NORAH Noise-related annoyance, cognition and health
Transportation noise effects in the environs of the airport
Vol. 7: General Overview of the NORAH Research Project
NORAH Noise-related annoyance, cognition and health study NORAH General Overview
AUTHORS, PROJECT MANAGEMENT
Rainer Guski Ruhr-Universität Bochum, 44780 Bochum
Dirk Schreckenberg ZEUS GmbH, Zentrum für angewandte Psychologie, Umwelt- und Sozialforschung, 58093 Hagen
PUBLISHER, CLIENT
Gemeinnützige Umwelthaus GmbH Rüsselsheimer Str. 100 65451 Kelsterbach
OVERALL COORDINATION OF THE NORAH STUDY
Rainer Guski Ruhr-Universität Bochum, 44780 Bochum
Dirk Schreckenberg ZEUS GmbH, Center for applied Psychology, Environmental and Health Research, 58093 Hagen
All of the scientific texts, graphics, tables and other content contained in this document are protected by copyright. They may not be copied, altered, duplicated or published, either in full or in part, without the prior written consent of the author or the publisher. Any publication or use of even part of this document is only permissible with the consent of the author or publisher with reference to the full source.
7 October 2015
NORAH Noise-related annoyance, cognition and health study NORAH General Overview
INTERNAL QUALITY ASSURANCE
Ali Erdogan Justus-Liebig-Universität, Gießen
Caroline Herr Bavarian Health and Food Safety Authority, Munich
August Schick University of Oldenburg
Rudolf Schuemer FernUniversität in Hagen
Enno Swart Otto-von-Guericke-Universität Magdeburg
Berthold Vogelsang Ministry for the Environment, Energy and Climate Protection of Lower Saxony, Hanover
Hajo Zeeb Leibniz Institute for Prevention Research and Epidemiology – BIPS GmbH
SCIENTIFIC ADVISORY BOARD – EXTERNAL QUALITY ASSURANCE
Mark Brink ETH Zurich
Erland Erdmann University of Cologne
Kerstin Giering University of Applied Sciences Trier, Environmental Campus Birkenfeld
Barbara Griefahn Leibniz Research Centre for Working Environment and Human Factors, TU Dortmund (IfADo)
Jürgen Hellbrück Catholic University of Eichstätt-Ingolstadt
Wolfgang Hoffmann University Medicine Greifswald
Christian Maschke State Office of Environment, Health and Consumer Protection Brandenburg, Potsdam
Lothar Ohse Hessian State Office for Environment and Geology, Wiesbaden
Georg Thomann Office for Nature and the Environment, CH Graubünden
Irene van Kamp Rijksinstituut voor Volksgezondheid en Milieu, NL-Bilthoven
Joachim Vogt Darmstadt University of Technology
NORAH Noise-related annoyance, cognition and health study NORAH General Overview
VOLUMES RELEASED ON THE NORAH STUDY – TRANSPORTATION NOISE EFFECTS IN THE ENVIRONS OF THE AIRPORT
Vol. 1 Cognitive development and quality of life of children
Vol. 2 Registration of the traffic noise exposures
Vol. 3 Annoyance and quality of life
Vol. 4 Aviation noise and sleep
Vol. 5 Blood pressure monitoring
Vol. 6 Health risks
Vol. 7 General overview
NORAH Noise-related annoyance, cognition and health study Summary NORAH General Overview
Contents Contents ...... 4 Summary ...... 6 1 Introduction ...... 10 1.1 The diverse health landscape: from wellbeing to heart attack ...... 10 1.2 The NORAH joint research project...... 12 2 Aims of the joint research project ...... 15 2.1 Primary aims ...... 15 2.2 Aims of module 1, "Annoyance and quality of life" ...... 16 2.3 Aims of module 2, "Health" ...... 17 2.3.1 Aims of the sub-study "Secondary data-based case-control study with detailed survey on health risks" ...... 17 2.3.2 Aims of the sub-study "Blood pressure monitoring" ...... 17 2.3.3 Aims of the sleep study ...... 18 2.4 Aims of module 3, "Cognitive development and quality of life of children" ...... 18 2.5 Aims of the inter-module "Registration of traffic noise exposures" ...... 19 3 Working groups ...... 21 4 Methodology ...... 23 4.1 Scientific results depend on the research question and the investigation method .... 23 4.2 Research questions of the sub-studies ...... 25 4.2.1 Research questions of the module "Annoyance and quality of life" ...... 25 4.2.2 Research questions of the secondary data-based case-control study with detailed survey 26 4.2.3 Research questions of the blood pressure monitoring ...... 27 4.2.4 Research questions of the sleep study ...... 27 4.2.5 Research questions of the child study ...... 28 4.3 Analysis approaches and sampling in the sub-studies ...... 28 4.3.1 Investigation in residential areas in the environs of the airport...... 28 4.3.2 Study region, spatial reference ...... 29 4.3.3 Recruitment of participants, sampling in the sub-studies ...... 37 4.3.3.1 Sampling in the sub-study "Annoyance and quality of life" ...... 37 4.3.3.2 Data acquisition in the sub-study "Secondary data-based case-control study with detailed survey" ...... 42 4.3.3.3 Sampling in the sub-study "Blood pressure monitoring" ...... 43 4.3.3.4 Sampling in the sleep study ...... 44 4.3.3.5 Sampling in the sub-study "Cognitive development and quality of life of children" ...... 45
NORAH Noise-related annoyance, cognition and health study Summary NORAH General Overview 4.4 Investigation methods in the sub-studies ...... 48 4.4.1 Module 1: "Annoyance and quality of life" ...... 48 4.4.2 Secondary data-based case-control study ...... 49 4.4.3 Blood pressure monitoring ...... 49 4.4.4 Sleep study ...... 51 4.4.5 Child study ...... 53 5 Central results ...... 54 5.1 Annoyance and quality of life ...... 55 5.1.1 The main results on Annoyance and quality of life ...... 55 5.1.2 Summary evaluation of the results of the studies on annoyance and quality of Life 69 5.2 Secondary data-based case-control study with detailed survey ...... 70 5.2.1 The main results for the individual diseases ...... 71 5.2.1.1 Cardiovascular disease risks (myocardial infarction, stroke, heart failure) ..... 71 5.2.1.2 Consideration of the nocturnal maximum levels in the case of cardiovascular disease risks ...... 73 5.2.1.3 Breast cancer risks ...... 73 5.2.1.5 Distortion of results due to unidentified or residual confounding? Results of the detailed survey ...... 76 5.2.1.6 Sub-analysis of insured persons with information on the social status available in the health insurance data ...... 77 5.2.1.7 Significance of the noise history: constant residential addresses over many years and "cumulative noise years" ...... 77 5.2.2 Summary evaluation of the results of the NORAH case control study ...... 77 5.3 Blood pressure monitoring ...... 80 5.4 Sleep study ...... 81 5.5 Cognitive development and quality of life of children ...... 84 6 Generalizability of the statements ...... 86 6.1 To whom do the NORAH statements apply? ...... 86 6.2 What influence to response rates have on the NORAH results? ...... 87 6.3 What influence do refusals to participate have on the NORAH results? ...... 88 7 Strengths of the NORAH joint research project ...... 90 7.1 General strengths: ...... 90 7.2 Strengths of the study "Annoyance and quality of life" ...... 91 7.3 Strengths of the secondary data-based case-control study ...... 92 7.4 Strengths of the blood pressure monitoring ...... 93 7.5 Strengths of the sleep study ...... 93 7.6 Strengths of the child study ...... 94 8 Challenges for the NORAH joint research project ...... 95
NORAH Noise-related annoyance, cognition and health study Summary NORAH General Overview 9 Conclusion ...... 96 References ...... 100
Summary
Aims: The multidisciplinary research project NORAH (Noise Related Annoyance, Cognition and Health) was aimed at providing a broad and scientifically reliable description of the effects of air, road and rail traffic noise on the health and life quality of affected residents. The following content areas were considered: annoyance and quality of life (module 1), health (module 2, including blood pressure regulation, cardiovascular diseases, breast cancer, depression and sleep disturbances) as well as cognitive development of school children (module 3). Where possible, effects of the changes due to the start-up of a new runway at Frankfurt International Airport were considered, as well as the implementation of a night curfew between 23:00 and 05:00 hrs and the rerouting of flights.
Methods: According to the research questions, different methods were applied. In the case of noise annoyance and quality of life, systematic surveys were conducted: a panel study made up of three waves between 2011-2013 in the area of Frankfurt Airport, cross-section studies in the vicinity of the airports Cologne/Bonn, Berlin-Schönefeld, and Stuttgart. Cross- section studies were also carried out to compare the effects of road, rail and air traffic noises and on the combination of air and road traffic and air and rail traffic noise.
In the case of cardiovascular health risks, breast cancer, and depressive episodes, a secondary data-based case-control study with detailed survey was performed in the administrative district of Darmstadt, the rural districts of Mainz-Bingen and Alzey-Worms, as well as in the cities of Mainz and Worms.
With respect to the long-term effects of traffic noise on blood pressure regulation, daily self- administered blood pressure measurements were registered for three weeks in two waves (2012 and 2013) with residents in the vicinity of the Frankfurt Airport ("blood pressure monitoring").
In order to study the short-term effects of night-time air traffic noise on the sleep of residents, sleep quality investigations were carried out in the years 2011-2013 in the homes of residents in the vicinity of Frankfurt Airport.
The effects of chronic exposure to aircraft noise on the cognitive performance and quality of life of school children near Frankfurt Airport were studied by means of performance tests (especially reading tests) with children, as well as surveys with children, parents, and teachers.
Address-specific equivalent continuous sound levels of different reference times for air, rail and road traffic noise were available for all study participants (except for participants in the sleep study where the participants' own measurements were used) (to some extent, maximum levels as well as the numbers of loud events) and were used in the evaluations.
NORAH Noise-related annoyance, cognition and health study Summary NORAH General Overview
Main results:
1. At all four airports studied, the percentage of persons highly annoyed by air traffic noise at comparable noise levels was larger than would be expected from the so- called "EU standard curves" (Miedema & Oudshoorn, 2001). In the vicinity of Frankfurt Airport, in 2011 (before the implementation of a new north-west runway) higher annoyance responses were observed than during a comparable survey performed in 2005 (Schreckenberg & Meis, 2006). The annoyance response increased in 2012 (after the implementation of the new runway), and decreased marginally in 2013. In the cross-sectional studies, it turned out that aviation noise was associated with higher noise annoyance than with road or rail traffic noise at comparable long- term levels. The height of road and rail noise annoyance was very similar at comparable noise levels. In the cross-sectional studies on noise combinations (aviation plus road traffic noise, or aviation plus rail traffic noise) it was observed that the total annoyance followed mainly the aircraft noise-related annoyance. 2. With respect to noise-related health risks, the largest risks connected to the 10-dB level increase were observed for unipolar depressive episodes – statistically significant with all three traffic noise sources. With respect to cardiovascular health risks, the effects of rail and road traffic noise on heart failure, myocardial infarction, and stroke were more clearly seen as compared to the effects of aviation noise. Road traffic noise showed the highest (statistically significant) risk increase per 10-dB level increase with depressive episodes (4.1%), myocardial infarction (2.8%), heart failure (2.4%), and stroke (1.7%). Rail traffic noise showed the highest (statistically significant) risk increases with a 10-dB level increase on depressive episodes (3.9%), heart failure (3.1%), and stroke (1.8%). Air traffic noise showed the highest (statistically significant) risk increases with 10- dB level increase on depressive episodes (8.9%), and heart failure (1.6%). The use of indoor noise levels partially showed a statistically significant increase of health risks, as compared to outdoor noise levels, but it should be kept in mind that indoor noise levels were estimated rather roughly. Breast cancer showed a statistically significant association with aviation noise levels during the night (23-05 h). 3. Residents who were exposed to long-term aviation noise levels <40 dB but had night- time maximum levels >50 dB, showed higher health risk estimates – statistically significant with respect to stroke and heart failure. Results of this type indicate that the consideration of night-time maximum levels may be relevant for estimating the health risks of aviation noise. On the other hand, such results need further tests from independent studies. 4. The mean systolic and diastolic blood pressure values of residents increased slightly (statistically not significant) with increased aircraft noise levels. Railway noise levels showed a slight (statistically not significant) increase of the systolic blood pressure. There was no statistically significant relation between traffic noise levels and pulse frequency, blood pressure amplitude, hypertension, and 10-year heart attack risk (PROCAM-Score).
NORAH Noise-related annoyance, cognition and health study Summary NORAH General Overview
5. The sleep study showed a diminished aircraft noise related probability of physiological night-time awakening associated with the introduction of the night curfew at Frankfurt Airport for a group being in bed during 22:00-22:30 hrs until 06:00-06:30 hrs). On average, the number of awakenings decreased from 2.0 to 0.8 (2011 vs. 2012). This shows that the curfew had a positive overall effect on sleep. In general, there was a significant effect of the number of aircraft noise events on the number of aircraft noise-related awakenings which lead to a fragmentation of the sleep (diminished continuity), without shortening the total sleep time. In a second group, being in bed from 23:00-23:30 hrs until 07:00-07:30 h, an average aircraft noise- associated awakening frequency of 1.9 was observed in 2012. The difference to the former group is due to the longer time (one hour) of aircraft noise exposure in the morning hours. At background noise levels of 28.8 dB(A), the odds of awakening in- creased by 23 % with an increase per 10 dB increase of the maximum level of an aircraft overflight. Total sleep time, sleep latency, sleep efficiency, waking time after falling asleep, and the percentage of waking after 04:30 hrs did not differ statistically significantly between 2011 and 2011. 6. Persons with a positive attitude towards air traffic did show less (objectively) measured sleep disturbances. The direction of causality is unclear, i.e., the question whether a disturbed sleep is due to negative attitudes to air traffic, or the other way round, could not be determined. The (subjective) evaluations of the residents with respect to sleepiness and tiredness in the morning are in a medium range in all of the three groups observed between 2011 and 2013. The self-assessed habituation to aircraft noise, the loudness of the residential area, the age as well as the chronotype of the participants all show a statistically significant influence on the individual assessment of sleepiness and tiredness. The subjective assessment of a good sleep diminished in spite of the introduction of the night curfew statistically significantly between 2011 and 2013 by 5 % and 11 % respectively, independently of the aviation noise exposure. This effect is also true for those participants who took part in all of the three measurement waves. This effect is probably due to factors not assessed in the study. 7. In the child study a statistically significant decrease of reading performance was observed with increasing aircraft noise levels: a one-month retardation of reading performance was observed for a 10 dB increase of the equivalent continuous sound level. The teachers in highly exposed schools concordantly report about considerable disturbance of the classes by aviation noise. In addition, there were statistically significant connections - although of lower effect size - between equivalent continuous sound levels and less positive assessments of the physical and mental well-being and children's attitudes towards school.
Seen from an interdisciplinary perspective, two parallel results seem remarkable:
1. Medically diagnosed depressive episodes show the strongest increase with increasing aircraft noise levels, and self-assessments of mental quality of life (Mental Component Summary (MCS), including depressive tendencies) were lowest in the panel group expecting or experiencing an increase of noise levels in 2011-2013.
NORAH Noise-related annoyance, cognition and health study Summary NORAH General Overview
2. Both the physiological sleep measurements and the survey on annoyance and quality of life found an improvement of sleep through the night after the implementation of the night curfew during 23:00-05:00 hrs, although the night-time equivalent sound
level LpAeq,22-06h decreased by only 0.6 dB between 2011 and 2012 in the panel sample. At the same time, both studies showed an increase of negative evaluations of the participants with respect to the morning time (e.g. disturbances of late sleep, tiredness and sleepiness).
NORAH Noise-related annoyance, cognition and health study Introduction NORAH General Overview
1 Introduction
1.1 The diverse health landscape: from wellbeing to heart attack
When scientists talk about "health", they often use different definitions. First of all we have the definition in the preamble to the constitution of the World Health Organization from 1946, in which health is described as a "complete state of physical, mental and social well- being, and not merely the absence of disease or infirmity" (see WHO 2014, p. 1). Sometimes they will then recall the regulatory-physiological definition proposed by Rudolf Virchow in 1869: "Disease begins at the moment when the regulatory equipment of the body no longer suffices to remove the disturbances. Not life under abnormal conditions as such engenders a disease, but rather disease begins with the insufficiency of the regulatory apparatus" (in Sudhoff, 1922, p. 261). Both definitions are unsatisfactory from our present-day point of view, because both of them ignore cultural and subjective considerations regarding wellbeing and disease. Klaus Hurrelmann (2010, p. 146) offers a modern definition of health: "Health is the state of balance of risk factors and protection factors that occurs when an individual succeeds in overcoming both the internal (physical and mental) as well as the external (social and material) challenges." If this balance breaks down, this leads to a state of "relative" illness, which can be returned to "relative" health by self-control or external help. For noise researchers this definition has the practical advantage that it is very similar to the definition of stress – but more about that later. Here we want to establish first of all that noise, i.e. the exposure to loud noise from road, rail and air traffic, has been recognized for many years now as a threat to human health which must be taken seriously (see Berglund & Lindvall 1995; WHO 2011).
The use of any health definition becomes complicated when we look at the breadth of the possible risks and their relative evaluation. We can ask, for example, whether the fact that somebody feels considerably annoyed and disturbed by noise is less significant in terms of health than the heart disease to which the noise may have contributed. And how do we evaluate sleep disruption due to noise, or the worry about the value of the house near the source of the noise, or the inhibition of the cognitive development of our children? As long as it is just one person who is, for example, experiencing considerable annoyance or suffering from heart disease, then the order of priority of the risks would appear relatively clear: we will regard the heart disease as more important than the annoyance. But this order of priority soon becomes shaky if we take into consideration how many people are affected by one or the other risk. Undoubtedly there are a lot more people suffering annoyance due to noise than are exposed to the risk of heart disease, but should we still evaluate the suffering of people experiencing annoyance as being less important than that of the heart disease patients?
NORAH Noise-related annoyance, cognition and health study Introduction NORAH General Overview
The WHO (2011) tried to bring a series of health risks due to environmental noise exposures down to a common denominator: (disability-adjusted life years, DALYs). The organization estimates that environmental noise in Europe costs 60,000 healthy life years per year per 1 million resident life years due to heart disease, 45,000 years due to the inhibition of cognitive development of children, 903,000 years due to sleep disruption, 21,000 years due to tinnitus and 587,000 years due to annoyance (see fig. 1-1). Road traffic noise is cited as the main cause of these losses.
Fig. 1-1: Lost healthy life-years (according to WHO 2011).
(Abb.: Schlafstörungen = sleep disturbance Kognitive Beeinträchtigung bei Kindern = impairment of cognitive development of children Herzerkrankungen = heart diseases Belästigung = annoyance Tinnitus = tinnitus)
These calculations are based on noise impact studies in which the various effect areas, e.g. heart disease and annoyance, were investigated separately, and it is still unclear whether there are connections between the effect areas and, if yes, what they look like. Only in terms of sleep disturbance and elevated blood pressure does the international literature provide any empirical data regarding a connection with cardiovascular diseases (see Cappuccio et al., 2011).
NORAH Noise-related annoyance, cognition and health study Aims of the joint research project NORAH General Overview
1.2 The NORAH joint research project
NORAH ("Noise-Related Annoyance, Cognition and Health") is one of the most demanding interdisciplinary research projects every carried out for the investigation of the effects of traffic noise on the population. The project proceeded on the basis of a stress-theory approach, where stress (as a process) is defined as when an acute environmental challenge exceeds the psychological and physiological regulation options of a human being, in particular in situations which include unpredictability and uncontrollability (see Lazarus & Launier 1978; Koolhaas et al., 2011). We must, however, make a distinction between short-term and long-term stress – for example between acute sleep disturbances with the corresponding negative consequences for wellbeing and performance the next day, and the constant repetition of these and other disturbances over a prolonged period. The short-term stress situations can trigger a wide range of physiological and psychological reactions, e.g. in the brain activity, in the cardiovascular and hormonal systems, in the emotional wellbeing and in behaviour. Such short-term exposures can be called stress – and relevant to health – if they are not balanced out within a short time. If these stress situations occur repeatedly without adequate options for compensation, we refer to chronic stress.
The severity of the stress-related psychological and physiological changes in the individual depends, among other things on (a) the frequency, distribution over time and intensity of the acoustic exposure, (b) personal factors (e.g. noise sensitivity and ability to cope with noise), (c) attitudes to the source of the noise and to the institutions and persons perceived as responsible for the reduction of noise, (d) social factors (e.g. media reporting), (e) situational factors (living conditions, ventilation, existing noise protection) and (f) other risk factors (e.g. exposures not related to noise, health behaviour) of the persons concerned. Particularly in the case of constant noise exposure, when protection and regulation options are no longer sufficient, the repeated acute noise reactions combined with constant annoyance can lead to the occurrence or aggravation of diseases, in particular cardiovascular diseases, and can also impair cognitive development in children.
Somatic and mental illnesses have a double function in this context: on the one hand, previous illnesses limit the resources of the individuals concerned to deal adequately – mentally and physically – with environmental burdens such as transportation noise. On the other hand, they can also be the result of the constant chronic physical and mental stress that occurs as a reaction to chronic acoustic burdens.
NORAH Noise-related annoyance, cognition and health study Aims of the joint research project NORAH General Overview
Fig. 1-2: Framework model of the effect of environmental noise
(Abb. Geräusch = acoustic exposure Informationsverarbeitung, akute Reaktionen, Stressverarbeitung = information processing, acute reactions, stress processing Physik = Physics Emission = emission Transmission = transmission Immission = immission Belastung = burden Schall = sound Medizin, Psychology = Medicine, psychology Lärm = noise Perzeption & Wirkung = perception & effect Geräuschbewertung = noise rating Interferenzen, Störungen = interferences, disturbances physische Stressreaktionen = physical stress reactions Kontrollmöglichkeiten, Lärmbewältigungsvermögen = control options, ability to cope with noise Neubewertung, resultierender Stress = reassessment, resulting stress Intervenierende Faktoren (personelle, situative, soziale) = intervening factors (personal, situational, social) Erholung = recovery? Weitere mögliche gesundheitliche Folgen = Other adverse health effects psychisch = mental körperlich = physical)
NORAH Noise-related annoyance, cognition and health study Aims of the joint research project NORAH General Overview
Special reactions were also expected in the case of significant changes in the acoustic exposure, such as occurred, for example, at Frankfurt Airport with the opening of the new runway and the subsequent operational changes. In such cases the resulting reactions can only be inadequately assessed on the basis of exposure-effect functions based on existing data. The expansion of the airport could possibly have its own health-relevant effect in addition to or independent of how severe the air traffic noise immissions actually are for the residents concerned.
