AHI vs RDI: Why Two Numbers Measure Your Sleep Apnea

If you've ever compared your CPAP machine's nightly number to the index on your sleep-study report and found they don't match, you're not imagining things. Your home device reports an AHI, while your lab report may quote an RDI — and the RDI is almost always the higher of the two. Understanding the difference explains a common, frustrating scenario: a "normal" AHI that still leaves you exhausted.

This is a data-education topic that sits at the heart of Understanding Sleep Apnea, so let's break down exactly what each number counts and why one can hide a problem the other misses.

What AHI counts

AHI stands for the Apnea-Hypopnea Index — the average number of two specific breathing events per hour of sleep:

• Apneas — a near-total or complete pause in airflow (typically a 90%+ drop) lasting at least 10 seconds.
• Hypopneas — a partial reduction in airflow (a shallow-breathing event) lasting at least 10 seconds, paired with a drop in blood oxygen or a brief arousal from sleep.

Add those two event types together, divide by your hours of sleep, and you get your AHI. It's the single most widely used number for measuring sleep apnea severity, and it's the headline figure on your CPAP report.

The conventional severity bands look like this:

For people already on therapy, a residual AHI below 5 events per hour is the widely used benchmark for effective CPAP — the AASM defines an "optimal" titration as bringing the AHI under 5. Some clinicians aim lower (for example, under 1 to 2) when it's comfortably achievable, but there's no formal guideline establishing "below 2" as a standard, and the right target is individualized with your provider. (More on this in What Is a Good AHI on CPAP?.)

One important caveat about how hypopneas get scored: the AASM offers two rules. Rule 1A (recommended) counts a hypopnea on a 3% oxygen desaturation or an arousal. Rule 1B (acceptable, and the one used by U.S. Medicare/CMS) requires a 4% desaturation. The same night can produce a different AHI depending on which rule the lab uses — one reason numbers don't always line up cleanly.

What RDI adds (RERAs)

RDI stands for the Respiratory Disturbance Index. It counts everything AHI counts — apneas and hypopneas — plus a third category: RERAs.

A RERA (Respiratory Effort-Related Arousal) is a more subtle event. Your airway narrows enough that you have to work harder to breathe, and that increased effort fragments your sleep with a brief arousal — but the event doesn't meet the threshold to be scored as an apnea or hypopnea. There may be no significant drop in oxygen at all. You don't fully wake up or remember it, yet it pulls you out of deep, restorative sleep.

So the relationship is simply:

RDI = AHI + the RERA index (RERAs per hour of sleep)

RERAs are the events that "fall through the cracks" of AHI. They represent real, disruptive airflow limitation — your body straining against a partly collapsed airway — that the apnea-hypopnea count was never designed to capture. We go deeper on these in RERA and Flow Limitation: The Events That Don't Show in Your AHI and in CPAP Event Types Decoded.

Why RDI is always at least as high as AHI

Because RDI is built by adding RERAs on top of the apneas and hypopneas already in your AHI, it is mathematically impossible for RDI to be lower than AHI.

• If you have zero RERAs, your RDI equals your AHI.
• If you have any RERAs, your RDI is higher than your AHI.

That's the whole rule: RDI is always equal to or greater than AHI. It's never the other way around.

This matters because the gap between the two numbers is itself a clue. A small gap means RERAs aren't contributing much. A large gap — say, an AHI of 4 but an RDI of 18 — tells you that arousal-causing, effort-related events are the dominant disturbance in your sleep, even though the apnea-hypopnea count looks reassuringly low.

When a normal AHI hides an elevated RDI (UARS)

Here's the scenario that confuses so many people: your AHI sits comfortably in the normal range (under 5), but your RDI is elevated — because the RERAs that drive it up simply aren't counted in AHI.

