• Accessibility
• Email alerts
• RSS feeds
• Contact us

The neglected pollutants: the effects of artificial light and noise on human health Contents

---

Chapter 2: Scientific evidence of health effects

5.This chapter summarises the scientific research basis for the current understanding of the impacts of artificial light and noise on human health. It outlines the existing literature and some uncertainties that need to be addressed by further research.

Figure 1: Schematic of light and noise pollution impacts on health and wellbeing

Source: Schematic prepared by Professor Russell Foster, outlining definitions of light and noise pollution and some mechanisms by which they can impact health and wellbeing.

Noise

6.‘Noise’ generally refers to unwanted sound. Sound is characterised by acoustic properties including pitch and volume. Volume is usually measured in decibels (dB), on a logarithmic scale; when a sound is perceived to double in loudness, this corresponds to an increase of roughly 10 dB, a tenfold increase in power (see Table 1).4 Sounds can be continuous or intermittent; the timing and duration of a sound are also relevant to its potential effects on human health.

Table 1: The Decibel Scale

Decibel measure (dB)	Common sound
30	Leaves rustling/whisper
40–50	Average room noise
60	Background music
70	Average office noise
80	Inside an aeroplane or underground carriage
90	Hairdryer
110	Nightclub or rock concert
135	Jet engines

Source: Common sounds and their relative volume in decibels. Levels of sound will vary depending on the distance from the source of the sound. Hearing Health Foundation, ‘Decibel Levels’: https://hearinghealthfoundation.org/decibel-levels [accessed 30 June 2023]

Evidence for the health impacts of noise

7.In 2018, the World Health Organization (WHO) published environmental noise guidelines for the European region, which were based on systematic reviews of the scientific literature.5 Professor Anna Hansell, Professor of Environmental Epidemiology and Director of the Centre for Environmental Health and Sustainability at the University of Leicester, summarised its findings:

“There is obviously good evidence for annoyance and for sleep disturbance. There is now good evidence for impacts on cardiovascular disease; the strongest evidence is on ischaemic heart disease, that is, heart attacks, in relation to road traffic noise. There is some evidence on metabolic impacts, for example diabetes.”6

8.One method of quantifying the health effects of environmental noise is by estimating the overall disease burden. The WHO and European Environment Agency estimated in 2018 that more than 100 million people were exposed to harmful levels of environmental noise pollution. They estimated that this contributed to “48,000 new cases of heart disease and 12,000 premature deaths every year in Europe. In addition, 22 million people suffer chronic high annoyance, and 6.5 million suffer chronic high sleep disturbance.”7 Research from the UK Health Security Agency (UKHSA) suggests that in 2018, 130,000 healthy life years were lost in the UK due to noise pollution; and that 40% of the population were exposed to harmful levels of noise pollution from road traffic.8

9.The mechanisms behind these health impacts remain a subject of research; we heard that there are multiple pathways by which noise can affect health, summarised by Professor Hansell as “noise annoyance, sleep disturbance, fight or flight reaction, and non-specific stressor.”9

Sleep and circadian rhythm disruption

10.Both light and noise can influence health by disrupting circadian rhythms. Professor Kenneth Wright, Director of the Sleep and Chronobiology Laboratory at the University of Colorado Boulder, described negative health effects from circadian rhythm disruption as including “insomnia, prescription of hypnotic drugs in older adults, obesity, type 2 diabetes, heart disease, elevated blood pressure, depression and cancer.”10 However, Professor Dr Manuel Spitschan, Professor of Chronobiology and Health at the Technical University of Munich, emphasised that “good light exposure [supports] the circadian system”.11

11.Professor Shantha Rajaratnam, Professor of Sleep and Circadian Medicine at Monash University, discussing the effects of artificial light exposure in occupational settings, noted that, while evidence is still building towards scientific recommendations for healthy light exposure, for “particularly vulnerable populations, such as in hospitals, age care settings, and so on, we should make recommendations early”.12 Night-shift workers are particularly vulnerable owing to the persistent disruption to their circadian rhythms. Professor Rajaratnam said that there was an “urgent need for studies in that area.”13

12.We heard of studies which estimate the approximate cost of sleep disruption to economies. One study by the RAND Corporation suggested “that in the UK this is around 1.86% of GDP ($50 billion/£42 billion).”14 Some researchers have set up large-scale studies on sleep in the UK, such as the UK Sleep Census, but these are independent research projects rather than public health actions.15

Emerging evidence for health impacts of noise

13.Some studies have examined possible associations between environmental noise and other health effects. Professor Charlotte Clark, Professor of Epidemiology at St. George’s, University of London, described one such area, noting: “We think that children’s cognition generally is affected, but we do not see it consistently across all the sources”.16 Owing to the lack of confidence around these emerging health effects, they have not always been included in burden-of-disease calculations.

