Free Athens Insomnia Scale (AIS) — Online Self-Check

Your AIS score in context

The AIS is a snapshot of how the past month slept, not a fixed measurement of your sleep capacity or insomnia identity. Sleep is highly responsive to circumstance — a stressful month, a stretch of shift work, a new caregiving demand, illness in yourself or someone close, a recent move, a change in medication, a change in alcohol or caffeine patterns, a partner with a different sleep schedule, jet lag — can push the score meaningfully upward during the window you happened to take the screen. A genuinely settled month can pull it downward. The 30-day recall window the AIS uses was designed to balance two competing needs: long enough to smooth out individual rough nights, short enough to detect change before a difficult pattern becomes entrenched at the three-month threshold that defines chronic insomnia disorder in DSM-5 / ICSD-3.

One distinction worth pinning down: the AIS measures self-reported sleep complaints across the past month, with a frequency stem (at least three times per week) that the instrument uses to keep the screen specific to clinically meaningful sleep disturbance. The frequency stem matters more than it may sound — it is the same threshold the World Health Organization uses to define nonorganic insomnia in ICD-10, and it is what separates the AIS from instruments that pick up occasional rough nights. If a particular complaint did not occur at least three times per week during the past month, the AIS treats it as not a problem; if it occurred several times per week, the AIS counts it. That design choice is also why the AIS is reasonably specific against ICD-10 diagnosis (85% specificity at the ≥6 cutoff) rather than over-flagging anyone who slept poorly on a few nights.

On the positive-screen cutoff: a total of 6 or higher is the Soldatos 2003 validated threshold for ICD-10 nonorganic insomnia, with 93% sensitivity and 85% specificity against clinician diagnosis in the original validation sample. That cutoff is widely reproduced in research and clinical use of the AIS, and it is the boundary between the no-clinical-signal band and the mild band shown above. On the severity split: the 4-band breakdown shown here (0–5 none, 6–9 mild, 10–15 moderate, 16–24 severe) is the Okajima 2020 calibration against the Insomnia Severity Index. The two interpretations are complementary rather than competing — Soldatos 2003 gives you a binary positive-screen signal; Okajima 2020 layers severity stratification on top.

When you retake the AIS, the change is more informative than any single number. No formal minimum clinically important difference (MCID) has been established for the AIS per Brown 2024 systematic review, so progress is best described in terms of band-shift (none → mild, severe → moderate) or sustained directional change across several administrations rather than a fixed point-change threshold. A single high score is information; a sustained pattern across multiple administrations is a different and stronger signal. The AIS was designed for repeated administration at intervals matched to its 30-day recall window — every 4 weeks is the natural cadence. Retesting more often produces noise; retesting much less often loses tracking sensitivity. If you are working with a clinician, they will typically look at AIS scores across several administrations rather than making decisions off a single result.

How to bring this to a clinician

The Athens Insomnia Scale is well-known in sleep medicine and primary-care research literature, and most clinicians will either recognize the instrument by name or recognize the ICD-10 nonorganic insomnia construct it was designed to map onto. You do not need to explain the AIS in detail — bringing the score gives the conversation a concrete starting point that descriptions of sleep otherwise lack, because feeling like sleep is not working is one of the more semantically diffuse complaints in clinical conversation and a numeric anchor on the 0–24 scale (with the Soldatos 2003 ≥6 positive-screen cutoff and Okajima 2020 4-band severity context) shortens the path to specifics.

What to bring:

• The total score on the 0–24 AIS scale (the number shown on your result above)
• Which items felt heaviest — the item pattern often tells a clinician as much as the total. AIS items split into a nocturnal cluster (items 1–5: sleep induction, awakenings, early waking, total duration, overall quality) and a daytime cluster (items 6–8: sense of well-being, physical and mental functioning, daytime sleepiness). Which side dominated for you is useful clinical information
• How long the sleep difficulty at this level has been present (best guess in weeks or months) — the three-month threshold matters diagnostically because it separates acute insomnia from chronic insomnia disorder in DSM-5 / ICSD-3
• Whether the pattern is more about sleep onset, maintenance (awakenings during the night), or early morning waking — these three sub-patterns sometimes respond differently to treatment
• What has been happening in your life in the same window — major recent events, a job change, shift work, a caregiving role, a relationship rupture, a health diagnosis, a medication change, alcohol or caffeine pattern shifts — that may be contributing
• Sleep environment and behavioural patterns — bed and wake time consistency across the week, time spent in bed vs. time spent asleep, what you do when you cannot sleep, screen use before bed, caffeine and alcohol patterns
• Whether a bed partner has noticed loud snoring, witnessed pauses in breathing, gasping or choking during sleep, or excessive limb movement at night — these point toward sleep apnea or movement disorders that may need separate evaluation
• Whether daytime sleepiness is severe enough to affect driving, working, or daily activity — that level of impact warrants prompt evaluation in its own right
• Whether mood-, anxiety-, or stress-related concerns are also part of the picture. If yes, bringing a PHQ-9, GAD-7, or PSS-10 result to the same conversation usually shortens the path to a useful plan, because comorbidity is the rule rather than the exception at moderate or severe AIS bands

A two-line opening you can use as-is:

I took the Athens Insomnia Scale at home and scored [X] on the 0–24 scale, which is above the validated positive-screen cutoff of 6. The items that felt heaviest were [item descriptions]. I'd like to talk about what to do next.

