Sleeplessness in Society:

A Critical Appraisal
Jim Horne
 British Medical Journal 1894:
editorial –Sept 29th

“The subject of sleeplessness is once more under public

discussion. The hurry and excitement of modern life is
held to be responsible for much of the insomnia of
which we hear; and most of the articles and letters are
full of good advice to live more quietly and of platitudes
concerning the harmfulness of rush and worry. The pity
of it is that so many people are unable to follow this
good advice and are obliged to lead a life of anxiety and
high tension.”
 Insomnia - ‘Medicalising’
Sleeplessness
• By 1890s new drugs effective in treating sleeplessness, now called
‘insomnia’.
• BMA campaigned against opiates, cannabis (and cocaine) from
tonics and non-prescription medicines.
• Physicians responsible for prescribing; apothecaries had to cede to
the professional pharmacists.
• These new drugs (paraldehyde, chloralamide, chloral hydrate, sulphonal, potassium
bromide) provided ‘real cures’.
• Insomnia increasingly attributed to ‘brain dysfunction’: little regard
for psychological cause.
• Patients diagnosed as ‘insomniacs’ rather than belittled for ‘lacking
moral fibre’, on bizarre remedies or addicted to opium.
 Bradbury JB ‘Some points connected with sleep
sleeplessness and hypnotics’ (1899 - BMJ July 15th
p134).
the reason [for insomnia] is that the cerebral
cells have assumed an irritable condition and
it is necessary to depress their activity to bring
them back to a more natural state…it is here
that hypnotics are of such great value”.
Sawyer J ‘Causes and Cure of Insomnia’, (1900 BMJ
Dec 1 – p551),

‘occurs mostly in persons who are members of what are

known as the upper and middle classes… mostly in
persons of high mental endowment and of neurotic
temperament’

(He probably never saw the working class patient, who

would not have afforded his fees !)
 ‘Sleep Debt’ in Today’s Society

• Many claims that we sleep less today.

• Google ‘sleep debt’ - many 1000s of hits
• Seemingly, 100y ago we slept 9h a night,
whereas today this is only 7h
• Also popular belief that we ‘need 8h sleep’
• Thus we are chronically sleep deprived
• This ‘Societal Insomnia’ apparently leads
to obesity and related disorders
 Are We Chronically Sleep
Deprived/Sleepy?
• Is typical 6.5-7.5h sleep/night insufficient ?
• Is chronic ‘sleep debt’ endemic?
• Are many healthy adults unaware of their
apparent chronic sleepiness ? Is it of real
concern to society ?
• Tiredness not the same as sleepiness
• Insomnia as a ‘disorder of wakefulness’ - not
necessarily one of sleep loss
 No solid evidence that:

i) we used to sleep for longer than today

ii) that we all need 8h sleep
• Claims that 9h sleep was the norm comes from
study by Terman & Hocking (1913) based on 9-17 y
schoolchildren, not adults
• Even these authors conceded that 9h was longer
than found by other contemporary studies
 Misinterpretations
Claims (Bliwise et al 1996) that in the 1930s, we used to
sleep for longer, are based not on sleep duration
itself, but on historic records of people’s
responses as to whether they felt: ‘rested’, ‘had
trouble functioning’, ‘had stamina’, ‘were
energetic’.
Such vague terms only allude to excessive
sleepiness and sleep loss.
 Average sleep duration has not
changed in UK
• UK population studies over the last 40 y show
average daily sleep for adults is 7-7.5h (Tune 1969;
McGhie & Russell 1962; Palmer et al., 1980; Reyner & Horne 1995;
Groeger et al 2004)

• More acceptable now to admit to being

tired/sleepy
• Fears over ‘sleep debt‘ increases worries and
unfounded health concerns - more demand for
hypnotics.
 Adult sleep duration over last 40y – only
small changes in other countries
• UK: last 40y consistent 7-7.5h (Tune 1969; McGhie & Russell 1962; Palmer et al.,
1980; Reyner & Horne 1995; Groeger et al 2004)

• USA : 1.7% more adults sleep <6h/day since 1975 (Knutson et al 2010)

• Scandinavia: 5.5min less daily sleep per 10y since 1972 - no change
in extremes (Reven et al 2010)

• Swedish women over the last 36y - 15min less sleep (Kronholm et al 2007)

