who sleep disturbance lares

8/8/2019 WHO Sleep Disturbance Lares

1/22

WHO LARES

Final reportNoise effects and morbidity

By:Dr Hildegard Niemann / Dr Christian Maschke

Int erdisciplinary research netw ork N oise and Health


2/22


3/22

EUR/04/5047477page 1

CONTENTS

Introduction..................................................................................................2

Data processing ............................................................................................ 2 Logistic regression ....................................................................................................... 2

Control variables .......................................................................................................... 3

Effect of noise annoyance .............................................................................. 3 Adults ......................................................................................................................... 5

Children ...................................................................................................................... 8

Elderly....................................................................................................................... 11

Effect of noise induced disturbed sleep.......................................................... 12 Adults ....................................................................................................................... 13

Children .................................................................................................................... 14

Elderly....................................................................................................................... 15

Discussion................................................................................................... 15

Conclusions................................................................................................. 18

Literature 20


4/22

EUR/04/5047477page 2

Noise effects and morbidity

Introduction

Environmental noise, irrespective of its source road, rail or air traffic, or neighbourhoodactivities remains a key issue in most European countries.The predominant health effect of noise is auditory damage which includes hearing loss. This isfrequently caused by loud work noise or by loud music (i.e. from continuous high exposure towalkmans and discotheques) as well as by loud fireworks. But besides these auditory noiseeffects attention must also be paid to the non-auditory effects of noise. Non-auditory effects of noise appear to occur at levels far below those required to damage the hearing organ.

Environmental noise acts as a stressor at night by disturbing sleep and via annoyance (orbothering) during the day.

The effect of environmental noise on sleep and annoyance were included as were severalmorbidities in the LARES study. The final report examines the strength of the associationbetween noise induced annoyance or noise induced sleep disturbances with stress mediateddiseases.

Data processing

The examination of the statistical association between noise annoyance as well as noise inducedsleep disturbances and diseases must do justice to the fact, that both diseases and risk factors areaffected by many other factors other than noise. Therefore, for a meaningful statistical analysismultiple methods must be applied, orientated by the knowledge of noise effect research, theunderlying pathogenesis model and the scale level of the collected variables. With respect to thedifferent health end-points the control variable set has to be regarded as less complete. Thereforea careful interpretation of the statistical effects is necessary.

Logistic regression

In all analysis, multiple logistic regressions were used. Normally, a logistic regression is used toanalyse an event which may or may not occur (e. g. fall ill or not, medical treatment or not). Bymeans of the logistic regression an estimator (odds ratio, OR) for the relative risk of theoccurrence of an event depending on the examined factor (e. g. sleep disturbance) was calculatedtaking into consideration other influence factors (control variables).Odds Ratios (OR) were calculated as an estimator of the relative risk (RR) of the occurrence of an event (e. g. prevalence of a disease) and its dependents on the examined factor (disturbances)and maybe other influencing factors (Confounder, moderator).


5/22

EUR/04/5047477page 3

Control variables

In total, 16 control variables were included in the logistic regression, which, in the case of the

diseases, are assumed to influence the result. The well known control variables age, gender,consumption of alcohol, smoking behaviour, sports, body mass index, and the social-economic state were completed by housing factors. To identify housing factors from the largehousing und health questionnaire a dimension testing was carried out in the form of a factoranalysis (SPSS 11.0). According to the factor analysis six relevant housing dimensions wereextracted.In Table 1 the six housing dimensions are introduced by factor labels together with the variablesused in the statistical calculations to control the housing factors. The presented control variableswere suggested by the competent expert groups. Otherwise the variable correlating most stronglywith the extracted factor was used.

Table 1: Housing and health dimensions extracted by factor analysis and control variables used in the statisticalcalculations

Factor labels Control variables

Inner housing factor: dampness in dwelling Total mould rating (Mouldy2)

Inner housing factor: air quality in dwelling Air quality in dwelling (aq_1)

Outside housing factor: green areas Visible vegetation in housing environment(he_6_1)Outside housing factor: Satisfaction with residential areas Residential areas (iesc4gr)

Inner housing factor: temperature and heatingin winter Problem with temperature in winter (t_7_1)

Inner housing factor: daylight in dwelling Need to turn on light (li_1)

Furthermore the different European cities of the LARES study were taken into account asadditional control variable (city_nr) as well as their averaged temperatures (avc_yea ) andhumidities (rh_av).

Effect of noise an noyance

A central effect of noise is annoyance. Annoyance is defined as a feeling of discomfort which isrelated to adverse influencing of an individual or a group by any substances or circumstances.Annoyance express itself e. g. by malaise, fear, threat, trouble, uncertainty restricted libertyexperience, excitability or defencelessness. With chronically strong annoyance a causal chainmay exist between the three steps health - annoyance - disease.

By means of the questions Thinking about the last 12 month, when you are here at home, howmuch would you say that noise from following sources bothers or annoys you?, the individualannoyance by out-door and indoor noise sources was determined. The percentage frequency of different noise sources which bothers or annoys is presented in including the intensity of theeffect.

To fill in the LARES-questionnaire regarding noise annoyance, it is necessary to be aware of theannoyance. That is not always the case for sleeping persons and a general problem for younger


6/22

EUR/04/5047477page 4

children. Annoyance therefore has a closer link with day time noise and the interpretation couldbe difficult in young children.

The individual annoyance from road noise, airplane noise, train noise, as well as parking noise,were summarized to examine the effect of traffic noise altogether, because traffic noise is one of the predominant noise source in the living environment. The second major noise source is theneighbourhood. General annoyance from neighbourhood noise was therefore additionallysummarized by neighbour flat, staircase, playing children, as well as by noise within dwelling.Both sum scores were subdivided into 3 categories (with the expressions: not at all, moderatelyand strongly (cut-off point 8)). The 3-scale variables, simply called traffic annoyance andneighbourhood annoyance were subsequently used for the analyses carried out with disturbanceand annoyance. The percentage frequency of each 3-scale variable is presented in Figure 2.