14.The WHO guidelines included cognitive impairment of children and tinnitus in its burden-of-disease calculations.17 However, we heard from Professor Hansell that health effects outside those currently included by the WHO had a “much lower weight of evidence, but there are suggestions that there might be impacts outside the cardiovascular system. Some studies have looked at cancer, although the findings have been quite inconsistent. A few studies have looked at noise and respiratory disease.”18

15.We heard that the evidence base for the health effects of environmental noise is growing rapidly. Dr Benjamin Fenech, Noise and Public Health Group Leader at the UKHSA, noted that, although the WHO guidelines were based on evidence published up to 2015, when UKHSA did its own review in 2022 half of the studies used were published in the last five years. This expansion of the evidence base meant that UKHSA felt the need to commission more up-to-date systematic reviews, “to make sure that we are capturing the latest evidence to inform decision-making”.19

16.There is increasing epidemiological evidence of the harmful effects of noise on human health. Aggregated over the whole population, even small effects on the individual can be a significant public health concern. New evidence is likely to change the understanding of these effects, for example the role played by intermittency. DEFRA should work with the UK Health Security Agency and other organisations to assess the significant, growing evidence on the health effects of noise. This should include refining existing estimates for the disease burden from well-established health impacts of noise pollution, for example on the cardiovascular system. UKHSA should also assess whether health effects for which evidence is emerging, such as on the metabolic system, meet the evidentiary threshold for policy action.

Areas for further research

17.The evidence for the negative health impacts of noise is predominantly epidemiological, as laboratory-based studies for population health are difficult to conduct.20 The evidence base consists of plausible biological mechanisms with some laboratory-based studies and epidemiological studies which find a statistical association between exposure and health effects.

18.We heard that this evidence base could be expanded with additional types of study. On laboratory studies, Professor Clark said “it will be a good way to look at the mechanisms in a bit more detail, which is an area that traditionally has not been well researched.”21 Professor Hansell acknowledged the limitations of the epidemiological evidence when compared to the evidence for harm from other pollutants, noting “With air pollution, we tend to use much more complex models than we do in noise so far, because … there are handfuls of studies on noise, compared with the thousands of studies on air pollution.”22

19.One of the main ways by which noise affects health is through annoyance.23 Annoyance generally rises with noise level, but also depends on the type of noise, with differing results found for aircraft, road and railway noise.24 Dr Antonio Torija Martínez, Reader in Acoustic Engineering at the University of Salford, noted that “high-frequency or high-pitched sounds are perceived to be more annoying.”25 Annoyance does not depend only on the properties of the sound but can also depend on one’s noise tolerance.26

20.Annoyance is also influenced by non-acoustic factors. Professor Clark said that “sound accounts for only a small proportion of the annoyance we measure”, with non-acoustic factors including “your age, your biological sex, if it is airport noise your attitude to the airport”.27 Dr Torija Martínez said perhaps as little as “30% of noise annoyance [is] related to acoustic factors”,28 but more research is needed “to account for non-acoustic factors”.29 Dr Fenech said “non-acoustic factors have a big role to play in improving health outcomes.”30

21.The standard metrics used for measuring exposure to environmental noise involve averaging the noise levels over time.31 However, intermittent loud noises, with a high peak volume, might have different health impacts compared with a more constant, average, quieter noise. Professor Clark described Swiss studies that have defined metrics for intermittency and which found that the degree of intermittency of noise can explain differences in annoyance reactions.32

22.Dr Fenech mentioned the “need [for] research using different noise exposure metrics” due to the limitations of long-term exposure metrics currently in use.33 Asked whether DEFRA’s noise modelling included intermittency, Dr Bill Parish, Deputy Director for Air Quality and Industrial Emissions at DEFRA, confirmed it did not, and said “We will need to develop how we apply a more sophisticated approach to those scenarios.”34

23.Witnesses were asked what kinds of studies should be conducted to improve the evidence base for the impacts of environmental noise on health. Areas mentioned included:

