1. Measuring Sleepiness Murray Johns
Epworth Sleep Centre, Epworth Hospital, Melbourne, Australia,
and
Sensory Neuroscience Laboratory, Swinburne University of
Technology, Melbourne, Australia

2. Sleepiness: Normal and Abnormal
Excessive daytime sleepiness (EDS)
involves an abnormal tendency to become
drowsy and fall asleep at times and under
circumstances when the intention and expectation
is to remain awake
Excessive daytime sleepiness (EDS) is a common
symptom among patients with sleep disorders
We cannot measure EDS without being able to
define and measure normal sleepiness

3. Measuring Normal and Abnormal Sleepiness
Over the past 30 years many different
methods have been proposed for measuring
sleepiness, objectively and subjectively
There is confusion about what each test
measures and how the results of one test are
related to those of a different test in the
same subjects

4. Sources of Confusion About Measurements of Sleepiness
Several different meanings of the word sleepiness
have been adopted by clinicians and researchers in
sleep medicine over the past 30 years, often
without a clear understanding of the differences:
state of drowsiness, sleep propensity, sleep drive
Some confusion has also arisen because one test (MSLT)
has been designated as a gold standard, inappropriately
in this writer’s opinion
Different tests measure sleepiness on different time-
scales

5. A: Objective Tests of Sleepiness - 1
Sleep Latency (how long it takes to fall asleep)
Multiple sleep latency test (MSLT)
Maintenance of Wakefulness Test (MWT)

6. A: Objective Tests of Sleepiness - 2
Objective measurements of variables that reflect the presence of the drowsy state
EEG frequency, power (alpha, theta-waves, microsleeps, etc)
Evoked potentials (auditory, visual)
Pupillometry
Eye movements (saccadic velocity, slow eye movements etc)
Eyelid movements (duration, amplitude-velocity ratios)
Performance tests (reaction time, tracking tasks,
divided attention tasks, key tapping/Osler Test, etc)
Palmar sweat-gland activity (electrical resistance of skin)

7. B: Subjective Tests of Sleepiness - 1
Reports of Feelings and symptoms at the time
(“Subjective sleepiness”)
Karolinska Sleepiness Scale (KSS)
Stanford Sleepiness Scale (SSS)
Visual analogue scale (VAS)
eg. of alertness/drowsiness

8. B: Subjective Tests of Sleepiness - 2
Retrospective Reports of Dozing Behaviour
Epworth Sleepiness Scale (ESS)
Sleep-wake Activity Inventory (SWAI)

9. Different Time-scales for Measuring Sleepiness
Instantaneous SP (ISP): a subject’s SP at a particular time, whatever the circumstances
Situational SP (SSP): a subject’s usual SP when in the same situation repeatedly
Average SP (ASP): a hypothetical construct based on the subject’s average SP in the different situations of his/her daily life

10. The Epworth Sleepiness Scale © MW Johns, 1990-97
How likely are you to doze off or fall asleep in the following situations, in contrast to feeling just tired? This refers to your usual way of life in recent times. Even if you have not done some of these things recently, try to work out how they would have affected you. Use the following scale to choose the most appropriate number for each situation.
0 = no chance of dozing
1 = slight chance of dozing
2 = moderate chance of dozing
3 = high chance of dozing
It is important that you answer each question as best you can 10

11. The Epworth Sleepiness Scale © MW Johns, 1990-97
Situation
Chance of Dozing
Sitting and reading
Watching TV
Sitting inactive in a public place (e.g. a theatre or a
meeting)
As a passenger in a car for an hour without a break
Lying down to rest in the afternoon when circumstances
permit
Sitting and talking to someone
Sitting quietly after a lunch without alcohol
In a car, while stopped for a few minutes in traffic 11

12. Analysis of Total ESS Scores in Four Different Groups of Subjects (n=990) With and Without Sleep Disorders
Group Total ESS scores No.Ss
sleep clinic patients with OSA, narcolepsy, etc /
industrial workers /
university students (Swinburne University) /
4th year medical students (Monash University) /
F(3, 988) = 38.67, p<0.0001

13. sitting quietly after lunch
ESS Item Scores in 990 Subjects With and Without Sleep Disorders
One-way ANOVA p<0.0001
sitting reading
watching TV
sitting public place
car passenger
lying down
sitting talking
sitting quietly after lunch
car stopped

14. Factorial ANOVA of ESS Item-scores from 990 Subjects in 4 groups
Factor p Effect-size
(partial eta- squared)
ESS item-number <
Group of subjects <
Group x ESS item-number =

15. Ranking ESS item-scores
When ESS item-scores were ranked from highest
to lowest within each subject, a consistent
pattern emerged
That pattern can be largely explained in terms of
the somnificity of the different activities described
in the ESS

16. Somnificity
The general characteristic of a posture, activity and situation that reflects its capacity to facilitate sleep-onset in the majority of subjects
Johns MW, 2002

17. There were 6 significantly different levels of somnificity
ESS Item No.
Situation
Somnificity Mean SD 5
Lying down to rest in the afternoon when circumstances permit 2
Watching TV 1
Sitting & reading 4
As a passenger in a car for an hour without a break 3 7
Sitting, inactive in a public place
Sitting quietly after a lunch without alcohol 6 8
Sitting & talking to someone
In a car, while stopped for a few minutes in the traffic
The somnificity of ESS situations in 990 subjects, with and without sleep disorders
There were 6 significantly different levels of somnificity