0 5 10 15 20 25 30 35 40 45

noise - airplane

staircase use

noise within dwelling

animals/birds

noise - parking

neighbour flat

noise - traffic

Slightly

Moderately

Strongly

Extremely

0 5 10 15 20 25 30 35 40 45

noise - train

lift

ventilation system etc.

playgrounds etc.

playing children

commercial etc. sites

surrounding area

Slightly

Moderately

Strongly

Extremely

Figure 1: Percentage frequency of different sources which bothers or annoys adults, children and elderly (in 4scales: slightly, moderately, strongly, extremely) N = 7949


7/22

EUR/04/5047477page 5

47,7

37,9

14,4

0 10 20 30 40 50 6

strongly

moderatly

not at all

percentage brothered or annoyed by general traffic noise

0

52,4

35,2

12,4

0 10 20 30 40 50

strongly

moderatly

not at all

percentage brothered or annoyed by general neighbourhood noise60

Figure 2: Percentage frequency for general traffic noise (road, airplane, train, and parking noise) (N = 7817) aswell as for general neighbourhood noise (neighbour flat, staircase, playing children, and noise withindwelling) (N=7789) for all valid cases (adults, children and elderly)

Adults

In the following figures the Relative Risk of diseases (Odds Ratio) including the 95% confidenceinterval are represented for such adults (18-59 years) who indicated annoyance by traffic orneighbourhood noise, in comparison with a reference group. For annoyance by traffic noise thereference group are adults without traffic noise induced annoyance, for annoyance byneighbourhood the reference group are adults without neighbourhood noise induced annoyance.The Relative Risks are statistically significant if the drawn confidence interval did not includethe value 1. Note that for adults the correlation between traffic noise induced annoyance andneighbourhood noise induced annoyance is significant but low (r=0.29). The variables aretherefore not statistically independent but the overlapping variance is less than 9%.

The evaluation of the effect of moderate annoyance by general traffic noise shows for asthma,diabetes, and migraine slight but significantly elevated disease risk (see Figure 4) . With thecomplexities of arthritic, respiratory and cardiovascular (CV) symptoms, disease symptoms areadditional collected for the musculoskeletal and the respiratory system as well as for the

cardiovascular system. Furthermore, the complex of SALSA considers the "trend to depression".


8/22

EUR/04/5047477page 6

Adults: general traffic noise which bothers or annoys moderatelyrelated to diseases

1,1 1,11,21,1

1

1,31,31,11,11,71,11,11,21,61,1 1,1 1,1

0

1

2

3

4

R e f e r e n c e

# A l l e r g

y N = 2 2 3

A r t h. s y

m p t. N = 2 0 5

# A s t h m a N =

6 6

# B r o n c

h i t i s N = 7 3

C V s y m

p t o m s N = 2

2 0

# D e p r e

s s i o n N = 1

1 0

# D i a b e

t e s N = 4 6

# G a s t r

i c u l c e r N =

7 9

# H e a r t

a t t a c k N =

1 4

# H y p e

r t e n s i o n N = 2

1 2

# M a l i g

. T u m o r N =

2 0

# M i g r a

i n e N = 1 8 1

R e s p. s

y m p t. N = 1

3 8

S A L S A N = 3

1 8

# S k i n d

i s e a s e N =

6 9

# S t r o k e N

= 8

# diagnosed by physician

A d j u s t e

d O d d s

R a

t i o

Figure 3: Relative disease risks for adults who indicated moderate noise induced annoyance by general trafficnoise within the last 12 months, in comparison with adults without noise induced annoyance. Adjustedfor "age", "gender", "socio-economic state", "consumption of alcohol", "smoking behaviour", "sports","body mass index", dampness in dwelling, air quality in dwelling, winter temperature in dwelling,daylight in dwelling, green areas, satisfaction with residential areas, "cities", "averaged outdoortemperature", and "averaged outdoor humidity (N=3279).

The evaluation of the effect of strong annoyance by general traffic noise reveals that annoyancecould be a serious risk factor for many of the examined health endpoints (see Figure 4) . ForAsthma, diabetes, gastric ulcer, heart attack, malignant tumour, skin disease and stroke there isno significant increased risk.

Adults: general traffic noise which bothers or annoys stronglyrelated to diseases

1,11,81,4 1,7 1,9 1,5 2,2 1,3 1,4 1,6 1,6 2,4

1

1,9 0,82,0 1,2

0

1

2

3

4

R e f e r e n

c e

# A l l e r g

y N = 1 1

2

A r t h. s y m

p t. N = 1 0

5

# A s t h m

a N = 2 8

# B r o n c

h i t i s N =

5 3

C V s y m

p t o m s N

= 1 2 7

# D e p r e

s s i o n N

= 7 1

# D i a b e

t e s N = 1

4

# G a s t r

i c u l c e r

N = 3 7

# H e a r t

a t t a c k N

= 9

# H y p

e r t e n s i o

n N = 1 2

3

# M a l i g

. T u m o r

N = 1 0

# M i g r a

i n e N = 9

7

R e s p. s

y m p t. N

= 9 4

S A L S A

N = 1 8 0

# S k i n

d i s e a s e

N = 2 8

# S t r o k

e N = 4


A d j u s

t e d O d d s R

a t i o


9/22

EUR/04/5047477page 7

Figure 4: Relative disease risks for adults who indicated strong noise induced annoyance by general traffic noisewithin the last 12 months, in comparison with adults without noise induced annoyance. Adjusted for"age", "gender", "socio-economic state", "consumption of alcohol", "smoking behaviour", "sports","body mass index", dampness in dwelling, air quality in dwelling, winter temperature in dwelling,

daylight in dwelling, green areas, satisfaction with residential areas, "cities", "averaged outdoortemperature", and "averaged outdoor humidity (N=2163).