This pattern is the signature of UARS — Upper Airway Resistance Syndrome. People with UARS have an airway that narrows and increases breathing effort throughout the night, fragmenting sleep with arousals, yet rarely collapse far enough to trigger scored apneas or hypopneas. The result is the classic disconnect:

• AHI looks normal, so a basic screening can call sleep apnea "ruled out."
• The person feels anything but normal: unrefreshing sleep, daytime fatigue, brain fog, sometimes insomnia or sensitivity to light and sound.
• The RDI — if it's measured — reveals the real burden of disturbance.

If you have a clean AHI but still feel wrecked, an unmeasured RERA load is one possibility worth raising with your clinician. Our guide to UARS, the diagnosis your AHI misses covers this in depth, and Still Tired on CPAP With Good Numbers? walks through other explanations too.

There's a catch, though, that explains why RDI doesn't appear everywhere.

Why CPAP machines report AHI, not RDI

Your CPAP machine reports AHI — not RDI — and there's a concrete technical reason: scoring RERAs requires equipment your machine doesn't have.

A RERA is defined by an arousal from sleep. Detecting an arousal reliably requires EEG (brain-wave) sensors, the kind used in full in-lab polysomnography (PSG). RERA scoring needs that EEG arousal data plus measures of breathing effort. A bedside CPAP machine has none of it. It infers events from airflow and pressure signals alone — it can't see your brain waves, so it genuinely cannot count RERAs as a RERA index.

This is also why device-reported AHI and lab-scored AHI differ in general: your machine estimates events from flow patterns, with no EEG and no arousal scoring, while a lab technician scores from a full sensor array. (For how labs and home tests differ, see In-Lab PSG vs Home Test.)

What home devices can see is flow limitation — the flattened, restricted breath shapes that RERAs and UARS are physiologically built on. The machine can't label these as RERAs, but the underlying flow-rate waveform tells a story your AHI alone won't. This is where reviewing your own data pays off: SomniCharts imports ResMed, Philips Respironics (including the encrypted DreamStation 2), and Löwenstein prisma data and explains it in plain language automatically — surfacing the flow-limitation patterns and helping make sense of why your home-device numbers and your lab report can diverge. Pairing that picture with How to Read the CPAP Flow Rate Waveform and CPAP Flow Limitation: The Hidden Metric Beyond AHI gives you a far richer view than the single AHI digit.

One trust-building note while you're reading your data: a large mask leak can invalidate or under-report your AHI, because the machine may miss events when too much air is escaping. If your numbers look suspiciously good on a high-leak night, check the leak rate before trusting the index — see CPAP Leak Rate: What's Acceptable.

Putting it together

The short version: AHI is the everyday number your machine gives you; RDI is the fuller, lab-only number that adds RERAs. RDI is always equal to or higher than AHI, and that gap is exactly where conditions like UARS live. A normal AHI is reassuring — but if you still feel unrested, it isn't the final word.

And remember the bigger picture: a single night's index is noise. Trends across weeks are what reveal whether your therapy is genuinely working — see Why Does My AHI Change Night to Night? for what normal variation looks like.

Frequently asked questions

Is RDI always higher than AHI?

RDI is always equal to or higher than AHI, never lower. RDI = AHI + the RERA index, so if you have any RERAs, RDI is higher; if you have none, they're equal.

What is a normal RDI on a sleep study?

RDI is interpreted on the same severity scale as AHI: an RDI under 5 events per hour is generally considered the normal range, 5 to under 15 is mild, 15 to under 30 is moderate, and 30 or more is severe. Interpretation should always be done with your sleep clinician, who weighs the number against your symptoms.

Why doesn't my CPAP machine show RDI?

Because counting RERAs requires EEG (brain-wave) sensors to detect arousals, which only in-lab polysomnography has. Your CPAP estimates AHI from airflow and pressure alone and cannot score RERAs.

Can I have sleep apnea with a normal AHI?

You can have significant sleep-disordered breathing — most notably UARS — with a normal AHI, because the arousal-causing RERAs that drive it up aren't counted in AHI. An elevated RDI, or persistent symptoms despite a low AHI, is the clue. Bring this to your clinician rather than trying to interpret it alone.