• Additional studies of railway and neighbour noise35
• Large-scale longitudinal epidemiological studies36
• Experimental and quasi-experimental (natural experiment) studies37
• Studies into the impacts of noise on mental health.38

24.There are also limitations in understanding of indoor noise exposure; Professor Hansell noted that “there is little information at population level on sound indoors, which will depend on building characteristics, what else is going on in the house”.39 This may be of concern because of socioeconomic disparities in the quality of housing stock, which could lead to “higher noise exposures from outdoor noise penetrating indoors.”40

25.Witnesses agreed more research was needed into the impact of interventions to reduce noise exposure on health.41 Professor Clark said: “we do not have good studies of interventions where we change the noise exposure, or we try to, and then we assess how that impacted people’s sleep, annoyance, cardiovascular responses.”42

26.Witnesses mentioned that researching the health effects of noise is intrinsically multidisciplinary and that this can cause problems in securing funding.43 Dr Torija Martínez said: “We need some mechanisms to facilitate interdisciplinary research to do this. For example, it is difficult to work within different research councils.”44

27.More laboratory and field studies are needed to supplement epidemiological evidence and to establish the mechanisms by which noise might affect health. The current metrics used to characterise noise pollution are mostly long-term average intensity (decibel) metrics, which do not capture peak volume, pitch and intermittency. The latter influence annoyance and may correlate more closely with health outcomes but are not widely measured. Quantifying the health effects of interventions to reduce exposure is important for cost-benefit analyses.

28.The UK should seek opportunities to collaborate with similar countries, sharing research data and methodologies. Alongside these efforts, DEFRA should commission and fund a research programme into noise and health. This should include:

• large-scale epidemiological studies, including long-term longitudinal studies, which can make use of international big data;
• laboratory-based studies establishing mechanisms for health impacts;
• field studies establishing the indoor exposure to noise, which can contribute to mapping the indoor exposure to noise;
• interdisciplinary studies to understand the variation in response caused by non-acoustic factors;
• modelling and experimental studies into the health effects of interventions to reduce noise; and
• whether alternative metrics for noise, including pitch and intermittency, should be measured and used to better understand health outcomes.

An independent advisory panel for noise

29.Some witnesses were concerned that there was no clear channel in place by which their evidence could inform Government policy. Professor Anna Hansell praised the work of the UKHSA noise team for calculating the “burden of disease from noise for all local authorities in England”, but noted that “there is no expert scientific advisory group for noise … as there is for air pollution”.45 She said “there is no clear policy group to go to.”46

30.Although UKHSA has a noise and health team which summarises research in this field for policymakers, there is no advisory group as there is for air pollution. An interdisciplinary, independent advisory panel should be established to provide independent advice to the Government and a forum for new evidence, particularly on emerging health effects and technologies, to be assessed.

Light

Evidence for the health impacts of artificial light at night

31.Artificial light is characterised by properties, including the intensity of the light or its wavelength. Intensity can be measured using the SI unit, lux, while wavelengths are often characterised using the colour temperature.47 The eye operates over a wide range of intensities in lux (see Table 2).

32.Artificial light at night could influence human health by disturbing sleep and circadian rhythms.48 The science on how light exposure influences circadian rhythms is evolving. Professor Rajaratnam noted many properties of light exposure influence circadian rhythms, including: “the duration … intensity and the wavelength composition … as well as the timing of the light exposure and the history of the light exposure”. He suggested that these properties should be measured in “large-scale studies”.49

Table 2: Common light scenarios and their measurements in lux

Lux	Comparison
0.0001	Starlight
0.25–1	Full moon
80	Typical indoor lighting
400	Sunrise or sunset
500	A well-lit office
1000	Overcast day outdoors
10,000	Daylight
100,000	Intense, direct sunlight

Sources: Trong-Hop Do, ‘Performance Analysis of Visible Light Communication Using CMOS Sensors’, ResearchGate (February 2016): https://www.researchgate.net/publication/296477842_Performance_Analysis_of_Visible_Light_Communication_Using_CMOS_Sensors ; James Druzik, ‘Guidelines for Selecting Solid-State Lighting for Musums—Figure 13—The scale of light intensities from moonlight to candlelight to sunlight’, ResearchGate (December 2015): https://www.researchgate.net/figure/The-scale-of-light-intensities-from-moonlight-to-candlelight-to-sunlight-and-the-range_fig5_287207122; Torchspot, ‘Lumens, Lux and Candela Explained—Lux Comparison’: https://www.torchspot.com/lumens-lux-and-candela/#Lux_Comparison_Chart; and The Electrical Counter, ‘What are Lux levels?’: https://www.electricalcounter.co.uk/lux-levels-chart [accessed 10 July 2023]