The evaluation of the effect of general neighbourhood noise show that not only traffic noisecould influence the health risk, and that the risk pattern for strong annoyance is nearly the sameas for traffic noise.With moderate annoyance by general neighbourhood noise the relative risks for allergy, arthriticsymptoms, CV-symptoms and hypertension are significantly but slightly increased (see Figure5), whereas by traffic noise the risks asthma, diabetes, and migraine are slightly elevated.

Adults: general neighbourhood noise which bothers or annoysmoderately related to diseases

1,31,31,3 1,2 1,0 1,3 1,1 0,9 1,4 1,3 0,8 1,2

1

1,2 1,2 0,91,3

0

1

2

3

4

R e f e r e n

c e

# A l l e r g

y N = 2 3 0

A r t h. s y m

p t. N = 1 8

5

# A s t h m

a N = 5 5

# B r o n c

h i t i s N =

6 3

C V s y m

p t o m s N

= 2 2 5

# D e p r e

s s i o n N

= 9 6

# D i a b e

t e s N = 3

0

# G a s t r

i c u l c e r

N = 7 6

# H e a r t

a t t a c k N

= 1 8

# H y p e

r t e n s i o n

N = 2 1 4

# M a l i g

. T u m o r

N = 1 4

# M i g r a

i n e N = 1

5 8

R e s p. s

y m p t. N

= 1 4 7

S A L S A

N = 2 9 1

# S k i n

d i s e a s e

N = 6 1

# S t r o k

e N = 8


A d j u s

t e d O d d s

R a

t i o

Figure 5: Relative disease risks for adults who indicated moderate noise induced annoyance by generalneighbourhood noise within the last 12 months, in comparison with adults without neighbourhood noiseinduced annoyance. Adjusted for "age", "gender", "socio-economic state", "consumption of alcohol","smoking behaviour", "sports", "body mass index", dampness in dwelling, air quality in dwelling,winter temperature in dwelling, daylight in dwelling, green areas, satisfaction with residentialareas, "cities", "averaged outdoor temperature", and "averaged outdoor humidity (N=3402).

Strong annoyance by general neighbourhood noise is associated with significant and highincreased risks for many diseases. For diabetes, gastric ulcer, heart attack, malignant tumour, andstroke there is no statistical association and significance is just missed for Asthma and skindisease.


10/22

EUR/04/5047477page 8

Adults: general neighbourhood noise which bothers or annoysstrongly related to diseases

1,7 0,62,31,6

1

1,80,91,72,01,31,81,61,91,61,4 2,3 1,3

0

1

2

3

4

R e f e r e n

c e

# A l l e r g

y N = 5 6

# A r t h. s

y m p t. N

= 8 5

# A s t h m

a N = 2 9

# B r o n c

h i t i s N =

5 6

C V s y m

p t o m s N =

8 7

# D e p r e

s s i o n N

= 6 0

# D i a b e

t e s N = 1

5

# G a s t r

i c u l c e r

N = 3 3

# H e a r t

a t t a c k N

= 8

# H y p e

r t e n s i o n

N = 8 5

# M a l i g

. T u m o u

r N = 4

# M i g r a

i n e N = 8

5

R e s p. s

y m p t. N

= 8 0

S A L S A

N = 1 7 8

# S k i n

d i s e a s e

N = 3 0

# S t r o k e

N = 2


A d j u s t e

d O d d s

R a

t i o

Figure 6: Relative disease risks for adults who indicated strong noise induced annoyance by general neighbourhoodnoise within the last 12 months, in comparison with adults without neighbourhood noise inducedannoyance. Adjusted for "age", "gender", "socio-economic state", "consumption of alcohol", "smokingbehaviour", "sports", "body mass index", dampness in dwelling, air quality in dwelling, wintertemperature in dwelling, daylight in dwelling, green areas, satisfaction with residential areas,"cities", "averaged outdoor temperature", and "averaged outdoor humidity (N=2227).

Children

Many function systems of little children (e.g. the nervous system and the cognitive system) aresubject to rapid growth and development. The development of the child is not designed tocompensating for high environmental noise exposure. Therefore children could be a risk groupregarding the effect of noise annoyance on health.In the following figures the Relative Risk of diseases (Odds Ratio) including the 95% confidenceinterval are represented for such children (1-17 years) who indicated annoyance by traffic or

neighbourhood noise, in comparison with a reference group. For annoyance by traffic noise thereference group are children without traffic noise induced annoyance, for annoyance byneighbourhood noise the reference group are children without neighbourhood noise inducedannoyance. The Relative Risks are statistically significant if the drawn confidence interval didnot include the value 1.Hypertension, heart attack, gastric ulcer, malignant tumour and stroke are diseases whoseprevalence increase with age and therefore in children only appear rarely. The low number of cases did not allow a meaningful logistical regression. A second problem is the annoyance ratingfrom younger children. We assume that the annoyance rating by children is meaningful if children are 10 years or older. Therefore, for the younger children we use the averagedannoyance rating of all adults in the household.

Note that for children the correlation between traffic noise induced annoyance andneighbourhood noise induced annoyance is greater than for adults (r=0,366). The variablestherefore are not independent but the overlapping variance is less than 13%.


11/22

EUR/04/5047477page 9

Moderate annoyance by general traffic noise in Children (0-17 years) is associated with elevatedrisks for respiratory symptoms (RR=1.8) and SALSA (trend of depression, RR=1.8) ( Figure 7) .

