The harmful effects of CFLs and LEDs to skin and eyes

Health effects of CFLs, LEDs and halogen lamps

  1. General notions about UV radiation
  2. Use of the IEC62471 standard to evaluate LED-based products
  3. Harmful effects of CFLs
  4. Harmful effects of LEDs
  5. Depression, a common effect of CFLs and LEDs
  6. Harmful effects of halogen lamps
  7. Discussion concerning the answer given by the European Commission to my question

More about recent books about CFLs, LEDs and diseases such as AMD can be found here.

Please help LightAware to stimulate investigation into how artificial light affects human health and wellbeing, and to create a vital resource for people who cannot tolerate new forms of lighting.

1. General notions about UV radiation

UV light radiation
  • UVC: from 100 to 280 nm. (The radiations with wavelengths from 10 to 180 nm propagate only in vacuum and thus are of litte value for our investigation. (Navy Environmental Health Center, Ultraviolet Radiation Guide, 1992 'UVRG1992')) The radiation is transmitted through air and through quartz but absorbed by ordinary glass. (UVRG1992)
  • UVB: from 280 to 315 nm; The radiation is transmitted through air and through quartz but absorbed by ordinary glass. (UVRG1992)
  • UVA: from 315 to 400 nm; is easily transmitted through air and glass. (UVRG1992) A distinction is made between
  • visible radiation: from 380 to 800 nm

The absorption of UVB induces direct DNA damage: basis changes, induced covalents (abnormal formation of chemical bonds), strand breaks (DNA strands drift apart). UVA radiation and visible light are not absorbed by DNA but by exogenous or endegenous chromophores that can, in an excited form, degrade the genome. (p. 78)

Source: Case Western Reserve University, Ultra Violet Radiation Safety (UVRS)

Special attention should be given to children and special groups:

Special attention should be given to children because of the transparency of their crystalline lens and both aphakics (with no crystalline lens) and pseudophakics (with artificial crystalline lenses) who consequently either cannot or can only insufficiently filter short wavelengths (especially blue light); (ANSES_report 2010, p. 73-74:) In young people (before 10 years), the lens passes virtually all blue light (80%), especially waves between 430 and 440 nm, the most dangerous to the retina (peak absorption at 360 nm). With age, the lens becomes yellow and absorbs shorter wavelengths. This change in age-dependent transmission protects the retina from the blue light and reduces scotopic vision (vision at night) significantly (about 33% at 50 years compared to 5 years). At the age of 50, protection against UVA, UVB and blue light grew 80%. SCENIHR, Health Effects of Artificial Light, March 19, 2012 gives the following figures:

Wavelength Age < 9 years Age 10 years Age 60-70 years
320 nm 2 - 5 % reaches the retina No UV of 320 nm reaches the retina No UV of 320 nm reaches the retina
400 nm 15 % reaches the retina 15 % reaches the retina 1 % reaches the retina
460 nm 65 % reaches the retina 60 % reaches the retina 40 % reaches the retina

Overview of biological affects caused by UV radiation

Hazard Biological Affects: Exposure to Skin Biological Affects: Exposure to Eyes
  • Suntans are related to UVA exposure. They do not cause sunburns because of their lower energy than UVB or UVC.
  • Commercial UVA in the form of a black light emits long wave radiation with very little visible light.
  • The long waves of UVA generates free radicals and causes indirect DNA damage which is responsible for malignant melanoma.
  • Since UVA penetrate deeper they damage collagen fibers and destroy vitamin A.
  • UV-A passes through the cornea to the lens and overexposure contributes to the formation of cataracts by creating oxidants that cause accelerated formation of cataracts.
  • Corneal damage is possible since UVA passes through it to get to the lens.
  • Erythema or "sunburns" are related to UVB exposure. Symptoms depend on the intensity and or length of the exposure.
  • Skin cancer, the most deadly form malignant melanoma, is caused by indirect DNA damage from UVB.
  • Direct photochemical damage to DNA also causes skin cancers.
  • One positive affect of moderate doses of UVB is that in induces the production of vitamin D and vitamin K.
  • Photokeratitis, Welders Flash, or Arc Eye is literally burning of the cornea by intense exposure to UVB.
  • Cataracts can form as described with UVA affects.
  • Inflammatory, invasive and proliferating lesions called pterygia can form on the cornea.
  • Pinguecula or yellowish deposits between the cornea and sclera can occur.
  • The most common injuries of UVC are corneal burns and erythema or severe skin burns.
  • UVC burns are painful, but most injuries are short lived.
  • Excessive exposure to UVC causes skin cancers as UVA and UVB.
  • Although literature on UVC damage is scarce since it is relatively benign in the natural form, it is the most dangerous form industrially. It can cause damage to eyes in as little as 3 seconds and DNA damage to all biological surfaces.
  • Photokeratitis is prevalent documented injury.
  • Chronic exposures to acute intense UVC can lead to cataract formation and retinal damage.