After a public hearing in September 2010 of the Hessian state parliament, the Hessian government resolved to commission a comprehensive investigation of the health effects of public transport noise. After a public tendering procedure and the corresponding offers, in April 2011 a group of ten scientific institutions headed by the Ruhr University of Bochum received the commission to carry out these investigations. At the same time, a scientific advisory board was installed in which experienced and independent experts from all of the disciplines concerned are responsible for the external quality assurance. They oversaw the careful execution of all parts of the study and ensured that all of the quality standards of the disciplines concerned were observed.
The researcher group (see chapter 3) called the joint project "NORAH" (Noise Related Annoyance, Cognition and Health) and organized itself with 5 sub-studies in 4 modules:
- "Annoyance and quality of life" (Module 1) - "Health" (Module 2) - "Cognitive Development and Quality of Life of Children" (Module 3).
This means that – apart from tinnitus - NORAH covers all of the effect areas cited by the WHO (2011). The investigation area is mainly the Rhine-Main region (Fig. 3), but also includes – for sub-studies – the airports Berlin-Schönefeld, Cologne/Bonn and Stuttgart. The basis for the description of the acoustic exposures was the individual registration of the source-specific acoustic data (equivalent sound level, maximum level) for air, road and rail traffic noise.
This general overview is meant to offer a summary of the whole NORAH research project, including a classification of the project and its results in the current state of scientific research. For space reasons, a lot of important information had to be left out of this summary, but can be found in the specific reports for the sub-studies.
NORAH Noise-related annoyance, cognition and health study Aims of the joint research project NORAH General Overview
2 Aims of the joint research project
2.1 Primary aims
The NORAH joint research project was aimed at achieving a broad and scientifically founded description of the effects of noise from air, rail and road traffic on the health and quality of life of the residential population concerned.
Two essential elements of the NORAH project were (a) the main investigation region and (b) the time of the investigation:
The Rhine-Main region as the main investigation region is characterized by high population and traffic density, and has the largest airport in Germany. The investigation period began in summer 2011, i.e. before the opening of a new runway, which was followed by the introduction of a night flight curfew, and ended in spring 2014. Some of the flight routes had already been changed, however, in March 2011 before the NORAH investigations began (adjustment of downwinds approaches, shifting of the banking ranges and lowering of the airspace).
The primary aims of the investigation were thus as follows:
- to answer the question as to whether the noise effects in the Rhine-Main region can, in principle, be compared with those in other German regions, and what influence road, rail and air traffic noise has on the population
- and to answer the question as to which influence the changes in airport operations in autumn 2011 had on the health and quality of life in the environs of the airport, and whether the annoyance experienced by the residential population near airports in expansion situations differs from that experienced by people near steady-state airports.
Neither of these primary questions could be answered by the analysis of existing scientific literature, and this is why ten prominent research and technical groups from medicine, psychology, social science and acoustics joined to form a research consortium (see chapter 3) and, within the framework of a Europe-wide tendering procedure, made a joint research offer to the Land of Hessen, which was accepted in April 2011.
NORAH Noise-related annoyance, cognition and health study Aims of the joint research project NORAH General Overview
2.2 Aims of module 1, "Annoyance and quality of life"
One of the main aims of the module "Annoyance and quality of life" was the compilation or updating of so-called exposure-response relationships between acoustic and psychological variables – i.e. parameters of chronic acoustic exposures to noise from road, rail and air traffic on the one hand, and the answers of the persons concerned with regard to annoyance due to noise, disturbances experienced and quality of life on the other.
This was connected with other aims which included various comparisons:
- the comparison of the extent of aviation noise annoyance, experienced noise-related disturbances by day and by night with the health-related quality of life in the Rhine- Main region before and after the opening of the North-West runway in October 2011
- the comparison of aviation noise annoyance, reported disturbances and health- related quality of life in the Rhine-Main region with those at the airports Berlin- Schönefeld, Cologne/Bonn and Stuttgart, as well as the comparison of the annoyance, reported sleep disturbances and quality of life reductions due to air traffic noise with those due to road and rail noise
- at the same time the study wanted to find out what effects the various types of traffic noise sources combined as a multiple burden have on the annoyance reports of the persons concerned
- finally there was also an attempt to identify relationships between the effect areas, in particular between the noise annoyance assessment and the reported health-related quality of life.
In each case, the influence of personal, social and situational factors on the effect of traffic noise was also to be taken into consideration.
NORAH Noise-related annoyance, cognition and health study Aims of the joint research project NORAH General Overview
2.3 Aims of module 2, "Health"
The main aim was to answer the question as to which extent chronic noise from air, rail and road traffic has an influence on the health of the persons concerned – particularly on the adult residents of the Rhine-Main region.
2.3.1 Aims of the sub-study "Secondary data-based case-control study with detailed survey on health risks"
The general aim was to investigate the effects of chronic acoustic exposure to noise from road, rail and air traffic on the health of the residential population in the Rhine-Main region. The sub-study focussed on cardiovascular diseases (myocardial infarction, heart failure, stroke), mental illnesses (in particular unipolar depressive episodes) and cancer (especially breast cancer in women) as they are registered in the accounting and prescription data of health insurance companies.
The primary aim of the (additional) detailed survey was to assess the significance of important confounders – here in particular the health behaviour (e.g. body mass index, smoking, alcohol) and social status. To do this, the risk estimates without consideration of these important confounders had to be compared with the risk estimates with consideration of these confounders registered in the survey. This comparison allows important statements on the extent and direction of possible distortions of the traffic noise-related risk estimates in the (exclusively) secondary data-based case-control study; the detailed survey thus allows more reliable statements on the level of the "true" risk estimates in the secondary data- based analysis.
A secondary aim of the detailed survey was to compare the connections between cardiovascular diseases and traffic noise-related indoor noise levels with those of outdoor levels.
2.3.2 Aims of the sub-study "Blood pressure monitoring"
The blood pressure monitoring was carried out to register the blood pressure as a measurable physical reaction to stress in a sample group that is chronically exposed to air, rail and road traffic noise. It was assumed that a chronic traffic noise exposure leads to a higher rate of disease, complaints and measureable physical reactions in the area of the autonomous nervous system which are reflected in the blood pressure level. This is based on the fact that cardiovascular diseases are closely linked with blood pressure regulation.
NORAH Noise-related annoyance, cognition and health study Aims of the joint research project NORAH General Overview A further aim was to investigate the influence of chronic acoustic exposure to road, rail and air traffic noise on the overall cardiovascular risk and the closely linked 10-year heart attack risk.
2.3.3 Aims of the sleep study
Extensive investigations of the nightly sleep of residents near the airport Cologne/Bonn, which handles many mainly freight flights at night, have been used in the noise abatement concepts of various airports. It was not possible to examine up to now, however, whether the exposure-effect curves compiled for Cologne/Bonn could be transferred to airports with different night-time operations. This is why an important aim of the sleep study was to compare the exposure-effect curves for air traffic noise-related wake-up reactions between the airports Cologne/Bonn and Frankfurt/Main. Unlike the other sub-studies of the NORAH joint research project, "exposure" here refers primarily to the acute acoustic exposure to air traffic noise measured in the bedroom of the residents in the proximity of the airport.
In addition to this, the exposure-effect curves for air traffic noise-related wake-up reactions at Frankfurt/Main airport over the course of time between 2011 and 2012 were to be investigated. In October 2011 changes in the nocturnal operations of the airport came into effect (introduction of the night flight curfew and partial shifting of flight movements to the daytime or night-time shoulder hours), and this is why measurements of sleeping patterns were carried out before and after the operational changes.
In connection with the operational changes at Frankfurt Airport, the question arose as to whether these had a different or similar effect on the sleep of people who habitually go to bed or get up relatively early or relatively late. A third aim was, therefore, to compare the air traffic noise-related wake-up reactions at Frankfurt Airport between persons who go to bed earlier or later and get up accordingly earlier or later.
A further aim of the sleep study was to develop and field-test a method which – unlike the complicated polysomnography (PSG) – allows the analysis of air traffic noise-related wake-up reactions (the greatest possible disturbance of sleep) in a larger number of persons with the exclusive measurement of the electrocardiograph (ECG) and the body movements. This vegetative-motor method (VMM) has the advantage that – unlike PSG – it is not necessary for a project worker to visit the study participant every evening and every morning to help them to fit and remove the equipment. In addition to this, the evaluation of the data is fully automatic.
2.4 Aims of module 3, "Cognitive development and quality of life of children"
Module 3 investigated air traffic noise effects – concentrating on cognitive development and quality of life – on children while the other modules all dealt with adults. The child study looked primarily at the acquisition of reading skills and the development of reading-relevant
NORAH Noise-related annoyance, cognition and health study Aims of the joint research project NORAH General Overview speech skills as well as the quality of life of primary school children who were exposed to acoustic burdens due to air traffic noise both at school and at home. An important aim was the establishment of exposure-response relationships between the extent of the chronic noise exposure and the various cognitive development parameters or measures of the quality of life of the children.
2.5 Aims of the inter-module "Registration of traffic noise exposures"
The registration of the traffic noise exposure formed the foundation for the NORAH joint project. The basis for all effect analyses in the aforementioned sub-studies was the most realistic, source-specific description of the exposure to air, road and rail traffic noise in the investigation area of Frankfurt and in the comparison locations of Berlin-Schönefeld, Cologne/Bonn and Stuttgart for the residential addresses of the participants in the NORAH sub-studies.
The main aim was connected with a series of secondary aims:
- Determination of the address-specific, long-term noise exposures (with reference to a year or several months) in all study regions by calculation of various noise level measures for different times of the day: equivalent continuous sound level and maximum sound level for road, rail and air traffic as well as maximum level frequency statistics for air traffic, among other things for the definition of so-called NAT values (number above threshold, e.g., number of flight movements above a defined maximum level), as well as for outdoors and to some extent indoors at the residential addresses of the participants. For all of the sub- studies with the exception of the sleep study, the noise exposure was determined by calculation of the acoustic parameters. In the sleep study the sound pressure level and the noise data were recorded continuously at the ear of the sleeper for the whole time that the participant lay in bed, and this was used to derive several acoustic parameters.
- Determination of a "noise history" for study participants: this secondary aim referred to the sub-study "secondary data-based case-control study with detailed survey". The purpose of this was to calculate for all available addresses in the study region of the case-control study in the Rhine-Main region the source-specific traffic noise exposure over several years, in this case for the years 1996 to 2010.
NORAH Noise-related annoyance, cognition and health study Aims of the joint research project NORAH General Overview
By combining this with the address and change-of-address data of the study participants, this was to be used to compile an individual "noise history" of participants, even if they had moved, insofar as any change of address was within the study region.
To our knowledge for the first time in a noise impact study: estimation of the uncertainties or reliability of the calculated acoustic characteristic values by a detailed error analysis taking into account the uncertainties of the input and model parameters separately for the investigated noise source types. The influence of the uncertainty on the results of the exposure-effect relationship was to be shown as an example.
NORAH Noise-related annoyance, cognition and health study Working groups NORAH General Overview
3 Working groups
The joint research project NORAH ("Noise-Related Annoyance, Cognition and Health") is an interdisciplinary research project to investigate the effects of traffic noise on the population. It was carried out between 2011 and 2015. The main region for the investigations was the Rhine-Main region. A total of ten scientific working groups from the fields of acoustics, medicine and health sciences, sleep research, epidemiology and psychology were involved in the project. The participating institutions are listed below in alphabetical order: - German Aerospace Center (Deutsches Zentrum für Luft- und Raumfahrt/DLR Cologne), Institute of Aerospace Medicine, dept. of flight physiology: Effects of air traffic noise on nightly sleep (Uwe Müller, Daniel Aeschbach, Eva-Maria Elmenhorst, Alex Hoff, Franco Mendolia, Julia Quehl, Iris Rieger, Stefan Schmitt, in cooperation with Wilma Littel, Gießen-Marburg University Hospital). - Hörzentrum Oldenburg GmbH, in collaboration with TU Kaiserslautern: Cognitive development and quality of life of children (Markus Meis). - Justus-Liebig University in Gießen, Institute for Hygiene and Environmental Medicine, in collaboration with the Institute for Medical Informatics: Blood pressure monitoring, field work for sleep quality and case-control study (Thomas Eikmann, Anja zur Nieden, Susanne Harpel, Azita Lengler, Doreen Ziedorn, Marcus Bürger, in collaboration with Jan Spilski (TU Kaiserslautern). - Justus-Liebig University in Gießen, Institute for Medical Informatics, in collaboration with the Institute for Hygiene and Environmental Medicine: Analyses on blood pressure monitoring, field work for case-control study (Jörn Pons-Kühnemann, Karin Römer, Abed Atiq, Helge Hudel). - Möhler und Partner Ingenieure AG, Munich, in cooperation with SoundPLAN GmbH, Backnang, and AVIA Consult, Strausberg: Registration of the traffic noise exposures for all modules (Ulrich Möhler, Manfred Liepert, Maximilian Mühlbacher, Alfred Beronius, Mar- tin Nunberger (all Möhler und Partner), Gerd Braunstein, Michael Gillé, Jochen Schaal (SoundPLAN) and Rüdiger Bartel (Avia Consult)). - Dresden University of Technology, Institute and Outpatient Clinics of Occupational and Social Medicine in collaboration with the data collection centre, Justus-Liebig University in Gießen: Secondary data-based case-control study with detailed survey (Andreas Seidler, Mandy Wagner, Melanie Schubert, Patrik Dröge, Janice Hegewald). - Kaiserslautern University of Technology, Faculty of Social Sciences, in collaboration with Hörzentrum Oldenburg: Cognitive development quality of life of children (Maria Klatte, Kirstin Bergström, Jan Spilski, Jochen Mayerl, Thomas Lachmann). - Ruhr University in Bochum, Workgroup for Environmental and Cognitive Psychology, in collaboration with Zeus GmbH: Scientific direction and project coordination (Rainer Guski, Christin Peschel, Jördis Wothge). - Sozialwissenschaftliches Umfragezentrum GmbH (SUZ Duisburg), in collaboration with ZEUS GmbH: Annoyance and quality of life (Frank Faulbaum, Lars Ninke).
NORAH Noise-related annoyance, cognition and health study Working groups NORAH General Overview - ZEUS GmbH, Hagen, in collaboration with Ruhr University in Bochum and SUZ Duisburg and in cooperation with TU Kaiserslautern: Annoyance and quality of life as well as project coordination (Dirk Schreckenberg, Christin Peschel, Jördis Wothge, Jan Spilski).
NORAH 20 Noise-related annoyance, cognition and health study
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
4 Methodology
4.1 Scientific results depend on the research question and the investigation method
Every scientific study has to have at least one specific investigable research question and at least one specific investigation method, and both have to be coordinated with each other. This is why studies with different research questions and/or investigation methods are not directly comparable.
For research questions with reference to noise effects, for example, we have to distinguish between acute and chronic noise effects: when, in response to an acute noise, our blood vessels contract and the muscle tension increases, this is certainly an acute noise effect, but it does not in itself tell us anything about the danger to health of chronic noise exposure. To answer this question we would have to know how the body responds to repetition of the acute noise exposure and different relax times. A distinction also has to be made between effect variables that are dependent on or independent of the noise context: we know that a person usually gives higher noise annoyance values in surveys if it is clear that it is a noise impact study than he or she would if they did not know this (see Bodin, Björk, Öhrström, Ardo & Albin, 2012). We also have to take into account the situation of the data collection: it has been shown in sleep studies, for example, that people display stronger noise effects in the sleep laboratory than at home (see Basner et al., 2004).
In addition to this, the transfer of the results of a study to other research questions and/or other situations requires an especially careful discussion of the conditions. Even the apparently simple comparison of effects of road traffic noise with those of rail traffic noise has to take into consideration that the temporal structure of the two noises is generally very different – in the first case we are usually dealing with relatively frequent, short and medium-loud events, in the second case the noise is less frequent but louder and more prolonged. Even if both noise sources have the same equivalent continuous noise level, the exposure-effect relationships differ both in terms of the annoyance and of sleep disturbance (see, for example, Miedema & Oudshoorn, 2001; Elmenhorst et al., 2012; Giering, 2010; Müller, 2010). The transfer of the results from one airport to another is also problematic if, for example, the airports have different operating times and/or fleet compositions.
In the field of applied research (e.g. noise impact research) we encounter an additional problem: those causing the problem are generally different persons or groups to those affected by the problem; they often have at least different interests in the results. While one side may be more concerned about weighing up health risks against the economic benefits, the other side usually wants to know which maximum health risks can occur – regardless of the probability of their occurrence (see Hastie & Dawes, 2001). It is thus to be expected
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview that any study will be subject to criticism of its research questions and methodology by at least one interest group.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
This is why the independent scientific quality assurance which accompanies the NORAH project in all of its research, from the study planning to the final report, is an important instrument for the discussion and consideration of all research processes under exclusively scientific criteria. The scope of scientific quality assurance realized here sets new standards for applied research.
4.2 Research questions of the sub-studies
4.2.1 Research questions of the module "Annoyance and quality of life"
In general, the module "Annoyance and quality of life" asked the question as to which statistical relationships exist between the acoustic exposure to noise from air, rail and road traffic on the one hand, and ascertainable noise effects on the other. Alongside general annoyance, the identifiable noise effects included annoyance at different times of the day, disruption of communication, quiet, sleep, as well as psycho-vegetative disorders (headaches, irritability/nervousness, acoustic startle responses, health-related quality of life, environment-related quality of life).
For these general questions and for most of the specific questions cited below it was also asked which personal and social factors moderate the noise effects and to what extent. This includes in particular:
- expectations regarding the situation after the opening of the new runway (perceived changes so far, future residential situation, expectations of the consequences), perception and assessment of active noise abatement measures - social moderators: attitude towards the three noise sources (registered on a source- specific basis), trust in those responsible for aviation noise, perceived procedural fairness - personal moderators: noise sensitivity and ability to cope with noise.
Specific questions referred to comparisons of the statistical relationships between acoustic exposure and identifiable noise effects at different times, in particular before and after the start-up of the North-West runway at Frankfurt Airport, but also between the results of the RDF study (Schreckenberg & Meis, 2006) and the study carried out in 2011-2013 in the Rhine-Main region.
Other questions referred to:
- the comparison of the air traffic noise effects at Frankfurt Airport with those at the airports Berlin-Schönefeld (airport under renovation with planned new opening),
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview Cologne/Bonn (airport with a relatively high proportion of night flights) and Stuttgart (airport with night flight curfew from 0:00 to 06:00 hrs). Here it was primarily a matter of comparing the air traffic noise effects at an airport in a "change situation" (Frankfurt Airport and - planned - Berlin-Brandenburg Airport) with those at an existing, steady-state airport where there are no changes in the noise situation, e.g. due to expansion or daytime capacity changes in flight operations (Airports Cologne/Bonn, Stuttgart). The changes at Frankfurt Airport included mainly: start-up of the North-West runway, adjustment of flight routes, introduction of a night flight curfew from 23:00 to 05:00 hrs, introduction of noise-optimized landing and take-off procedures and fleet upgrading within the framework of an active noise abatement measures program. The changes at Berlin-Schönefeld Airport (planned) referred to: expansion of the regional airport and the opening as Berlin-Brandenburg International Airport. - the comparison of the identifiable noise effects between the three traffic noise sources (road, rail, air traffic) and, finally, - the question as to what effect the combinations of two traffic noise sources (here: air plus road traffic or air plus rail traffic) has on reported annoyance.
4.2.2 Research questions of the secondary data-based case-control study with detailed survey
The questions here were as follows:
- What statistical connection is there between the chronic acoustic exposure to road, rail and air traffic noise on the one hand, and cardiovascular diseases (myocardial infarction, heart failure, stroke), mental illnesses (especially episodes of a unipolar depressive disorder) and cancer (in particular breast cancer in women) on the other? - Are the risk estimates of the (exclusively) secondary data-based case-control study distorted by the limited identifiability of important confounders? If yes: to what extent and in what direction?
- Do the risks differ from each other for individual time slices (LpAeq) for the time from 22.00 to 08.00 hrs? - Which health risks do person have who are exposed to traffic noise for long periods (5 years, 10 years)? - Do the health risks change if the various traffic noise sources are all included in an "analysis model"? - Do the exposure risk courses for indoor noise levels differ from those for outdoor noise levels? (Only for the detailed survey.)
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
4.2.3 Research questions of the blood pressure monitoring
The blood pressure monitoring addressed the influence of chronic (i.e. regularly recurring) exposure to air, rail or road traffic noise on the blood pressure and the risk of cardiovascular disorders (here: estimated 10-year heart attack risk). The following questions were addressed:
- Were higher blood pressure measurement values recorded in residents in the environs of Frankfurt Airport in relation to increasing chronic exposure to air, rail or road traffic noise? - Do these residents have a higher 10-year heart attack risk due to increasing chronic exposure to air, rail or road traffic noise? - Was it possible to identify changes in the blood pressure values over time?
4.2.4 Research questions of the sleep study
The sleep study concentrated on the influence of the aviation noise exposure on sleep. The questions here were as follows:
- Does the bundling of the flight movements at Frankfurt Airport in the shoulder hours as a result of the curfew cause an increase in acute sleep disturbances, i.e. problems falling asleep before the curfew time starts and frequently waking up too early when it ends? - Has sleep improved or deteriorated over the whole sleeping time compared with the basic survey in 2011? - Can the exposure-effect relationships of Cologne/Bonn Airport be transferred in terms of physiological wake-up reactions to Frankfurt Airport with curfew? - Do persons with a positive attitude towards aviation react less strongly in sleep to aviation noise than persons with a negative attitude? - Is decreasing nocturnal air traffic noise exposure associated with decreasing reported sleep disturbances (e.g. with regard to quality of rest, subjective wake-up frequency as well as tiredness and sleepiness in the morning)?
