Blog overview
8 min readSanoLabs Editorial

MOS Sleep Scale: what the 12 questions ask, how the score works and what it can show

How the 12-question MOS Sleep Scale is scored, what the 0-100 Sleep Problems Index means, why it has no cut-off and how Sam tracks it each month.

sleepmos-sleep-scalesleep-problems-indexquestionnairesself-tracking
On this page

The MOS Sleep Scale asks 12 questions about how you slept over the past four weeks and turns the answers into scores from 0 to 100. You can fill it in on paper or in Sam's monthly sleep check-in, where your scores can sit next to the average sleep score Sam builds from your nights in Apple Health. This guide explains the questions, the scoring and what a score can and cannot tell you.

Unlike a sleep diary, which you fill in every morning, the MOS Sleep Scale looks back over a whole month. That makes it a tool for comparing months, not for judging one bad night.

What does the MOS Sleep Scale ask?

The scale comes from the Medical Outcomes Study, a US study run by RAND. Hays and Stewart described it in 1992. It was developed with patients who had chronic illnesses. The scoring manual reports data from 3,445 people in that study.

All 12 questions refer to the past four weeks. We describe them in our own words, because the wording belongs to RAND:

  • One question asks how long you usually took to fall asleep, in five steps from 0-15 minutes to more than 60 minutes.
  • One question asks how many hours you slept on an average night.
  • Ten questions ask how often something happened, on six steps from "all of the time" to "none of the time".

The ten frequency questions cover restless sleep, waking up rested, and waking short of breath or with a headache. They also ask about feeling drowsy by day, trouble falling asleep and waking at night without getting back to sleep. The last four cover trouble staying awake by day, snoring, naps and getting the amount of sleep you needed.

How is the MOS Sleep Scale scored?

The RAND scoring manual by Spritzer and Hays (2003) uses two steps. First, each answer is recoded to a value from 0 to 100. Then the answers that belong to one scale are averaged.

For the frequency questions, the six answers become 0, 20, 40, 60, 80 and 100. For the time to fall asleep, the five steps become 0, 25, 50, 75 and 100. A higher value always means more of what the scale is named after. A scale can be scored as soon as one of its questions is answered.

Sleep disturbance

Four questions: time to fall asleep, restless sleep, trouble falling asleep, and waking without getting back to sleep. Higher means more disturbance.

Sleep adequacy

Two questions: waking up rested and getting the sleep you needed. Higher means more adequate sleep.

Somnolence

Three questions: drowsiness by day, trouble staying awake and naps. Higher means more daytime sleepiness.

Snoring

One question. Higher means more frequent snoring.

Short of breath or headache

One question about waking that way. Higher means it happened more often.

Sleep quantity

Your average hours per night, kept as a plain number. The manual also marks 7 or 8 hours as "optimal sleep".

Two summary scores combine several of these. Sleep Problems Index II averages nine questions: all except hours slept, snoring and naps. Sleep Problems Index I uses six of them. Both run from 0 to 100, and higher means more sleep problems. In both, the two adequacy questions are turned around, so too little sleep counts as a problem.

An example with invented answers: your nine recoded values are 25, 40, 60, 20, 40, 60, 40, 20 and 60. Together they make 365. Divided by nine, that is about 41.

The wellness score calculator on this site uses the shorter Index I. Its sleep number is therefore not the same as the one in Sam.

How do you read your MOS Sleep Scale score?

The RAND manual gives no cut-off that marks a score as normal or too high. Instead, it reports averages from the 3,445 people in the Medical Outcomes Study:

  • Sleep Problems Index II: average 29.2, standard deviation 18.0.
  • Sleep disturbance: average 29.2, standard deviation 23.4.
  • Sleep adequacy: average 60.7, standard deviation 25.4.
  • Somnolence: average 26.4, standard deviation 19.8.

These people lived with chronic illnesses. Their averages are a reference point, not a normal range. A score above them does not mean something is wrong. A score below them does not mean all is well.

The comparison that helps most is with yourself. Say your index was 41 in March and 28 in May. Then your May answers describe fewer sleep problems than your March answers. What that means for you is a question for your doctor.

The Insomnia Severity Index is a different questionnaire with its own score bands. Its numbers cannot be compared with MOS scores.

What can the MOS Sleep Scale not tell you?

The scale measures how you rate your sleep. It uses no sensor and no sleep lab. It cannot say what causes a problem.

One scale is less steady than the others. Somnolence reached an internal consistency of 0.75 in the original study group. In 1,011 US adults it was 0.63 (Hays et al., 2005). In the German version it was 0.61 (Viala-Danten et al., 2008). The 2008 study used 0.70 as its threshold. So read a single somnolence score with extra care.

Snoring and waking short of breath or with a headache are single questions. The scale counts how often they happened, not why. The NHS names loud snoring, gasping, snorting or choking noises, waking up a lot and a headache on waking as possible signs of sleep apnoea. It advises seeing a GP if you have them and suggests bringing someone who has seen your symptoms.

If your Apple Watch has shown a sleep apnea notification, the guide on what to do after that notification walks you through the next steps. Days when you wake with a headache can go into Sam's symptom diary in a few taps.

Why compare months instead of single scores?

Each score covers four weeks. A cold, a deadline or a holiday in that window can move it without a lasting change. Several scores in a row show whether a change holds.

In a clinical drug trial with 173 people who had nerve pain, the scales changed measurably over the course of the trial (Hays et al., 2005). So the scale can move when sleep changes. That does not make any number a target for you.