Children: general traffic noise which bothers or annoys moderately related todiseases

1

1.11.81.41.81.0 1.0

0

1

2

3

4

R e f e r

e n c e

# A l l e r g y

N = 7 8

# B r o n c h i t i s

N = 2 4

R e s p. S y m

p t. N = 8 9

M i g r a i n e

N = 6 2

S A L S A

N = 5 4

# S k i n d i s e

a s e N =

2 7


A d j u s

t e d O d d s

R a

t i o

Figure 7: Relative disease risks for children who indicated moderate noise induced annoyance by general trafficnoise within the last 12 months, in comparison with children without traffic noise induced annoyance.Adjusted for "age", "gender", "socio-economic state", "consumption of alcohol", "smoking behaviour","sports", "body mass index", dampness in dwelling, air quality in dwelling, winter temperature indwelling, daylight in dwelling, green areas, satisfaction with residential areas, "cities", "averagedoutdoor temperature", and "averaged outdoor humidity. Note: SALSA is not validated for children(N=1195).

The examination of strong annoyance by general traffic noise shows that children fall 2.6 timesmore ill with bronchitis, than not traffic noise annoyed children. The Relative Risks forrespiratory symptoms (RR=2.6) and SALSA are considerable increased but for SALSA (trend todepression) the significance is just missed (see Figure 8) .

Children: general traffic noise which bothers or annoys strongly relatedto diseases

1

2,01,12,61,1 2,6

0

1

2

3

4

R e f e r e n

c e

# A l l e r g

y N = 9

# B r o n c

h i t i s N =

6

R e s p. S

y m p t. N

= 1 1

M i g r a i n

e N = 7

S A L S A

N = 6


A d j u s t e

d O d d s

R a

t i o

Figure 8: Relative disease risks for children who indicated strong noise induced annoyance by general traffic noisewithin the last 12 months, in comparison with children without traffic noise induced annoyance. Adjustedfor "age", "gender", "socio-economic state", "consumption of alcohol", "smoking behaviour", "sports","body mass index", dampness in dwelling, air quality in dwelling, winter temperature in dwelling,daylight in dwelling, green areas, satisfaction with residential areas, "cities", "averaged outdoortemperature", and "averaged outdoor humidity. Note: SALSA is not validated for children (N=720).


12/22

EUR/04/5047477page 10

Moderate annoyance by general neighbourhood noise in Children (0-17 years) is associated withelevated risks for respiratory symptoms (RR=1.6) and skin disease RR=3.1), but the risk of skindisease are accompanied by a large confidential interval (see Figure 9) .

Children: general neighbourhood noise which bothers or annoysmoderately related to diseases

1,80,9 1,6 1,3 1,0

3,11

0

1

2

3

4

R e f e r e n

c e

# A l l e r g

y N = 8 2

# B r o n c

h i t i s N =

2 7

R e s p. s

y m p t. N

= 9 3

M i g r a i n

e N = 6 4

S A L S A

N = 3 9

# S k i n

d i s e a s e

N = 2 8


A d j u s

t e d O d d s

R a

t i o

Figure 9: Relative disease risks for children who indicated moderate noise induced annoyance by generalneighbourhood noise within the last 12 months, in comparison with children without neighbourhoodnoise induced annoyance. Adjusted for "age", "gender", "socio-economic state", "consumption of alcohol", "smoking behaviour", "sports", "body mass index", dampness in dwelling, air quality indwelling, winter temperature in dwelling, daylight in dwelling, green areas, satisfaction withresidential areas, "cities", "averaged outdoor temperature", and "averaged outdoor humidity. Note:SALSA is not validated for children (N=1155).

Strong annoyance by general neighbourhood noise is associated with significant andconsiderably high increased risks for bronchitis (RR=3.5), respiratory symptoms (RR=2.6),migraine (RR=2.2), SALSA (RR=3.3) and skin disease (RR=3.0), but the risk of bronchitis andskin disease are accompanied by large confidential intervals (see Figure 10) .

Children: general neighbourhood noise which bothers or annoys stronglyrelated to diseases

1

3.0

3.3

2.2

3.6

0.9 3.5

0

1

2

3

4

R e f e r e n

c e

# A l l e r g

y N = 1 8

# B r o n c

h i t i s N =

1 3

R e s p. s

y m p t. N

= 2 3

M i g r a i n

e N = 2 0

S A L S A

N = 2 1

# S k i n d

i s e a s e N

= 4


A d j u s

t e d O d d s

R a

t i o

Figure 10: Relative disease risks for children who indicated strong noise induced annoyance by general

neighbourhood noise within the last 12 months, in comparison with children without neighbourhoodnoise induced annoyance. Adjusted for "age", "gender", "socio-economic state", "consumption of alcohol", "smoking behaviour", "sports", "body mass index", dampness in dwelling, air quality indwelling, winter temperature in dwelling, daylight in dwelling, green areas, satisfaction with


13/22

EUR/04/5047477page 11

residential areas, "cities", "averaged outdoor temperature", and "averaged outdoor humidity. Note:SALSA is not validated for children (N=660).

Elderly

From general stress research it is well known, that the ability to cope with noise is decreased inthe elderly. Therefore elderly people could be a risk group regarding noise annoyance and health.In the following figures the Relative Risk of diseases (Odds Ratio) including the 95% confidenceinterval are represented for such elderly people (60 and older) who indicated annoyance bytraffic or neighbourhood noise, in comparison with an elderly reference group. For annoyance bytraffic noise the reference group are elderly people without traffic noise induced annoyance, forannoyance by neighbourhood noise the reference group are elderly people withoutneighbourhood noise induced annoyance. The Relative Risks are statistically significant if thedrawn confidence interval did not include the value 1. Note that the correlation between trafficnoise induced annoyance and neighbourhood noise induced annoyance is low for elderly people(r=0,271). The variables therefore are not independent but the overlapping variance is less than8%.