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview 4.2.5 Research questions of the child study
The child study investigated the connection between aviation noise and the cognitive development and quality of life of children with the following questions:
- Is it possible – with comprehensive control of other influencing factors – to verify negative effects of air traffic noise, such as occurs in the environs of Frankfurt Airport, on children's reading skills? - Is it possible to verify effects of air traffic noise on the linguistic-cognitive functions on which reading acquisition is based? - Is it possible to verify effects of air traffic noise on the health-related quality of life and the wellbeing of the children in school? - To what extent do children feel disturbed by air traffic noise in their school and domestic environment? - What concrete effects does air traffic noise have on the everyday school routine?
4.3 Analysis approaches and sampling in the sub-studies
4.3.1 Investigation in residential areas in the environs of the airport
All of the sub-studies in the NORAH joint research project were field-oriented, i.e. there were no lab studies, but exclusively investigations on people exposed to a greater or lesser extent to traffic noise for at least one year. All of the surveys were carried out in the environs of Frankfurt Airport and some of them in the environs of other airports (Cologne/Bonn, Berlin-Schönefeld, Stuttgart). The definition of airport environs as the investigation area stems from the fact that the expansion at Frankfurt Airport was the reason for the commissioning of the NORAH project, and the aim was to compare the air traffic noise in the Rhine-Main region around the airport before and after the start-up of the North- West runway in its effects on residents with other traffic noise sources and the noise situation at other airport locations.
Depending on the sub-study of the NORAH joint research project and the respective research questions, different approaches were taken in the detail of the surveys and the recruitment of participants.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
4.3.2 Study region, spatial reference
The study region for the sub-studies "Annoyance and quality of life", "Blood pressure monitoring" and "Cognitive Development and Quality of Life of Children" was defined in each case by an acoustic delineation criterion. For the sub-studies "Annoyance and quality of life" and "Blood pressure monitoring", first all of the residential addresses in the Rhine-Main region were registered that have an equivalent continuous sound level >= 40 dB during the day from 06:00 to 22:00 hrs (LpAeq,06-22h) or at night from 22 to 06 hrs (LpAeq,22-06h) in the aviation noise exposures. The allocation to the sample was determined with the aid of the function Max(LpAeq,06- 22h,LpAeq,22-06h), i.e., all persons were included whose daytime or nighttime level for aviation noise was >= 40 dB. The criterion was also applied at the airports Berlin-Schönefeld, Cologne/Bonn and Stuttgart for the sub-study "Annoyance and quality of life".
For the sub-study on the Cognitive Development and Quality of Life of Children, the same "40- dB criterion" was applied, but on the basis of the daytime equivalent continuous sound level
LpAeq,06-22h, as in this study the prevailing daily aviation noise exposure was the main factor for the selection of school locations.
The "40-dB criterion" was introduced essentially for two reasons:
1. In the range below 40 dB, the reliability of the calculated level values is considerably reduced, particularly at a high-traffic airport such as Frankfurt. This leads to a higher frequency of errors in the levels. Due to the higher error variance, an evaluation at such low levels would be more likely to cause an underestimation of possible effects. In residential areas with an aviation noise exposure of less than 40 dB in the equivalent continuous sound level it can be expected that other, possibly unregistered, noise sources dominate and the calculated noise level for an individual source (here: air traffic) falls below the background noise level. Measurements at the measurement stations at Frankfurt Airport showed that in the quietest hours of a 24-
hour day (between 0:00 and 04:00 hrs) background noise levels of LpA95,1h from 28 to 42 dB (average: 36 dB) are reached (Möhler et al., 2015). It can happen that rare noise events which add up to a equivalent continuous sound level which is lower than the background level can still be heard. If, however, as in the present study, we take the equivalent continuous sound level as a measure for the noise exposure, this takes us into the realm of "causal undecidability" if statements on exposure-effect relationships do not allow a distinction as to whether an effect is due to the mathematically calculated source-specific noise level or to the (higher) background level. The noise level calculations for the sampling were based on grid calculations. The noise levels calculated after sampling for the addresses of study participants were based on individual point calculations, some of which had equivalent continuous noise levels less than 40 dB.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
The reasons for this can be a. that there were changes in the input data between the calculations for the sampling and later for the study participants, particularly as the reference years of the calculations are different. b. that the equivalent continuous noise level contour calculated for the sampling of 40 dB was "generously" measured and includes "gaps" or "islands", i.e.
residential areas with a noise exposure of less than 40 dB in the LpAeq for day and night, and c. that differences in methodology in the grid and individual point calculation led in some cases to equivalent continuous noise levels less than 40 dB after the sampling.
2. The second reason concerns in particular those sub-studies that specifically looked at the common (continuous) covariation of cause and effect – here noise exposure and its effects. In the analysis of the effects of noise in the sense of the difference between results in an exposed group of persons compared with those of a – possibly unexposed – reference group, as was the case in the case-control study on health risks, the reference group was defined as having "no lower limit", e.g. as a group with an equivalent continuous noise level "less than 40 dB".
For the Case-Control Study the defined study region in the Rhine-Main region included the entire government district of Darmstadt, the Rhine-Hessen cities of Mainz and Worms and the regional districts Mainz-Bingen and Alzey-Worms.
The sleep study focussed exclusively on the effect of air traffic noise on sleep. Accordingly, the investigated residential areas in the Rhine-Main region were selected on the basis that the air traffic noise there prevailed as the dominant noise exposure and there was as little other (traffic) noise as possible heard in the bedrooms of the study participants. The primary criterion for the selection of the investigation areas was that from 2011 to 2013 they had a relatively high occurrence of nocturnal flight movements between 22:00 and 06:30 hrs and between 23:00 hours and 07:30 hours proceeding from Frankfurt Airport. For the participant recruitment the aim was a distribution of the maximum sound levels (LpAmax) for air traffic noise at the ear of the sleeper in the range of around 35 –70 dB. Then the following locations in the Rhine-Main region were included in the investigation: in the years 2011 and 2012 Gräfenhausen, Klein-Gerau, Nauheim, Offenbach Süd, Raunheim, Rüsselsheim, Wixhausen, Worfelden. These were joined in 2013 by Flörsheim, Frankfurt-Süd, Hochheim and Mörfelden.
The following images show the study regions of the individual sub-studies and the distribution of the respective participants.
NORAH Noise-related annoyance, cognition and health study
Methodology NORAH General Overview
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview Study regions of the sub-study "Annoyance and quality of life" a Participants Panel 2011 Participants Panel 2012 Participants Panel 2013 Participants Cross-sections studies rail, road Frankfurt combination 2012 am Main Wiesbaden
Offenbach
Mainz Flughafen Frankfurt a. M.
Untersuchungsgebiet
Darmstadt
Kartenhintergrund: "© OpenStreetMap-
b Participants (dark blue)
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
c
Participants (dark blue)
d Participants (dark blue)
Figure 4-1a-d: Study regions of the module "Annoyance and quality of life" in the Rhine-Main region (a) and at the airports Berlin-Schönefeld (b), Cologne/Bonn (c) and Stuttgart (d).
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview Study region of the sub-study "Secondary data-based case-control study with detailed survey"
Figure 4-2: Study region of the sub-study "Secondary data-based case-control study with detailed survey" in the Rhine-Main region.
NORAH Noise-related annoyance, cognition and health study Methodik Methodology NORAH General Overview
Participants BP1 Participants BP2
NORAH General Overview
Study region of the sub-study "Blood pressure monitoring“ Note: BP1 is the first survey wave in the Blood pressure monitoring, BP2 a follow-up survey on a section of the participants from BP1 – see section 4.4.3.
Figure 4-3: Distribution of the study participants in the study region of the sub-study "Blood pressure monitoring" in Rhine-Main region.
NORAH Noise-related annoyance, cognition and health study
Methodology NORAH General Overview
Methodology
NORAH General Overview
Study region of the sleep study
Figure 4-4: Study region of the sleep study in the Rhine-Main region around Frankfurt Airport. Light- green: areas for 2011-2013, blue: areas added in 2013.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
Study region of the sub-study "Cognitive Development and Quality of Life of Children"
Addresses of participants Primary schools
Figure 4-5: Distribution of primary schools and residential addresses of the participating schoolchildren in the study region of the module "Cognitive Development and Quality of Life of Children" in the Rhine-Main region
4.3.3 Recruitment of participants, sampling in the sub-studies
4.3.3.1 Sampling in the sub-study "Annoyance and quality of life"
In the sub-study "Annoyance and quality of life" surveys were carried out of adult persons living in the environs of the investigated airports. The surveys were carried out as telephone interviews or, optionally, online. The equivalent continuous sound level and the maximum sound level for the air, rail and road traffic noise at the address of each respondent were calculated.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
For the various questions of the study, several random samples were taken from the respective adult population within the study regions in the environs of the investigated airports. The random sampling was carried out in layers according to the noise levels of the respective traffic noise source concerned. For this purpose, the residential areas within the respective study regions were divided into noise level classes – depending on the question for air, rail and road traffic – and a random sample of adult persons taken on the basis of resident registration data.
For the comparison of the air traffic noise effect before and after the opening of the North- West runway at Frankfurt Airport, a panel study was carried out which involved multiple surveys of participating residents in the years 2011 to 2013 (Panel Study Rhine/Main).
- For this purpose the study region in the Rhine-Main region was divided into 2.5-dB classes of the maximum of the day and night equivalent continuous sound level for
aviation noise LpAeq,06-22h and LpAeq,22-06h >= 40 dB. The basis for this were flight movement figures from the data registration system of Frankfurt Airport from 2007 (DES 2007). - In addition to this, changes in the prognosis levels for the aviation noise levels for the year 2020 compared with the reference year 2007 were taken into account and a distinction made between three prognosis groups: residential addresses with a
predicted increase by more than 2 dB in the day-night level Ltn, with a decrease by
more than 2 dB in the Ltn and with relative constancy of the predicted noise levels
Ltn in a range of +/- 2dB relative to the reference year 2007. The day-night level Ltn is a 24-hour equivalent continuous sound level with a level supplement of 10 dB for the night period from 22:00 to 06:00 hrs. It was used in order to be able to take changes during the night period into account to a greater extent in the 24-hour observation with a single measure. At 8 hours, the night periods is shorter than the 16-hour period from 06:00 to 22:00 hours. - Then, from the three prognosis groups, a random sampling of adult persons was drawn within each of the noise level classes calculated from DES 2007. - The telephone numbers of these randomly selected persons were researched. The selected persons were also sent letters informing them about the study and, where no telephone number was registered, requesting them to provide one. The pool of registered and reported telephone numbers formed the total selection for the telephone interviews. Where participants chose the optional online survey with the same content, they were contacted and sent the access data via their given e-mail address.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
For the comparison of the aviation noise effect between various airports, a similar procedure was chosen for the random sampling at the other airport locations:
- The study regions at the airports Berlin-Schönefeld, Cologne/Bonn and Stuttgart were divided into classes of the maximum of the day and night equivalent continuous sound level for air traffic noise in the range >= 40 dB and a random sampling drawn of adult persons per level class on the basis of resident registration data. For the level class division at Berlin-Schönefeld Airport, which was to be expanded as Berlin- Brandenburg Airport and was scheduled to open at the start of the NORAH Study in 2012, noise levels on the basis of the prognosis DES for the reference year 2015 were used. The level values for the random sampling at Cologne/Bonn Airport were based on the DES of the airport for the reference year 2010.
- At Stuttgart Airport the residential addresses were drawn in layers according to noise level classes (on the basis of the DES with reference year 2010) with subsequent allocation of households to the addresses on the basis of telephone registers as, at the time of the sampling, no resident registration data were available. Within the households reachable by telephone, the survey participant was selected according to a random procedure.
- Just as in the Rhine-Main region, at Berlin-Schönefeld Airport changes in the air
traffic noise exposure (Ltn) were also taken into account in the sampling: the most up-to-date flight movement numbers from 2010 at the time of sampling (DES 2010) at Berlin-Schönefeld regional airport were used for the address-specific calculation of
Ltn. On the basis of the difference of these values to the prognosis levels, the residential addresses were then grouped according to expected increase, decrease and according to unexpected changes (over +/- 2 dB).
Further surveys were carried out in 2012 in the Rhine-Main region in the sub-study "Annoyance and quality of life", focussing on the other traffic noise sources and their combination with aviation noise. This included surveys on
a) road traffic noise
b) rail traffic noise
c) combination of air and road traffic noise
d) combination of air and rail traffic noise
For the cross-section (CS) studies on rail and road traffic noise, those residential addresses were included at which the rail and road traffic noise exposure was >= 40 dB in the maximum of the LpAeq,06-22h and of the LpAeq,22-06h. These residential addresses were grouped according to level classes in increments of 2.5 dB and a random sampling carried out per level class on the
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
basis of the available resident registration data. In these random samples care was also taken that the road or rail traffic noises were dominant in relation to the respective other two traffic noise sources, i.e. lie in a higher 2.5-dB class of the 24-hour equivalent sound level. With the 24-hour equivalent continuous sound level a level measure was selected for this dominance definition with which the source-typical differences in the day/night distribution are emphasized as little as possible (e.g. the night equivalent continuous sound level for rail traffic noise is often higher than the daytime equivalent continuous sound level and thus differs systematically from road traffic noise).
NORAH Noise-related annoyance, cognition and health study
Methodology NORAH General Overview
Methodology
NORAH General Overview
In the supplementary surveys on the effect of aviation noise exposure in combination with noise from another traffic noise source, additional random sampling was made of persons whose residential address is exposed to the same noise level by the respective two traffic noise sources, i.e. the noise levels of both sources lie in the same 2.5-dB class of the 24-hour equivalent sound level. The combination of the various survey data on air, rail and road traffic noises collected in 2012 in the Rhine-Main region should then allow an analysis of the effect of multiple exposure to air, rail and road traffic noise depending on the exposure level and the dominance of the traffic noises involved in the overall noise exposure.
Table 4-1 shows the number of persons selected and questioned for the sub-study "Annoyance and quality of life". The final report on the sub-study (Schreckenberg et al. 2015) shows the process of participant recruitment, the achieved response rate and a non- responder analysis.
Table 4-1. Participation in the sub-study "Annoyance and quality of life" Sample ContactedSurvey Pool yearof tel. Respon- nos. or e- dents 2011 2012mail 2013 Contac- Pool of Respon- addresses Contac- Pool of Respon- ted tel. nos. dents ted tel. nos. dents or e-mail or e-mail ad- ad- dresses dresses FRA Panel 169,460 65,230 9,244 -- -- 3,508 -- -- 4,867 (7%/17%#) 3,172 FRA Road 16,862 18,295 (24%) FRA Rail 3,307 17,653 19,082 (24%) FRA subsequent recruits 13,292 13,292 2,400 FRA air- (24%) road 342 2,109 2,109 FRA air- (22%) rail 292 1,452 1,452 BER (26%) 5,548 21,127 36,374 CGN (31%) 12,937 12,937 2,955 STR (29%) 13,015 13,015 1,979 (20%) Note: airports: FRA = Frankfurt Airport, BER = Berlin-Schönefeld, CGN = Cologne/Bonn, STR = Stuttgart. Percentages in brackets: response rate.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
4.3.3.2 Data acquisition in the sub-study "Secondary data-based case-control study with detailed survey"
The study region included the government district of Darmstadt as well as the Rhine-Hessen cities of Mainz and Worms and the rural districts Mainz-Bingen and Alzey-Worms. The study population was made up of all insured persons over 40 years old (N = 1,026,658) with three major statutory health insurance companies in the study region, corresponding to around 23.3% of the over-forty residents of the study region. On the basis of the accounting and prescription data of the participating health insurers for the report years 2005-2010, those insured persons in whom any of the above-listed disorders was newly diagnosed between 2005 and 2010 were allocated to a case group. Separately for the case groups myocardial infarction (N = 19,632), stroke (N = 25,495), heart failure (N = 104,145), depressive episodes (N = 77,295) and breast cancer (in women: N= 6,643), those insured persons in whom the respective disorder was not diagnosed in the stated period were allocated to the respective control group.
The precise, address-specific exposure to road, rail and air traffic noise was calculated for all of the insured persons surveyed. Odds ratios (OR) were calculated as risk estimates with the aid of logistical regression analysis, adjusted for age, gender, education and profession (from the social security number) and regional SGB II rate (basic social security rate according to the German Social Security Code 2). In the statistical analyses, the equivalent continuous sound levels for the individual traffic noise sources were used both as continuous values and grouped in 5-dB classes. Cases and control persons with a traffic noise equivalent continuous sound level <40 dB were allocated to the reference category; persons with an air traffic equivalent continuous sound level <40 dB and a nocturnal maximum level of 50 dB or more were allocated to a separate group. In addition to the categorized evaluations, the exposure-risk relationship between the equivalent continuous sound levels was examined as a continuous variable with a linear model and with a 3rd degree polynomial. In the connection between cardiovascular disorders and traffic noise- related indoor and outdoor levels.
Table 4-2: Participation in the detailed survey of the case-control study
Health insurer Health insurer Health insurer Total I II III Number of insured persons 122,812 15,856 15,400 154,068 contacted by letter
Number of participating insured 5,689 1,181 1,670 8,540 persons
Response in % 7.4 4.6 10.8 5.5
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
4.3.3.3 Sampling in the sub-study "Blood pressure monitoring"
The blood pressure monitoring involved mainly persons surveyed in the panel study "Annoyance and quality of life" who had agreed in the basic interviews to participate in the monitoring. Target persons of another random sample were drawn from the population data set of the study region in the Rhine-Main region (resident registration data combined with address-specific day and night equivalent continuous sound levels calculated for the reference year 2007). Both partial random samples were drawn subject to consideration of the air traffic noise exposure of the reference year 2007, i.e. it is a sample layered according to air traffic noise levels.
In total, 3,272 persons were contacted regarding participation in the study. This was made in three waves: The partial samples 1 and 2 (PS1, PS2) were drawn within the framework of the panel study of the module "Annoyance and quality of life" at different times (March 2012 and April 2013), and a further random sample (ZS) was contacted directly by the IHU. The willingness to participate lay between 18.6% (ZS) and 45.8% (PS1) or 52.3% (PS2) with reference in each case to the number of persons contacted. The blood pressure monitoring was carried out in two observation periods (OP); observation period 1 (OP1) took place from July 2012 to June 2013, observation period 2 (OP2) from July 2013 to June 2014. Of the 3,272 persons contacted, 1,371 took part in OP1, and 1,095 thereof in OP2.
Table 4-3: Participation in the blood pressure monitoring (total)
Contacted Participants Participants OP1 OP2 Persons drawn in layers according 3,202 1,301 1,043 to air traffic noise levels ("target persons") Self-selected persons 70 70 52
Total 3,272 1,371 1,095
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
4.3.3.4 Sampling in the sleep study
Similar to the procedure in the sub-study "Annoyance and quality of life", a panel study was carried out in the Rhine-Main region, i.e. a group of selected persons was investigated in 2011 before the start-up of the North-West runway and introduction of the night flight curfew from 23:00 to 05:00 hrs, then in 2012, one year later, and again in 2013, the second year after the introduction of the night flight curfew.
In the selected investigation areas in the Rhine-Main region (see Section 4.3.2), potential participants were addressed by information leaflets sent by post, advertisements in local newspapers, the NORAH website, the websites and Facebook pages of local communities and, sometimes, (in 2013) on radio.
A comprehensives preselection was then made from the potential participants which was mainly designed to ensure that the participants had normal hearing and did not suffer from any chronic disorders that would interfere with sleep, i.e. "healthy sleepers", so that the effects of nocturnal air traffic noise on sleep could be investigated with as little distortion as possible. The selection criteria were as follows:
- residential areas with air traffic noise as the dominant noise source, few other sounds that can be perceived in the bedroom - no chronic disorders that would interfere with sleep - duration of residence: at least one year at the investigation site - age: at least 18 - no regular consumption of sleeping pills - no drug/alcohol abuse - no shift work - no hearing difficulties - no small children in the home - no use of noise-protection aids such as ear plugs
This preselection was made on the basis of general surveys and medical history surveys. These were followed in 2011 and 2102 by night-time trial measurements (acoustic measurement and pulse oximetry) and visits to the homes in which, for example, the indoor noise and noise penetrating the bedroom from the outside, sound insulation and other specific features of the home and family were recorded, and the characteristic symptoms for heavy snoring or apnoea were registered.
In the years from 2011 to 2013 various groups of persons (bedtime groups) were recruited in this way for participation:
- Bedtime group 1 was recruited for participation in 2011 as a panel sample to investigate sleep disturbances by air traffic noise as well as the influence of the night flight curfew from 23:00 to 05:00 hrs introduced with the start-up of the North-West
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview runway in October 2011. The participants of this group agreed that they would go to bed between 22:00 and 22:30 hrs and get up between 06:00 and 06:30 hrs. - Bedtime group 2 was recruited in 2012 and differs from the first group in that the sleeping time is shifted by one hour, i.e. from 23:00 - 23:30 hrs to 07:00 - 07:30 hrs. While group 1 was affected both in the going-to-sleep phase and in the waking-up phase by nocturnal flight movements in the shoulder hours at either end of the curfew time (22:00 – 23:00 hrs and 05:00 - 06:00 hrs), group 2 only perceived flight movements in the waking-up phase, but then for longer. - The third group was formed in 2013 with newly recruited participants. On the basis of the data sets from 2011 and 2012, the NORAH team, in collaboration with scientists at the University of Pennsylvania, developed and validated a simplified method for the investigation of the influence of air traffic noise on physiologically measured, noise-related reactions in sleep (registration of vegetative-motor reactions by combined measurement of the heart frequency acceleration by ECG and the body movements by actimeter). This method makes it possible to carry out the investigation on a larger number of persons and was used in the third participant group.
A total of 202 adult persons with healthy sleeping habits between the ages of 18 and 78 were examined in the study from 2011 to 2013. In 2011 49 persons and in 2012 83 persons were examined using the complicated method of polysomnography (PSG) over three nights each time. In terms of the number of participants, the NORAH sleep study is thus the largest survey ever carried out worldwide on the acute effects of air traffic noise on the sleep of residents measured by polysomnography. In 2013, 187 persons were again examined over three nights with a less complicated method for the detection of vegetative-motor reactions in sleep. 39 persons took part in all three years, 36 in two years. In 2011 the participants went to bed at 22:00-22:30 hrs and got up at 06:00-06:30 hrs (bedtime group 1), in 2012 and 2013 another group with the sleeping time shifted by one hour later (bedtime group 2) was examined.