• Overall review – 12 studies - no consistent decrease in adults over

40y (Bin et al 2012).
 Distribution of Today’s Adult Sleep

Sleep Length Distribution - 2000 UK Adults

(Groeger et al 2004)

30

25
Percentage

20
MEN
15
WOMEN
10

0
1 to 2 to 3 to 4 to 5 to 6 to 7 to 8 to 9 to 10 to 11 to 12 to
2 3 4 5 6 7 8 9 10 11 12 13

Sleep Duration h
‘Hypnogram’
 ‘DEEP SLEEP’
Delta EEG activity – ‘stage 4’
Only part of lost sleep is regained after
sleep loss: mainly ‘deep sleep’

Before After 72h SD After 10h Sleep

‘Hypnogram’
Dreaming
• Keeps brain preoccupied
- distracted from waking
• ‘Cinema of the mind’
• Concentrated at end of
sleep when sleep less
‘necessary’ ?
• Close to wakefulness
• ‘Screensaver’
 ‘Short sleep’ - often overlooked:

• Sleep quality as important

• Subjective estimates – often only 1 question
• Confuse with time in bed ?
• Daytime napping ?
• Sleep is litmus test of physical/mental health
• Laboratory tests of sleepiness (Reaction Time, &
Multiple Sleep Latency Test) ‘supersensitive’ ?
• Correlations – obesity and short sleep just
symptom of common underlying cause?
 ‘Insufficient Sleep’ - Population Based
Study of 12,423 People (Hublin et al, 2001)
• 16% men, 24% women claimed they needed >1 h more daily
sleep
• Attributed to ‘insomnia’, excessive work hours, daytime
‘tiredness’
• 44% had same complaint 9 years later
– why – it is really a problem ?
– what has happened to all this lost sleep
– Huge (365 x 9 = 3285 h) accumulated sleep debt

IS LACK OF SLEEP THE REAL ISSUE ?

 Sleep Dissatisfaction:
sign of poor “well-being” - not sleep need ?
(Girardin et al -2000)

• 273 people – home ‘actimetry’

• Subjective ratings of: health, sleep, & ‘well-being’
• Sleep satisfaction not related to sleep
characteristics, but to ‘well-being’
• Those dissatisfied with sleep not made happier
just by focussing on improved sleep
Key Sign of Insufficient sleep:
excessive daytime sleepiness
SLEEPINESS -24h (Circadian) Rhythm
Daytime Alertness After: Good, Poor & Bad Night's Sleep

1 Very Alert
2
3 Alert Good Night
4
5 Neither
6
Poor Night

7 Sleepy
8
9
Very Sleepy
10 Bad Night
02 00
03 00
04 00
05 00
06 00
07 00
08 00
09 00
10 00
11 00
12 00
13 00
14 00
15 00
16 00
17 00
18 00
19 00
20 00
21 00
22 00
23 00
0
01 0

:0
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
Bad Night
Time (24h)
 ‘Tiredness’ is not same as
‘Sleepiness’
TIRED: tense, fatigue, bored dejected, sad,

SLEEPY: likely to fall asleep

‘MSWord Thesaurus’ – 2 x more synonyms for ‘tired’

compared with ‘sleepiness’

Many people with insomnia: tired but not sleepy

More acceptable today: to admit to being tired/sleepy

 Stanford Sleepiness Scale
ambiguous – ‘non-sleep’ terms (in red)
can give wrong impression of sleepiness