33.Scientists have defined a new metric, “melanopic lux”, which accounts for the wavelengths to which the circadian system is most sensitive. Professor Dr Manuel Spitschan told us that “when we are assessing or trying to measure the impact of light exposure on human health or the human circadian system, we need to take this wavelength preference … into account.”50 However, this metric is not widely used51 and the Circadian Neuroscience Group at the University of Oxford recommended that “all studies on light pollution use appropriate units (mel-EDI) when considering the non-visual health effects of light.”52

34.Guidelines are being developed for light exposure that promotes circadian health. Witnesses referred to the “expert consensus recommendations paper”53 known as the Manchester Guidelines, published in 2022.54 These values are based on “existing laboratory data for human dose-response curves to light” which characterise how the circadian clock shifts when it is exposed to light at different intensities.55 Whilst such laboratory-based studies are informative, they may not capture how people are exposed to light in reality.56

35.There are concerns that light pollution has been increasing rapidly in recent years. Evidence for the health effects of light pollution is at a less mature stage than noise pollution, but it may influence health by disrupting circadian rhythms and sleep. The Government should commission research to establish how light intensity, wavelength, duration, time of exposure, light history and age affect the circadian system. This should move beyond laboratory-based studies and investigate more realistic light exposure patterns for humans. Such knowledge would provide an evidence base for guidelines that could mitigate the harmful effects of light pollution on human biology, including the circadian system, mood and alertness.

Other health effects from artificial light

36.Artificial light can cause discomfort due to flicker and glare. Dr Christopher Kyba, Researcher at the Helmholtz Centre Potsdam, said that “flicker … is an issue for a lot of individuals—again, not everybody, but some are very sensitive to it.”57 Glare was described as “dangerous” for drivers in certain situations, and we were told it could “cause eye strain and headaches.”58

37.Surveys have indicated that dazzle from car headlights is a growing problem, which may be related to the rollout of LEDs,59 with studies suggesting they should be made dimmer in urban environments to aid contrast and avoid dazzle.60 Dr Kyba said: “There is very little discussion between the people who create headlights and the people who create street lighting”, leading to poor interactions between the lighting types.61

38.Concerns have been raised around the LED rollout.62 LEDs are cheaper and more energy-efficient than traditional lighting, but this means that “more people are using more lights when they do not necessarily need to”.63 Dr Luke Price, Principal Radiation Protection Scientist at UKHSA, told us “there is nothing intrinsically unhealthy about LEDs, but, if they are more energy efficient, we can use more of them and light more”.64 Dr Kyba told us that there is evidence they contribute to light pollution: “the number of stars that people report they are able to see has dramatically decreased” since the rollout.65

39.The UK Health Security Agency has a team that collates evidence for environmental noise and health, but there is no explicit team for light. Dr Luke Price is their light specialist and co-authored the Manchester Guidelines. He noted that they were “framed … specifically for policymakers”.66 Dr Edward Wynne-Evans, Director of the Radiation, Chemicals and Environmental Hazards Directorate at UKHSA, said that “On light, we want to work with other bodies to expand that [evidence] base” to quantify risks and costs from artificial light, but the evidence base was not yet good enough to carry out a cost-benefit analysis as can be done with traffic noise.67

40.Flicker, glare and dazzle can cause visual disturbance. There is not clear evidence that LEDs cause ill-effects in healthy people when used properly. However, there is widespread concern that the LED rollout has been associated with poor lighting practice and over-lighting. Research should be carried out in order to establish the level of risk from glare, flicker, and dazzle, for example in night-time driving.