Source: Case Western Reserve University, Ultra Violet Radiation Safety (UVRS)

2. Use of the IEC62471 standard to evaluate LED-based products

These two standards (IEC 62471 and EN 62471) are identical, with exception of the limit values (Threshold Limit Values or TLVs). The international standard IEC 62471 sets limit values on the exposure to optical radiation, while the NF EN 62471 refers to the exposure limit values of Directive 2006/25/EC (DIRECTIVE 2006/25/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 5 April 2006 on the minimum health and safety requirements regarding the exposure of workers to risks arising from physical agents (artificial optical radiation)). Limit values (TLVs) are calculated for exposure to a source for a working day of 8 hours. However, the TLVs in force relating to the risk of blue light, cannot evaluate the risk of chronic daily exposure, especially because they do not take into account the long-term risks and cumulative exposures. (ANSES_Report 2010, p. 103)

Photobiology is the interaction of optical radiation with living organisms. Optical radiation is defined as electromagnetic radiation having wavelengths between 100 nm in the deep ultraviolet (UV) to 1 mm in the far infrared (IR). However, this range is often restricted for practical purposes to 200-3000 nm due to atmospheric absorption below 200 nm, and the negligible effect of low-energy photons in the far IR. (Leslie Lyons, Part 1)

Photobiological risk corresponding to the risk factors of this standard are risks associated with the ultraviolet, visible light and infrared. In the visible range, the risk for vision are the risks associated with blue light. The standard suggests a risk of retinal damage. The standard specifies that for blue light, "the photochemical reaction initiates a chain of biological reactions apparently centered on the retinal pigment epithelium," and they are "misunderstood pathophysiological mechanisms." (ANSES_Report 2010, p. 114)

Scope of IEC62471

  • Considers exposure to the skin and eye through consideration of six hazards over the wavelenght range 200 nm to 3000 nm. (Lyons_pdf)
  • Long-term exposure not addressed nor account taken of abnormal photosensitivity or the presence of photosensiters (Lyons_pdf)
  • It covers six hazards to the skin and eyes for exposure up to eight hours.
  • It defines a classification structure, from Exempt (no risk) to Risk Group 3 (hazardous even with momentary exposure).

Table 1: The six photobiological hazards to the skin and eye (Source: LEDs Magazine, October 2011, p. 32)

N.B.: photokeratitis means 'arc eye/snow blindness'.

  • When light of specific wavelengths causes chemical changes in tissue, the damage mechanism is considered 'photochemical'. In a photochemical interaction, light of a specific wavelength (and therefore energy) excites electrons in cellular molecules, leading to the breaking or reorganization of chemical bonds therein. This may have direct consequences to DNA, whereby base pairs are bound together, creating a disruption in the DNA strand. Indirectly, an excess of highly-reactive free radicals may be produced. These can interact with DNA to cause structural reorganization, and with other cells such as retinal photoreceptors to cause deterioration of cellular function and cell death. Importantly, damage to DNA, if not repaired, has the potential to give rise to cancer. (...) Photochemical processes are dose dependant, meaning that low-level, long-term exposure gives rise to the same damage as high-level, short-term exposure. (Leslie Lyons, Part 1)
  • When light of larger wavelengths causes local tissue heating, the damage mechanism is considered 'thermal'.