For elderly people (60 years and older) the examination shows that strong annoyance by trafficnoise is a significant risk factor only for arthritic symptoms (RR=2.1) and stroke (RR=2.7). Thatis nearly the same for strong annoyance by neighbourhood noise (RR=2.1; 2.4), but withneighbourhood noise the risks for gastric ulcer (RR=3.7) and depression (RR=2.0) are inadditional significantly elevated (see Figure 11) . It seems that in comparison with adults forelderly people other factors than noise annoyance is more decisive for the appearance of a

disease. Elderly: general neighbourhood noise which bothers or annoys

strongly related to diseases

3,7

2,11,6 1,1 1,2 1,3 2,0 0,7 1,2 0,9

2,1

0,8

1

1,1 1,0

2,4

0,9

0

1

2

3

4

R e f e r e n

c e

# A l l e r g

y N = 1 5

A r t h. s y m

p t. N = 5 2

# A s t h m

a N = 6

# B r o n c

h i t i s N =

1 8

C V s y m

p t o m s N =

5 8

# D e p r e

s s i o n N

= 1 7

# D i a b e

t e s N = 1

3

# G a s t r

i c u l c e r

N = 2 3

# H e a r t

a t t a c k N

= 1 3

# H y p e

r t e n s i o n

N = 4 9

# M a l i g

. T u m o r

N = 5

# M i g r a

i n e N = 1

3

R e s p. s

y m p t. N

= 2 2

S A L S A

N = 2 8

# S k i n d

i s e a s e N

= 5

# S t r o k e

N = 1 1


A d j u s t e d O d d s

R a

t i o

Figure 11: Relative disease risks for elderly people who indicated strong noise induced annoyance by generalneighbourhood noise within the last 12 months, in comparison with elderly people without neighbourhood noiseinduced annoyance. Adjusted for "age", "gender", "socio-economic state", "consumption of alcohol", "smokingbehaviour", "sports", "body mass index", dampness in dwelling, air quality in dwelling, winter temperature indwelling, daylight in dwelling, green areas, satisfaction with residential areas, "cities", "averaged outdoortemperature", and "averaged outdoor humidity (N=936).


14/22

EUR/04/5047477page 12

Effect of noise in duced disturbed sleep

Sleep is an essential condition for humans and can be severely disturbed by noise. Acute sleepdisturbances affect the subjective wellbeing and with an individual latency, also affect qualitativeor quantitative performance. Chronic forms of sleep disorders are frequently classified as ahealth risk.It is well known that sleep can be disturbed by noise exposure. The effect of noise induced sleepdisturbances was analysed by means of the question Has your sleep been disturbed by noiseduring the past 4 weeks?. The percentage of sleep disturbances regarding different noisesources is presented inFigure 1 2.

2,4

4,1

5,1

2,7

9,8

9,5

3,4

0 2 4 6 8 10

noise within dwelling

staircase use

animals/birds

surrounding area

noise - parking

neighbour flat

noise - traffic

percentage of sleep disturbances12

0,7

0,8

1,1

1,2

1,2

2,1

1,4

0 2 4 6 8 10

playgrounds etc.

ventilation system etc.

playing children

lift

noise - train

commercial etc. sites

noise - airplane

percentage of sleep disturbances12

Figure 12: Percentage frequency of noise sources which induced sleep disturbances N = 8519 (adults, children andelderly)


15/22

EUR/04/5047477page 13

The percentage distribution of noise induced sleep disturbances is presented in Figure 13.

23,3

76,7

0 10 20 30 40 50 60 70 80 90

No

Yes

n o

i s e

i n d u c e

d s

l e

e p

d i s t u r

b a n c e

percentage

Figure 13: Percentage distribution of noise induced sleep disturbances in the whole sample (N = 8325)

Adults

In the following figure the Relative Risk of diseases (Odds Ratio) including the 95% confidenceinterval are represented for such adults (18-59 years) who indicated noise induced sleepdisturbances by noise, in comparison with adults without noise induced sleep disturbances. TheRelative Risks are statistically significant if the drawn confidence interval did not include thevalue 1.The noise induced sleep disturbances are associated for adults with significantly elevated risksfor the vast majority of diseases. Only for diabetes, malignant tumour, skin disease and stroke noremarkable association were calculated. Furthermore, significance is missed at heart attack (seeFigure 14) .

Adults: Noise induced sleep disturbances related to diseases

2,3

1,2

1,21,61,6

1

0,61,51,41,61,51,41,41,41,71,5 1,6 1,1

0

1

2

3

4

R e f e r e n

c e

# A l l e r g

y N = 1 8 3

A r t h. s y m

p t. N = 1 8

8

# A r t h r o

s i s N = 1

6 1

# A s t h m

a N = 5 2

# B r o n c

h i t i s N =

7 5

C V s y m

p t o m s N =

1 9 9

# D e p r e

s s i o n N

= 1 1 1

# D i a b e

t e s N =

3 2

# G a s t r

i c u l c e r

N = 7 5

# H e a r t

a t t a c k N

= 1 7

# H y p e

r t e n s i o n

N = 1 9 1

# M a l i g

. T u m o r

N = 1 0

# M i g r a

i n e N = 1

6 2

R e s p. s

y m p t. N

= 1 3 8

S A L S A

N = 3 5 1

# S k i n d

i s e a s e N

= 4 3

# S t r o k

e N = 8


A d j u s

t e d O d d s

R a

t i o

Figure 14: Relative disease risks for adults which indicated noise induced sleep disturbances within the last 4 weeks

in comparison with adults without noise induced sleep disturbances. Adjusted for "age", "gender", "socio-economicstate", "consumption of alcohol", "smoking behaviour", "sports", "body mass index", dampness in dwelling, airquality in dwelling, winter temperature in dwelling, daylight in dwelling, green areas, satisfaction withresidential areas, "cities", "averaged outdoor temperature", and "averaged outdoor humidity (N=4309).


