Health effects of large LED screens on local residents

In 2011, three high-definition outdoor LED video billboards were erected around a city sports stadium. The residents of the surrounding community are exposed to the lighted images of these screens when they are illuminated. A representative of a community organization advocating to remove the video screens has asked the Chief Medical Health Officer of the local health department whether the office agrees that these screens pose a health concern to the local residents.

Background of the Case Study
Exposure to Light Sources
Potential Health Effects
Requirements for Risk Assessment
Conclusions
Acknowledgements
References

Questions:

• What are the potential health effects of exposure to large outdoor LED billboards for nearby residents?
• What further information is required to assess risk in this case?

Background of the Case Study

The light-emitting diode (LED) screens were erected as part of an overall upgrade to the sports stadium. A total of three screens were installed, each measuring 30 feet by 50 feet. Two of the screens are illuminated daily from 6:00 am to 11:00 pm. Software adjusts the intensity of the screens over the course of the day. From 9:00 am to 7:00 pm, these two screens operate at 50% capacity; at all other times, they operate at 25%. The third screen operates on a seasonal schedule and is illuminated from sunrise to sunset with the exception of “event” nights such as soccer or football games. From 9:00 am to 5:00 pm, the third screen operates at 50%; on event nights, this period is extended to the start of the event. It otherwise operates at 20%. All three screens are illuminated until 15 minutes after the particular event finishes.

Exposure to Light Sources

The light spectrum includes three wavelength groupings: ultraviolet (UV) light, visible (V) light, and infrared (IR) light. Visible light is perceived in the wavelength range between 390 nm and 780 nm. It can be produced artificially by different types of lamps, including incandescent lamps, electrical discharge lamps and solid-state lighting such as light-emitting diodes (LEDs). To quantify the intensity of visible light, two sets of quantities and units are used: radiometric and photometric.

Radiometric quantities describe the absorption of radiant energy in biological tissue. Hence, exposure limits are expressed in terms of radiometric quantities such as irradiance (Watts per meter squared, W/m²), radiant exposure (Joules per meter squared, J/m²), effective radiance (Watts per meter squared per steradian—the SI unit of a solid angle, W/(m².sr)), and time-integrated radiance (Joules per meter squared per steradian, J/(m².sr)). Radiometric measurements are usually carried out in laboratories by means of spectroradiometers for calibration and output assessment of small and large viewing monitors.

Photometric quantities, luminance and illuminance, describe the relative brightness of visible light to the human eye and are measured by photometers. Luminance is defined as the light brightness of a source (Candela per meter squared, cd/m²) and illuminance as the intensity of light falling perpendicularly on a surface (Lumen per meter squared, lm/m² or Lux), both as perceived by the human eye. The human eye is adapted to comfortably view a wide range of photometric intensities. The average indoor lighting has illuminances of 100 to 1,000 lux, and average outdoor sunlight is 50,000 lux. The sky on a cloudy day has a luminance of about 3,000 cd/m²; this increases to 6,000–10,000 cd/m² on a clear day.¹

Case Study Exposure

The manufacturer and technical specifications were not immediately available and thus the actual photometric measurements are unknown. In this case, the contribution of artificial light from the outdoor LED screens compared to other sources is unknown. LED lights flicker at a rate dependent on fluctuations in their power frequency. However, no information is available on the flicker rate for the LED screens. Also, the screen’s maximum luminance is not known, but the company who provided the outdoor screens lists their maximum luminance as 5,000–6,000 cd/m².

Potential Health Effects

The following possible health effects of artificial light from LED sources were identified and are discussed below.

• Photosensitive Epilepsy
• Retinal Damage
• Stress and Annoyance
• Circadian Rhythm and Sleep Disruption

Most of the research examines indoor lighting, personal screen use, or occupational exposures to industrial lighting and thus may not be generalizable to the scenario of exposure to a large outdoor screen.

Photosensitive Epilepsy

Photosensitive epilepsy (PSE) is characterized by epileptic seizures induced by certain visual stimuli in susceptible individuals. The prevalence of PSE varies depending on age and is more common in children. A British study estimated the prevalence of PSE to be rare at 1.2 per 100,000 in the overall population, but 5.7 per 100,000 in children aged 7 to 19 years.² PSE can be triggered by a variety of flickering lights, malfunctioning fluorescent lighting, light filtering through blinds or fan blades, rolling escalators, striped walls, and even clothing.

The most common visual stimuli causing seizures are flashing lights and striped, grid-like, or checkerboard patterns. The key features of provoking stimuli have been determined through EEG changes during controlled exposure to visual stimuli.³ These are:

• Flicker frequency: Flashes at 5–30 Hz are most likely to stimulate seizures.
• Light intensity and contrast: Alternating dark and bright images with a difference of 20 cd/m² are of greatest concern.
• Area of stimulus: The larger the visual image, the greater the portion of the brain is stimulated. This variable also depends on the distance of the viewer from the source.
• Number of repetitions of a pattern: For a static image, more than eight pairs of stripes increases the risk of seizure. If the image is moving, more than five pairs of stripes increases the risk of seizure.
• Colour: Flashing from a deep, saturated red to another colour is more likely to induce seizures.