Table 2: Description of the IEC62471 standard and risk groups

The 62471 standard concerning the photobiological safety of lamps and devices using lamps suggests maximum exposure limits for radiation from light sources commonly used for lighting, and provides a method of classification based on radiance and actual irradiance together with a method for measuring these values. This standard covers all photobiological hazards for the eye (thermal and photochemical hazards), for ultraviolet to infrared wavelengths. (Source: ANSES_Opinion 2010, p. 6)

Risk Group Basis
Exempt (Risk Group 0) The product involves no photobiological risk.
Risk Group 1 (Small risk) The product involves no risk in terms of maximum exposure limits under normal conditions of use.
Risk Group 2 (moderate risk) The product involves no risk in terms of aversion response to very bright light sources or due to thermal discomfort.
Risk Group 3 (high risk) The product may involve a risk even during momentary or short exposure periods.

Update: See viewpoint of LED manufacturers

3. Harmful effects of CFLs

  • Harmful effects as a consequence of broken CFLs
  • Flicker
    • High frequency flickers are hardly noticeable to the naked eyes, but research has proven that certain levels of exposure to flickering lights could become hazardous to health. (*)
    • Low-frequency flicker can induce seizures in people with photosensitive epilepsy, and the flicker in magnetically ballasted fluorescent lamps used for office lighting has been linked to headaches, fatigue, blurred vision, eyestrain, and reduced visual task performance for certain populations. (**)
    • Flicker can also produce hazardous phantom array effects - which may lead to distraction when driving at night, for example - or stroboscopic effects, which may result in the apparent slowing or stopping of moving machinery in an industrial setting. (**)
    • When discussing the potential human impacts of flicker, it is important to understand the difference between sensation and perception. Sensation is the physiological detection of external conditions that can lead to a nervous system response, while perception is the process by which the brain interprets sensory information. Some sensory information is not perceived, and some perceptions do not accurately reflect the external conditions. As a result, some people who suffer from flicker sensitivity may not be aware that flicker is the reason they are suffering, or even that the light source responsible for their suffering is flickering. Furthermore, not all human observers are equally sensitive to the potential effects of flicker. Populations that tend to be more susceptible to the effects of flicker include children, people with autism, and migraineurs. While the sizes of some specific at-risk populations have been characterized - approximately 1 in 4,000 humans suffer from photosensitive epilepsy, for example - most have not. (**)
    • Unfortunately, most consumers' primary concern is whether or not the pricing fits their profile, without understanding that many existing manufacturers would lower their cost and quality and potentially put consumers' health at risk, just to cope with their price demands in exchange for market dominance. (*)

    (*):, 1 January 2015


  • Harmful effects to the skin
    • 2013: 16:9 Shedding Some Light on compact fluorescent bulbs. A very good video on the current state of research. It gives more information about ultraviolet radiation and electromagnetic emissions.
    • Harmful Effects of CFL Bulbs to Skin A team of Stony Brook University researchers published in the June issue of the journal of Photochemistry and Photobiology the study "The Effects of UV Emission from CFL Exposure on Human Dermal Fibroblasts and Keratinocytes in Vitro".Stony Brook researchers collected CFL bulbs purchased from different locations across Suffolk and Nassau counties, and then measured the amount of UV emissions and the integrity of each bulb’s phosphor coatings. Results revealed significant levels of UVC and UVA, which appeared to originate from cracks in the phosphor coatings, present in all CFL bulbs studied.

      At Stony Brook’s Advanced Energy Research and Technology Center (AERTC), the team took the same bulbs and studied the effects of exposure on healthy human skin tissue cells, including: fibroblasts, a type of cell found in connective tissue that produces collagen; and keratinocytes, an epidermal cell that produces keratin, the key structural material in the outer layer of human skin. Tests were repeated with incandescent light bulbs of the same intensity and with the introduction of Titanium Dioxide (TiO2) nanoparticles, which are found in personal care products normally used for UV absorption.

      "Our study revealed that the response of healthy skin cells to UV emitted from CFL bulbs is consistent with damage from ultraviolet radiation," said Professor Rafailovich. [Cells exposed to CFLs exhibited a decrease in the proliferation rate, a significant increase in the production of reactive oxygen species, and a decrease in their ability to contract collagen. (Abstract)] "Skin cell damage was further enhanced when low dosages of TiO2 nanoparticles were introduced to the skin cells prior to exposure." Rafailovich added that incandescent light of the same intensity had no effect on healthy skin cells, with or without the presence of TiO2.