4.3.3.5 Sampling in the sub-study "Cognitive development and quality of life of children"
In the sub-study on the cognitive development and quality of life of children, primary schools were selected within the investigation area in the Rhine-Main region delineated by the 40-dB contour of the daytime equivalent continuous sound level which differ in terms of the air traffic noise exposures, but are as similar as possible in terms of other factors (e.g. sociodemographic factors, German-language proficiency of the children). Schools exposed mainly to other noise sources (road, rail, construction work etc.) were not included. At the time of the surveys the participating children were in second grade.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
The investigation area was divided into 5-dB classes of the air traffic equivalent continuous sound level for daytime from 06:00 to 22:00 hrs: 40 to below 45 dB, 45 to below 50 dB, 50 to below 55 dB, 55 dB and higher. For the preselection, all 297 primary schools in the investigation area were contacted and the principals requested to complete a questionnaire. The questionnaire included questions on, among other things, the school and the catchment area, the size and composition of the second grade classes, the migration background of the schoolchildren (for a well balanced selection of schools) as well as on the exposures of the school to different noise sources. Questionnaires returned by 160 primary schools could be taken into consideration for recruitment. The further selection of the school from the
individual LpAeq,06-22h level classes was carried out in a matching procedure which attempts to achieve the best possible comparability of the primary schools in the various level classes according to the following criteria:
1. proportion of second-grade pupils (if not available, of the pupils in the whole school) with a migration background 2. proportion of second-grade pupils with poor German-language skills (this proportion could not be taken into consideration for schools which did not have the corresponding data) 3. no very high exposure to other noise sources (road, rail, construction sites) 4. number of second-grade pupils per school > 40 5. assessment of the social status in the catchment area of the school 6. broad spatial distribution of the selected schools in the investigation area 7. a largely positive assessment of the importance of the study by the school
Within the framework of the matching procedure, 29 primary schools were then selected that fulfilled the above criteria. The selected schools and the parents of the second-grade pupils were provided with written information material on the study, and information events were held with the teachers and the parents (parent-teacher evenings). The letters to the parents were translated into different languages according to the place of origin of the parents. Alongside information on the study, the letters to the parents also contained a parent questionnaire and a declaration of consent to be completed for the participation of the children in the study. 1,309 parents (77.3%) signed the declaration of consent. Ultimately, as a result of moves, illnesses and individual revocations of declarations of consent, a total of 1,246 primary school children aged 7 to 10 took part. The data for three integration children were not used due to existing development issues so that the data from 1,243 children were eventually used for the data analysis.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
Alongside the parents and children, teachers in the participating schools were questioned about the lessons, the atmosphere in the class and possible problems due to noise. Table 4-4 shows the participation of schools, teachers, parents and schoolchildren.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
Table 4-4: Participation in the sub-study "Cognitive Development and Quality of Life of Children"
Sample Number Schools 29 Teachers 85 Parents contacted 1,694 Parents agreed to participate 1,309 Parents questionnaire 1,185 Children participation 1,246 Children data used 1,243
4.4 Investigation methods in the sub-studies
The investigation methods used included questionnaires (all sub-studies), special person- specific measurements (blood pressure monitoring, sleep study and child study) as well as secondary data analyses (case-control study).
4.4.1 Module 1: "Annoyance and quality of life"
Most of the surveys in this module were carried out by telephone, but also online. They were carried out mainly in the residential population in the Rhine-Main region, and also at three other airports which differed from each other, among other things, in the number of flight movements and their distribution over the day. Another distinguishing feature was that two of the airports (Cologne/Bonn, Stuttgart) are steady-state airports where there were no scheduled or announced changes in flight operations, while the other two airports (Frankfurt, Berlin-Schönefeld) are so-called "change airports" which were being expanded or about to be expanded or newly opened. The concept of the investigation plan was based on methods of quasi-experimental study design (Cook & Campbell, 1979; Shadish & Cook, 2009). By layering random samplings according to acoustic conditions at various airport locations and with reference to various traffic noise sources (see Section 4.3.3.1), (exposure) groups to be compared with each other were defined and the observed noise effects investigated and compared. Precise, address-specific data on the acoustic exposure to air, road and rail noise were available to determine the exposure-effect curves.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
4.4.2 Secondary data-based case-control study
The secondary data-based case-control study with detailed survey examined the effects of traffic-related noise on the health of the residential population in the Rhine-Main region on the basis of routine data from statutory health insurers. A precise address-specific allocation of the exposure to air, rail and road traffic noise was made for all of the insured persons.
The concept of the case-control study was to compare the occurrence and the exposure level to air, road and rail traffic noise in persons who had fallen ill ("cases") and in persons in whom the relevant disorders had not been diagnosed in the study period 2005-2010 ("controls"), taking into account any further possible influencing factors, and to calculate the relative health risks (odds ratios). Insurance data for insured persons aged at least 40 from three health insurers in the government district of Darmstadt were used for this purpose (see Section 4.3.3.2).
This secondary data-based case-control study was complemented by a detailed written survey of selected insured persons of the participating health insurers. This was to gain information on individual, ascertainable risk factors and confounder variables in order to examine their influence on the effect of traffic noise on the health risks. The survey also allowed the investigation of the effects of traffic noise-related indoor noise levels versus outdoor noise levels as the indoor levels could be estimated from the outdoor traffic noise levels on the basis of acoustically relevant survey data (see 4.3.3.2). Although indoor levels appear biologically more relevant than outdoor levels, they are subject to greater uncertainties than the calculated outdoor levels.
4.4.3 Blood pressure monitoring
In the blood pressure monitoring, adult persons (from 18 years) in the area around Frankfurt Airport measured their own blood pressure every morning and evening over three weeks. In addition to this there was also a personal survey on current health condition, lifestyle and subjective factors (e.g. noise sensitivity). The study participants were given telemedical blood pressure measuring devices to carry out their own blood pressure measurements and trained in the use of the devices. With the telemedical sets, consisting of upper arm blood pressure measurement device and mobile phone, the measurement values (systolic/diastolic, heart frequency) are transferred to the mobile phone immediately after the measurement and from there via a secure connection to the data station, where the incoming measurements were regularly checked.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
The acoustic exposure to air, rail and road traffic noise was precisely allocated to the
participants on an address-specific basis. The combined evening and night level LpAeq,18-06h for all traffic noise sources with reference to outdoors and indoors was used primarily for the evaluations.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
4.4.4 Sleep study
In order to attribute wake-up reactions and changes in other sleep parameters to the air traffic noise, this study examined persons with healthy sleeping habits. These sleep parameters describe so-called "primary effects". It is not possible to generate any causal links with possible long-term health consequences from the results. It is possible, however, to use the results to derive plausible hypotheses for the occurrence of disease profiles in persons chronically exposed to noise which can then be tested in epidemiological longitudinal studies.
The sleep investigations took place in three phases, partly with different methods: in the years 2011 – before the start-up of the North-West runway and the introduction of the night flight curfew in October 2011 – and 2012, adult residents with healthy sleeping habits in the environs of Frankfurt Airport were examined on three successive nights by polysomnography, i.e. the recording of brain activity (EEG), muscle tensions, oxygen-saturation of the blood, eye, breathing and heart activities (ECG). At the same time, body movement data were recorded with the aid of a motion sensor. In 2013 a simpler method for the measurement of vegetative-motor physical reactions (ECG + body movement) was used. Both measurement methods were supported by the continuous recording of the sound pressure level and the recording of the noises in the bedroom with the aid of a calibratable sound level meter (Class 1) to ensure the clear identification of the overflight noises. In addition to the objective noise measurements, the participants assessed their own subjective strain by air traffic noise on a daily basis.
Table 4-5: Measurement periods for the NORAH sleep study in the years 2011 – 2013
Year Measurement period 2011 25 July – 8 October 2012 21 May – 9 November 2013 17 June – 28 November
As polysomnographic methods require that all study participants are wired at night and "de- wired" in the morning by a trained (and strange) person, it was only possible to take care of and examine just a few persons at a time.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
In 2013, instead of polysomnography, a method was used for the measurement of vegetative-motor reactions which allows the study persons to attach two electrodes to their own bodies and operate a small device for the measurement of the ECG and the body movement. With the aid of this technology, it is only possible to register vegetative-motor physical reactions, but these parameters are possibly relevant for the occurrence of cardiovascular diseases after many years of nocturnal noise exposure. With the same study budget a much larger number of participands can be examined than with the polysomnography method.
NORAH Noise-related annoyance, cognition and health study Methodology NORAH General Overview
4.4.5 Child study
In spring 2012 performance surveys were carried out in 85 school classes at 29 primary schools with different levels of air traffic noise exposure in the Rhine-Main region as well as surveys of the children, parents and teachers. The performance measurements included in particular the standardized reading comprehension test for first to sixth grade pupils, tests on listening comprehension, on phonological short-term memory, on phonological awareness and speech perception as well as on non-linguistic skills. The auditive test items (syllables, words, sentences) were supplied by a wireless headset system to ensure that all of the children could hear the test items without interference. The surveys of the children and the questionnaires for the parents referred to the school- related, physical and mental wellbeing of the children as well as the school-related and address-related annoyance by noise. Influencing factors not related to noise, in particular social status, migration background and German language proficiency of the children as well as the exposure to other noise sources (road and rail noise, noise insulation and room acoustics in the classrooms) were controlled with the greatest possible care. At the time of the survey the children were at the end of the second grade. For all of the children, the air traffic noise levels at the school locations and at the specific residential addresses were provided for the period of 12 months before the start of the data collection.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
5 Central results
It must first be stated that the unweighted, source-specific, 24-hour equivalent continuous
sound level (LpAeq,24h) with reference to the loudest façade of the address of the participants served as the acoustic basis for many exposure-effect relationships in the NORAH project, usually with reference to a prolonged period (e.g. 12 months), often supplemented by so- called time slices (e.g. day, evening, night or school times) and in some cases by maximum sound levels and frequencies of loud air traffic noises (NAT values). Knowledgeable readers
may ask why the Lden (day/evening/night level) used in many European laws and administration regulations was not taken as the basis for the exposure-effect calculations.
The answer is that the Lden already contains a daytime weighting, i.e. it already assumes that the noise effect is stronger in the evening than during the day, and stronger again at night. That may often be the case, but, from a scientific point of view, has to be tested from case to case.
In the various NORAH sub-studies, the equivalent continuous sound level at the façade with reference to the residential address (or school in the case of the child module) in the Rhine- Main region varied to a greater or lesser degree depending on the noise source:
- With reference to air traffic between o 35.0 and 61 dB (LpAeq,24h, module "Annoyance and quality of life") o 35.0 and >60 dB (LpAeq,24h, "Case-control study") o 30.7 and 56.9 dB (LpAeq,18-06h, "Blood pressure monitoring") o 39.1 and 58.9 dB (LpAeq,06-18h, "Child study") - With reference to road traffic between o 35.0 and 81.7 dB (LpAeq,24h, module "Annoyance and quality of life") o 35.0 and >70 dB (LpAeq,24h, "Case-control study") o 30.0 and 72.3 dB (LpAeq,18-06h, "Blood pressure monitoring") o 36.5 and 61.5 dB (LpAeq,06-18h, classified in 8 levels, "Child study") - With reference to rail traffic between o 35.0 and 81.8 dB (LpAeq,24h, module "Annoyance and quality of life") o 35.0 and >70 dB (LpAeq,24h, "Case-control study" o 30.0 and 77.0 dB (LpAeq,18-06h, "Blood pressure monitoring") o 31.5 and 61.5 dB (LpAeq,06-18h, classified in 3 levels, "Child study") It is clear that road and rail traffic noise levels in the NORAH studies varied more than the air traffic noise levels. This is not an effect of the random sampling, but corresponds rather to the residential situation in Central Europe's population centres (see European Environmental Agency, 2014).
The following results generally do not state any effect size because the statistical models in which the effects of confounders were also taken into consideration do not allow this.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
Where effect sizes are stated (e.g. in the case of annoyance data), they refer to non- adjusted models.
5.1 Annoyance and quality of life
5.1.1 The main results on Annoyance and quality of life
For the various random samples of the sub-study "Annoyance and quality of life" the
investigated range in the exposure to air traffic noise during the day (LpAeq,06-22h)) and with
reference to a 24-hour day (LpAeq,24h) was mainly between ca. 35 dB and 62 dB, and at night
(LpAeq,22-06h) between less than 35 dB to ca. 57 dB – see Table 5-1. It should be noted here that the 24-hour equivalent continuous sound levels with reference to air traffic at Frankfurt Airport were in some cases considerably lower than the corresponding daytime levels due to the relatively low proportion of nocturnal flights, but relatively few of the people surveyed had 24-hour equivalent continuous sound levels <40 or >60 dB. At Cologne/Bonn Airport the maximum equivalent sound levels were just under 10 dB. Overall, the noise level range at Frankfurt Airport has remained about the same since the start-up of the North-West runway. This is due to the fact that there were noise reductions as well as additional exposures, and for around three quarters of those surveyed in the Rhine/Main panel study the changes in the equivalent sound levels (day or night) in 2012 compared with 2011 lie in the range of +/- 2 dB.
The noise exposure due to road and rail traffic covered a larger range than that due to air traffic. For the noise exposure to road and rail traffic, the range in the sub-study "Annoyance and quality of life" covered equivalent sound levels from less than 35 dB up to around 80 dB. For rail traffic the day and night differences were considerably less than for the other types of traffic; in some cases the nocturnal noise exposures were higher than during the day. There was a significant statistical connection – expressed in correlation values (see below) – between the equivalent sound levels of the traffic noises and the assessments of the persons surveyed of their source-specific noise annoyance and the reported traffic noise-related sleep disturbances. The noise annoyance generally correlated to a somewhat greater extent with the 24-hour equivalent sound levels and with the daytime equivalent sound level than with the night-time level. This applies for all traffic noise types. The Cologne/Bonn Airport is an exception; here the air traffic noise annoyance correlated more strongly with the night-time than with the daytime equivalent sound level. This may be possibly due to the relatively highest proportion of nocturnal flight movements at Cologne/Bonn Airport (30%) compared with the other airports investigated (Berlin-Schönefeld: 12%, Frankfurt: 7 – 10%, Stuttgart: 7%).
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
Correlation coefficients between two data series describe their linear statistical connection. They can assume values between -1 and +1, whereby negative values indicate that the increase in one variable means a reduction in the other variable, while positive values signal a rise/fall in the same direction. The closer the value is to zero, the smaller the linear statistical connection. The correlation coefficients between noise annoyance and acoustic
exposure (LpAeq,24h, LpAeq,06-22h and LpAeq,22-06h) lie between 0.35 and 0.59, and thus indicate, according to Cohen (1988), medium to strong effects of the acoustic exposure. The correlation coefficients between reported sleep disturbances and acoustic exposures lay between 0.23 and 0.44; the effect sizes according to Cohen (1988) are medium.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
Table 5-1. Descriptive statistics for traffic noise exposure in the sub-study "Annoyance and quality of life".
Equivalent sound level Correlation r with ... Descriptive statistics location (dB) Noise annoyance Sleep M SD Min Max disturbances Air traffic LpAeq,24h FRA 2011 0.473 0.376 47.8 6.1 35.8 61.0 LpAeq,06-22h FRA 2011 0.470 0.369 49.1 6.2 36.6 62.2 FRA 2011 0.414 0.400 41.9 6.2 56.7 LpAeq,22-06h ≤ 35.0 n = 9,243 n = 9,232 n = 9,244
LpAeq,24h FRA 2013 0.466 0.393 47.2 6.3 ≤ 35.0 60.3 LpAeq,06-22h FRA 2013 0.466 0.393 48.6 6.3 35.9 61.7 FRA 2013 0.425 0.364 41.6 5.9 54.5 LpAeq,22-06h ≤ 35.0 n = 3,508 n = 3,505 n = 3,508
LpAeq,24h BER 0,501 0.442 42.9 6.4 ≤ 35.0 59.3 LpAeq,06-22h BER 0,502 0.441 43.9 6.7 ≤ 35.0 60.6 BER 0.481 0.434 39.8 5.1 54.7 LpAeq,22-06h ≤ 35.0 n = 5,548 n = 5,537 n = 5,548
LpAeq,24h CGN 0.419 0.343 46.5 7.1 ≤ 35.0 72.6
LpAeq,06-22h CGN 0.409 0.334 46.4 7.2 ≤ 35.0 74.4 CGN 0.420 0.363 46.3 7.2 65.7 LpAeq,22-06h ≤ 35.0 n = 2,954 n = 2,951 n = 2,955
LpAeq,24h STR 0.591 0.429 43.9 7.5 ≤ 35.0 61.0 LpAeq,06-22h STR 0.594 0.428 45.2 7.7 ≤ 35.0 62.4 STR 0.555 0.415 38.8 4.6 53.8 LpAeq,22-06h ≤ 35.0 n = 1,979 n = 1,975 n = 1,979 Road traffic
LpAeq,24h FRA 0.353 0.227 57.4 9.6 < 35.0 81.7 LpAeq,06-22h FRA 0.353 0.227 58.9 9.6 36.5 83.3 FRA 0.348 0.232 50.9 9.5 < 35.0 73.4 LpAeq,22-06h n = 3,172 n = 3,162 n = 3,172 Rail traffic
LpAeq,24h FRA 0.432 0.307 58.3 8.7 35.0 81.8 LpAeq,06-22h FRA 0.430 0.303 57.8 8.7 35.0 81.3 FRA 0.428 0.308 58.8 9.0 35.0 82.6 LpAeq,22-06h n = 3,307 n = 3,267 n = 3,307
Legend: The exceeding probability value for all correlation values p < 0.001. The abbreviations FRA, CGN, BER and STR correspond to the IATA code for the airports Frankfurt, Cologne/Bonn, Berlin-Schönefeld and Stuttgart
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
Overall, the correlations of the reported sleep disturbances with the equivalent sound levels were lower than those of the noise annoyance; in some cases the reported sleep disturbances correlated somewhat more strongly with the daytime than with the night-time level (Frankfurt 2013, Berlin-Schönefeld, Stuttgart with reference to air traffic noise) – see Table 5-1. Both suggest that it is difficult for respondents to make an integrative assessment in retrospect about concrete disturbances during the sleeping times (sleep onset, deep sleep, sleeping through the night). In most cases, supplementary analyses (not represented here in tables) showed that the reported sleep disturbances correlated at least equally with the nocturnal maximum level, in some cases somewhat more strongly than the equivalent sound levels. This does not apply to the noise annoyance. This could indicate that nocturnal disturbances due to individual noise events with higher levels are more easily remembered and it is easier to make an assessment of them. The correlations with the number of flight movements were all lower for the noise annoyance and the sleep disturbances so that it can be assumed that, at least for the reported sleep disturbances in this study, the height of the noise levels is more relevant than the frequency of noisy events.
We can also reach this conclusion if we compare the correlations for the traffic noise source types: the respective correlations of the noise annoyance and the reported sleep disturbances with the noise levels were lower for road traffic than for air and rail traffic. Road traffic noise differs from air and rail traffic noise in that there is generally a higher volume of traffic with relatively moderate noise levels from the individual vehicles. Another possible reason for the lower correlation with road traffic is that the noise annoyance and disturbance assessments can be influenced by other factors related to road traffic, e.g. traffic-related air pollution, which are not covered by this study. In the surveyed personal and situation-related context factors, there were no abnormalities that would explain the lower correlation level for road traffic.
With regard to the change over time of the aviation noise annoyance in the first (2012) and second (2013) year after the start-up of the North-West runway compared with the situation previously (2011), there was a slight shift of the exposure-effect relationship for the proportion of persons subject to high levels of noise annoyance (%HA share) in 2012, in particular in lower/middle equivalent sound level ranges up to 50 dB LpAeq,06-22h – see Fig. 5-1. In the NORAH project, persons were defined as "highly annoyed" (HA) who chose one of the two uppermost steps on the 5-step annoyance scale. The exposure-effect curve for 2012 is higher, i.e. at the same daytime equivalent sound level LpAeq,06-22h the proportion of highly annoyed persons was higher than in 2011. In the level ranges from 40 to 45 dB, the difference was 5%, at ca. 50 dB 3%, in higher level ranges the difference was close to 0%. The %HA curve from 2013 lies between that of 2011 and 2012. The shift towards a higher proportion of annoyed persons in the context of expansion projects is known in the research literature on noise impacts and is referred to as "change effect" or "excess response" (inter alia Brown & van Kamp, 2009).
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview In short, this is an effect where, in cases in which the noise situation changes due to interventions (e.g. traffic-calming measures or, conversely, expansion projects), the noise- related annoyance changes "excessively" in the direction of the noise change.
FRA 2011 to 2013 are data of the NORAH Study before the North-West runway was opened (2011), in the 1st (2012) and 2nd year (2013) thereafter. For the air traffic noise from the RDF Annoyance Study (Schreckenberg et al. 2010) two versions of newly calculated equivalent sound levels were used: according to AzB-08 on the basis of the data collection system (DES) 2005 and on the basis of STANLY radar data on flight movements from the year 2005. The diagram shows the model-based curves of the exposure-effect relationships including 95% confidence intervals. The exposure-effect relationships are based on logistic regressions of the HA proportion on the air traffic daytime equivalent levels LpAeq,06-22h (the newly calculated noise levels for the RDF annoyance study were calculated separately for day and night; there are no newly calculated 24-hour equivalent levels.)
Figure 5-1. Proportion of highly annoyed (HA) persons in comparison (model-based representation). Basis: Data of the NORAH Study for the years 2011 to 2013 versus RDF Annoyance Study from 2005 (Schreckenberg et al. 2010).