A clinician will commonly follow up by taking a more detailed sleep history (the structured sleep interview is the cornerstone of insomnia evaluation), screening for conditions the AIS does not capture (sleep apnea, restless legs syndrome, circadian rhythm disorders, narcolepsy), assessing daytime impact (work, driving, mood, relationships), and ruling out medical or medication contributors. They may also ask about specific patterns the AIS captures less directly — sleep architecture, drowsy-driving incidents, witnessed apneas, substance use, suicidal thoughts — to fill in gaps the broad AIS total does not address. Mentioning these areas upfront, even briefly, can shorten the appointment and get you to a concrete plan faster than open-ended descriptions. Per current practice guidelines (American Academy of Sleep Medicine, European Sleep Research Society, American College of Physicians, NICE), the recommended first-line treatment for chronic insomnia at moderate or severe AIS bands is cognitive-behavioural therapy for insomnia (CBT-I), a structured multi-component intervention typically delivered across 4–8 sessions; a clinician will likely walk you through what CBT-I involves and whether in-person therapy, group therapy, or a digital CBT-I program is the right starting point for your situation. You can print this page or save it as PDF using your browser's print menu — the result, score, and items all carry through.

If you're reading this with someone who took the test

If you are a partner, parent, sibling, close friend, or housemate reading this result alongside the person who took the test, this section is addressed to you. Insomnia at the moderate or severe AIS band is often quietly carried — by the time someone takes an AIS, they may have been managing the underlying picture for weeks or months without making it fully visible to the people around them, partly because not sleeping well is socially expected to be pushed through and partly because describing what is happening with sleep can feel diffuse and hard to start. The score gives you a concrete starting point for a conversation that can otherwise be hard to begin. Ask them directly what they want from you before drawing your own conclusions from the number. Different people in different shapes of insomnia want different kinds of support, and a score does not tell you which they need.

Three things that consistently help: showing up steady and present, without trying to solve the sleep picture or explain it away. Insomnia does not respond to "have you tried [obvious sleep hygiene tip]" or "you should just go to bed earlier" even when those statements are well-intentioned; both tend to land as dismissive of the actual lived experience and to imply that the person has not already tried the obvious options. Practical help with tasks that the sleep deficit has made harder — taking on logistics or caregiving tasks during particularly bad stretches, watching the kids for an evening or morning, picking up groceries, preparing a meal, driving them to a clinician appointment — often meets the moment in a way words cannot, because it directly addresses the daytime capacity that the sleep loss has compressed. And asking calmly what kind of support feels most useful right now: a quiet presence on a particularly hard morning, help thinking through a specific decision (CBT-I vs. starting with sleep hygiene, in-person vs. digital, what to bring to the clinician appointment), or just knowing you are available if needed.

Three things that tend not to help: minimizing the sleep loss ("everyone has trouble sleeping sometimes," "at least you don't have it as bad as a new parent," "you'll catch up on the weekend") — these land as dismissive even when meant to be reassuring, and the catch-up framing is also factually wrong (weekend sleep does not fully repay accumulated sleep debt); offering have-you-tried suggestions for apps, supplements, teas, weighted blankets, sleep trackers, or specific sleep medications that the person has almost certainly already considered or tried; and pressing them to make big life changes (quit the job, end the relationship, move, have children differently) in the moment, when their capacity to evaluate big decisions is already compressed by the sleep loss itself. Most people in the moderate or severe AIS band have spent considerable time thinking about their own sleep; the help that lands tends to be smaller, more concrete, and more present than the help that tries to fix the whole picture.

If you share a bed with them, you may have noticed things they cannot self-observe — loud snoring, witnessed pauses in breathing, gasping or choking during sleep, excessive limb movement, talking or shouting in sleep, sleepwalking. These are valuable observations because they point toward sleep apnea, periodic limb movement disorder, or parasomnias that the AIS does not screen for and which the person who took the AIS cannot detect from their own side of the bed. Mentioning what you have observed — calmly, factually, not as a diagnosis — to them and to their clinician shortens the path to a useful evaluation. Sleep apnea in particular is commonly missed in self-report because the affected person does not remember the breathing pauses; a bed partner's observation is often the single most important piece of clinical information for triggering an evaluation.

One situation calls for specific care: if they mention thoughts of suicide or self-harm — even passively, in the form of not wanting to be here, wishing they could just stop, or feeling that others would be better off without them — that is information to take seriously rather than redirect away from. Chronic severe insomnia is a known independent risk factor for suicide in research populations, particularly when paired with depression, hopelessness, or escalating substance use. The most useful response is to stay calm, ask gently whether they have any specific plans or means available, and help them connect with support today. Suggesting they call or text 988 (US Suicide and Crisis Lifeline, free and confidential) while you sit with them is a concrete next step. If they feel unsafe or you feel they may not stay safe, an emergency department visit is appropriate. Asking about suicidal thoughts does not put the idea in someone's head; it makes it possible for them to talk about something they may have been carrying alone. The AIS itself does not assess suicidal thoughts, so an elevated score on this scale is not a substitute for a safety conversation when those thoughts are present.

If the sleep picture has reached a point where it is meaningfully compressing daily life — work suffering, basic self-care slipping, drowsy-driving incidents, important relationships strained because the energy is missing — helping them schedule and keep the first clinical appointment is one of the most concrete, high-leverage things you can do. Getting to the room is often the hardest single step, particularly when the sleep deficit itself has compressed the bandwidth required to handle logistics.

Other screens you might also take

The AIS is a brief ICD-10-aligned screen for nonorganic insomnia, calibrated on a 30-day recall window. Several patterns commonly co-occur with elevated AIS scores, and a more targeted second screen often clarifies the larger picture. Combinations are common at moderate and severe AIS bands and call for different responses than insomnia alone. The first card below is descriptive — the ISI guide on Symptomatik is a fair-use reference page rather than a working self-check tool — and the other three are working tools that pair naturally with the AIS at moderate or severe bands.