1. Feeling active, vital, alert, wide awake

2. Functioning at high level but not peak, able to
concentrate
3. Relaxed, awake but not fully alert
4. A little foggy, let down
5. Foggy, beginning to lose track, difficult to stay
awake
6. Sleepy, prefer to lie down, woozy
7. Almost in reverie, cannot stay awake, sleep onset
imminent.
Epidemiology
 Sleep duration & mortality:
a controversial study, but probably true
 Sleep Duration and Mortality
‘meta-analysis’ – Gallicchio & Kalesan 2009
• 23 studies of short sleep; 17 of long sleep
• Sleep length – self reports - only one Question
• Categories: ‘Short’ <7h, normal 7-8h, ‘long’ >9h sleep
• ‘Short’ contains <5h sleepers where there is risk
• Short sleep :(RR) all mortality = 1.10 (sig)
• Long sleep : all mortality = 1.23 (sig)
• Death accountable from short sleep = <2% (huge n)
• Cannot claim that 6-7h sleep is risky
• But AUs still claim: <7h sleep increase mortality risk
 Short sleep correlates with obesity:
too many assumptions ?
• Many epidemiological studies find association between
short sleep and obesity in adults & children.
• Also are claims that adults & children sleep fewer hours,
today, than several generations ago.
• Thus, inadequate sleep is seen to contribute to the
‘obesity epidemic’ in adults & children
• Acute sleep restriction lab studies seem to back this up
• BUT – these points assume that:
– i) correlation = causation
– ii) short sleep & obesity = effects of common cause ?
– iii) statistical significance = clinical significance
– iv) cant argue that short sleepers are sleep deprived
 Be wary of statistics !
• Stat sig effects may be too small to be
of real clinical, physiological or
psychological significance !
• Look at actual data
• Check findings behind :
– ‘odds ratios’ and ‘relative risks’
– Linear correlations
– Huge participant numbers
• Look for absolute risks
 Sleep Duration and Morbidity
adult hypertension - Gottlieb et al. 2006

Sleep outside 7-8h had sig (P<0.001) risk for hypertension n=5910

Sleep length (hours) Relative risk Systolic BP (mmHg)

≤6 1.7 132
6–7 1.2 130
7–8 (reference) 1.0 128
8–9 1.2 130
≥9h 1.3 133

• BP differences clinically small – especially 6-7h vs 7-8h

• Hypertension usually = >140mmHg – How many really hypert. ?
• 69% of those sleeping <6h had symptoms of insomnia (stress?)
• Would hypnotic treatment of insomnia improve BP?

.
 SHORT SLEEP & OBESITY
ADULTS: short sleep = slow, small (stat sig) weight
gains over 1y
Wanatabe et al (2010) n=31,447 men – 1y follow up - *sig

Sleep dur. <5h 6 to <6h 6 to <7h 7 to <8h 8 to<9h >9h

% of n 2.5% 15% 48% 27% 5% 0.5%
BMI 24.4* 24.1 23.8 23.6 23.6 23.5
BMI gain 1y 0.07* 0.05 0.03 0 -0.04 0.07*
Approx Kg 0.2 0.15 0.09 0 0.1 0.2 [4g/wk]

*200g fat burnt off by c5-6 h brisk walking

Short sleep associated with increased BMI
(n=1024 –multiple sampling) Taheri S et al 2004 PLoS1
NB - ALL BMI averages in ‘obese’ category
 ADULTS: 6-7h sleepers no greater risk
for obesity, diabetes, etc
• Claims of higher obesity/morbidity/ mortality in <7
h sleepers, includes <5h (is some risk)
• BUT: only 5–8% of adult population sleep <5h
• Most 6h good sleepers are not obese - only 3%
are diabetic1-3
• Few obese adults/children are short sleepers
• Few short sleepers obese3
1. Gangwisch JE et al. Sleep. 2005;28:1289–96.
2. Ayas NT et al. Diabetes Care 2003;26:380–4.
3. Horne JA Obesity Reviews 2011
 Metabolic (pre-diabetic’) syndrome & glucose
intolerance: short sleep unlikely cause*
• Generalisation from acute 4h/day sleep restriction studies, no
stress controls - effects only when compared with recovery sleep,
not normal baseline sleep

• 4h sleep intolerably short & unsustainable = severe sleepiness

• Some possible at risk only when chronic sleep atypically short

(5h/day) = 5–8% of adults, where there are other causes

*Horne (2011,2016)
 Obesity : more sleep Vs more exercise

• ‘Best’ evidence is that less than 6 h sleepers may add

<1.5kg fat/year due to ‘lost sleep’
• Want to lose 1.5kg over a year ? Then don’t increase
sleep by 1-2h/day
• 15 min min daily brisk walking = same effect
• ‘Dieting’ even better !
• Few obese adults/children short sleepers; few short
sleeping adults/children are obese or suffer obesity-
related disorders
 Exercise vs More Sleep

15 min brisk exercise is a far better

treatment for metabolic syndrome than
>1h extra sleep
 Controversy !
Is being ‘overweight’ (ie.BMI 25-29) unheathy ?