The "change effect" at Frankfurt Airport was generally relatively low (corresponds roughly to
3 dB LpAeq,24h at 25% highly annoyed), which roughly corresponds to the results of a longitudinal study at Amsterdam-Schiphol Airport (Breugelmans et al. 2007; Houthuijs et al. 2012) before and after the fifth runway went into operation. It must also be taken into consideration, however, a) that at Frankfurt Airport there were already changes in flight operations as of March 2011 and b) that in the course of the start-up of the fourth runway there were local increases and reductions of the noise burden. In-depth longitudinal analyses have shown that the surveyed residents with local noise reductions responded "excessively" with lower noise-related annoyance and, in the case of an increase in exposure, with higher noise-related annoyance.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
Alongside the change in the sound level, this was influenced in particular by
- the individually assessed ability to cope with noise, i.e. a person's own assessment of how well in general he/she deals with noise situations - the attitude towards aviation and - the expectations regarding the effects of future flight operations on the person's own living situation.
But a noticeable aspect of Figure 5-1 concerns not only the differences in the proportion of highly annoyed persons between 2011, 2012 and 2013, but the considerably larger differences of these three curves to the results of the annoyance study carried out in 2005 at Frankfurt Airport (so-called. "RDF Annoyance Study" commissioned by the Regional Dialogue Forum Frankfurt Airport (RDF)). For the comparison with the HA curves of the NORAH Study, the exposure-effect curve of the RDF Annoyance Study was newly calculated: sound levels were recalculated with the same calculation methods as in the NORAH Study (AzB 2008); the HA definitions used for the comparison in the RDF Annoyance Study and in the NORAH Study are identical. The "RDF curve" was already higher than the generalized exposure-effect curve used within the framework of the EU environmental noise directive for aviation noise-related annoyance by Miedema & Oudshoorn, 2001, but at about the same level as other European exposure-effect relationships on air traffic noise annoyance (Schreckenberg et al., 2010). Various authors discuss the fact that the overall air traffic noise-related annoyance at similar equivalent sound levels increased over the course of time (inter alia Gjestland et al. 2015; Guski, 2004; Janssen & Vos, 2009). It is still unclear whether this is due to the fact that, in recent years, in the European area, airports were mainly examined that were in a change situation, or that the air traffic increased while technology brought about a reduction in the noise levels of individual aircraft, and/or whether it is due to the limitations of the equivalent sound level in adequately representing the effect-relevant acoustic characteristics of the aviation noise exposure in terms of the annoyance.
The comparison of the exposure-effect curves for air traffic noise annoyance at Frankfurt Airport and Berlin-Schönefeld, Cologne/Bonn and Stuttgart airports (Figure 5-2) suggests, however, that the change situation at Frankfurt Airport alone cannot explain the changes in time. It was shown that despite all the differences between the investigated airports, at all airports the proportion of highly annoyed persons – with reference here to the 24-hour equivalent sound level - is higher than in the RDF Annoyance Study, and thus also higher than in the generalized exposure-effect curve by Miedema & Oudshoorn, 2001.
Figure 5-2 shows that the model-based exposure-effect curves differ in terms of the increase between the airports. These differences reflect the different correlation, i.e. the variably close connection between the 24-hour equivalent sound level and the air traffic noise-related annoyance. The correlation is strongest at Stuttgart Airport (associated with a steeper curve) and lowest at Cologne/Bonn Airport (flatter curve).
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
Both airports differ significantly in terms of the proportion of nocturnal flights (30% in Cologne/Bonn versus 7% in Stuttgart) and slightly in the level range of exposure (<35 -
>65 dB in Cologne/Bonn vs. <35 – 60 dB LpAeq,24h in Stuttgart).
The statistical analyses showed that, alongside the equivalent sound level, other – non- acoustic – factors influenced the annoyance. This included at all airports the individual noise sensitivity and the attitude towards aviation in general (as useful or as damaging to the environment). The results at Stuttgart Airport differ from those at the other airports to the extent that the non-acoustic factors have a generally lower influence on the noise-related annoyance. At 111,585, the total number of flight movements at Stuttgart Airport, on which the 12-month noise levels are based, was around a quarter of the number of flight movements at Frankfurt Airport (473,231 flight movements in the same period), but cannot explain the difference to Cologne/Bonn Airport (115,805 flight movements, on which the calculations were based). It is possible that the different mix of traffic at the airports is a reason for the differences in the annoyance levels, but this cannot be tested on the basis of the available data. It can be confirmed that a conclusive, exhaustive explanation of the precise causes for the difference in the proportion of highly annoyed residents is not possible on the basis of the available data, and it can only be established that in general, if at all, the exposure-effect curve for noise-related annoyance at an airport can only be transferred to other airports within broader tolerance limits.
Models adjusted according to survey mode, gender, age, residence duration, home ownership, SWI, migration, noise sensitivity, source-specific evaluation of the traffic as useful, convenient, damaging to the environment, LpAeq,24h – road traffic and rail traffic, interaction mode * age. Samples: Berlin-Schönefeld (BER): n = 5,548, Cologne/Bonn (CGN): n = 2,955, Stuttgart (STR): n = 1,979, Frankfurt (FRA): 3,508.
Figure 5-2. Model-based proportion of persons highly annoyed by air traffic noise (%HA), airports Berlin- Schönefeld, Cologne/Bonn, Stuttgart vs. Frankfurt 2013.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
If we compare the proportion of persons highly annoyed by air traffic noise with the proportions of persons highly annoyed by road and rail traffic noise, we see a clear difference between the source types (Figure 5-3). At the same 24-hour equivalent sound levels, substantially more people were highly annoyed by air traffic noise than by road or rail traffic noise. This difference between the source types confirms findings known for decades in noise annoyance research even though the extent of the difference between air traffic and the other traffic types is very high in the NORAH studies. The higher proportion of persons annoyed by air traffic compared with road and rail traffic was shown in the secondary analyses by Miedema & Vos (1998) and Miedema & Oudshoorn (2001), in which more than 50 original studies with around 54,000 participants (the numbers differ somewhat in the two publications) were processed.
The difference, however, between road and rail traffic is less pronounced and varied than, for example, in the secondary analyses by Miedema and colleagues or in original studies comparing road and rail traffic (e.g. Möhler et al. 2000). There it was shown that rail traffic was less of an annoyance than road traffic. In the NORAH Study this effect was shown to a lower extent below LpAeq,24h = 60 dB. Between 60 and 65 dB the curves intersect, and rail traffic at LpAeq,24h values greater than 65 dB proved to be a greater annoyance compared with road traffic. Lercher, de Greve, Botteldooren and Rüdisser (2008) report a similar result in the investigation of the effect of rail and road traffic noises in the Alpine region (Wipptal). Their study showed that rail traffic caused less annoyance than road traffic; above day- evening-night levels Lden from 60 to 65 dB, however, rail traffic causes greater annoyance.
Models adjusted according to survey mode, gender, age, residence duration, home ownership, SWI, migration, noise sensitivity, source-specific evaluation of the traffic as useful, convenient, damaging to the environment, LpAeq,24h of the respective other two traffic noise sources, interaction mode * age. Samples: air: panel participants, survey wave 2012: n = 3,508 (participants who took part in all waves), road: participants of the study "CS Road": n = 3,172, road: participants of the study "CS Rail": n = 3,162
Figure 5-3. Model-based percentage of persons highly annoyed by traffic noise (%HA; highly annoyed) in the Rhine-Main region.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
The lower annoyance due to road and rail traffic noise corresponds to the fact that the noise sensitivity in the road and rail samples is slightly lower than in the panel sample (by 0.1 to 0.2 points on the 5-step response scale), and rail and road traffic tend to be more positively regarded than air traffic, particularly as more useful, while rail traffic is also seen as less damaging to the environment. It is not possible to clarify whether these assessment differences or acoustic aspects such as, for example, a different steepness of the level increase are responsible for the different effects of the traffic noise source types as, apart from the equivalent continuous sound level and the mean maximum level, no other (psycho)acoustic parameters are available.
A higher proportion of persons reporting serious sleep disturbances due to air traffic noise was shown at Frankfurt Airport in 2011 in comparison with the other airports (Figure 5-4). In the second year after the introduction of the night flight curfew from 23:00 to 05:00 hrs, 2013, the exposure-effect curve for the sleep disturbances is shifted downwards and is roughly at the level of the exposure-effect curves of the sleep disturbances reported at Cologne/Bonn airport. The shift of the curve could be an indication of a change effect in a positive direction. This is contradicted, however, by in-depth analyses in which the reported sleep disturbances were examined separately according to sleep onset, deep sleep and REM phase. These show that there is a reduction of the reported sleep disturbances, in particular with reference to the deep sleep phase. Disturbances with reference to the sleep onset phase, however, remained largely unchanged, while the degree of reported disturbances in the REM phase tended to increase. The overall decline in the sleep disturbances would thus be only a reaction to the noise level reduction in the deep sleep phase and not a general "excessive" response in a positive direction.
Models adjusted according to survey mode, gender, age, residence duration, home ownership, SWI, migration, noise sensitivity, source-specific evaluation of the traffic as useful, convenient, damaging to the environment, LpAeq,24h – road traffic and rail traffic, interactions mode * age, attribute "air traffic = useful. Samples: Berlin- Schönefeld (BER): n = 5,548, Cologne/Bonn (CGN): n = 2,955, Stuttgart (STR): n = 1,979, Frankfurt (FRA 2011, 2013): 3,508 (participants who took part in all survey waves from 2011 to 2013). Figure 5-4. Model-based percentage of highly sleep disturbed persons (%HSD), airports Berlin- Schönefeld, Cologne/Bonn, Stuttgart vs. Frankfurt 2011, 2013.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
Models adjusted according to survey mode, gender, age, residence duration, home ownership, SWI, migration, noise sensitivity, source-specific evaluation of the traffic as useful, convenient, damaging to the environment, LpAeq,24h of the respective other two traffic noise sources, interactions mode * age, mode * assessment of traffic = useful. Samples: air: panel participants, survey waves 2011, 2013: n = 3,508 (participants who took part in all waves from 2011 to 2013), road: participants of the study "CS Road": n = 3,172, road: participants of the study "CS Rail": n = 3,162.
Figure 5-5. Model-based percentage of highly sleep disturbed persons (%HSD; highly sleep disturbed) in the Rhine-Main region.
The cross section (CS) study on the overall noise annoyance due to two combined noise sources (air traffic plus road traffic and air traffic plus rail traffic) focussed on three questions:
1. What connection exists between the level of the acoustic exposure (expressed as an energetic addition of the individual equivalent continuous sound level over 12 months) to two sources and the overall noise-related annoyance?
2. What role is played by the level ratio between the two sources for the overall noise- related annoyance?
3. What role is played by personal factors in the overall noise-related annoyance?
With respect to the first question it was shown that although the overall noise-related annoyance at comparable equivalent continuous sound levels significantly increases in both source combinations with the energetically summed total noise level (r = 0.385 for air+rail; r = 0.396 for air+road, i.e. a "medium" effect in each case), the overall annoyance values remained slightly below the individual annoyance due to air traffic noise at energetically equivalent sound levels. The influence of the second source on the overall annoyance was low. It appears that a different reference system applies for the overall annoyance than for the annoyance due to individual sources and that the overall annoyance is strongly influenced by the air traffic noise-related annoyance.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
To examine the second question, three sub-groups of roughly the same size were formed that differed in terms of the acoustic dominance (expressed in the 24-hour equivalent continuous sound level) of the two sources. In group (1) source 1 is dominant (>2.5 dB more than source 2), group (2) has no dominant source, and in group (3) source 2 is dominant (>2.5 dB more than source 1).
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview Note.
Source dominance:
road > air: LpAeq,24h for road traffic is more than 2.5 dB higher than for air traffic,
road = air: LpAeq,24h are identical for road and air traffic in the range +/- 2.5 dB,
air > road: LpAeq,24h for air traffic is more than 2.5 dB higher than for road traffic.
Figure 5-6 Total noise-related annoyance due to air traffic and road traffic in relation to the source dominance.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
It was found that the level dominance in the combination of air plus road traffic noise only plays a decisive role when the road traffic noise level is higher than the air traffic noise level; in this case the total noise annoyance is significantly below the corresponding values for the other two dominance categories. In the combination of air traffic plus rail traffic noise, the total annoyance values for the three dominance categories were more clearly separated, with the dominance of air traffic at the top and the dominance of rail traffic at the lower end, while the values for non-dominance of a source were in the middle.
In the explanation of the variance of the total noise-related annoyance on the sole basis of acoustic variables plus survey mode as a possible confounder, the models with separate individual levels showed the best adaptation compared with the models with energetic sum level plus dominance categories. The model adaptation was substantially improved in both cases when personal factors (e.g. noise sensitivity and source evaluation) were also included.
In the investigation of the connection between quality of life variables and air traffic noise levels it was found that although the two sum scores for quality of life, the mental (MCS) and physical (PCS) quality of life showed slight correlations with the 24-hour equivalent continuous sound level, the influence of the noise exposure, however, is mainly (negatively) conveyed by the air traffic noise-related annoyance. This applies to a greater degree for the mental quality of life than for the physical quality of life, and at all of the investigated airports. In addition to this, sub-group analyses at Frankfurt Airport showed that in the case of an increase of the exposure to air traffic noise after the start-up of the North-West runway, the correlation between the 24-hour equivalent continuous sound level and mental quality of life became greater.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview a
b
Note. SF8-Score MCS: mean value and 95% confidence interval; Germany = standard data (mean value ± 95% confidence interval) of the telephone health survey 2003 (GSTel03) by the Robert Koch Institute (see Ellert et al., 2005). (a) Frankfurt (FRA) 2011 – 2013, (b) Berlin-Schönefeld (BER), Cologne/Bonn (CGN), Stuttgart (STR).
Figure 5-7: Mental quality of life (MCS) grouped according to air traffic noise-related annoyance (a) in the Rhine-Main region and (b) at the comparison airports
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
5.1.2 Summary evaluation of the results of the studies on annoyance and quality of Life
The sub-study confirms a trend in aviation noise-related annoyance that has been described in recent years by several authors, namely, a shift of the exposure-effect relationship, i.e. an increase in the annoyance at the same equivalent sound levels over the course of time (Brooker, 2009; Guski, 2004; Janssen & Vos, 2009; van Kempen & van Kamp, 2005). It was assumed that this is, inter alia, a result of excessive responses due to expansion projects (in the planning stage or after implementation) at the airports that were investigated in recent years. In this study also, this was assumed with reference to the expansion of Frankfurt Airport, specifically with reference to the start-up of the North-West runway. And, in fact, in the first year after the start-up of the new runway, an increase in the noise annoyance compared with 2011 at the same equivalent continuous sound level was established. However, it was also shown that at the other investigated airports the aviation noise annoyance was higher than some years beforehand in the RDF Annoyance Study at Frankfurt Airport in 2005 (Schreckenberg & Meis, 2006). The chronological trend of the increase of the aviation noise annoyance from 2005 to 2011 before the start-up of the North-West runway was stronger than the difference before and after the start-up of the new runway 2011 versus 2012/2013. This increase in annoyance could be a combined effect of expansion planning and actual operational changes as of March 2011, i.e. a combination of change expectations and experienced exposure changes. Both aspects are part of the so-called change effect, i.e. the excessive responses in the change in the noise annoyance in the course of measures-related changes in the noise exposure. This occurs not only after a change in the noise exposure itself, but also already in advance due to changes in expectations as soon as interventions which will have an effect on the noise exposure are announced.
With the start-up of the North-West runway in October 2011, a curfew on night flights from 23:00 to 05:00 hrs was introduced. At first glance, the changes in the reported sleep disturbances in 2012 and 2013 compared with 2011 suggest that the introduction of the curfew has a positive effect on the reported, aviation noise-related sleep disturbances. However, further-going analyses showed that this only applies for the deep sleep phase, but not for the sleep onset and REM phases, which, according to the respondents' own reports lie between 22:00 and 23:00 hrs (most frequently reported time for going to bed) and 06:00 and 07:00 hours (most frequently reported getting-up time).
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
The fact that traffic noise, in particular in terms of noise annoyance, has an effect on the (general) health-related and, in particular, mental quality of life, has also been shown in earlier studies (van Kamp et al., 2007; Dratva et al., 2010; Schreckenberg et al., 2010). Overall, the effect of traffic noise on the quality of life in our sub-study was marginal. However, the result that the negative correlation between the air traffic noise exposure and mental quality of life increases when an increase in the noise exposure has occurred suggests that, in particular, the (negative) change in the air traffic noise exposure in the residential environment is noticed and possibly leads to a perceived loss of control (Hatfield et al., 2002), combined with a deterioration in the mental quality of life. A possible cause could lie in the increase in depressive tendencies (inter alia La Torre et al., 2007; Wallston et al., 1987). With respect to the effect of traffic noise combinations (air plus road or air plus rail traffic) on the total annoyance, it was found that the purely energetic summing up of the two source levels underestimates the combined effect on the annoyance – in particular when the air traffic noise level is below the levels of the respective second source. A better approach appears to be the weighting of the two levels according to the degree of annoyance of the respective source types, whereby it has to be taken into consideration that the average annoyance of the population for both air traffic and rail traffic noise has changed today compared with the so-called EU standard curves (Miedema & Ousdhoorn, 2001).
5.2 Secondary data-based case-control study with detailed survey
The secondary data-based case-control study examined the statistical correlations between the acoustic exposure to road, rail and air traffic noise in the period from 2005 to 2010 and the risks of suffering
- myocardial infarction - stroke - heart failure - breast cancer - episodes of a unipolar depression (here also referred to as “depression”).
A summary overview of the results of the secondary data-based case-control study is given in the following. A detailed representation of the results – as with the other sub-studies of the NORAH project – can be found in the final report of the secondary data-based case- control study. Odds ratios are given in each case as risk estimates for the relative health risks. An odds ratio of 1.05, for example, means that in the corresponding exposure category, the health risk is increased to 1.05 times that of the "non-exposed" persons in the reference category. In other words: the relative health risk compared with non-exposed persons is increased by 5 %. Odds ratios below 1 represent reduced health risks.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
The odds ratios are always statistically significant when the value 1 does not lie within the 95-% confidence interval. The following graphs represent the linear course of the risk per 10 dB increase in the 24-hour equivalent continuous sound level. Insofar as the exposure-risk relationship is adequately represented by a linear model, the course of the risk is represented in the subsequent graphs with a continuous line, supplemented by the 95-% confidence interval. In those cases in which the linear model does not correspond to the optimum degree with the data, the risk estimates are represented on individual level steps with individual confidence intervals. The risk estimates cited in this chapter refer in principle to the model ("fully") adjusted for age, gender and social status (education, profession, regional social welfare recipient rate).
In general, the highest risk increases per 10 dB increase in the equivalent continuous sound level for all traffic noise sources in the NORAH case-control study were found in depressive episodes. It was also shown that the connections between cardiovascular diseases and the noise levels of the road traffic (applies in particular to heart failure) and rail traffic (applies for myocardial infarction and stroke as well as for heart failure) have a more significant trend than those with air traffic.
5.2.1 The main results for the individual diseases
5.2.1.1 Cardiovascular disease risks (myocardial infarction, stroke, heart failure)
In general there was no statistically significant increase in the myocardial infarction per 10
dB level increase (see Figure 5-8). Air traffic noise levels as of 60 dB LpAeq,24h were (statistically not significantly) linked with an increased myocardial infarction risk (odds ratio = 1.42 [95% CI 0.62-3.25]) and an increased stroke risk (odds ratio = 1.62 [95% CI 0.79- 3.34]; see Figure 5-9). When only myocardial infarction patients who were deceased by the year 2014 are included in the analysis, air traffic levels as of 60 dB showed a statistically significant increase in the risk estimate (odds ratio = 2.70 [95% CI 1.08 – 6.74]). In the case of stroke, the risks did not change significantly when the analysis was restricted to the deceased patients. In the case of air traffic noise, there was no positive linear course of the myocardial infarction risk or the stroke risk. In the case of heart failure, however, air traffic (also) showed a statistically significant risk increase by 1.6 % per 10 dB equivalent continuous sound level increase (odds ratio per 10 dB = 1.016 [95% CI 1.003-1.030], see Figure 5-10) in the sense of a linear exposure-risk relationship.
NORAH Noise-related annoyance, cognition and health study
Central Results NORAH Gesamtbetrachtung Central results NORAH General Overview
Notes:
Linear risk change (violet line with shaded 95-% confidence interval):
Total -0.7% per 10 dB (95% CI -3.4% - 2.0%; not sig.)
Men -0.7% pro 10 dB (not sig.) Women -0.7% pro 10 dB (not sig.)
Maximum levels were not considered in the Figure 5-8: Air traffic noise levels representation of the linear risk change (LpAeq,24h) and myocardial Legend:infarction Inheart order attack to be able risks to use the same OR scale for all figures, ORs<0.9 and >1.3 are not shown (OR=1.42 for >60 dB) Notes:
Linear risk change (violet line with shaded 95-% confidence interval):
Total -2.4% per 10 dB (95% CI -4.7% - 0.0%; not sig.)
Men -0.3% per 10 dB (not sig.) Women -4.1% per 10 dB (stat. sig.)
Maximum levels were not considered in the Figure 5-9: Air traffic noise levels representation of the linear risk change
(LpAeq,24h) and stroke risk Legend: In order to be able to use the same OR scale for all figures ORs<0.9 and >1.3 are not shown.
Notes:
Linear risk change (violet line with shaded 95-% confidence interval):
Total 1.6% per 10 dB (95% CI 0.3% - 3.0%; stat. sig.)
Men 1.1% per 10 dB (not sig.) Women 2.1% per 10 dB (stat. sig.)
Maximum levels were not considered in the representation of the linear risk change
Figure 5-10: Air traffic noise levels
(LpAeq,24h) and heart failure risk
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
On the basis of the 24-hour equivalent continuous sound level the correlation between cardiovascular diseases and road traffic noise levels (Figure 5-11 to Figure 5-13; applies in particular for heart failure) and rail traffic levels (Figure 5-14 to Figure 5-16; applies for myocardial infarction and stroke as well as for heart failure) was higher than the correlation with air traffic levels: the risk for a myocardial infarction, stroke or heart failure showed a statistically significant increase predominantly as of road traffic noise levels and rail traffic noise levels of 50 dB, but also in some cases already as of 55 dB, and increased continuously with higher noise levels. It must be pointed out, however, that there was also a significantly increased risk of 14 % for stroke in the case of rail traffic noise level of 45 to <50 dB when the rail-related maximum level was at least 20 dB above this (so-called "emergence analysis", see final report of the case- control study).