Mapping light pollution and human exposure to artificial light at night

41.Although there is a statutory requirement to map noise pollution, there is no such requirement for light. Emma Marrington of the charity CPRE told us that the 2016 Night Blight mapping, which they created “with consultants that looked at satellite data”, was “the baseline and currently the most detailed map for England”, but there was “a need to remap light pollution using the latest technology.”68

42.Ruskin Hartley, CEO of the International Dark-Sky Association, explained that “scientists have estimated, based on the satellite data, that light pollution has been growing maybe by 2% or 3% per year”,69 but this was contradicted by citizen-science projects that involve counting visible stars, which suggest that light pollution and sky glow is growing by 10% every year.70 The reason for this discrepancy is that satellite measures are not always a reliable proxy for on-the-ground exposure.71

43.Asked about the idea of a central light map, Rebecca Pow MP, Minister for Environmental Quality and Resilience, said “We are doing it for some projects … the CPRE does some of its own light mapping. There are a lot of limitations to our ability to do that right now … We need more research … to establish … the methodology, the metrics and what we are measuring.”72

44.We heard that, although it is possible to calculate a “burden of disease” estimate for noise pollution, this cannot yet be done for light. This is partially because of a lack of data about the light levels that people are exposed to at night. Dr Luke Price of UKHSA described “a series of longitudinal studies in Japan that measured the light in people’s environment, which we lack in the UK”.73 Concerns were also raised about specific occupational exposures to light, such as in hospitals, which are not being measured.74

45.Regulating light pollution is difficult if it is not measured; our current approach is inadequate. It is also difficult to assess the health implications if it is not known how people are exposed to light pollution, particularly indoors at night. DEFRA should establish a standard methodology for tracking, monitoring and reporting on light pollution. This should be in place by the next five-year Environmental Improvement Plan cycle. The Government should commission a regular survey to track light pollution once the methodology is agreed. The research should aim to understand both indoor and outdoor exposure to artificial light at night, so its health impact can be quantified.

Expert advisory group on circadian rhythms and light

46.There is no expert advisory group for the impacts of light pollution on human health, or on the environment more generally.75 Ms Pow told us that “a review was done of the light issues after the Royal Commission on Environmental Pollution report” in 2009, but that it concluded “there was not enough evidence to do anything to change the way we regulate”. She conceded that the “evidence is building and changing all the time.”76 However, it is unclear who the Government would consult to assess the evidence base on the health impacts of artificial light at night, or for circadian science in general.

47.The Government should have a team of experts in UKHSA, on circadian rhythms and impacts of light on health to act as a single point for evidence gathering and co-opting external expertise. As the field develops, it may be appropriate to set up an independent advisory panel, as for noise and air pollution.

Issues affecting noise and light

Possible beneficial effects of light and noise on human health

48.Professor Shantha Rajaratnam told us that, given the link between sleep and circadian rhythm disruption and mental health, there is “a unique opportunity to intervene and prevent” mental health conditions.77

49.Light boxes can be used as treatment for mental ill-health. This involves exposing the patient to bright illumination (in excess of 10,000 lux) early in the morning to help reset the circadian clock. Professor Pierre Geoffroy, Professor of Psychiatry at Université Paris Cité, cited a meta-analysis that showed “light therapy is as efficient as antidepression treatment [for] both seasonal and non-seasonal depression.”78

50.Professor Geoffroy said that “the combination of the two treatments—light and antidepressants—is clearly superior to antidepressants alone.” He said that the “level of evidence … depends on the disorder” and called for “larger, good-quality randomised controlled trials”, in particular long-term follow-up studies and effects of treatment on people with different natural sleep-wake cycles, or chronotypes.79 The NHS website mentions light boxes as a possible treatment for seasonal affective disorder, but the NHS does not currently prescribe them.80 Professor Geoffroy noted that research into light boxes was “much less supported compared to that for drugs, where pharmaceutical companies conduct or sponsor very large-scale studies.”81

51.Social prescribing is an approach to care that seeks to connect people to activities, groups and services in their community that can address their needs.82 Green social prescribing involves nature-based interventions and activities, such as walking-for-health schemes or gardening projects.83 Dr Alison Greenwood, CEO of A Dose of Nature, told us that its schemes had “over 800 referrals from GPs”.84 Natural light and sounds may play a role in the positive psychological impacts of green social prescribing.85

52.However, witnesses agreed that more research was needed to understand the precise effect sizes and mechanisms of these positive impacts, with Dr Greenwood noting that it was “difficult to look at the effects of nature in a randomised controlled trial”.86

53.The National Institute for Health and Care Excellence should review evidence for the effectiveness of therapies such as light boxes that might promote improved circadian rhythms and therefore physical and mental health.

54.The National Institute for Health and Care Research should commission research to establish the mechanisms by which green social prescribing may affect health.

---

• A-Z index
• Glossary
• Contact us
• Freedom of Information
• Jobs
• Using this website
• Copyright