      "Despite their large energy savings, consumers should be careful when using compact fluorescent light bulbs," said Professor Rafailovich. "Our research shows that it is best to avoid using them at close distances and that they are safest when placed behind an additional glass cover."

    • The following skin deseases are linked to ultraviolet radiation: (Source: ANSES_Report 2010, p. 99-101)

      N.B.: Ranking of evidence (Source: SCENIHR 2008)

      • level A: sufficient evidence
      • level B: some evidence
      • level C: inadequate evidence
      • level D: anecdotal evidence
      • level E: no reported effects

      • Idiopathic photodermatoses:
        • The most common photodermatose is PLE (Polymorphic Ligth Eruption). It appears in spring and early sommer as a rash, itching on areas exposed to sunlight. This type of reaction disappears during winter months. It is estimated that 10-20% of Europeans can be affected from the first three decades with a female prevalence. It is estimated that this response to aggression by UVA, mainly solar, without sunburn, depends on the production of abnormal proteins in the epidermis and dermis. (SCENIHR 2008: It is possible that in the most severely affected, CFL could produce the eruption [Evidence level C].) (SCENIHR 2012: In the absence of clinical data, it is reasonable to assume that in a small minority of individuals, provocation of PLE may follow artificial light UV exposure.)
        • Chronic actinic dermatitis: most patients have a long history of contact allergies and their skin is abnormally sensitive to UVA and visible radiation. This disease affects predominantly men aged over 50 years (in Scotland, 17 subjects per 100 000 inhabitants). This type of injury is sometimes associated with young adults with atopic dermatitis. Photosensitivity can be very severe. (SCENIHR 2008: Degree of photosensitivity suggests there may be a problem with CFL (Moseley 2008) [Evidence level C].) (SCENIHR 2012: Severe and perhaps even moderately affected individuals with this condition may, when exposed to artificial UV or visible light, experience induction of CAD.)
        • Actinic prurigo: this condition rarely reaches European populations of Caucasian or Asian. Subjects are affected in the first decade. Subjects emphasize deterioration of the skin condition occurring in spring and summer. Solar prurigo is displayed as edematous erythema sclerosis reinforced by papules from sun exposure. One can trigger this type of injury by repeating the UVA provocation test. (SCENIHR 2008: Severe cases may potentially be at risk from CFL (Moseley 2008) [Evidence level C].) (SCENIHR 2012: ... and other UV emitting sources.)
        • Solar urticaria: it is a rare skin disorder which affects men and women, particularly in the first 4 decades of life. This condition is persistent and there is no known treatment for one third of the subjects. It results from exposure to UVA extending into the visible. The incitation of this type of injury is simply using a slide projector. (SCENIHR 2008: It is possible that some patients could be at risk from CFL. It should be noted that incandescent light sources also cause problems in some patients [Evidence level C].) (SCENIHR 2012: Severely affected patients may be at risk from CFL and unfiltered halogen sources producing UV/visible radiation. It should be noted that incandescent light sources also cause problems in some patients.)
      • Photosensitivity due to chemicals and drugs: Many drugs are known for their ability to induce photosensitive skin reactions. For most phototoxicity is triggered provided that appropriate dose of light is applied. Among the most frequently photosensitizers, amiodarone (heart antidisrythmique agent), the derivatives of phenothiazine and fluoroquinolone antibiotics are primarily responsible for reactions to UVA. It should be noted that these patients showed no abnormal reactions to incandescent sources. (SCENIHR 2008: Given the degree of photosensitivity, it is not anticipated that drug induced photosensitivity to the above will be a particular problem when patients are exposed to CFL vs. incandescent sources [Evidence level C].) (SCENIHR 2012: With photofrin, photosensitivity might be expected to occur with CFL and LED sources to a greater extent than that currently seen with incandescent lighting. This is due to a combination of greater sensitivity of porphyrins to blue light (soret band), coupled with an enhanced blue light emission of these sources.)