16/22

EUR/04/5047477page 14

Children

The sleep structure and the sleep quality change dramatically in the course of childhood.

Therefore children could be a risk group regarding noise induced sleep disturbances.In the following figure the Relative Risk of diseases (Odds Ratio) including the 95% confidenceinterval are represented for such children (1-17 years) who indicated noise induced sleepdisturbances, in comparison with children without noise induced sleep disturbances. We assumefor the analysis that the rating for noise induced sleep disturbances is more or less reliable for allchildren. The Relative Risks are statistically significant if the drawn confidence interval did notinclude the value 1.With children noise induced sleep disturbances are statistically associated with a highly elevatedrisk of medical treatments due to bronchitis (RR=3.7) as well as with elevated risks forrespiratory symptoms (RR=1.9) and for trend of depression (SALSA, RR=3.5). For headachesand migraines, it is well known that in the case of children physicians are consulted only rarely.Therefore, in the following Figure 15 noise induced sleep disturbances with self reportedmigraine were examined. The amount of cases of self reported migraines was more than doubledin comparison to migraines diagnosed by a physician. Children with noise induced sleepdisturbances fall 2.2 times more ill with migraines. The strongest association was calculated forarthritic symptoms. Children with noise induced sleep disturbances have 7.3 times more arthriticsymptoms, but the risk is accompanied by an very large confidential interval (CI=2.4-22.2).

Children: Noise indused sleep disturbances related to diseases

1,5

0,8 1,3

3,7

1,9

3,5

2,2

1

0

1

2

3

4

R e f e r e n

c e

# A l l e r g

y N = 3 0

A r t h. s y

m p t. N =

1 0

# A s t h m

a N = 1 5

# B r o n c

h i t i s N =

2 4

R e s p. s

y m p t. N

= 5 0

S A L S A

N = 3 8

M i g r a i n

e N = 4 0

# S k i n d

i s e a s e N

= 9


A d j u s

t e d O d d s R a

t i o

7,3

Figure 15: Relative disease risks for children which indicated noise induced sleep disturbances within the last 4weeks in comparison with children without noise induced sleep disturbances. Adjusted for "age", "gender","socio-economic state", "consumption of alcohol", "smoking behaviour", "sports", "body mass index",dampness in dwelling, air quality in dwelling, winter temperature in dwelling, daylight indwelling, green areas, satisfaction with residential areas, "cities", "averaged outdoor temperature", and"averaged outdoor humidity. Note: SALSA is not validated for children (N=1274).


17/22

EUR/04/5047477page 15

Elderly

Sleep quality decreases strongly in elderly people. Therefore elderly people could be a risk group

regarding noise induced sleep disturbances.In the following figure the Relative Risk of diseases (Odds Ratio) including the 95% confidenceinterval are represented for such elderly people (1-17 years) who indicated noise induced sleepdisturbances, in comparison with elderly people without noise induced sleep disturbances. TheRelative Risks are statistically significant if the drawn confidence interval did not include thevalue 1.With elderly people noise induced sleep disturbances are statistically associated with arthriticsymptoms (RR=1.6), asthma (RR=2.0), gastric ulcer (RR=2.2) migraine (RR=1.7) and SALSA(RR=1.4).

Elderly: Noise induced sleep disturbances related to diseases

2,2

1,61,4

2,0

1,3 1,0 1,3 0,9 1,2 0,9 1,1 1,7

1

1,3 1,4 0,61,3

0

1

2

3

4

R e f e r e n

c e

# A l l e r g

y N = 4 5

A r t h. s y m

p t. N = 1 7

6

# A s t h m

a N = 2 3

# B r o n c

h i t i s N =

5 6

C V s y m

p t. N = 1 7

3

# D e p r e

s s i o n N

= 4 5

# D i a b e

t e s N = 5

8

# G a s t r

i c u l c e r

N = 5 3

# H e a r t

a t t a c k N

= 3 7

# H y p e

r t e n s i o n

N = 1 5 5

# M a l i g

. T u m o r

N = 1 5

# M i g r a

i n e N = 5

9

R e s p. s

y m p t. N

= 7 0

S A L S A

N = 1 2 0

# S k i n d

i s e a s e N

= 1 9

# S t r o k

e N = 1 4


A d j u s t e

d O d d s

R a

t i o

Figure 16: Relative disease risks for the elderly which indicated noise induced sleep disturbances within the last 4weeks in comparison with elderly people without noise induced sleep disturbances. Adjusted for "age","gender", "socio-economic state", "consumption of alcohol", "smoking behaviour", "sports", "body massindex", dampness in dwelling, air quality in dwelling, winter temperature in dwelling, daylight in

dwelling, green areas, satisfaction with residential areas, "cities", "averaged outdoor temperature", and"averaged outdoor humidity (N=1404).