The International Telecommunications Union (ITU) has produced guidelines for broadcasters to reduce the risk of photosensitive seizures.⁴ The recommendations limit five features of broadcast:

1. Frequency: Flashes with frequency greater than 3 Hz are prohibited.
2. Changes in luminance: Flashes with changes in luminance greater than or equal to 20 cd/m² are prohibited.
3. Area of flashes: Flashes greater than one-fourth of the screen area are prohibited.
4. Color: Flicker from saturated red light is prohibited.
5. Striped patterns: Colour, size, and duration of striped images are limited.

Patients with PSE are advised to avoid stimuli and other factors which make them more susceptible to seizures (e.g., fatigue, alcohol use). Should an exposure occur, they are advised to decrease the amount of the visual field and thus the brain that is stimulated by increasing their distance from the potential source or by occluding one eye until the stimuli pass.²

Retinal Damage

Because the eye focuses visible light, the retina may be damaged if sufficient energy is absorbed to raise its temperature and denature retinal proteins. The aversion response to bright light—contraction of the pupil, turning the head, closing the eyes—can protect against this damage, but in the context of very powerful light or prolonged exposure, the aversion response may not be sufficient to prevent acute injury. In animal studies, the retina appears to be particularly sensitive to blue light with wavelengths between 400 and 460 nm. Blue light has a unique photochemical effect on the retina compared to colours of other wavelengths, making the retina more sensitive.

The International Commission on Non-ionizing Radiation (ICNIRP) has established limits for light sources to protect the human retina: Light sources with luminance levels below 10,000 cd/m² are not hazardous and radiometric measurements are not necessary.⁵ If the luminance exceeds 10,000 cd/m², radiometric surveys around the light source are needed to ensure the light’s effective radiance is below the limit of 100 W/m².sr for long exposures (e.g., greater than 10,000 seconds, equivalent to 2 hours and 45 minutes).

The European Union’s Scientific Committee of Emerging and Newly Identified Health Risks (SCENIHR) conducted a review of the health effects of artificial lighting on the general population.⁶ This review, however, examined artificial lighting used for visibility purposes (e.g., lamps, light bulbs) rather than other sources of artificial light such as LED screens, televisions, or tablet computers. Based on the ICNIRP guidelines, the review concludes that typical artificial lighting for visibility would not cause acute damage to the eye, and there is no evidence that chronic exposure to visible light (in the absence of UV light) has long-term effects.

Stress and Annoyance

Features of lighting associated with annoyance or stress have been primarily studied in indoor, occupational settings. A wide variety of workplaces have been studied, including office and health care settings.⁷ Flickering fluorescent lights are a common source of complaint in this literature; headaches, eyestrain and general stress are attributed to flicker. Replacing magnetic fluorescent ballasts with digital ballasts is highly effective at reducing visual discomfort and self-reported stress among office workers.8 Most LEDs are designed to control for these fluctuations, and perceptible flicker occurs only when the LED is deeply dimmed.⁹

No studies examining physiological stress responses to unwanted outdoor light in humans were found. However, environmental parameters that are not in the control of an individual such as heat, noise or light, may induce stress and annoyance. For instance, in the office setting, an inability to control the lighting levels is associated with increased worker stress and decreased productivity.⁷

Circadian Rhythm and Sleep Disruption

Almost all organisms, including humans, have 24-hour circadian and biological rhythms, which have profound effects on physiological and behavioural processes. In humans, the circadian rhythm is regulated by several genes, hormones, and the environment. Light is the major environmental modulator of the human circadian rhythm through its effect on the hormone, melatonin. Melatonin has many functions in humans including the regulation of the sleep-wake cycle, immune system, and metabolism, and has anti-cancer properties such as scavenging free radicals and activating antioxidative pathways.¹⁰ Exposure to artificial light at times of normal darkness, “light at night,” suppresses melatonin and disrupts the natural circadian rhythm. Retinal stimulation by light at certain wavelengths between 420 and 520 nm (“blue” spectra light) appears to have strongest inhibitory effects on melatonin production and, although this effect is independent of the technology generating the light, some light sources may have stronger spectral power distributions in this range (e.g., LED vs tungsten incandescent).¹¹ If sleep is disrupted, associated effects may include changes in appetite and energy balance, body mass index, blood glucose regulation, and increased risk of developing type 2 diabetes.^12-14 These physiological and behavioural changes associated with sleep deprivation can be more significant for the young, elderly, those with chronic conditions, and shift workers. However, these effects have yet to be studied in relation to exposure to outdoor light at night, and no causal links have been established.