      • Genophodermatoses: This group (250 000 people in Europe) includes xeroderma pigmentosum (XP) and Bloom and Rothmund-Thomson syndrome. These pathologies are the result of a system of excision repair of deficient DNA with marked sensitivity to UVB. This results in the early onset of skin cancers. (SCENIHR 2008: It is possible that unfiltered CFL could be associated with increased disease activity. Patients are currently advised to avoid unfiltered fluorescent lighting. There could be assumed to be a similar problem with other members of the group [Evidence level C].) (SCENIHR 2012: UV radiation from artificial light sources is associated with an increased skin cancer risk in XP. Patients are currently advised to avoid all sources emitting UVB/A wavelengths. These would include CFLs and unfiltered halogen bulbs.)
      • Porphyrias: These pathologies are related to the presence of porphyrins in the skin and have as symptom denominator intolerance after an exposure of a few minutes to visible light. The erythropoietic protoporphyria develops from childhood. It should be noted that the cutaneous porphyrias are particularly sensitive in the blue region of the visible. This disease affects about 2 persons on 1 000 inhabitants. (SCENIHR 2008: CFL in extremely sensitive patients could possibly produce a slight increase in the problem compared to tungsten light sources, although there is published evidence against this (Chingwell et al, 2008, in press) [Evidence level C].) (SCENIHR 2012: Artificial, visible light sources which would include incandescent bulbs may produce skin reactions in the most sensitive patients.) More about Erythropoietic Protoporphyria (EPP) or Protoporphyria
      • Lupus erythematosus: It is a chronic autoimmune disease often exacerbated by sun exposure. This pathology affects 30 subjects out of 100 000 inhabitants. Some patients understood artificial light as an incentive agent. It is an erythematous rash affecting the face forming an aspect of butterfly wing on the cheeks and cheekbones. The long UVA (360-400 nm) are the causal element. (SCENIHR 2008: Through their UV component, chronic exposure to CFL could possibly be a problem. Systemic lupus is an important condition in that skin flares can be associated with internal disease activity [Evidence level C].) (SCENIHR 2012: It seems reasonable to assume that at least some LE patients, and particularly those with SLE [Systemic lupus erythematosus], are at risk from chronic UV exposure from some low energy emitting lamps such as CFLs and unfiltered halogen bulbs. In this context it is noted that LE support groups are already advising the use of double rather than single envelope CFLs.)
      • Skin cancer: Ultraviolet radiation is a recognized risk factor. As a result, UV radiation, associated with any artificial light source must be minimized. Although carcinogenic doses emitted by fluorescent light sources used in the home or as a light source at work are minor (less than 1%), it was calculated that these could represent up to 30% of the total exposure of workers inside. These doses represent no more than 3 or 4% for workers outside. It has been shown definitively there was no increased risk of melanoma during exposure to sources of fluorescent light. (SCENIHR 2008: Fluorescent lamps do not contribute significantly to the melanoma risk [evidence level A] and by analogy CFL will not [Evidence level B]. Fluorescent lamps, including CFL, are estimated to contribute insignificantly to UV doses effective in causing skin carcinomas [Evidence level B].)


