67 percentage points
• 92.9% of participants’ weekly estimates fell within 5 percentage points
• 95.7% fell within 10 percentage points
• the statistical limits of agreement ran from −11.4 to +12.8 percentage points

Those last two lines should change how you hold the number. A 6% week and a 2% week may not be a real difference. It’s a tool for a trend over months, not for resolving a four-point gap between two Mondays.

So what is a good AFib burden?

There isn’t a passing score and I’m not going to invent one. I’m a software developer, not a doctor.

The honest reading of a 4% estimate is: the watch estimated AFib during roughly 4% of the hours you wore it last week, give or take the spread above. Whether that matters is a question for your cardiologist, who knows your history, your medications, and what your number was six months ago.

It also isn’t comparable to anyone else’s. Different wear habits, different weeks, different everything. Anyone online who hands you a threshold without knowing your history is guessing.

Three things the number doesn’t cover

It isn’t an alert. Apple states it plainly: “AFib History does not notify you when you’re experiencing AFib.” It’s a weekly summary, after the fact.

It isn’t every irregular rhythm. Apple says AFib History measures atrial fibrillation only — not other atrial arrhythmias such as atrial flutter or atrial tachycardia.

It isn’t meant to steer treatment. Apple’s Instructions for Use says the feature “is not intended to be used to aid in medical management of AFib” — not for changing medications, not for your doctor to adjust treatment. The support page does encourage sharing it with your physician for richer conversations. Both at once: material for a conversation, not an input to a dose.

The setting you weren’t asked about

When you set up AFib History, your watch turned off irregular rhythm notifications. Automatically. Apple’s reason: those notifications aren’t intended for people who already have an AFib diagnosis, and setup asks you to confirm you have one.

It catches almost everybody out. You went from a watch that occasionally tapped you to one that says a single thing a week. Nothing broke — you traded a real-time nudge for a weekly summary, and it’s reversible.

Longer piece on that setting, including how to switch back: Turning On AFib History Switched Off Your Irregular Rhythm Alerts.

What to actually do with it

Write it down every week. The percentage, plus roughly how many hours a day you wore the watch. Thirty seconds. Without the wear hours, the percentage can’t be interpreted later.

Bring the trend, not the snapshot. One week says very little; twelve in a row is a shape. Health app → Search → Heart → AFib History. After six weeks, Highlights adds a breakdown of the percentage of time spent in AFib by day of the week, and by four-hour block of the day — useful if you suspect nights, with the standing caveat that it only knows the hours you wore it. Show Life Factors overlays exercise minutes, sleep, weight, alcohol consumption and mindful minutes.

Ask one specific question. Not “is this bad.” Ask: “What would you want to know about between now and my next visit?” Write the answer at the top of wherever you keep records. That turns anxious data-collecting into a task with an end.

Getting it off the phone

Apple will export all of it as a PDF, free: Health app → Search → Heart → AFib History → scroll to Options → Export PDF → Share. If the burden chart is all you need, use it — it costs nothing and it works.

Where it stops being enough is context. Apple’s export has the burden chart. It doesn’t have what you felt, what you’d been doing, or which medications you took. That gap is what I built My AFib Companion for — it reads the AFib burden your watch already recorded, sits it next to episodes you log yourself, and makes one report you can hand across a desk. It’s $9.99/month or $59.99/year after a 30-day free trial, no free tier after that.

Use Apple’s free export first. If a percentage on its own turns out to be enough for your appointments, you’ve saved sixty dollars — and I’d rather you heard that from me.

---

Sources for the device behaviour above: Apple Support, “Track your AFib History with Apple Watch” (support.apple.com/en-us/108375, published 15 September 2025) and “Heart health notifications on your Apple Watch” (support.apple.com/en-us/120276, published 15 September 2025), plus Apple’s Atrial Fibrillation History Feature Instructions for Use (document 099-31365, revision K, July 2023). Checked 5 August 2026 — Apple changes these features, so check the current documentation if something here doesn’t match what you’re seeing.

---