18. Somnificity and EDS
Somnificity is a characteristic of particular postures, mental and physical activities, and environmental situations. It is not a characteristic of people or their sleep disorders
Differences of somnificity are presumably mediated by the sensory nervous system which then influences the wake-drive
Activities with a high somnificity induce less wake-drive and therefore facilitate sleep onset

19. Sleepiness in Different Situations: Subjective Measurements by ESS Item-scores
The mean of 28 Spearman correlation coefficients
between ESS item-scores in 990 subjects was 0.39
(range = 0.33 – 0.57, all p<0.001)
Sleepiness measured in one situation shared only
about 15% of variance with that measured in a
different situation

20. Sleepiness in Different Situations: Objective Measurements by Sleep Latencies
The mean sleep latency measured in the MSLT gives an objective
measure of one particular SSP, the MSLT-SSP
The mean sleep latency measured in the MWT gives an objective
measure of a different SSP, the MWT-SSP
When Sangal et al (1992) performed MSLTs and MWTs in the same
patients on the same day, their mean sleep latencies were only
moderately correlated, r=0.41, n=258 , p<0.001
For comparison, the mean sleep latency in the MSLT has a test-
retest reliability of about 0.7 when the same subjects are retested on
different days.

21. Sleep latencies in MSLT’s and MWT’s on the same day
MWT MEAN (MIN)
Sleep latencies in MSLT’s and MWT’s on the same day
in 258 subjects (r = 0.41) (Sangal RB et al, 1992) M S L T E A N ( I ) 5 10 15 20 30 40

22. Sleepiness in Different Situations
Even allowing for differences in the somnificity of different activities and situations, we cannot always predict accurately how sleepy a person will be in one situation from measurements of their sleepiness in a different situation
This appears to be equally true of both objective and subjective methods for measuring sleepiness
This is presumably due to subject x situation-specific interactions, as demonstrated by the ANOVA of ESS item-scores

23. Initial Conclusions
With EDS, all situational sleep propensities tend to
be increased as a general characteristic of the
person, eg. SL in patients with OSA, narcolepsy, etc.
However, each situational sleep propensity also reflects
the general effects of the person’s posture and activity
at the time (somnificity), as well as subject x
situation-specific effects
It is not possible to measure a person’s sleepiness
accurately without reference to the situation in which
it is measured

24. The total ESS score gives a measure of the
Total ESS scores
The total ESS score gives a measure of the
subject’s average sleep propensity in eight
situations that have different somnificities.
The total ESS score is a proxy measure for the
subject’s average sleepiness in daily life (ASP)
which cannot be directly measured at present

25. Control of sleep and wakefulness
Currently accepted models of sleep and wakefulness
are based on
Process – C (time of day, circadian rhythm, SCN)
Process – S (“homeostatic”, duration of wakefulness)
Process – W (“sleep inertia”)

26. Need for a new model of sleep and wakefulness
I have suggested that all inputs to the sensory nervous
system, both somatosensory and exterosensory, continuously
affect the wake-drive that opposes the sleep-drive in the control
of sleep and wakefulness
Hypothetically, the wake-drive has two components
primary wake-drive (Process - C)
from spontaneous neuronal activity in the SCN etc.
secondary wake-drive (Process - A)
from the effects of all sensory inputs to the central nervous system,
integrated (? by the thalamus) over periods of a few minutes
I suggest that we need a new model of sleep and wakefulness that includes Process - A

27. Which Test of Sleepiness to use?
There is no one test of sleepiness that fulfills all our requirements under different circumstances.
We do not have a gold standard.
Many currently used tests, including the MSLT and MWT, measure a single SSP. They do not address the issue of the general effects, or the subject x situation-specific effects that different postures and activities have on sleepiness.
The total ESS score gives an estimate of the subject’s ASP, but it has the limitations of all subjective tests.

28. Tests for Measuring ISP
Objective measurements at some particular times
Individual naps in the MSLT,MWT, Osler Test, etc.
Performance tests (PVT, tracking, etc)
Pupillometry
Objective measurements made continuously
EEG, EOG, Video camera images (PERCLOS)
Infrared reflectance oculography (Optalert)
Subjective measurements at some particular times
KSS, SSS, VAS (alert – drowsy)

29. Tests for Measuring Particular SSPs
Objectively
Mean sleep latency in the MSLT or MWT
Repeated performance tests
Subjectively
ESS item-scores
Specific questions
(eg about drowsiness while driving)

30. Tests for Measuring ASP
Objectively
There are no tests at present
Subjectively
Total ESS scores (0 - 24)

31. Conclusions - 1 A person’s sleepiness cannot be measured
accurately without reference to what they are doing
at the time – their posture, physical and mental
activity, and environmental situation
A person’s average level of daytime sleepiness in
daily life (ASP) is a hypothetical construct for which
we currently have only proxy methods of
assessment, such as total ESS scores

32. Conclusions - 2 It may be unwise to rely on any one test of
sleepiness eg. when determining the fitness of
a patient with OSA to have a driver’s license
Existing models of the control of sleep and
wakefulness do not adequately explain
sleepiness in real-life situations
They should be expanded to include the role of the sensory nervous system (Process-A)

33. Thank you for your attention