    • Progress in the knowledge of the penetration of visible radiation in the skin, and studies on the absorption spectrum of endogenous and exogenous chromophores in the skin concluded with the existence of various biological effects exerted by the visible portion of the spectrum of non-ionizing radiation. Besides the possibility to induce erythema or pigmentation and thermal damage, the production of reactive oxygen species in the skin is a reality. Through the generation of free radicals, visible light can induce indirect damage to DNA, thus contributing to a possible photo carcinogenicity by cumulation to the effects of ultraviolet radiation present during sun exposure. (Source: ANSES_Report 2010, p. 99-101)
    • SCENIHR 2008: Peer reviewed definitive test data comparing incandescent vs. CFL in these diseases is required to provide a clear answer to the question being asked in this report.
      • There is sufficient evidence to show that UV and in some cases visible radiation from lamps can provoke a clinically significant skin reaction in light-sensitive patients [Evidence level A].
      • Fluorescent lamps, including CFL emit UV radiation that may be harmful to a subset of particularly sensitive patients [Evidence level C].
      • CFL may be harmful when in close proximity to the skin (around 20 cm or less) [Evidence level B].
  • Harmful effects to non-skin pre-existing conditions
  • With evidence A:

    • It is unlikely that any EMF emitted from CFL or other fluorescent lamps would contribute to electromagnetic hypersensitivity [Evidence level A].
    • There is sufficient evidence [Evidence level A] that the conditions of patients with Irlen-Meares syndrome are not influenced by CFL.(p. 22) But the conclusion of the chapter 'Irlen-Meares' was much more limited: It [is] has been shown that dyslectics and Irlen-Meares patients tend to have difficulties detecting flicker. Therefore, flicker from fluorescent tubes should not be a problem [Evidence level A]. An extrapolation from fluorescent tubes to CFLs is not permitted, since CFLs are more dangerous, due to the cracks in the phosphor coatings.
    • There is evidence showing that flicker can cause seizures in patients with photosensitive epilepsy [Evidence level A]
    • Migraine can be induced by flicker [Evidence level A]
    • Fibromyalgia: Light conditions do not play a role in fibromyalgia [Evidence level A]. Problems with fluorescent lamps are not investigated but are very unlikely [Evidence level E].

    With evidence B, C

    • Blue light can aggravate retinal diseases in susceptible patients [Evidence level B]
    • It cannot be excluded that Photophobia is induced or aggravated by different light conditions, but it is not even mentioned in self-reports [Evidence level C].
    • It is unlikely that fluorescent lamps can cause snow-blindness or cataracts [Evidence levels B, C].

    With evidence D

    • Blue light can possibly aggravate migraine [Evidence level D].
    • There are self-reported indications that the condition [regarding dyslectics and Irlen-Meares patients] is aggravated by mainly UV and blue light [Evidence D].
    • Myalgic encephalomyelitis (Chronic Fatigue Syndrome): Symptoms may be aggravated by many factors, including light conditions as stated by self-reporting [Evidence level D]. There is no evidence for a link between chronic fatigue syndrome and fluorescent lighting [Evidence level E].

    With evidence E: not mentioned here

Read the evidence of one of the many light sensitive persons living in the United Kingdom. This article was published in Chronicle (Newcastle upon Tyne) on 15 April 2013.

Regarding Irlen-Meares patients, read the alarming report Student eye checks after shock tests.

General conclusions: (SCENIHR 2008)

  • SCENIHR did not find suitable direct scientific data on the relationship between energy saving lamps and the symptoms in patients with various conditions (i.e xeroderma pigmentosum, lupus, migraine, epilepsy, myalgic encephalomyelitis, Irlen-Meares syndrome, fibromyalgia, electrosensitivity, AIDS/HIV, dyspraxia, and autism).

    Therefore, SCENIHR examined whether three lamp characteristics (flicker, electromagnetic fields, and UV/blue light emission) could act as triggers for disease symptoms. Due to lack of data on CFLs, existing data on traditional fluorescent tubes were extrapolated to situations when compact fluorescent lamps may be used. (Emphasize added)

    While for some conditions either flicker and/or UV/blue light could exacerbate symptoms, there is no reliable evidence that the use of fluorescent tubes was a significant contributor. Of all compact fluorescent lamps properties, only UV/blue light radiation was identified as a potential risk factor for the aggravation of the light-sensitive symptoms in some patients with such diseases as chronic actinic dermatitis and solar urticaria.

    The committee wishes to draw attention of the Commission Services to the fact that it has been observed that some single-envelope CFLs emit UVB and traces of UVC radiation. Under extreme conditions (i.e. prolonged exposures at distances <20 cm) these CFLs may lead to UV exposures approaching the current workplace limit set to protect workers from skin and retinal damage.

    Due to the lack of relevant data, the number of all light-sensitive patients in the European Union, who might be at risk from the increased levels of UV/blue light radiation generated by CFL is difficult to estimate. However, a preliminary rough estimation of the worst-case scenario yields a number of around 250,000 individuals (0.05% of the population) in the EU.