Discussion

For noise effect research the disease as a pathological health end-point is of high importance andcan only be examined in epidemiological studies. In epidemiological studies, like the LARESstudy, however only statistical associations can be uncovered. The assessment, whether acalculated association reflects a causal relation has to be carried out according to differentcriteria. After exclusion of chance, bias and confounding, the strength of the association,expressed by the relative risks, as well as the biological plausibility are main topics of thesecriteria. Whereby biological plausibility means that there are credible mechanisms which explain


18/22

EUR/04/5047477page 16

how the noise exposure can contribute to the development of the illness. Furthermore it has to bedemanded that the confidence intervals should be small for a save point-estimation. The causalinterpretation of the association is finally supported by finding a dose effect relationship and

comparable findings in independent studies.The presented results fulfils not all but most of the causality-criteria. Change as well as bias canbe excluded with high probability due to the sampling frame and due to the quality controlwithin the study.The evaluation of the data was carried out multiple with an extensive control variable set, takinginto account the housing environment, regarding their influence on health. The large controlvariable set is, with regard to the variety of the examined diseases, not always to be consideredcomplete. Furthermore, well known control variables are missing as one noise sensitivity andoccupational noise exposure. Also the family history could not be taken into account. Airpollution, which has the same origin as traffic noise, could have a confounding impact on theresults, but the effect could not differ essentially, because the risk pattern for traffic noise and

neighbourhood noise does not differ dramatically. Altogether, the control process must beclassified as sufficient according to comparable examinations.The health effects of noise refer in all analyses to the cardiovascular system, the respiratorysystem, and the musculoskeletal system as well as on psychic disturbances. The biologicalplausibility for these effects is given due to experimental examinations. With central nervousprocesses, noise stress influences the neuro-endocrine system either directly or indirectly aboutemotional experience (disturbances, annoyance) or disturbed sleep. In this way noise exposurescan lead to an inadequate and dangerous neuro-endocrine reaction pattern and finally toregulation diseases (see [Frankenhuser et al. 1976]. Pathological changes can manifestthemselves in the cardiovascular system (see [Babisch 2001]), in the respiratory system (see[Langewitz et al. 2003]) as well as in the musculoskeletal system (see [Eich 2003]) or can appearas psychic disturbances (see [Maschke et al. 2003]).

On the other hand pathological changes reduce more or less the mental well being and thereforemay increase the susceptibility to noise. From this point of view the epidemiological associationbetween annoyance and health-endpoints could be interpreted as caused by the diseases. For theassociation between sleep disturbances and health-endpoints it is nearly the same. Sleep could bedisturbed by diseases, for example by bronchitis or asthma.

That well being as well as sleep could be influenced by individual state of health is well known,but the thesis that the associations in the LARES-study are essentially caused by the state of health is not very probable. Most of the examined diseases are heavily dependent on age, with aconsiderably higher occurrence among the elderly (with the exception of allergy). If theepidemiological associations are essentially caused by diseases, the frequency of noise inducedannoyance as well as of noise induced sleep disturbance in the elderly people must beconsiderably higher than in adults. That is not the case for noise induced annoyance and noiseinduced sleep disturbances. The relative frequencies of noise induced annoyance as well as of noise induced sleep disturbance in elderly people are remarkably less than in adults. Therefore,we assume that the epidemiological associations are essentially caused by noise annoyance aswell as by noise induced sleep disturbances, although also influenced by the effect of theindividual state of health.

This assumption of a causal relation between noise induced annoyance and health-endpoints issupported by dose effect considerations. The relative risks for pathological changes in the

cardiovascular system (hypertension, CV-symptoms) were clearly increased in adults atchronically strong annoyance (RR=3.0; 3.1), in comparison to adults indicated by moderateannoyance. At moderate annoyance the Relative Risks are no different than in the reference


19/22

EUR/04/5047477page 17

group or only slightly increased (general neighbourhood noise, RR=1.3). This also applies forpathological changes in the respiratory system (Bronchitis, respiratory symptom), for arthritisand for depression.

To summarize, it can be concluded that the results of the LARES study confirm increased healthrisks at chronically strong noise annoyance as well as for noise induced sleep disturbances on anepidemiological level.

The health risks are dependent on the age groups. Few differences can be recognized within anage group regarding the pathogenesis mechanism of annoyance and sleep disturbances. At strongannoyance the risk pattern between general traffic noise and general neighbourhood noise arevery similar within the age groups (see Table 2) . With children the uncertainty of the annoyancerating is, however, to be taken into account (see chapter 0). The surprisingly high relative risksfor neighbourhood noise partly can be explained by the high information content of neighbourhood noise.

If the risk pattern of noise induced sleep disturbances compared with the risk pattern of strongannoyance induced by noise (see Table 2) , a good agreement is visible. In the group of elderlypeople a remarkable feature is highlighted. A significant increase of the risk of stroke is evidentonly at strong noise induced annoyance and only in elderly people.

Table 2: Significantly Odds Ratio for diseases calculated in the LARES-study

Significantly OR for diseases Adults Elderly Children

Hypertension 1,588 ---- ----CV symptoms 1,545 ---- (5,455)*

Stroke ---- 2,718 ncAsthma ---- ---- ----Bronchitis 1,861 ---- 2,624Resp. Sympt. 1,969 ---- 2,563Arthr. Sympt. 1,754 2,066 ----Depression 2,229 ---- nc

general traffic noisestrongly annoyed

SALSA 1,879 ---- ----Hypertension 1,706 ---- ncCV symptoms 1,601 ---- ----Stroke ---- 2,415 ncAsthma ---- ---- ----Bronchitis 1,907 ---- 3,453Resp. Sympt. 1,572 ---- 3,562Arthr. Sympt. 2,346 1,885 ----Depression 1,78 1,989 nc

general neighbourhood noise

strongly annoyed

SALSA 2,276 ---- 3,322Hypertension 1,485 ---- ----CV symptoms 1,449 ---- ----Stroke ---- ---- ncAsthma ---- 2,019 ----Bronchitis 1,455 ---- 3,674Resp. Sympt. 1,632 ---- 1,943Arthr. Sympt. 1,598 1,617 (7,308)*

Depression 1,466 ---- ----

sleep been disturbed by noise

SALSA 2,26 1,413 3,413( )* very large confidence intervalsnc = not calculable


20/22

EUR/04/5047477page 18

Regarding the different age groups pathological changes appear in adults both in thecardiovascular system, the respiratory system, the musculoskeletal system and in depression (seeTable 2) . Unlike this, with older people significantly increased risks exist only for the

musculoskeletal system (arthritic symptoms) independent of the noise source. With strongannoyance, furthermore the risk is increased significantly for stroke among the elderly. Theresults of the LARES study do not confirm the thesis that older persons are subjects of anincreased noise induced health risk.