Requirements for Risk Assessment

In order to conduct a thorough assessment for this case study, further information is required. More data on exposure is needed, including the screen manufacturer, flickering frequency, and maximum luminance. Data would also need to be collected on the population at risk, including the distance of nearby residential buildings and the line of sight to the screen. Characteristics of the populations are also needed. The risk assessment should accommodate the population of greatest risk from the LED screen exposure including those with photosensitive epilepsy. Filling the current data gaps would help identify any appropriate health endpoints with consideration of their prevalence and severity. A population-based survey that provides population demographics as well as information regarding self-reported effects (health or nuisance) from the community to the LED screen could also inform this process.

Conclusions

• Additional information regarding the LED screen, exposure, and population characteristics is needed to fully assess the issue of potential health effects associated with residing nearby the LED billboard screens.
• The risk of PSE should be minimal if images on the LED screen comply with the International Telecommunications Union recommendations to prevent photosensitive seizures.
• The screen’s maximum luminance should be less than 10,000 cd/m² to avoid retinal damage. Given that the screen is operating at less than full capacity at all times, it seems unlikely to exceed the 10,000 cd/m² threshold. If the luminance of the screen exceeds this limit, a more detailed exposure assessment is required. Although the possibility of retinal damage seems limited, no research has been done specifically on large LED screen exposure and retinal damage.
• As a bright light source from which the residents have no control, the LED screen may promote annoyance and stress to residents living nearby. The LED technology should eliminate perceptible flicker, which has itself been associated with annoyance.
• The screens contribute to the light associated with urban space. However, we are not able to determine the degree to which extraneous light from the LED screens contributes to the disruption of sleep or circadian rhythms, if any, in local residents; light is only one of many factors affecting these biological rhythms.
• The lack of information on exposure characteristics of the LED video billboards and the affected population, along with the lack of scientific evidence informing the process, precludes undertaking a current risk assessment.

Acknowledgements

We would like to thank Joanna Oda, Daniel Fong, Abderrachid Zitouni, and Tom Kosatsky for research and write-up. Helen Ward and Anne-Marie Nicol provided valuable input and review of this document.

References

1. Rogers SP, Spiker VA, Cicinelli J. Luminance and luminance contrast requirements for legibility of self-luminous displays in aircraft cockpits. Appl Ergon. 1986 Dec;17(4):271-7.
2. Erba G. Shedding light on photosensitivity, one of epilepsy’s most complex conditions. Landover, MD: Epilepsy Foundation; [cited 2012 July 14, 2012].
3. Wilkins A, Emmett J, Harding G. Characterizing the patterned images that precipitate seizures and optimizing guidelines to prevent them. Epilepsia. 2005;46(8):1212-8.
4. International Telecommunication Union. Guidance for the reduction of photosensitive epileptic seizures caused by television. Recommendation ITU-R BT.1702. Geneva, Switzerland: ITU; 2005.
5. International Commission on Non-Ionizing Radiation Protection. Guidelines on limits of exposure to broad-band incoherent optical radiation (0.38 to 3 µM). Health Physics. 1997;73(3):539-54.
6. Scientific Committee on Emerging and Newly Identified Health Risks. Health effects of artificial light. Brussels, Belgium: European Commission; 2012.
7. Cummings J. Wind farm noise and health. Lay summary of new research released in 2011. Santa Fe, NM: Acoustic Ecology Institute; 2012 Apr.
8. Wilkins AJ, Nimmo-Smith I, Slater AI, Bedocs L. Fluorescent lighting, headaches and eyestrain. Light Res Technol. 1989 March 1, 1989;21(1):11-8.
9. Bullough JD, Sweater Hickcox K, Klein TR, Narendran N. Effects of flicker characteristics from solid-state lighting on detection, acceptability and comfort. Light Res Technol. 2011;43(3):337-48.
10. Blask DE. Melatonin: chronobiological and chronopharmacological role in cancer prevention and treatment. Alternative Medicine Alert. 2005;8(10):109-13.
11. Bartholomew RE. Re: "Epidemic hysteria: a review of the published literature". Am J Epidemiol. 2000 Jan 15;151(2):206-7.
12. Taheri S, Lin L, Austin D, Young T, Mignot E. Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased body mass index. PLoS Med. 2004 Dec;1(3):e62.
13. Van Cauter E, Spiegel K, Tasali E, Leproult R. Metabolic consequences of sleep and sleep loss. Sleep Med. 2008 Sep;9 Suppl 1:S23-8.
14. Chapman S, Simonetti T. Summary of main conclusions reached in 17 reviews of the research literature on wind farms and health. Sydney, Australia: School of Public Health, University of Sydney; 2012 Jan.

January 2013