    The committee notes that the use of double-envelope energy saving bulbs or similar technology would largely or entirely mitigate both the risk of approaching workplace limits on UV emissions in extreme conditions and the risk of aggravating the symptoms of light-sensitive individuals. (Source: SCENIHR (Scientific Committee on Emerging and Newly-Identified Health Risks), Light Sensitivity, 23 September 2008, p. 4.)


  • Photofrin and other anti-cancer photodynamic therapy (PDT) agents: With photofrin, photosensitivity might be expected to occur with CFL and LED sources to a greater extent than that currently seen with incandescent lighting. This is due to a combination of greater sensitivity of porphyrins to blue light (soret band), coupled with an enhanced blue light emission of these sources.
  • (p. 81) The previous SCENIHR opinion on Light Sensitivity (SCENIHR 2008) identified that some pre-existing conditions (epilepsy, migraine, retinal diseases, chronic actinic dermatitis, and solar urticaria) could be exacerbated by flicker and/or UV/blue light. However, at that time there was no reliable evidence that compact fluorescent lamps (CFLs) could be a significant contributor. This conclusion needs updating as more recent studies indicate a negative role for certain CFLs and other artificial light sources (including sometimes incandescent bulbs) in photosensitive disease activity.
  • (p. 81) it seems reasonable to assume that the UV, and in some cases the blue radiation component of artificial lighting in an as yet undefined number of patients, may contribute to the aggravation of symptoms related to their skin disease, and in the case of lupus erythematosus possibly also to the aggravation of their systemic disease.
  • (p. 21) Single envelope CFL classified as RG1 [see explanation] may be hazardous to a photosensitive patient if used closer than 20 cm to the skin.
  • (p. 83) While a second envelope undoubtledly reduces the UV component of any particular lamp, the currently available data, however, documents the high variability of UV and blue light emissions due to different internal design parameters.

My comment:

  • The committee notes that the use of double-envelope energy saving bulbs or similar technology would largely or entirely mitigate ... What is the scientific meaning of 'largely or entirely'? What if double-envelope CFLs are not the desired solution to their problems?
  • Double-envelope CFLs do not meet the required standards! See newspaper reports, May 2012.
  • SCENIHR is playing a dangerous game. The committee stated in the conclusion that they did not find suitable direct scientific data on the relationship between energy saving lamps and the symptoms in patients with xeroderma pigmentosum, lupus, migraine, epilepsy, ...while the contrary is ascertained explaining each case. Also, the committee has the opinion that as long as no studies are available that prove a direct harm on the health of the citizens, the modern lighting is fully permitted.
  • Due to lack of data on CFLs, existing data on traditional fluorescent tubes were extrapolated to situations when compact fluorescent lamps may be used. It is known that a CFLs differ from linear fluorescent tubes on an essential point: when phosphors are applied to the CFL's tight coils, it cracks so that UV leaks can originate. This was not investigated by SCENIHR.
  • It is known that some medications can induce photosensitivy. SCENIHR is lacking expertise by stating that drug induced photosensitivity will not be a particular problem when patients are exposed to CFL vs. incandescent sources [Evidence level C]. Evidence C is not sufficient to eliminate all risks!

Attention should be given to this statement from the NCBI website:

"A real problem for the public health aspect is that we have really insufficient knowledge about the actual exposure [to UV radiation],"says Mattsson, who chaired the SCENIHR in 2012. "Emissions are not the same as knowing the exposure."

That’s an important distinction, says Brian Pollack, an assistant professor of dermatology and pathology at Emory University, because "UV radiation is carcinogenic. The bottom line is if these [bulbs] are emitting UV radiation of any amount, it needs to be defined, and it needs to be prevented."

  • Effect on the eyes
  • Report published in Am. J. Public Health (December 2011) Helen L. Walls, Kelvin L. Walls, Geza Benke, Eye Disease Resulting From Increased Use of Fluorescent Lighting as a Climate Change Mitigation Strategy.

    Abstract: "Increased use of fluorescent lighting as a climate change mitigation strategy may increase eye disease. The safe range of light to avoid exposing the eye to potentially damaging ultraviolet (UV) radiation is 2000 to 3500K and greater than 500 nanometers. Some fluorescent lights fall outside this safe range. Fluorescent lighting may increase UV-related eye diseases by up to 12% and, according to our calculations, may cause an additional 3000 cases of cataracts and 7500 cases of pterygia annually in Australia. Greater control of UV exposure from fluorescent lights is required. This may be of particular concern for aging populations in developed countries and countries in northern latitudes where there is a greater dependence on artificial lighting."