• BMI of 25 is an arbitrary 30year old cut-off

for ‘overweight’
• Flegal et al 2013 JAMA – BMI of 25-30
“overweight ...associated with significantly
lower all-cause mortality (created much debate !)
• Is BMI <30 really ‘unhealthy’?
 Hypnotics to treat obesity, metabolic
syndrome & type 2 diabetes ?

• No evidence that hypnotics cause weight loss or

useful in treating diabetes
• Unlikely to increase sleep length by more than 20
min
• 20 min brisk walking far better for weight
management
How much should we eat & drink?

• Enough to stop all feelings of hunger &

thirst ?
• Can eat & drink without being hungry &
thirsty - without physiological need
• Can eat when bored, drink for pleasure

Same applies to sleep ?

ABILITY TO TAKE MORE SLEEP

• Around 7 hours daily sleep is about right for

most people with few symptoms of excessive
daytime sleepiness.

• So, why can we easily take an extra 1 – 2 h

sleep daily (e.g. 9h/24h) ?
 Acute sleep extension in humans:
confinement to bed

• Aserinsky (1969) 20h sleep during 30h bed-rest

• Aserinsky (1973) 32h sleep during 54h bed-rest
• Campbell (1984) 28h sleep during 60h bed-rest
• Gagnon et al (1985) 12h sleep during 15h bed-rest

• Ss recovering from ‘sleep debt’ or just bored ?

• Can we sometimes sleep with minimal sleepiness if
bored and lying on a bed?
 More Sleep – Animal Studies

• Cats - sleep longer with isolation & overfeeding

• Baboons - sleep less in wild than in lab.
• Cows - sleep more in stable than in field
• Ponies - sleep 30% more in stable than in field
• Sloths in zoos sleep >4h longer than in wild
• This extra sleep excess of need - “luxury”/“optional” sleep
• Why can’t we do the same ?
 Not long ago in human evolution

Wherever seasonal daylight changes

+ our poor night vision

= gradual seasonal changes to sleep duration

= adaptability of sleep duration

 Do we really want more sleep?
strength of desire for more sleep – 11,000 adults
(Anderson & Horne 2008 )

• Questionnaire - perceived sleep deficits, no leading questions

• SPT, ESS, desired SPT, ESS, choice of daytime activities for ‘free
hour’ (including sleep), ‘stressful lifestyle ?’

FINDINGS
• Half desired more sleep.
• ‘Deficit’ unrelated to ESS for any age group.
• Irrespective of deficit, few opted for ‘sleep’
• Stressful lifestyle’ related to deficit
• More sleep synonymous with more ‘time out’ ?
• Extra sleep not the only anodyne for sleep deficit.
Should we strive for zero tolerance of
daytime sleepiness ?
(even only mid-afternoon)
or

Accept that some sleepiness (eg. mid-

afternoon) is natural ?
and
Sleep can also be for enjoyment ?!
 ‘Sleep Debt’ in Adults - Summary
• 6 ½ -7 ½h daily sleep is OK for most adults – little real sleepiness

• Critical sign of insufficient sleep = excessive daytime sleepiness

• Has not changed historically

• Can sleep extra 1–2 h daily: sign of sleep debt, or pleasure/boredom ?

• Sleep with minimal sleepiness if bored and lying on a bed?

• Laboratory tests of sleepiness (RT & MSLT) ‘supersensitive’ ?

• Eat and drink more than physiological needs – why not sleep ?

• “Tiredness” confused with “sleepiness”

• Today:socially acceptable to admit to being tired/sleepy

 Children’s Sleep
Plus ca change
“evil of insufficient sleep in children is
widespread ... “a loss equivalent to one
night in four in the youngest children, and
one night in five among those of
intermediate ages”.
Crichton-Brown J, 1908. Nature, 2036: 28
 Terman & Hocking 1913
Widely cited survey of children’s sleep

“Sleep is but one of the many needs of children,

and it is foolish to make it the scapegoat for all
kinds of physical and mental evils as hygienists
have so often done. It is possible that the
quantity of sleep is less important than its
quality, and that when disturbances of the latter
occur they are more likely to be the effect of ill-
health than its cause..... sleep cannot be
accurately measured in units of time alone....”
 CHILDREN -little change in sleep
over 100 years
• 51 studies of secular trends in children’s sleep, only
“limited scientific basis” 1 for sleep reduction, “repeated
references to the same sources of evidence, reference
to secondary sources, mis-referencing, and a failure to
cite contrary evidence”.
• Linear regression2 – maybe 45 sec less sleep per year
over last 100 years
• 1913 study3 reported av. 11.2h sleep/24h for 6-8y year
olds – same as today4
•
1-Matricciani et al 2011; 2-Matricciani et al 2012;
3-Terman & Hocking 1913; 4-Carter et al 2011
 Children - short sleep ‘doubles’
obesity
• Locard et al (1992)
7.7% of ≤10h sleepers obese vs 3.6% of ≥10h.
BUT 92.3% ≤10h not obese vs 96.4% for ≤10h sleepers

