How to Read OSCAR Charts in Plain English (AHI, Leaks, Pressure)

You downloaded OSCAR, popped in your SD card, and... now you're staring at a wall of squiggly graphs that look like a seismograph during an earthquake. You're not alone. OSCAR (the Open Source CPAP Analysis Reporter) is the gold-standard free desktop tool for reading your own CPAP data, but it hands you raw charts with no plain-language explanation. This guide walks through the daily view one graph at a time — AHI, leaks, pressure, and the flow rate waveform — so each row finally means something.

The OSCAR daily view layout

The daily view is OSCAR's main screen and where you'll spend almost all your time. It shows a single night (technically a single CPAP session, or set of sessions, from one sleep period) broken into stacked, time-aligned graphs.

From top to bottom, the typical layout includes:

• A statistics panel (usually top-left) with your headline numbers: AHI, usage hours, and pressure summaries.
• Event flags — colored tick marks showing exactly when each apnea, hypopnea, or other event was scored.
• Leak rate — how much air escaped your mask seal over the night.
• Pressure — what your machine actually delivered, second by second.
• Flow rate — the raw breathing waveform, the most detailed and most intimidating chart of all.

The key idea: every graph shares the same horizontal time axis. If you see a spike in leaks at 3:14 a.m., you can look straight down (or up) to the flow rate and event rows at 3:14 a.m. and see what your breathing was doing at that exact moment. That vertical alignment is the whole point of OSCAR — it lets you connect cause and effect.

If this layout already feels like a lot, that's normal. For a gentler orientation to the underlying metrics before you dive in, see our plain-English guide to reading CPAP data.

Reading the AHI and event flags

AHI stands for Apnea-Hypopnea Index — the average number of breathing events per hour of sleep. It's the single most-quoted number in CPAP therapy, and OSCAR puts it front and center.

OSCAR (like your machine) flags several event types, color-coded along the events row:

What's a good AHI?

A residual AHI below 5 events per hour is the widely used benchmark for effective CPAP therapy — the AASM defines an "optimal" titration as reducing the AHI to fewer than 5, which is also the normal, non-apneic range. Some clinicians aim lower (under 1–2) for fuller symptom control when it's comfortably achievable, but there's no formal guideline establishing "below 2" as a standard target, and goals are individualized with your provider. For the full picture, see what counts as a good AHI on CPAP.

A critical caveat about device-reported AHI

The AHI OSCAR shows is your machine's estimate, not a lab-scored number. Your CPAP has no EEG to detect arousals, so it can't score events exactly the way a sleep lab does. Labs use AASM Rule 1A (the recommended rule — a hypopnea needs a 3% oxygen desaturation or an arousal) or Rule 1B (the acceptable/CMS rule — a 4% desaturation). Your device just estimates from airflow. So your home AHI and a lab AHI can legitimately differ.

When central events climb

If you notice your clear airway (CA) count rising over time on CPAP, that can signal treatment-emergent central sleep apnea (TECSA), sometimes called complex sleep apnea. CPAP is built to splint the airway open and reliably treats obstructive apneas, but it doesn't directly correct the unstable breathing drive behind central events. TECSA shows up in roughly 5–15% of PAP titrations and often resolves on its own within weeks to a few months of continued CPAP (reported spontaneous resolution around 60–80%). Persistent central events — especially with returning daytime sleepiness, deep oxygen drops, or new medications (like opioids) or heart conditions — deserve a clinician's review, which may lead to BiPAP or ASV. The one thing not to do is raise your own pressure to chase centrals; higher pressure doesn't fix them and can sometimes provoke them. We cover this in depth in central apneas showing up on CPAP and what "clear airway" (CA) events mean.

The leak graph

The leak graph shows how much air escaped your mask system over the night — and it's arguably the most important row for trusting every other number on the screen.

Here's why: a large leak corrupts your AHI and your flow-limitation reading. When air gushes out of a bad seal or an open mouth, the machine's flow sensor gets noisy data. It can miss real events or misclassify them, so a night with high leaks gives you an AHI you simply can't trust. This is the single biggest reason to glance at leaks before you celebrate (or panic over) an AHI number.

A few brand specifics matter when reading the leak row:

• ResMed machines report excess leak (leak above the expected vent flow). The widely cited threshold is 24 L/min of excess leak at the 95th percentile — above that, ResMed considers your data unreliable.
• Philips Respironics machines report total leak (intentional vent flow plus unintended leak), so the baseline is different and the numbers aren't directly comparable to ResMed's.

Don't compare a Philips leak number to a ResMed threshold — they're measuring different things. To dig into acceptable ranges and fixes, see what's an acceptable CPAP leak rate and how to fix CPAP mask leaks.