Questions worth asking about your own series:

  • Does it move in one direction over several months?
  • Does a jump line up with something you remember?
  • Does a high score stay high after that event has passed?

Bring those observations to your doctor.

How does Sam use the MOS Sleep Scale?

In Sam, the questionnaire belongs to the Sleep program, one of Sam's lifestyle programs. You set a nightly sleep score goal. Sam asks the MOS Sleep Scale when you start and then every 30 days. When a check-in is due, Sam reminds you once, and again three days later if it is still open. You can take an extra check-in in between without moving the monthly rhythm.

Sam asks all 12 questions with the same answer steps. The English wording closely follows RAND's. The German wording is Sam's own translation. You pick your hours slept on a slider, in whole hours.

From your answers, Sam calculates one number: Sleep Problems Index II, from 0 to 100. Lower means fewer problems. The separate scales, such as snoring or somnolence, are not shown as scores. Next to the number, Sam shows one of four labels: few, some, frequent or very frequent sleep problems reported. The steps sit at 25, 50 and 75. These labels are Sam's own words to describe your answers, not published cut-offs.

Unless they all look strong, your answers also point to one area: falling asleep, staying asleep or feeling awake during the day. Sam suggests one habit for it until the next check-in.

Under "Your Surveys", every score sits on a chart. When enough nights are recorded, a second line shows your average sleep score from the 30 days before each check-in. From there you can export a PDF with up to your 12 latest scores, Sam's labels, the change between check-ins and your latest answers. The monthly report adds a page with your latest score and the change since your first check-in. The monthly report guide explains that page.

The Exercise program works the same way with a different questionnaire, the PAVS.

Keep your MOS Sleep Scale in Sam and bring it as a PDF

Your monthly sleep scores stay in one place, next to the nights behind them, ready for your next appointment.

  • Instead of a paper form: Sam asks the 12 questions every 30 days, calculates the index and reminds you when the next check-in is due.
  • Instead of Apple Health alone: once enough nights are recorded, your scores sit on one chart with your average sleep score from the 30 days before each check-in.
  • Instead of memory at the appointment: the PDF lists your scores, the change between check-ins and your latest answers.

Sam records your answers and shows them over time; what they mean is for you and your doctor to discuss. More on sleep tracking in Sam.

Try Sam for free

Sam is a wellness companion, not a medical device. The app does not diagnose, treat, cure or prevent any condition or health state. Always consult qualified professionals for medical decisions.

Sources

Updated: October 2026

  • Spritzer KL, Hays RD: MOS Sleep Scale: A Manual for Use and Scoring, Version 1.0. Los Angeles, November 2003 (scoring rules, scales, averages of the MOS sample, n = 3,445)
  • RAND: MOS Sleep Scale survey form. Copyright 1986 RAND. Hays RD, Stewart AL: Sleep measures. In: Stewart AL, Ware JE (eds.): Measuring functioning and well-being: The Medical Outcomes Study approach, pp. 235-259. Duke University Press, Durham NC, 1992.
  • Hays RD, Martin SA, Sesti AM, Spritzer KL: Psychometric properties of the Medical Outcomes Study Sleep measure. Sleep Medicine 2005;6(1):41-44, doi:10.1016/j.sleep.2004.07.006
  • Viala-Danten M, Martin S, Guillemin I, Hays RD: Evaluation of the reliability and validity of the Medical Outcomes Study sleep scale in patients with painful diabetic peripheral neuropathy during an international clinical trial. Health and Quality of Life Outcomes 2008;6:113, PMC2637242
  • NHS: Sleep apnoea (last reviewed 11 May 2026)

Frequently Asked Questions

What does the MOS Sleep Scale measure?+

It asks 12 questions about your sleep over the past four weeks: how long you take to fall asleep, how many hours you sleep, and how often you had restless sleep, daytime sleepiness, snoring and similar problems. It measures how you rate your own sleep. It is not a diagnostic test.

What is the Sleep Problems Index?+

It is a summary score from the RAND scoring manual. Index II averages 9 of the 12 questions after recoding them to 0 to 100. It leaves out hours slept, snoring and naps. Index I uses 6 questions. In both, a higher score means more sleep problems. Sam tracks Index II.

Is there a cut-off for a bad MOS Sleep Scale score?+

The RAND scoring manual names no cut-off for individual scores. It reports averages from the original study group, and those are not a normal range. The most useful comparison is with your own earlier scores. Your doctor can put them in context.

Is there a German version of the MOS Sleep Scale?+

Yes. A German version was tested during an international drug trial in people with painful diabetic nerve damage (Viala-Danten et al., 2008). Its daytime sleepiness scale reached an internal consistency of 0.61, below the 0.70 the study used as its threshold. The German questions in Sam are Sam's own translation.

How often should I fill in the MOS Sleep Scale?+

The scoring manual sets no schedule. Each answer covers four weeks, so once a month gives you one score per month without overlap. In Sam, a check-in is due every 30 days, and you can add an extra one in between.

Can the MOS Sleep Scale tell me if I have insomnia or sleep apnoea?+

No. It records how you rate your sleep, and Sam records and shows those answers over time. Whether a sleep disorder is behind them is for your doctor to assess.

Does Sam still use the Insomnia Severity Index (ISI)?+

No. Sam's Sleep program uses the MOS Sleep Scale. The ISI runs on a different 0 to 28 scale, and there is no accepted way to convert one score into the other. Sam does not show older ISI results next to the MOS scores.