• Von Kries et al (2002)

n=6862, 5-6y. 14.5% ≤10h sleepers overweight/obese vs 10% for
11h sleepers: ‘duration of sleep has a clear effect on the children’s body
composition’ (greater influence was parental obesity)

• Avon prospective study (Reilly et al 2005)

14,000 UK children from birth. 8 factors @ 3y linked to obesity at 8y:
parental obesity, TV ≥8h/week, short (≤10.5) sleep.
BUT @ 8y 89.7% of s.slprs normal BMIs vs 93.2% for ≥12h slprs
Obese children ‘shorter sleepers’
• Lumeng et al (2007)
n=785. 9y to 12y. Sig sleep diffs bet overweight vs normal = 8.78h
and 9.02h = ONLY 14.4min, inc. 7min later bed-time

• Agras et al (2004)
9y prospective study - 38 of 150 children overweight; sig less sleep.
BUT 25min less in day and 5 min less at night. Other factors: parent
overweight, child temper-ament, low parental concern over child
weight, food tantrums
•
 Obesity & sleep - longitudinal studies
Children: newer findings – stat. vs clin. sig.
• 1h less sleep/night from 3y to 5y - stat.sig. increase in BMI of 0.56 at
7y (0.48kg fat). But took years – 100s hours of apparently ‘less sleep’
Carter et al 2011 BMJ 242: d2712

• 3800 children from 0y to 7y - sleep duration didn’t predict obesity at

any period. Hiscock et al 2011. Arch Dis Child 96(8):735-9

• 8 sig. early life risk factors for obesity, inc. short sleep (OR=1.45).
10% short sleepers obese vs 7% for normal sleepers

• But - 90% of short (<10.5h) sleepers not obese, vs 93% ‘non-

obesity’ for longer (11-12h) sleepers - only a 3% absolute difference.
Reilly et al 2005 BMJ 330: 1357-1363
 Children - short sleep & weight gain
Recent metanalysis of longitudinal studies
Magee & Hale 2012 (Sleep Med Rev)

• Shorter sleep can predict only some weight

gain in children (but inconsistent for adults)
 Conclusions - children
• Where there is chronic sleep loss – probably
symptom of other problems not rectified just by
increasing sleep duration.

• Rather than extending sleep as a putative way of

reducing obesity, a short period of moderate
daily exercise is far better (and improved diet)
 FINAL CONCLUSIONS
Short sleep is a minor cause of obesity &
related morbidity in adults & children

•Sleep confounded by: ‘time in bed’ & naps

•Wide categorisation – eg <7h vs >7h sleep
•5h sleepers= 8% pop
•Acute 4h sleep restriction = impossible in real life
•Stat sig not same as clin sig
•Fat increase v slow<1.5kg/year in adults=550h ‘lost sleep’
•No solid evidence that more sleep = reduces fat
•1h/week brisk walking instead
•Poor sleep = poor health = obesity.
•More sleep = unwarranted use of hypnotics
•KIDS-short sleep=problems not overcome by sleep
•More subtle effects of REM sleep ?
Thanks For Listening
jimhorne.co.uk

‘Sleeplessness’
2016 palgrave- macmillan
THAT’S ALL FOLKS !

THANKS
EXTRA SLIDES
 Chronic sleep extensions to 9h/night
(upto 3 weeks)

• Main outcome = elimination of afternoon dip

• Moderate pm dip is ‘natural ?
• 15 min pm nap = >1h night sleep extension
• Mean MSLT improvement usually <4 min
• Morning alertness on arising not improved
• RT improved by >20msec
• Are they really making up for sleep debt ?
• Is extended sleep “supersaturated” sleep – and it eliminates dip
• ‘Sleep debt’ not as large as >1h per night - more easily rectified
by this short nap ?