The pressure graph

The pressure graph shows what your machine actually delivered, in centimeters of water (cmH₂O), throughout the night.

• On a fixed CPAP, this line is nearly flat — one prescribed pressure all night.
• On an APAP (auto-adjusting), the line moves up and down as the machine responds to events and flow limitation, so you'll see a staircase pattern.

The number most people care about is the 95th-percentile pressure — the pressure your machine reached or stayed below 95% of the night. It's a robust way to describe "how hard your machine was working" without letting a single brief spike skew the average. Many people find their 95th-percentile pressure is the most useful figure to bring to a provider conversation. We break down 95th-percentile vs median pressure separately.

Reading the pressure row helps you have an informed conversation with your provider — for example, if your APAP is constantly pinning at its maximum, or if pressure climbs right when leaks spike. Use the data to ask better questions; don't adjust prescribed pressure on your own. If you're wondering whether your pressure is off, is your CPAP pressure too high or too low walks through the signs.

The flow rate waveform

The flow rate waveform is the raw, breath-by-breath trace of air moving in and out of your airway — and reading it manually is the most intimidating part of OSCAR for almost everyone. It's also where the richest information lives.

Here's the orientation:

• Above the center line = inhalation. Below = exhalation.
• A healthy breath looks like a rounded, sine-wave-like hump on the inhale side.
• You have to zoom way in (OSCAR lets you scroll down to a few breaths at a time) to read it — at full-night zoom it's just a solid blue block.

What people look for once they zoom in:

• Flat-topped inhalations. A breath whose top is chopped flat instead of rounded is a classic sign of flow limitation — your airway is narrowing even though no full apnea is scored. This is the "hidden metric beyond AHI."
• A clear flatline during a flagged apnea — confirming airflow truly stopped.
• Periodic, waxing-and-waning breathing (a crescendo-decrescendo pattern) can indicate periodic breathing or Cheyne-Stokes respiration, which is worth flagging to a clinician.

This is where OSCAR rewards patience, but it's also where most people give up — pattern-recognition on a waveform is a skill. Our dedicated guide on how to read the flow rate waveform and flat-top breaths goes breath by breath, and flow limitation: the hidden metric explains why a "good" AHI can still leave you tired.

One number, many nights

Whatever the waveform shows on any single night, remember: single-night numbers are noise. A bad AHI after a glass of wine or a stuffy nose tells you very little. Trends over weeks are what matter — a leak that creeps up over ten nights, or flow limitation that's worse every night you sleep on your back. Read your charts for patterns, not verdicts. Why your AHI changes night to night is essential context here.

Or skip the decoding entirely

OSCAR is powerful and free, but it asks a lot: a desktop install (no Chromebook support), no auto-scoring, and a learning curve that lands squarely on the flow rate waveform. For many people, that's the wall they hit.

Or let SomniCharts read it for you. SomniCharts shows you the same charts — AHI, leaks, pressure, flow rate — but interprets them automatically in plain language, in the cloud, with nothing to install. Upload your SD card and it tells you whether a leak invalidated your AHI, whether those flat-topped breaths add up to flow limitation, and how this week compares to last — the trend view that actually matters. It imports ResMed, Philips Respironics (including the encrypted DreamStation 2, which OSCAR can't read), and Löwenstein prisma data.

If you want to weigh the desktop and cloud routes side by side, see our OSCAR CPAP software guide and the easier cloud alternative, the broader CPAP Data Tools & Apps hub, and the tool-by-machine support matrix.

FAQ

What does AHI mean in OSCAR? AHI is the Apnea-Hypopnea Index — the average number of apneas and hypopneas your machine scored per hour. A residual AHI below 5 is the standard benchmark for effective therapy, but it's your machine's estimate, not a lab-scored number, so it can differ from a sleep study.

Why is my flow rate graph just a solid block? You're zoomed out to the whole night. Scroll and zoom in until you can see individual breaths — only then can you spot flat-topped (flow-limited) inhalations or true flatlines during apneas.

Can a high leak make my AHI look better or worse? Both. A large leak feeds the machine noisy data, so it can miss real events or invent false ones. Always check the leak graph before trusting your AHI — on ResMed, excess leak over 24 L/min at the 95th percentile means the night's data is unreliable.

Should I change my pressure based on what I see in OSCAR? No. Use OSCAR to understand your therapy and have an informed conversation with your sleep clinician. Don't self-adjust prescribed pressure — and never raise pressure to chase central (clear-airway) events.

Is one bad night a problem? Usually not. Single nights are noisy; trends over weeks are what reveal real issues like a worsening mask seal or persistent flow limitation.