A comparable result is published by Maschke [2004] for the Spandau Health-Survey (SHS). TheSHS shows a stronger association between treatments of hypertension and the nightly noise levelfor adults less than 60 years in comparison with elderly people.

Adults, older than 60 years

1,00

1,38

1,52

0

1

2

3

4

5

< 50 dB(A ) 50-55 dB(A ) > 55 dB(A )

equivalent s ound level during night (8h)

R e

l a t i v e

R i s i k ( O d d s

R a

t i o

Adults, 20 to 60 years

2,87

2,52

1,00

0

1

2

3

4

5

< 50 dB(A) 50-55 dB(A) > 55 dB(A)

equivalent sound level during night (8h)

R e

l a t i v e

R i s i k ( O d d s

R a

t i o )

Figure 17: Statistical association between nightly road traffic noise and hypertension treatments in the 9th repetitioncycle. Adjusted for "age", sex, "consumption of alcohol", "consumption of tobacco", "professionalmobility", "sports", "body mass index", "social economic index", "loss of spouse", "hearing ability", "noisesensitivity" and "season of the medical examination".

With children the effects of noise is to be seen primarily in the respiratory system (see Table 2) .Strongly annoyed children fall 2-3.7 times more in ill with respiratory diseases than non noiseannoyed children. The same effect exists for noise disturbed sleep with approximately the samerange of risk. In contrast, allergy, one of the most frequent illnesses of children, does not seem torespond to noise. According to the presented results, children have to be classified as particularlysensitive to noise with regard to the respiratory system.

Conclusions

The result confirms the thesis that for chronically strong annoyance a causal chain existsbetween the three steps health strong annoyance increased morbidity.

For adults with strong annoyance, significantly elevated relative risks exist both in thecardiovascular system, the respiratory system, and the musculoskeletal system as well as bydepression. Elderly people with strong annoyance respond less than adults with the exception of stroke. With children the effects of noise is to be seen primarily in the respiratory system.


21/22

EUR/04/5047477page 19

A further effect of noise is sleep disturbances. Sleep can be severely disturbed by noise. Acutesleep disturbances effect performance and in the long run health. For adults significantlyelevated relative risks in the cardiovascular system, the respiratory system, and the

musculoskeletal system as well as with depression exist with noise induced sleep disturbances.Many of these diseases increase with age and therefore in children only appear rarely. Withchildren significantly elevated relative risks in the respiratory system as well as in migraines(self reported) are of great importance.

The association between noise induced sleep disturbances and morbidity are approximately inthe same range as the consequences of noise induced strong annoyance. By noise induced sleepdisturbances more people are effected (23.3%) than by noise induced strong annoyance (12.4%neighbourhood noise; 14.4% traffic noise). Therefore, particular attention must be paid to nightlyenvironmental noise exposure.

Little attention was paid to neighbourhood noise till now and therefore pathological effects are

considerably underestimated. The health effect of neighbourhood noise induced annoyance isapproximately in the same range as the health effect of traffic noise induced annoyance. Theresults point out that it is necessary to improve the sound insulation in residential buildings.

The respiratory system also reacts to neighbourhood noise with increased relative risks. Incontrast to traffic noise, with neighbourhood noise, air pollution cannot be responsible as analternative reason for the increased health risks in the respiratory system. The noise has to beclassified as an independent risk factor for the respiratory system.


22/22

EUR/04/5047477page 20

Literature

Babisch, W.; Ising, H. (2001): Noise induced stress is a risk factor in cardiovascular disease. In:The 2001 International Congress Noise Control Engineering. The Hague, TheNetherlands

Eich, W. (2003): Chronische Polyarthritis and Fibromyalgie (Chronic polyarthritis andFibromyalgie). In: Adler, RH.; Herrmann, JM.; Khle, K.; Langewitz, W.; Schonecke,OW.; Uexkll von, T.; Wesiak, W. (eds): Uexkll Psychosomatische Medizin- Modellerztlichen Handelns. 6. Auflage. Urban & Fischer, Mnchen, Jena, 941- 946

Frankenhuser, M.; Lundberg, U. (1976): Psychological reactions to noise as modified bypersonal control. Reports from the Department of Psychology, U. Stockholm Vol. 471

Langewitz, W.; Solr, M. (2003): Asthma bronchiale (Asthma bronchial). In: Adler, RH.;Herrmann, JM.; Khle, K.; Langewitz, W.; Schonecke, OW.; Uexkll von, T.; Wesiak,W. (eds): Uexkll Psychosomatische Medizin- Modelle rztlichen Handelns. 6. Auflage.Urban & Fischer, Mnchen, Jena, 891- 907

Maschke, C.; Wolf, U.; Leitmann, T. (2003): Epidemiologische Untersuchungen zum Einflussvon Lrmstress auf das Immunsystem und die Entstehung von Arteriosklerose(Epidemiological examinations to the influence of noise stress on the immune system andthe emergence of arteriosclerosis). WaBoLu-Hefte 01/03 Umweltbundesamt, Berlin

Maschke, C. (2004): Health Effects of nocturnal traffic noise sound level at subjects homeversus self-reported sleep disturbances. In: 30. Jahrestagung der Deutschen Gesellschaftfr Akustik (DAGA) (Proceedings of the 30th Annual Acoustical Congress of theGerman Society of Acoustics), Strasbourg, France (in press)

who sleep disturbance lares

Documents