[Anmerkung des Übersetzers: Abbildungen 5-11 – 5-16 siehe separate Datei “EN_Grafiken Seite 66_67_68.docx]
5.2.1.2 Consideration of the nocturnal maximum levels in the case of cardiovascular disease risks
As an important new finding, the NORAH case-control study pointed up the importance of the consideration of the nocturnal maximum levels in the estimation of the cardiovascular disease-related air traffic noise effects.: On the one hand, the lowest sound level category of <40 dB included persons who were exposed to nocturnal maximum levels of >50 dB. Increased risk estimates were observed in many analyses for this group. The inclusion of these persons in the reference category would thus led tendentially to an underestimation of the air traffic-related health risks. On the other hand, these – in the case of stroke and heart failure statistically significant (see Figure 5-9 and Figure 5-10) – increased risk estimates indicate that aviation noise-related health risks at air traffic maximum levels over 50 dB can occur even if the equivalent continuous sound level is under 40 dB. This finding has population-specific relevance but must first be backed up scientifically in further studies.
5.2.1.3 Breast cancer risks
In the case-control study a correlation was observed between nocturnal air traffic noise levels and the diagnosis of breast cancer; the correlation reached statistical significance for the time between 23:00 and 05:00 hrs ("mediation night") albeit at low field assignments (odds ratio = 2.98 [95% CI 1.31-6.79]; Figure 5-17). No correlation with the breast cancer risk was detected for road and rail traffic noise.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
Notes:
Linear risk increase (violet line with shaded 95-% confidence interval):
Women 0.3% per 10 dB (95% CI -4.2% - 5.0%; not sig.)
Figure 5-17: Air traffic noise level
(LpAeq,24h) and breast cancer risk for women
Legend: In order to be able to use the same OR scale for all figures, ORs<0.9 and >1.3 are not shown.
5.2.1.4 Depression risks
The NORAH case-control study was able to establish a correlation between all three traffic noise sources (air, road and rail traffic; Fig. 5-20 – 5-22) and the diagnosis of an episode of unipolar depression. The risk increase per 10 dB level increase was greater for air traffic noise at 8.9 % (Figure 5-18) than for road traffic (4.1%; Figure 5-19) and rail traffic (3.9%; Figure 5-20). However, in the case of higher air traffic noise levels and higher rail traffic noise levels, the depression risk estimates sank again (in the sense of a reversed "U" shape), and the linear model does not adequately represent the exposure-risk relationship for air or rail traffic.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
Notes:
Linear (not optimally adjusted) risk increase
Total 8.9% per 10 dB (95% CI 7.4% - 10.4%; stat. sig.)
Men 8.6% per 10 dB (not sig.) Women 9.2% per 10 dB (stat. sig.)
OR is 0.94 (95% CI 0.85-1.05) in the not represented category <40 dB, maximum level Figure 5-18: Air traffic noise level (LpAeq,24h) and ≥50 dB risk of an episode of unipolar depression
Notes:
Linear risk increase (violet line with shaded 95-% confidence interval
Total 4.1% per 10 dB (95% CI 3.2% - 5.0%; stat. sig.)
Men 4.0% per 10 dB (stat. sig.)
Women 4.0% per 10 dB (stat. sig.)
Figure 5-19: Road traffic noise level (LpAeq,24h) and risk of an episode of unipolar depression
Notes:
Linear (not optimally adjusted) risk increase:
Total 3.9% per 10 dB (95% CI 2.9% - 4.9%; stat. sig.)
Men 4.0% per 10 dB (stat. sig.)
Women 3.9% per 10 dB (stat. sig.)
Figure 5-20: Rail traffic noise level (LpAeq,24h) and risk of an episode of unipolar depression
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
5.2.1.5 Distortion of results due to unidentified or residual confounding? Results of the detailed survey
In order to assess a possible distortion of results due to confounders not or only incompletely contained in the health insurance data, 8,540 insured persons were given an exhaustive questionnaire with questions, for example, on their education, profession, income, tobacco consumption, height and weight, alcohol consumption, shift work, noise at work and physical activity. It was only possible, of course, to include insured persons who had not yet died at the time of the contact letter (up to ten years after the incident diagnosis) in the detailed survey. In the case of myocardial infarction and stroke – both diseases with a relatively high mortality rate – the exclusion of the deceased led to a clear increase in the secondary data-based risk estimates so that a considerable selection distortion for myocardial infarction and stroke had to be assumed. This selection distortion was further reinforced to some extent by the low response rate of 6% on average and the differential response behaviour obviously associated with this. In the case of heart failure, however, the selection of the non-deceased insured persons from the secondary data set did not lead to any essential change in the risk estimates: the secondary data-based heart failure risk estimates for the detailed survey respondents did not differ substantially from the heart failure risk estimates for the group as a whole. This is why the detailed survey of the cases with heart failure (but not the cases with myocardial infarction or stroke) and the corresponding control persons could be used to answer the question as to a distortion of the secondary data-based results due to unidentified or residual confounding – this was the primary aim of the detailed survey. As a result, the heart failure risk estimates based solely on the secondary data did not change substantially when an additional adjustment was made for the individual social status (represented by the Scheuch-Winkler Index), tobacco consumption, body-mass index, alcohol consumption, shift work, noise at work and physical activity. Thus, it can be assumed that there is no substantial distortion of the results for heart failure due to inadequate consideration of the social status and lifestyle.
5.2.1.6 Sub-analysis of insured persons with information on the social status available in the health insurance data
In order to check for a possible distortion of the results for myocardial infarction, stroke, depressive episodes and breast cancer, an additional sub-group analysis included only those insured persons in the secondary data-based evaluation for whom there was individual health insurance information regarding social status (education, profession). No substantial change in the effect estimates was established so that it can be assumed that the aforementioned disease groups are not subject to any substantial distortion due to inadequate consideration of the social status as a possible confounder.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview 5.2.1.6 Sub-analysis of insured persons with information on the social status available in the health insurance data
In order to check for a possible distortion of the results for myocardial infarction, stroke, depressive episodes and breast cancer, an additional sub-group analysis included only those insured persons in the secondary data-based evaluation for whom there was individual health insurance information regarding social status (education, profession). No substantial change in the effect estimates was established so that it can be assumed that the aforementioned disease groups are not subject to any substantial distortion due to inadequate consideration of the social status as a possible confounder.
5.2.1.7 Significance of the noise history: constant residential addresses over many years and "cumulative noise years"
The elaborate "reconstruction" of historical noise exposures for the years 1996 to 2010 allowed the consideration of a long-term "constant" residential address as well as – for the first time possible in a traffic noise study – the consideration of so-called "cumulative noise years". The traffic noise-related total exposure over a defined period was used to calculate the "cumulative noise years". A restriction of the analysis to cases with a long-term, "constant" residential address in the case-control study led to tendentially higher risk estimates for several disease profiles: the aviation noise-related risk estimates for a stroke in the highest equivalent continuous sound level category ≥60 dB increased in the case of a residential duration of more than 5 years. For a residential duration of more than 10 years the risk estimates increased further, but did not reach (at low field assignment) statistical significance. The analysis of the cumulative noise years in the last 5 years before the diagnosis year or (in the case of control persons) before the year 2008 showed continuously increasing heart failure risk estimates with increasing cumulative noise years which reached statistical significance for the highest quartile of the cumulative noise years. In the case of breast cancer, the highest equivalent continuous sound level category ≥60 dB showed a clear risk increase to a (statistically not significant) odds ratio of 3.96 for women with a residential duration at the self-reported index address of more than 5 years. Compared with this, the results for air and rail traffic after consideration of the cumulative noise years were inconspicuous. For the depressive episodes there were statistically significant risk estimates in the two highest categories (3rd and 4th quartile) of the air traffic-related noise years.
5.2.2 Summary evaluation of the results of the NORAH case control study
Overall, the results of the secondary data-based case-control study with detailed survey indicate a correlation between the traffic noise exposure and the occurrence of myocardial infarction, stroke, heart failure and depressive episodes.
It was not possible to establish any statistically significant correlation with the equivalent continuous sound levels of the three traffic noise sources for breast cancer in women.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
A statistically significant increase in the breast cancer risk could only be shown for air traffic noise levels as of 55 dB in the time between 23:00 and 05:00 hrs.
With reference to the 10-dB level increase, the highest traffic noise-related health risks were shown for the diagnosis of a depressive episode – with statistical significance for all three traffic noise sources. With regard to the cardiovascular diseases, it is noticeable that the effects of the road and rail traffic on myocardial infarction, stroke and heart failure were more pronounced than those of aviation noise. In the case of road traffic noise, the highest (statistically significant) risk increases per 10 dB were shown for depressive episodes (4.1%), myocardial infarction (2.8%), heart failure (2.4%) and stroke (1.7%). In the case of rail traffic noise, the highest risk increases per 10 dB level increase for depressive episodes amounted to 3.9 % (albeit not adequately represented by a linear model), heart failure 3.1% and stroke 1.8%. In the case of air traffic noise, the highest risk increases per 10 dB level increase were found for depressive episodes (8.9%) (albeit not adequately represented by a linear model) and heart failure (1.6%).
In the comparison of current reviews by Babisch (2014) and Vienneau et al. (2015) on cardiovascular disease risks with the NORAH results, the latter indicate somewhat lower traffic noise-related risk estimates: as a result of the pooled analysis, Babisch (2014) finds a risk increase of 8% per 10 dB increase in the road traffic noise level. Vienneau et al. (2015) find a risk increase of 4% per 10 dB increase in the road traffic noise level (Lden). Per 10 dB increase of the air traffic noise level (Lden) Vienneau et al. find a risk increase of 6%. Compared with this, the risk increase per 10 dB for all traffic noise sources in all investigated incidences of cardiovascular disease in the NORAH case-control study lies below 4 %. Our results on the connection between traffic noise and depressive episodes are basically in harmony with the findings up to now.
Analogously with earlier investigations (e.g. Huss et al., 2010; Floud et al., 2013) the NORAH results indicate the importance of taking the duration of the noise exposure into consideration. The calculation of the "cumulative noise years" represents a promising approach here, which has never before been used in noise impact research. However, the aforementioned risk increases should generally be interpreted with some reserve as it is not possible to distinguish clearly between a "real" effect of a traffic noise exposure over several years and a selection effect (e.g. moving away of noise-sensitive persons).
The NORAH Study went to a lot of trouble to check for any unidentified or inadequately considered confounding: 8,540 persons included in the secondary data-based case- control study took part in a detailed survey. For the largest case group by far of heart failure it was easy to reproduce among these participants the risk estimates of the (solely) secondary data-based analysis. On the basis of the detailed survey it was possible to rule out any significant distortion of the secondary data-based heart failure risk estimates due to social status or lifestyle-related factors.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
So-called emergence analyses are a special feature of the NORAH case-control study. Here the individual classes of the 24-hour equivalent continuous sound pressure levels were divided in two according to the difference to the highest maximum level within the 24 hours: <20 dB difference and ≥20 dB difference. The influence of this emergence was examined for all three types of traffic. Considerably higher risk estimates were established here in particular for the connection between rail traffic noise levels and myocardial infarction in the higher level classes, while the contribution of the emergence for other traffic types and other diseases was not uniform. Nonetheless, the trend of the results indicates the significance of the maximum level for the health risk due to air and rail traffic levels.
Special attention is due to the fact that the highest risk increases per 10 dB level increase for all traffic noise sources were found with respect to episodes of a unipolar depression. This cannot be substantiated by pointing out that depressive episodes occur statistically more frequently than other diseases (total for Germany ca. 8%, for women 10 %, see RKI 2013) and to some extent have a reputation of being just a disorder label for complaints that are otherwise difficult to describe in concrete terms. As the NORAH case-control study is concerned with specific connections with the calculated, address-specific acoustic exposure, the increase of the depressive episodes could also be an expression of the fact that with increasing exposure the persons concerned realize that the problem is not getting better and that they are not capable of doing anything effective against it (loss of control). This assumption is supported by a finding from the sub-project "Annoyance and quality of life" on the sum score of the mental quality of life, which also covers emotional, depressive tendencies. When the noise exposure increased after the opening of the North-West runway at Frankfurt Airport, there was also an increase in the (negative) correlation between air traffic noise level and mental quality of life – this is a variable, which also reflects depressive tendencies. It is possible – particularly after experiencing an increase in the noise exposures – that a loss of control is perceived and the noise seen as reducing the mental quality of life. As the sub-project "Annoyance and quality of life" clearly showed the mediating role of the noise-related annoyance, it should also be examined which statistical correlation there is with the noise annoyance for the episodes of unipolar depression.
General evaluation of the effect size: in line with the scientific literature it can be established that the traffic noise-related cardiovascular disease risks found in the case-control study are substantially lower than the risks for "known" influencing factors such as tobacco consumption or obesity. Nonetheless, a large proportion of the population is exposed to traffic noise levels which, according to the NORAH case-control study, are associated with – albeit low – risk increases for cardiovascular diseases as well as depressive episodes. Due to the population-based frequency of the traffic noise exposures as well as the diseases investigated, population-based significance has to be attributed to even small risk increases.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview 5.3 Blood pressure monitoring
First it was examined to what extent the master samples subsample 1 (TS1), subsample 2 (TS2) and additional sample (ZS) differed from each other in terms of demographic characteristics, the blood pressure measurement values, the health behaviour and the prevalence of certain diseases. It was found that there were large deviations between the samples (in particular between TS1 und TS2), inter alia, in terms of the proportion of women, age, the Scheuch-Winkler Index, the proportion of hypertension sufferers and the average systolic and diastolic blood pressure values. This is why it was decided not to pool the three samples for the evaluation, but to limit the main analyses to TS1 of the original design (N = 844 with 493 women and 351 men) and to only carry out sensitivity analyses on the other samples.
The age of the participants in sample TS1 was between 19 and 82 years (median = 49 years), the average systolic blood pressure (in the morning) was between 91 and 157 mmHg (median = 116 mmHg), the average diastolic blood pressure (in the morning) between 48 and 102 mmHg (median = 72 mmHg).
The statistical models for the analysis of the correlation between traffic noise levels and the target values of average systolic blood pressure measurement value, average diastolic blood pressure measurement value, heart frequency, blood pressure amplitude, hypertension and 10-year heart attack risk (PROCAM score) were adjusted in the sample TS1 for the pre- specified variables of age, gender and Scheuch-Winkler Index as well as for tobacco consumption and physical activity.
These analyses did not show a statistically significant correlation for any of these target values with the air traffic noise level LAeq.18-06h. Although there were slight increases of the systolic and diastolic blood pressure values with increasing noise level (0.9 and 0.7 mmHg respectively at a level increase of 10 dB), these increases were so small that they hardly go beyond the range of unavoidable measurement fluctuations (+/- 5mmHg). The additional analyses stratified according to age provided some indications that older people react more strongly in the blood pressure values to air traffic noise than younger people. Further additional analyses indicated that persons with average noise sensitivity reacted more strongly in the blood pressure values to air traffic noise than persons with higher or lower noise sensitivity, and that hypertension sufferers react somewhat more strongly in the mean diastolic blood pressure to air traffic noise than non-hypertensive persons.
The correlations between the road traffic noise level LpAeq.18-06h and the blood pressure target values were also not statistically significant. The analysis stratified according to age indicated that older persons react more strongly to road traffic noise. The strongest connections were found for the end points of systolic blood pressure, diastolic blood pressure and heart frequency in the participants over 64 years of age.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
The analysis stratified according to noise sensitivity showed that persons with a high noise sensitivity react more strongly to exposure to road traffic noise. Higher risk estimates were identified in this group for systolic and diastolic blood pressure, heart frequency and blood pressure amplitude. In persons with low or medium noise sensitivity, the correlations were predominantly negative.
The correlations between the rail traffic noise level LpAeq.18-06h and the blood pressure target values were also not statistically significant. The biggest effect was shown in systolic blood pressure at 0.8 mm HG per 10 dB level increase. There was also a significant increase of the heart frequency in persons with high noise sensitivity. The stratified analysis according to residential duration indicated that participants who live less than 14 years in the same place react more strongly in systolic and diastolic blood pressure to rail traffic noise than persons with a longer residential duration.
Overall, the results of the blood pressure monitoring – low positive risk estimates without statistical significance – are in line with most of the research up to now, in particular with reference to air traffic noise (see inter alia Huang et al., 2015) and rail traffic noise (see Dratva et al., 2012). Significant correlations between air traffic noise and hypertension as in the HYENA Study (Jarup et al., 2008) were not established in the NORAH Study. A direct, quantitative comparison between the two studies is only possible to a restricted extent due to the differences in methodology.
5.4 Sleep study
The probability of being woken up by an environmental noise depends to a great extent on the maximum level of this noise. The effect of the maximum level also depends, however, on the current background level. The sleep study of the NORAH project measured acute air traffic noise maximum levels and background levels at the ear of the sleeper and an odds ratio of 1.23 was shown for a 10-dB increase in the maximum level – at a mean background level of 28.8 dB (median of the minute level directly before the overflight in the sleep samples 2011 and 2012). This means that, at a background level of 28.8 dB, the odds of waking up increased by 23 % per 10 dB increase in the maximum level of the air traffic event. Similar wake-up probabilities were shown in the years 2011 and 2012.
It was also shown that the introduction of the night flight curfew (23-05 hrs) in autumn 2011 at Frankfurt Airport had different effects on the sleep quality of the study participants. In the sleep measurements in 2011 and 2012 before and after the introduction of the curfew in October 2011 there were no statistically significant changes for the overall sleeping time, the sleep latency, the sleep efficiency, the waking time after falling asleep and the percentage of waking time after 04:30 hrs.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
If the whole sleeping time is observed, the reduction in the number of flight movements led to a lower number of nocturnal, air traffic-related wake-up reactions compared with the basic survey in 2011. This "objective", summarily observed improvement in the sleep quality was not, however, reflected in the assessments of the participants themselves in terms of their tiredness and sleepiness next day – these values tended to increase from 2011 to 2013. This is in line with the results of the Annoyance and quality of life study, according to which, the averaging of the reported sleep onset, deep sleep and REM sleep disturbances results in an overall reduction of the reported sleep disturbances.
On the other hand, results on the reported, long-term disturbances of sleep onset and deep sleep with reference to the previous 12 months in the sub-project "Annoyance and quality of life" showed that, in 2012, the persons concerned reported to a greater extent disturbances in the morning phase of their sleep. Accordingly, the relief brought about in the deep sleep phase is shown in both sub-studies. The lack of relief in the morning hours is more apparent on the longer-term psychological level than on the physiological-acute level. These two levels cannot be directly compared with each other as they are based on different time periods (integrated over 12 months versus acute in a few nights) and different response ranges (verbal versus physiological).
It also appears remarkable that those participants with a negative attitude towards aviation had a statistically significantly extended sleep latency. They spent statistically significantly more time awake in the time between going to sleep and getting up, and their sleep efficiency was significantly lower. Participants with a negative attitude towards aviation and a low to medium assessment of the necessity of aviation spent statistically significantly less time in the deep sleep phase. It was not possible in this study to clarify if there is a causal relationship here, i.e. whether the more disturbed sleep leads to the negative attitude or vice versa. It was not possible to establish any difference between the two groups with reference to the wake-up frequency.
In the comparison of the sleep-related results between the airports Cologne/Bonn and Frankfurt, it was found that a series of sleep quality parameters at the Cologne/Bonn Airport are below those for Frankfurt Airport. The sleep efficiency, total sleeping time and deep sleep time per overall sleeping time were statistically significantly lower in the Cologne/Bonn data set (from the years 2001/2002), while the sleep latency and the sleep fragmentation were statistically significantly increased compared with NORAH 2012. The probability of being woken up by an overflight noise at the same maximum level was higher in the Cologne/Bonn data set than in NORAH 2012 – e.g. at 45 dB(A) maximum level 5.0% higher. Possible reasons for these differences could be the higher nocturnal traffic, the bundling of fights in the times 23:00-01:00 hrs and 03:30 – 05:00 hours and the higher proportion of old aircraft (freight aircraft) in the older study at Cologne/Bonn Airport. It cannot be ruled out, however, that some of the differences are due to the different participant groups, going-to- bed times or evaluators between the two studies.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
These results suggest that, taking into account the above restrictions in the interpretation of the data, the exposure-effect relationship observed in 2001/2002 at Cologne/Bonn airport cannot be transferred with any certainty to Frankfurt Airport with its night flight curfew.
Both in 2011 and in 2012 vegetative-motor reactions occurred at the same maximum level with higher probability than wake-up reactions measured in the EEG. Vegetative-motor reactions thus appear to be another, more sensitive measure for noise-induced disturbances whose observation could be important for the understanding of longer-term health consequences in the future.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
5.5 Cognitive development and quality of life of children
As its main result, the study showed statistically significant negative effects of air traffic noise levels on reading acquisition in a linear exposure-effect relation. The increase of the noise level by 10 dB is associated with a reduction in reading performance by a T-value point (i.e. 1/10 of the standard deviation) (Figure 5-21). This corresponds to a delay in reading development by one month. The difference in the reading performance between the children in the investigation area with the highest and the lowest exposure to air traffic noise corresponds to a delay of two months. This effect is somewhat smaller than other factors that have an influence on learning to read. For example, children who own a lot of books were four months ahead in reading compared with children who did not own their own books. The investigated factors cannot, however be directly compared because they have very different significance for the persons concerned: they can, to a certain extent, influence the degree of reading promotion in the household, but not the level of air traffic noise.
Figure 5-21: Correlation between air traffic-related equivalent continuous sound level (LpAeq,08-14h) and overall reading performance.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
Other important results: - It was not possible to establish any negative effects of the air traffic noise levels on the phonetic skills important for reading acquisition (short and long-term memory, phonological awareness, speech perception). - Although the assessments of the health-related quality of life of the children themselves were generally positive, there were significant negative effects of the air traffic noise levels on the assessments (physical and mental wellbeing of the children according to the parents' assessments, physical wellbeing and sleep quality according to the children's assessments). - While the children's own assessments of their school-related wellbeing were generally positive, it was possible to verify a statistically significant negative effect of the air traffic noise levels on the assessments. - The teacher survey showed a substantial burden on classes due to air traffic noise in the relatively highly exposed schools (LpAeq,08-14h: 55 to 59 dB). The teachers in these schools unanimously assessed the burden on classes due to air traffic noise as high to very high. The assessments were highly correlated with the air traffic noise levels at the schools (r = 0.85). The burden on classes in the highly exposed schools results inter alia from frequent interruptions of the teacher-pupil conversation and noticeable distractions of the children during overflights. In addition to this, most of the teachers from these schools reported that the air traffic noise can be heard often to very often in the classrooms even when the windows are closed.
