A normal AHI (apnea-hypopnea index) is fewer than 5 events per hour in adults. This means your breathing stops or significantly decreases fewer than 5 times per hour while you sleep. For children, the threshold is much lower: a normal AHI is less than 1 event per hour.
AHI is the primary number used to diagnose sleep apnea and gauge its severity. Understanding where your score falls, and what can affect its accuracy, helps you make sense of a sleep study result.
AHI Severity Scale for Adults
The classification system is straightforward:
- Normal: fewer than 5 events per hour
- Mild sleep apnea: 5 to fewer than 15 events per hour
- Moderate sleep apnea: 15 to fewer than 30 events per hour
- Severe sleep apnea: 30 or more events per hour
Each “event” is either an apnea (your airway closes completely and breathing stops for at least 10 seconds) or a hypopnea (your airflow drops by at least 30% for 10 seconds or more, paired with a dip in blood oxygen or a brief awakening). Your AHI is simply the total number of these events divided by the hours you were asleep.
Someone with an AHI of 25, for example, stops or nearly stops breathing roughly once every two and a half minutes throughout the night. At the severe end, an AHI of 30 or above means a breathing disruption at least every two minutes.
Children Have a Much Lower Threshold
An AHI of 1 or higher is enough to diagnose obstructive sleep apnea in a child. The pediatric scale looks very different from the adult one:
- Normal: fewer than 1 event per hour
- Mild: 1 to fewer than 5 events per hour
- Moderate: 5 to fewer than 10 events per hour
- Severe: 10 or more events per hour
A child with an AHI of 3 would be considered mild, while that same number in an adult would be completely normal. Adolescents and teenagers sometimes fall into a gray area, and some clinicians apply the adult scale for older teens.
Why AHI Doesn’t Tell the Whole Story
AHI is useful as a starting point, but it has real limitations. A large historical cohort study found that after accounting for other health risk factors, AHI alone was not a strong independent predictor of heart attack, stroke, or heart failure. What mattered more were related factors like how much time a person spent with low oxygen levels, how often they woke up, and their overall sleep duration.
This is why your sleep report often includes another number called the oxygen desaturation index (ODI), which counts how many times per hour your blood oxygen drops significantly. AHI and ODI frequently overlap, but comparing the two reveals something important: when your ODI is nearly as high as your AHI, it means most of your breathing events are causing meaningful oxygen drops, which signals a stronger hypoxic (low-oxygen) burden on your body. Two people can have the same AHI but very different oxygen profiles, and that difference matters for long-term health.
Your AHI can also vary depending on your sleep stage. Breathing disruptions are often far more common during REM sleep (the stage associated with dreaming) than during other stages. Research has shown that a high AHI specifically during REM sleep is linked to worse executive function and language skills, even when the overall AHI looks moderate. So a “normal” overall number could mask a problem concentrated in one part of the night.
Home Tests vs. Lab Studies
The number you get can depend on how it was measured. An in-lab polysomnography tracks your brain waves, sleep stages, breathing, oxygen, heart rate, and leg movements all at once. It knows exactly when you’re asleep, so it can calculate a true AHI based on actual sleep time.
A home sleep apnea test primarily monitors breathing and oxygen. Most home devices cannot determine when you’re asleep versus lying awake, so instead of dividing events by sleep time, they divide by total recording time. If you spent an hour awake in bed, that dilutes your score and can make your AHI appear lower than it actually is. Home tests identify obstructive sleep apnea about 90% of the time in people with clear signs and symptoms, but they can come back negative or inconclusive even in someone who truly has the condition. When that happens, doctors typically follow up with an in-lab study.
If your home test AHI sits right around 4 or 5, close to the diagnostic cutoff, it’s worth knowing that the real number could be somewhat higher once actual sleep time is factored in.
What Counts as an Event
The scoring rules for each breathing event matter because they directly affect your AHI number. The American Academy of Sleep Medicine defines a hypopnea as a 30% or greater reduction in airflow lasting at least 10 seconds, accompanied by either a 3% drop in blood oxygen or a brief arousal from sleep. Some labs use a stricter standard that requires a 4% oxygen drop and ignores arousals. The stricter rule typically produces a lower AHI for the same night of sleep, because fewer events meet the threshold.
If you’re comparing AHI scores from two different sleep studies, or from a CPAP machine readout versus your original diagnostic study, differences in scoring criteria can account for some of the variation. Your breathing didn’t necessarily change; the measuring stick did.
What Your AHI Means in Practice
An AHI under 5 is reassuring, but it isn’t the only factor that determines whether you feel rested or face health risks. If you’re under 5 but consistently exhausted, snoring heavily, or waking with headaches, your doctor may look at oxygen levels, sleep fragmentation, or upper airway resistance that falls just below the hypopnea threshold.
For those with an AHI in the mild range (5 to 14), treatment decisions often depend on symptoms. Someone at an AHI of 8 who sleeps well and feels alert during the day may not need intervention. Someone at the same AHI who falls asleep at red lights is a different situation entirely. In moderate and severe ranges, treatment is generally recommended regardless of how you feel, because the cumulative oxygen stress and sleep disruption carry long-term cardiovascular and metabolic consequences even in people who don’t notice daytime symptoms.