It is worth noting that the main result of the NORAH child study, i.e. the deterioration of reading performance with increasing air traffic noise exposure, largely agrees with results of the international scientific literature, even though some of the underlying conditions were different: in the RANCH Study (Stansfeld et al., 2005; Clark et al., 2012) the children were on average two years older than the children in the NORAH Study, the air traffic- related noise exposure was up to 18 dB higher than in the NORAH Study, and the tests used are not entirely comparable due to the age difference of the children. Nonetheless, there are similarly large effects on the reading performance (1/10 standard deviation per 10 dB increase in noise exposure) in both studies.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
6 Generalizability of the statements
6.1 To whom do the NORAH statements apply?
As already reported in chapter 4.3, the NORAH team carried out their investigations mainly in the residential population around Frankfurt Airport, but also in the environs of the airports Cologne/Bonn, Stuttgart and Berlin/Schönefeld. They "only" took samples but nonetheless believe that they can make statements that also apply to many persons who were not included in the samples.
To what extent this belief is justified depends on several factors:
1. on the respective population definition, i.e. the population group that is supposed to be represented in the respective sample, 2. the quality of the sample, i.e. the success of the attempt to represent the population concerned. This success depends, among other things, on the willingness of the population to take part in studies, 3. the internal validity of the study results, i.e. the validity for the particular sample in which the data were collected, 4. to what extent it is possible to statistically control those aspects of the sample that could have a distorting influence on the results 5. on the external validity, i.e. similarities and differences between the investigated samples and the populations to which certain sample results are to be transferred or extrapolated.
The individual NORAH sub-studies referred to different populations: the NORAH Child Study, for example, was aimed at investigating children in second grade primary school who were exposed to more or less severe air traffic noise from Frankfurt Airport, but to little road or rail traffic noise. The participants in the sleep study were specifically selected with reference to the nocturnal air traffic noise exposure at Frankfurt Airport and, as far as possible, were not to be exposed to any other traffic noise source. The module "Annoyance and quality of life" had different population targets depending on the sub-study; the primary target in the panel study was the representation as uniformly as possible of the variance of the air traffic-related noise exposure in the adult residential population. The Blood pressure monitoring selected mainly persons from the basic survey of the panel sample for the module "Annoyance and quality of life". The secondary data-based case- control study, on the other hand, was aimed at making population-specific statements on health risks for persons who are living in the Rhine-Main region (government districts Darmstadt and Rheinhessen) and are over 40 years old. It was only in this case that the acoustic exposure to air traffic or other traffic noises did not play any role in the definition of the population.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
If we assume that sample quality and internal validity are secured and possible distortions of the NORAH sub-studies and investigation results are controlled for, we can ask to what extent the NORAH results apply not only for the investigated sample, but also for broader population groups. The answer will differ depending on how tightly the sample is defined and on the research question of the study.
We see, in particular, two restrictions on the extrapolation of the NORAH results:
1. The air traffic-related results refer mainly to Frankfurt Airport. This airport underwent a) in October 2011 substantial operational changes and b) has more flight movements than any other German airport, whereby these flight movements have been restricted to the time between 05:00 and 23:00 hrs. Whether the NORAH results can be applied to Cologne/Bonn Airport with its lower number of flight movements and its high occurrence of nocturnal traffic must be examined from case to case. On the other hand, with the help of the surveys at comparison airports, the module "Annoyance and quality of life" makes it possible to examine similarities and differences between the various airports. With recourse to earlier studies, the sleep study can also make comparisons between Frankfurt and Cologne/Bonn airports. 2. The NORAH evaluations referred mainly to the currently used energy-equivalent continuous sound level as a description of the acoustic exposure. In some cases the maximum level, the number of noise events (NAT values) and the emergence (defined here as the difference between maximum and equivalent continuous sound level) were also included. Although energy-equivalent continuous sound levels are the acoustic variables most frequently used internationally to describe acoustic exposure, it must nonetheless be examined in each case whether a
certain equivalent continuous sound level X (e.g. 50 dB LAeq,24h) with a different distribution of maximum levels and/or event frequencies should be regarded as comparable in terms of the effects.
6.2 What influence to response rates have on the NORAH results?
Several sub-studies within the NORAH joint research project have a relatively low response rate (the lowest rates are 5 – 7%), and the recruitment of participating persons or institutions (e.g. health insurance companies) proved to be complicated and difficult from the start. The achieved low response rates are not unusual nowadays – Miller et al. (2014) also report on 12.1% and 35.1% response rates for a telephone and written survey respectively at three American airports. With respect to the importance of the response rate, the scientific literature has, in the meantime, accepted the finding that different non-response reasons have to be evaluated differently (e.g. non-contact, refusal, inability, other obstacles), whereby in many study designs refusal cannot always be separated from non-contact (see Faulbaum, 2014). Refusal should, where possible, be examined in terms of the danger of selection bias because this is what is most likely to lead to low response rates (see below).
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview The question as to how high the response rate has to be (or the non-response rate can be) in order to allow valid statements is not possible to answer either generally or specifically. The answer depends, among other things, on the research question and on the study design. It must also be established that response rates in themselves are not decisive parameters of the study quality as long as a random sample was taken and the absolute number of study subjects is sufficient in order to a) answer the specific research questions and b) there are enough demographic data in order to statistically control for possible distortions with regard to the desired population at least subsequently (see Curtin et al., 2000). The decisive problem that can arise in connection with low response rates concerns selection distortion, i.e. distortions that occur because the missing data are associated with the content of the investigation. If, for example, the people who do not participate in a noise investigation are those that do not feel affected by noise, this would lead to a distortion of the statements on the noise effects.
6.3 What influence do refusals to participate have on the NORAH results?
If there are indications of a high degree of refusals to participate, it has to be examined whether there is a selective distortion of the sample (see Groves et al., 2004). In the case of an investigation on the effects of environmental noise, it is conceivable that those people who refuse to participate tend to have relatively low acoustic exposure, and thus do not regard the issue of "noise" as important enough to take the trouble of participation.
Breugelmans et al. (2004) followed up this suspicion with a special non-responder analysis: when, in 2002, a survey was carried out, partly in writing and partly by telephone, on residents around Amsterdam's Schiphol Airport, 46% of the persons contacted took part. (That is a relatively high response rate by today's standards; however, it appeared suspicious to the clients at the time). Overall, there was an increase of the response rate at increasing noise exposure. A (mainly telephone) survey was also carried out on 255 non-participants which showed that the "refusers" reported slightly lower annoyance values at similar acoustic exposure than the non-refusers.
A similarly differentiated non-responder analysis was carried out in the NORAH module "Annoyance and quality of life" for the telephone interviews at all four airports: if a person contacted by telephone refused to give the interview, he/she was asked about the reasons and, as of 2012, also about the extent of the annoyance due to the three traffic noise sources. They were also asked to provide demographic details. Of a total of 48,042 non- responders contacted at all airports, most of them gave no reasons, or "neutral" reasons with regard to noise: 30.2% hung up without a word; 39.4% said they "didn't want to/had no interest", 11.6% cited "lack of time", 9.6% "refused interviews in principle". It is not clear what proportion of the non-specified reasons or reasons classified as neutral are noise- related.
NORAH Noise-related annoyance, cognition and health study Central results NORAH General Overview
9.4 % of the non-responders gave reasons that have something to do with noise or the noise study, thereof more than half said that they were not annoyed by noise, but just under one third expressed a resigned attitude towards noise-related annoyance (e.g. "the survey won't change anything"). Even the discussion about the NORAH Study itself in the Rhine-Main region led to refusals and revocations, particularly in the year 2012 (1.7% of the explicitly cited refusal reasons). The relative proportion of the explicit reference to the study as a reason for refusal is low, but it is not known how high the hidden proportion of the noise-specific or NORAH-specific refusals is in the group of refusal reasons classified as "neutral".
Only 6 to 10% of the non-responders answered the questions on annoyance and on the demographic details at the four airport locations. It would not be very reliable to generalize the few complete answers for all non-responders within a noise level class; this is why this analysis is of little practical use.
NORAH Noise-related annoyance, cognition and health study Strengths of the NORAH joint research project NORAH General Overview
7 Strengths of the NORAH joint research project
7.1 General strengths:
The NORAH joint research project is one of the most extensive studies ever carried out on the impact of traffic noise on humans. Its strengths include:
- The entire project is based on a holistic concept that looks at health in a multidimensional manner, i.e. is not restricted to individual effect areas, but includes annoyance, quality of life, cognitive development of children, sleep quality, blood pressure regulation and health risks, and examines them respectively in relation to the strength of the noise exposure due to air, rail and road traffic. - The enormous scope of the acoustic calculations and the necessary effort for data collection, data processing and data maintenance are unique in Germany up to now. The varying quality and up-to-dateness of the input data for the acoustic calculations posed a major challenge, as did the lack of established calculation methods for the maximum levels of road and rail traffic noises. - The accuracy of the calculations, in particular for the individual air traffic noise levels of the participants in all of the sub-studies, has, as far as we know, never been achieved before in any other German study. Within the shortest time at the start of the project a method was developed which solved the problem of the too-short flight paths in the data collection system of Frankfurt Airport as well as the problem of the missing data collection systems for the years from 1996 to 2010 with the aid of radar data. This was made possible by cooperation with the German Air Traffic Control Authority (Deutsche Flugsicherung). - For the first time, all three traffic noise source types (road, rail and air traffic noise) were investigated in parallel in a field study. - For the first time with reference to the calculation of the acoustic exposures in a large field study, uncertainty and reliability observations were made, and the uncertainty of the acoustic values stated for all three traffic noise sources taking into consideration the uncertainties of the input and model parameters. The influence of the uncertainty on the results of the exposure-effect relationship was examined using the Child study as an example. It was shown here that the inclusion of the reliability of the traffic noise exposure in the interdisciplinary evaluations leads to an increase in the overall uncertainty, but has only an insignificant influence on the course of the exposure- effect relationship.
NORAH Noise-related annoyance, cognition and health study Strengths of the NORAH joint research project NORAH General Overview
7.2 Strengths of the study "Annoyance and quality of life"
The module "Annoyance and quality of life" is a combined socio-acoustic longitudinal and cross-section study on the impact of air, rail and road traffic noise in the Rhine-Main region and at three further airport locations on the noise-related annoyance, reported sleep disturbances and health-related quality of life. Its strengths can be summarized as follows: - The panel study at Frankfurt Airport in 2011-2013 made it possible to observe the development of self-reported air traffic noise effects in 2011-2013 in relation to different times and the operational changes at the airport. - The inclusion of further airport locations (multicentric study design) allowed statements on the generalizability of the results on air traffic noise effects beyond an individual airport in comparison with other airport regions with different flight movement numbers, different distributions of flight movements over the day and different contexts (airport undergoing changes versus existing, established airport). - The data collected was used to investigate a large bandwidth of individual questions, including the noise impact contexts o when various acoustic characteristic values are used (equivalent continuous level, maximum level, number of movements); o when a differentiated observation is made of various times of the day, which allowed, among other things, an evaluation of the effect of restrictions on operation time (night flight curfew); o when (non-acoustic) context factors are taken into account. - Alongside the air traffic area, other traffic noise source types (road and rail traffic) were taken into account in the evaluations not merely for the statistical control of possible confounder effects. Rather, especially executed cross section surveys made it possible to examine the effects of traffic noise from all sources with the same methodology and to compare the results. - Comparisons were also made between different traffic noise source combinations (air plus road traffic, air plus rail traffic). - The use of standardized questionnaires with internationally used questions and answer scales (e.g. registration of annoyance in accordance with the recommendations of ICBEN/ISO TS 15666) makes it easier to compare the results with other international studies. - With the extensive psychometric examination of the surveyed question complexes on the basis of the cross section and longitudinal data, a differentiated picture was produced of the quality of the collection instruments in terms of reliability (accuracy), validity and objectivity of the data collected, also with respect to the stability of the quality over time.
NORAH Noise-related annoyance, cognition and health study Strengths of the NORAH joint research project NORAH General Overview
- The longitudinal data were used to carry out extensive, in-depth analyses on the links between noise level, potential co-determinants (other influencing factors) and the observed target values (noise effects) to allow better differentiation of o moderating factors (moderators), o between noise levels and influencing factors that mediate noise effect variables (mediators) and o variables regarded more as a secondary reaction. This reduces the risk of increasing the explanation strength (explained variance) of exposure-effect relationships simply by treating the consequences of the observed noise effects as their influencing factors on the basis of associations, because a correlation alone does not tell us anything about the causality or the direction of an effect. The robustness of the calculated exposure-effect relationships is regarded as high, even against the background of relatively low response rates. This assessment is based on the results of extensive sensitivity analyses and the use of bootstrapping (i.e. the frequent repetition of statistical analyses with different sub-samples from the same sample) in central exposure-effect models.
7.3 Strengths of the secondary data-based case-control study
The strengths of the secondary data-based case-control study include: - Both inpatient and outpatient diagnoses were taken into consideration. - As far as possible, incident (newly occurring) diagnoses were taken into consideration - Individual social status parameters were taken into consideration. - The sensitivity analysis taking into consideration only those participants in the secondary data-based analysis for whom the individual social status was known (about one quarter) clearly rules out any significant residual confounding by the social status. - The consideration of the Scheuch-Winkler Index from the detailed survey led to no significant change in the heart failure effect estimates. - Individual lifestyle parameters were taken into consideration (for heart failure). - After additional adjustment for tobacco consumption, BMI, alcohol consumption, shift work, noise at work, physical activity (in the fully adjusted model), there was no substantial change in the risk estimates compared with the basic model. - Maximum level analyses were carried out for all of the observed traffic noise sources, in particular for the air traffic noise during the night.
- Emergence analyses were carried out which observe the LpAeq,24h taking into consideration
the difference to the maximum level LpAmax,24h.
NORAH Noise-related annoyance, cognition and health study Strengths of the NORAH joint research project NORAH General Overview
- Residential history information was available for policy holders with a health insurance company and participants in the detailed survey. The duration of the chronic noise exposures ("residence years", "noise years") was taken into consideration here. - It was possible to compare mortality risks with morbidity risks.
7.4 Strengths of the blood pressure monitoring
- Project workers visited the participants at home and showed them how to measure their own blood pressure so that they could carry out the blood pressure measurements themselves every morning and evening over a period of three weeks. - During this instruction, a health survey was also carried out. Questions were asked regarding chronic illnesses and current state of health. Any medication being taken was registered and waist measurements were carried out. - The repeated self-measurement of the blood pressure values at home and their telemedical transmission to the study centre ensured reliable blood pressure values. - In addition to this, data was collected on noise sensitivity, noise-related annoyance due to the three traffic types and on health-relevant lifestyle aspects (physical activity, alcohol consumption, smoking) and on social economic status, and used in the final models. - The formulations and content of the questions corresponded to the current, nationwide health survey by the Robert Koch Institute. - Additional clinical parameters were registered in accordance with the questionnaire of the DETECT Study. - The scope and the implementation of the blood pressure monitoring were substantially more extensive than previous population-based studies on the subject of noise and blood pressure, which were restricted in terms of both target and exposure parameters.
7.5 Strengths of the sleep study
- This was the largest ever series of field studies on residents around an airport accustomed to air traffic noise and with otherwise healthy sleeping habits using the elaborate electro-encephalography method. - The noise exposure was determined very precisely in these field studies by continuous measurements at the ear of the sleeper. - The simultaneous registration of "objective" and "subjective" sleep parameters allowed a comparison of different methods for the description of sleep quality.
NORAH Noise-related annoyance, cognition and health study Strengths of the NORAH joint research project NORAH General Overview - The consideration of the attitudes of the participants towards air traffic made it possible to examine the question as to what extent such attitudes are related to "objective" sleep parameters. - The development of a vegetative-motor method (VMM) made it possible to work on larger samples. As the method measures heart-frequency accelerations, it could possibly be used to identify a mechanism which could cause cardiovascular diseases due to long-term nocturnal noise exposure. The VMM cannot, however, deliver any results on the change in the sleep structure, or precisely represent air traffic noise- associated wake-up reactions as is possible with the polysomnography method.
7.6 Strengths of the child study
- Previous studies on the effects of air traffic noise on children often failed to take sufficient account of other factors that can influence reading acquisition such as family background, language skills of the children and school-related factors. The major education studies such as PISA and IGLU show, however, how strong the influence of these factors is on the learning success of the children, most particularly on reading. The NORAH Study took account of these factors with the greatest possible precision, and filtered them out statistically in order to be able to determine the influence of the air traffic noises as precisely as possible. - The test material was adapted especially for the German language, and in some cases even newly developed. - A priori matching was successfully executed.
NORAH Noise-related annoyance, cognition and health study Strengths of the NORAH joint research project NORAH General Overview
8 Challenges for the NORAH joint research project
One problem was the low response rate for all of the modules except the Child study. Although the willingness to participate in scientific studies has also declined in other research areas, this has rarely been seen to the same extent as in the Frankfurt region.
At first glance, the difficulties in participant recruitment or the low response rate in the sub- study "Annoyance and quality of life", the detailed survey for the case-control study in the blood pressure monitoring and in the sleep study appear to correspond with each other. The comparison is tenuous, however, because the demands on the potential participants differed from study to study. It was only possible (to a limited extent) in the sub-study "Annoyance and quality of life" to ascertain motives for the non-participation, and it was shown here that mainly general, non-noise-related reasons for refusal were given. Insofar as noise-related reasons were given, either there was a lack of concern about noise or a sense of resignation (e.g. "the noise situation won't change") that played a role in the decision not to participate. Discussions in the Rhine-Main region on the NORAH Study itself may also have had an influence on the decision to participate.
Although selection effects cannot be entirely ruled out, it was at least possible in the module "Annoyance and quality of life" with the aid of sensitivity analyses to show that the represented exposure-effect relationships are robust with respect to potential sample effects.
Another problem was the varying availability of acoustic parameters for the three traffic noise sources (rail, road and air traffic): while the NORAH project had access in the case of air traffic to, for example, reliable maximum levels and event frequencies alongside the usual equivalent continuous sound levels, there were no certain calculation methods for road and rail traffic. New estimation methods had to be developed for this. This somewhat restricted the comparability of the study results between the three traffic noise source types in terms of analyses going beyond the equivalent continuous sound levels.
NORAH Noise-related annoyance, cognition and health study Conclusion NORAH General Overview
9 Conclusion
The NORAH joint research project is one of the largest research projects ever undertaken in Germany on the subject of noise impacts. Even in the international comparison it is a very large project, and similar, subject to some restrictions, to the "Health Impact Assessment“ (see Staatsen et al., 2004) carried out 1992-2005 before and after the opening of the fifth runway in Amsterdam-Schiphol. The expectations prevailing in the state of Hessen and in Germany as a whole are accordingly high and diverse. Even though it is clear that no single project can fulfil all of the possible demands, the NORAH team is certain that they have made a major contribution towards our understanding of the effects of air, road and rail traffic noise. In terms of their objectives and research questions, the studies are more extensive and more diverse than earlier investigations, and the care taken in planning, execution and analysis is (also thanks to the Scientific Advisory Board for Quality Assurance) exemplary. Special credit is also due to the care with which the acoustic exposures to the three traffic types were calculated on a precise, address-specific basis for all of the participants of all of the sub-studies (with the exception of the sleep study, which had its own acoustic measurements). NORAH covered more than the four largest areas in which the World Health Organization (WHO 2011) sees potential for the impairment of health due to environmental noise: sleep disturbances, annoyance, cardiovascular diseases and cognitive development of children. The NORAH Study added breast cancer and depression as "end points". A wide range of questions was processed in each area, including, in particular, the question as to the exposure-effect relationship between the level of the noise exposures and the level of the effect. If practical conclusions are to be drawn from a study, it is not sufficient to simply establish that a noise source, e.g. road traffic, is statistically significantly correlated, for example, with a certain health risk or severe annoyance. To answer the question as to whether there is a causal relationship between acoustic exposure and impairment or risk, and to implement the findings in the form of recommendations for action, an exposure- effect relationship is necessary. In the first case, the assumption of a causality between noise exposure and possible effect gains weight when it is established that the effect increases as the exposure increases. In the second case, these findings are necessary in order to be able to estimate to what extent the noise exposure has to be reduced in order to prevent or at least mitigate the effect. The exposure-effect relationships in the NORAH project offer information on the level and form of the statistical correlation between the acoustic exposure and its possible effects. They do not provide any information on the "lost healthy life years", let alone on "noise- related fatalities". They can tell us, however, whether there is actually a statistically significant noise effect and what happens on the effect side as the acoustic exposure increases. We do not want to try here to compare the various effects shown in the NORAH sub-studies with each other. This is due, on the one hand, to the varying frequency of the occurrence of the effects in the population and, on the other hand, their respective
NORAH Noise-related annoyance, cognition and health study Conclusion NORAH General Overview
significance in terms of content. But we can try to identify and evaluate the clearest effects in the various theme areas, whereby this evaluation can never be entirely objective.
Very strong effects of traffic noise occurred in the area of chronic annoyance – for all three types of traffic. This was most pronounced in the case of air traffic, followed at a far distance by rail and road traffic. Although this is in line with the international literature, the level of air traffic-related annoyance at Frankfurt Airport was extreme compared with the road and rail-traffic noise annoyance. The percentage of persons highly annoyed by air traffic noise around Frankfurt Airport before the opening of the North-West runway was already higher at similar equivalent sound levels than in 2005 at the same airport, and in 2012 it was higher than at several other German airports. It rose again somewhat after the opening of the runway and then fell slightly in 2013. In 2013 at 50 dB (24-hour equivalent continuous sound level) it amounted to around 57 % - compared with around 5 – 8% for rail and road traffic. The especially low level of annoyance due to rail and road traffic noise compared with air traffic may be due to the fact that the investigation areas were concentrated on airport regions in which there is greater focus on exposure to air traffic noise than there is outside of airport regions. The results for noise-related annoyance in the case of multiple exposures in the Rhine- Main region showed that the noise exposure due to air traffic dominated the total annoyance over the whole investigated equivalent sound level range. This result corresponds to the general expectations because air traffic noise is generally more annoying than road or rail traffic noise. However, the absolute level of annoyance due to air traffic noise at Frankfurt Airport is not directly transferrable to other airports, any more than the absolute level of annoyance due to road or rail traffic noise. For example, the annoyance due to rail noise at highly exposed railway lines such as those in the Middle Rhine valley is higher than in the NORAH Study (Schreckenberg, 2013). Nonetheless, this is still lower than the annoyance caused by air traffic noise. Even if we take into account that the air traffic-related annoyance at similar levels today is generally higher than it was around 20 years ago (see Janssen & Vos, 2009), the Frankfurt value for 2013 is higher than expected. At a 24-hour level of 50 dB, the proportion of highly annoyed persons lies between 40 and 55 percent, depending on the investigated airport. If we consider that in 2012 343,316 persons in the environs of Frankfurt Airport were exposed to air traffic noise levels >=50 dB LpAeq,24h, we would say that this is a result that must be taken very seriously. Even if annoyance is not generally as much of a threat in terms of health as the risk of disease – in terms of the number of persons affected, this is a serious result. It must also be considered that the German noise abatement act (German parliament 2007) only provides for noise abatement measures as of 60 dB daily level for existing airports. Picking out just a few of the many results on traffic noise-related health risks may appear presumptuous or perhaps arbitrary, but the report for the secondary data-based case- control study made it clear that it is not only air traffic that played the dominant role
NORAH Noise-related annoyance, cognition and health study Conclusion NORAH General Overview
everywhere as the cause or "amplifier" of diseases, but that air, rail and road traffic alternate depending on the disease area. Taking into consideration the relevant confounders, each of the three types of traffic held statistically significant risks in terms of their noise – in the case of unipolar depression all three. In the case of air traffic noise in the investigation period 2005-2010, the highest statistically significant risk increases per 10- dB level increase were found for episodes of unipolar depression (8.9%) und heart failure (1.6%). In the case of road traffic noise, the highest corresponding risk increases were shown for depressive episodes (4.1%), for stroke (2.8%) and for heart failure (2.4%). In the case of rail traffic noise, the highest corresponding risk increases were shown for depressive episodes (3.9%), myoradial infarction (3.8% including the deceased), heart failure (3.1%) and stroke (1.8%). It was not possible to clarify during the term of the project whether there is a connection between diagnosed depressive episodes and chronic annoyance as a result of the traffic noise exposure. It did become clear, however, that both health-specific end points (annoyance and depression) have strong statistical correlations with the 24-hour equivalent continuous sound levels of all three traffic noise source types, whereby the highest correlation was found in each case with air traffic. This raises the question as to which connection there is between noise-related annoyance and depression. Both end points refer to a person's subjective mental state; in the first case with direct reference to noise and in the second case independently of noise. If, however, there is a statistical correlation between noise exposures and depressive episodes, it is obvious to suspect a mutual influence between noise annoyance and depression. This was suggested for the first time by Tarnopolsky et al. (1980), and connected by Van Kamp (1990) with noise coping strategies, but not systematically investigated afterwards. Under certain circumstances, the self-assessed mental quality of life, which appears to depend on the noise-related annoyance, also plays a role. As increased traffic noise-related health risks were expected in the project planning, particularly in the area of cardiovascular diseases, it was a logical step to look for statistically significant effects in the area of blood pressure regulation. This was done within the framework of the blood pressure study, but it was not possible to establish any statistically significant increase in the blood pressure at rest in connection with increasing traffic noise exposures. Consistent tendencies of the increase were shown, however, in the case of air and rail traffic noises. This result does not mean that the blood pressure regulation is not affected by chronic exposure to traffic noise, but only that the increase of the average blood pressure values at increasing air and rail traffic noise levels was very weak and did not reach the conventional significance limit. The fact that in this sub-study rail traffic noise had a similarly noticeable effect as in the case-control study is remarkable, and may be due to the special characteristics of these noises (high maximum levels, high level-increase speed). The results of the sleep study are especially important for the evaluation of the "night flight curfew" (23:00-05:00 hrs) at Frankfurt Airport. Advantages and disadvantages were identified here: if we observe the total sleeping time, the reduction of the number of flight
NORAH Noise-related annoyance, cognition and health study Conclusion NORAH General Overview
movements brings about an improvement in sleep compared with the basic survey in 2011. In the module "Annoyance and quality of life" a parallel result was established: the longer- term overall assessments of the persons concerned regarding their own sleep disturbances decreased as of 2012. In the individual questions regarding sleep onset, deep sleep and REM sleep disturbances, there was a slight improvement in sleep onset disturbances, a clear improvement in the case of deep sleep disturbances, but an increase in REM sleep disturbances after the introduction of the night flight curfew (see chapter 5.4 for the discussion). The main result of the Child study, the statistically significant reduction in reading performance due to air traffic noise, was published, and has already led to a decision on the part of the Hessen government to improve noise insulation in schools that are subject to high exposures to air traffic noise. If we look at the commonalities and differences of the NORAH main results with reference to the three traffic noise sources, we notice that the strongest annoyance effects and effects on diagnosed unipolar depressive episodes were found in the Frankfurt region for air traffic, but the strongest somatic effects (cardiovascular disease risks and blood pressure monitoring) for rail and road traffic. Although the designs of the respective sub-studies are different, we can establish that the three noise source types were given equal consideration – at least in terms of the equivalent sound level - in these studies. The observed difference is difficult to interpret. It suggests, on the one hand, that the different effect areas respond differently to different traffic noise source types, but on the other hand, that the various traffic noise source types have different strengths of effect at the same equivalent continuous sound level. If the first interpretation applies, it becomes difficult to maintain a common noise impact theory for different effect areas. If the second interpretation applies, it appears necessary to evaluate the different traffic noise source types differently. With respect to the different effects of the three traffic noise source types, the NORAH results indicate the necessity of re-evaluating the effects of rail traffic noise. In contrast to the commonly held view that the noise of rail traffic at the same equivalent continuous sound level causes less annoyance than road traffic noise, the cross- section comparison in the module "Annoyance and quality of life" showed that there is hardly any difference between rail and road traffic noises in terms of their long-term annoyance and that, at higher equivalent sound levels, rail traffic noise can be even more of an annoyance than road traffic noise. There were also some remarkable results for cardiovascular disease risks with reference to rail traffic. The causes of this are not yet fully understood, but there is also a suspicion here that high maximum levels and steep level rises as a train drives by play an important role.
NORAH Noise-related annoyance, cognition and health study Literature NORAH General Overview
References
Babisch, W. (2014). Updated exposure-response relationship between road traffic noise and coronary heart diseases: A meta-analysis. Noise & Health, 16(68), 1-9. Basner, M., Buess, H., Elmenhorst, D., Gehrlich, A., Luks, N., Maaß, N., Mawet, L., Müller, E-W., Müller, U. Plath, G, Quehl, J., Rey, E., Samel, A., Schulze, M., Vejvoda, M. 0 Wenzel, J. (2004). Nachtfluglärmwirkungen (Band 1): Zusammenfassung. Köln: DLR. Berglund, B. 0 Lindvall, T. (Eds., 1995). Community Noise. Stockholm: Archives of the Center for Sensory Research. Bodin, T., Björk, J., Öhrström, E., Ardö, J. 0 Albin, M. (2012). Survey context and question wording affects self reported annoyance due to road traffic noise: a comparison between two cross-sectional studies. Environmental Health, 11(14). Breugelmans, O., Houthuijs, D., Kamp, I.v., Stellato, R., Wiechen, C.v. 0 Doornbos, G. (2007). Longitudinal effects of a sudden change in aircraft noise exposure on annoyance and sleep disturbance around Amsterdam airport. Proceedings of the 19th International Congress on Acoustics (ICA 2007), Paper env-04-002, Madrid, Spain: http://www.sea- acustica.es/WEBICA07/fchrs/papers/env-04-002.pdf. Breugelmans, O. R. P., van Wiechen, C. M. A. G., van Kamp, I., Heisterkamp, S. H. 0 Houthuijs, D. (2004). Gezondheid en beleving van de omgevingskwaliteit in de regio Schiphol: 2002. Tussenrapportage Monitoring Gezondheidskundige Evaluatie Schiphol [Health and quality of life near Amsterdam Schiphol airport: 2002. Interim report] (Vol. Report 630100001, pp. 128). Bilthoven (NL): RIVM. Brooker, P. (2009). Do people react more strongly to aircraft noise today than in the past? Applied Acoustics, 70, 747 -752 Brown, A.L. 0 van Kamp, I. (2009). Response to a change in transport noise exposure: Competing explanations of change effects. Journal of the Acoustical Society of America, 125 (2), 905-914 Clark, C., Crombie, R., Head, J., van Kamp, I., van Kempen, E. 0 Stansfeld, S. A. (2012). Does Traffic-related Air Pollution Explain Associations of Aircraft and Road Traffic Noise Exposure on Children’s Health and Cognition? A Secondary Analysis of the United Kingdom Sample From the RANCH Project. American Journal of Epidemiology, 176(4), 327–337. Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale (USA, NJ): Lawrence Earlbaum Associates. Cook, T.D. 0 Campbell, D.T. (1979). Quasi-Experimentation: Design and Analysis Issues for Field Settings. Chicago: Rand-McNally. Cappuccio, F. P., Cooper, D., Elia, L., Strazzullo, P. 0 Miller, M. A. (2011). Sleep duration predicts cardiovascular outcomes: a systematic review and meta-analysis of prospective studies. [10.1093/eurheartj/ehr007]. European Heart Journal. Curtin, R., Presser, S. 0 Singer, E. (2000). The Effects of Response Rate Changes on the Index of Consumer Sentiment. Public Opinion Quarterly, 64(4), 413-428. Deutscher Bundestag. (2007). Neufassung des Gesetzes zum Schutz gegen Fluglärm. Bonn: Bundesgesetzblatt 2007, Teil 1, Nr. 56. Dratva, J., Zemp, E., Dietrich, D. F., Bridevaux, P.-O., Rochat, T., Schindler, C. 0 Gerbase, M. W. (2010). Impact of road traffic noise annoyance on health-related quality of life:
NORAH Noise-related annoyance, cognition and health study Literature NORAH General Overview Results from a population-based study. Quality of Life Research: An International Journal of Quality of Life Aspects of Treatment, Care & Rehabilitation, 19(1), 37-46. Dratva, J., Phuleria, H., Foraster, M., Gaspoz, J. M., Keidel, D., Kunzli, N., Liu, L. J., Pons, M., Zemp, E., Gerbase, M. W. 0 Schindler, C. (2012). Transportation noise and blood pressure in a population-based sample of adults. Environmental Health Perspectives, 120(1), 50-55. doi: 10.1289/ehp.1103448. Ellert, U., Lampert, T. 0 Ravens-Sieberer, U. (2005). Messung der gesundheitsbezogenen Lebensqualität mit dem SF-8. Eine Normstichprobe für Deutschland. Bundesgesundheitsblatt, 12(1330-1337). Elmenhorst, E.-M., Pennig, S., Rolny, V., Quehl, J., Mueller, U., Maaß, H. 0 Basner, M. (2012). Examining nocturnal railway noise and aircraft noise in the field: Sleep, psychomotor performance, and annoyance Science of the Total Environment, 424, 48-56. European Environment Agency. (2014). Noise in Europe 2014. EEA Report (Vol. 10/2014). Copenhagen (DK): EEA. Faulbaum, F. (2014). Total Survey Error. In N. Baur 0 J. Blasius (Eds.), Handbuch Methoden der empirischen Sozialforschung (pp. 439-453). Wiesbaden (D): Springer Fachmedien. Floud, S., Blangiardo, M., Clark, C., de Hoogh, K., Babisch, W., Houthuijs, D., Swart, W., Pershagen, G., Katsouyanni, K., Velonakis, M., Vigna-Taglianti, F., Cadum, E. 0 Hansell, A. (2013). Exposure to aircraft and road traffic noise and associations with heart disease and stroke in six European countries: a cross-sectional study. Environmental Health, 12(1), 89. Giering, K. (2010). Lärmwirkungen -- Dosis-Wirkungsrelationen. In Umweltbundesamt (Hg.), UBA-Texte (Vol. 13). Dessau (D): Umweltbundesamt. Gjestland, T., Gelderblom, F. B., Fidell, S. A. 0 Berry, B. (2015). Temporal trends in aircraft noise annoyance. Paper presented at the Inter-Noise 2015, San Francisco (CA, USA) Groves, R. M., Presser, S. 0 Dipko, S. (2004). The role of Topic Interest in Survey Participation Decisions. Public Opinion Quarterly, 68(1), 2-31. Guski, R. (2004). How to forecast community annoyance in planning noisy facilities. Noise & Health, 6(22), 59-64. Hastie, R. 0 Dawes, R. M. (2001). Rational Choice in an Uncertain World: The Psychology of Judgement and Decision Making. 2nd ed. Thousand Oaks (USA, CA): Sage Publications. Hatfield, J., Job, R. F. S., Hede, A. J., Carter, N. L., Peploe, P., Taylor, R. 0 Morrell, S. (2002). Human Response to Environmental Noise: The Role of Perceived Control. International Journal of Behavioral Medicine, 9(4), 341-359. Houthuijs, D., van Kamp, I., Breugelmans, O., Ameling, C., Marra, M. 0 van Poll, R. (2012). Community response to aircraft noise: recent examples from the Netherlands. Paper presented at the Inter-Noise 2012, New York City (USA). Huang, D., Song, X., Cui1, Q., Tian, J., Wang, Q. 0 Yang, K. (2015). Is there an association between aircraft noise exposure and the incidence of hypertension? A meta-analysis of 16784 participants. Noise & Health, 17(75), 93-97. Hurrelmann, K. (2010). Gesundheitssoziologie. Eine Einführung in sozialwissenschaftliche Theorien von Krankheitsprävention und Gesundheitsförderung. 7.Aufl. Weinheim, München: Juventa Verlag. Huss, A., Spoerri, A., Egger, M., Röösli, M. for-the-Swiss-National-Cohort-Study-Group. (2010). Aircraft Noise, Air Pollution, and Mortality From Myocardial Infarction. Epidemiology, 21(6), 829-836.
NORAH Noise-related annoyance, cognition and health study Literature NORAH General Overview Janssen, S.A. 0 Vos, H. (2009). A comparison of recent surveys to aircraft noise exposure- response relationships. Delft, NL: TNO report, TNO-034-DTM-2009-01799. Jarup, L., Babisch, W., Houthuijs, D., Pershagen, G., Katsouyanni, K., Cadum, E., Dudley, M. L., Savigny, P., Seiffert, I., Swart, W., Breugelmans, O., Bluhm, G., Selander, J., Haralabidis, A., Dimakopoulou, K., Sourtzi, P., Velonakis, M., Vigna-Taglianti, F. 0 team, H. (2008). Hypertension and Exposure to Noise near Airports - the HYENA study. Environmental Health Perspectives, 116, 329-333. Koolhaas, J. M., Bartolomucci, A., Buwalda, B., de Boer, S. F., Flügge, G., Korte, S. M., Meerlo, P., Murison, R., Olivier, B., Palanza, P., Richter-Levin, G., Sgoifo, A., Steimer, T., Stiedl, O., van Dijk, G., Wöhr, M. 0 Fuchs, E. (2011). Stress revisited: A critical evaluation of the stress concept. Neuroscience & Biobehavioral Reviews, 35(5), 1291-1301. La Torre, G., Moscato, U., La Torre, F., Ballini, P., Marchi, S. 0 Ricciardi, W. (2007). Environmental noise exposure and population health: a cross-sectional study in the Province of Rome. Journal of Public Health, 15(5), 339-344. Lazarus, R. S. 0 Launier, R. (1978). Stress-related transactions between person and environment. In L. Pervin 0 M. Lewis (Eds.), Perspectives in interactional psychology (pp. 1-67). New York: Plenum Press. Lercher, P., De Greve, B., Botteldooren, D. 0 Ruedisser, J. (2008). A comparison of regional noise-annoyance-curves in alpine areas with the European standard curves. Proceedings of the 9th International Congress on Noise as a Public Health Problem (ICBEN). 562-570. Foxwoods (USA, CT). Miedema, H. M. E. 0 Oudshoorn, C. G. (2001). Annoyance from transportation noise: Relationships with exposure Metrics DNL and DENL and their confidence intervals. Environmental Health Perspectives, 109, 409-416. Miedema, H.M.E. 0 Vos, H. (1998). Exposure-response relationships for transportation noise. Journal of the Acoustical Society of America, 104(6), 3432-3445. Miller, N. P., Cantor, D., Lohr, S., Jodts, E., Boene, P., Williams, D., Fields, J., Gettys, M., Basner, M. 0 Hume, K. (2014). Research Methods for Understanding Aircraft Noise Annoyances and Sleep Disturbance. In Airport Cooperative Research Program (Ed.), Contractor’s Final Report for ACRP 02-35. USA. Möhler, U., Liepert, M., Schuemer, R., Griefahn, B. Differences between Railway and Road Traffic Noise, Journal of Sound and Vibration (2000) 231(3), pp.853-864. Möhler, U., Liepert, M., Mühlbacher, M., Beronius, A., Braunstein, G., Gille, M., Schaal, J. 0 Bartel, R. (2015). Erfassung der Verkehrsgeräuschexposition: Dokumentation zur Berechnung der akustischen Daten in NORAH. In Gemeinnützige Umwelthaus gGmbH (Hg.), NORAH (Noise related annoyance cognition and health): Verkehrslärmwirkungen im Flughafenumfeld (Bd. 2). Kelsterbach: Umwelthaus gGmbH. Müller, U. (2010). Verbundprojekt DEUFRAKO/RAPS: Wirkungsorientierte Bewertung unterschiedlicher Verkehrslärmarten. Teilvorhaben DLR: Metaaanalyse und Feldstudie. Abschlussbericht. Köln: DLR. Robert-Koch-Institut (Hg., 2013). Diagnose Depression: Unterschiede bei Frauen und Männern (GBE kompakt, Vol. 02/2013). Berlin: Robert-Koch-Institut. Schreckenberg, D. 0 Meis, M. (2006). Gutachten: Belästigung durch Fluglärm im Umfeld des Frankfurter Flughafens - Endbericht. Bochum + Oldenburg: ZEUS / Hörzentrum Oldenburg. Schreckenberg, D., Meis, M., Kahl, C., Peschel, C., Eikmann, T. (2010). Aircraft noise and quality of life around Frankfurt Airport. International Journal of Environmental Research
NORAH Noise-related annoyance, cognition and health study Literature NORAH General Overview and Public Health, 7, 3382-3405. http://www.mdpi.com/1660-4601/7/9/3382/, Abruf am 09.08.2015. Schreckenberg, D. (2013). Exposure-response relationship for railway noise annoyance in the Middle Rhine Valley. Proceedings of Internoise 2013, Paper No. 1003, Innsbruck, Austria. Schreckenberg, D., Faulbaum, F., Guski, R., Ninke, L., Peschel, C., Spilski, J. & Wothge, J. (2015). Wirkungen von Verkehrslärm auf die Belästigung und Lebensqualität. In Gemeinnützige Umwelthaus gGmbH (Hg.), NORAH (Noise related annoyance cognition and health): Verkehrslärmwirkungen im Flughafenumfeld (Bd. 3). Kelsterbach: Umwelthaus gGmbH. Shadish, W.R. & Cook, T.D. (2009). The renaissance of field experimentation in evaluation interventions. Annual Review of Psychology, 60, 607–629. Staatsen, B. A. M., Franssen, E. A. M., Van Wiechen, C. M. A. G., Houthuijs, D. & Lebret, E. (2004). HIA in Schiphol airport. In J. Kemm, J. Parry & S. Palmer (Eds.), Health Impact Assessment: Concepts, theory, techniques and applications (pp. 265-283). Oxford: Oxford University Press. Stansfeld, S. A., Berglund, B., Clark, C., Lopez-Barrio, I., Fischer, P., Öhrström, E., Haines, M. M., Head, J., Hygge, Staffan. van Kamp, Irene & Berry, B. F. (2005). Aircraft and road traffic noise and children's cognition and health: a cross-national study. The Lancet, 365(9475), 1942-1949. Tarnopolsky, A., Watkins, G. & Hand, D. J. (1980). Aircraft noise and mental health: 1. Prevalence of individual symptoms. Psychological Medicine, 10, 683-698. Van Kamp, I. (1990). Coping with noise and its health consequences. Dissertation. Rijksuniversiteit Groningen, Groningen (NL). Van Kamp, I., Houthuijs, D., Van Wiechen, C., Stellato, R. & Breugelmans, O. (2007). Environmental noise and mental health. Paper presented at the 19th International Congress on Acoustics (ICA2007), Madrid (E). Van Kempen, E.E.M.M. & Van Kamp, I. (2005). Annoyance from air traffic noise. Possible trends in exposure-response relationships. Report 01/2005 MGO EvK, Reference 00265/2005, Bilthoven (NL): RIVM. Vienneau, D., Schindler, C., Perez, L., Probst-Hensch, N. & Röösli, M. (2015). The relationship between transportation noise exposure and ischemic heart disease: A meta- analysis. Environmental Research, 138, 372–380. Virchow, R. (1922). Rede auf der Naturforscherversammlung 1869 in Innsbruck. In K. Sudhoff (Ed.), Rudolf Virchow und die deutschen Naturforscherversammlungen. Leipzig: Akademie Verlagsgesellschaft. Wallston, K.A., Strudler Wallston, B., Smith, S., Dobbins, C.J. (1987). Perceived control and health. Currenty Psychological Research & Reviews, 6, 5-25. World Health Organization. (1946). Constitution of the World Health Organization. New York: WHO. World Health Organization. (2011). Burden of disease from environmental noise. Quantification of healthy life years lost in Europe. Bonn (D): WHO. World Health Organization. (2014). Basic Documents, 48th edition. Geneva (CH): WHO.
NORAH Noise-related annoyance, cognition and health study