    Some excerpts: "The replacement of incandescent lamps with fluorescent lighting appears to be a global trend. However, this change in lighting sources may lead to an increase in eye diseases unless there is greater control of UV exposures from many of the fluorescent lights currently in use or technological advances enabling efficient lighting from other sources. For Australia alone, we estimate at least 10 000 additional cases of eye disease each year. Our estimates are conservative and crude in that they are limited by the poor information currently available with regard to the incidence and etiology of many eye diseases. We have not included in our estimates possible increases in AMD because there is not yet universal agreement in the literature regarding causality with UV radiation. But if a link between UV radiation and AMD is firmly established in the future, this would have significant public health implications.

    Kitchel also suggested that UV light causes irreparable damage over time to the human retina, especially in young children, a public health issue that has not been investigated.

    The safe range of light, to avoid exposing the eye to potentially damaging UV radiation, appears to be between 2000 and 3500K and a wavelength of greater than 500 nm. Some fluorescent lights currently fall outside this safe range. This may increase UV-related eye diseases by up to 12% (...) and result in unforeseen adverse public health consequences. There is a conflict between climate change mitigation through elimination of incandescent lights and the unregulated use of primarily fluorescent lighting.

    In response, we advocate for the use of incandescent and warm-white lamps instead of cool-white fluorescent lamps, as well as for further research into improving lighting from such sources. This public health issue may be of particular concern for aging populations, such as those of many developed countries and countries in northern latitudes where there is a greater dependence on artificial lighting."

    4. Harmful effects of LEDs

    A recent report has appeared: "Scientific Committee on Health, Environmental and Emerging Risks (SCHEER), Preliminary Opinion on Potential risks to human health of Light Emitting Diodes (LEDs), 6 July 2017. A critical view on this report can be consulted here.

    • Health effects of lighting systems using LEDs
    • A. Characteristics of LEDs relevant to risk assessment (Source: ANSES_opinion 2010)

      Given the technological constraints and the imperatives concerning electrical efficiency, currently the most widely-used method for producing white light uses a yellow luminophore to transform part of the light from a blue diode. Some characteristics of LEDs are:

      • Spectral imbalance within the blue: The light spectrum from white LEDs is largely made up of very weak emissions ranging between blue and yellow, but with a high proportion of blue light (a blue spike in the spectrum). These characteristics are highly specific to LEDs, and are not found in other, traditional types of lighting.
      • High luminance: (i.e. the high brightness density per surface unit emitted by these very small sources.) LEDs are point sources of light that can be aggregated in lighting units to achieve high luminous flux. Because the emission surfaces of LEDs are highly concentrated point sources, the luminance of each individual source produces very high luminance, at least 1 000 times higher (107 cd/m2) than that from a traditional lighting source.
      • Stroboscopic effect: Depending on their architecture, the electrical power supplied to LED lighting systems can vary, causing fluctuations in the intensity of the light produced that are more or less perceptible to the naked eye. (Source: ANSES_Opinion 2010)

      N.B.: Comparison of the spectra of the sun and some lamps

      Note: all incandescent and halogen bulbs have, by definition, a CRI (Color Rendering Index) close to 100.

      B. Identified health issues (Source: ANSES_Opinion 2010, p. 5 - 8)

      Potential damage

      Two properties of LEDs contribute to the hazard. Their small size causes high retinal irradiance (glare) and the wavelength which has the highest potential to cause damage coincides with the typical peak wavelength of the blue LEDs that pump the phosphor in white PC-LEDs.

      The blue peak of cool white LEDs is and significantly greater than that of warm white LEDs. It is the popular cool white LEDs that have a greater potential for damage.

      At present there are no concerns about acute exposure – today’s highest power LEDs are classified at the lower end of RG2. It would be necessary to look at the source from a distance of 200mm for 40-50 seconds before reaching exposure limit values.

      However, cumulative exposure over eight hours should be considered, and further research should be done into the reported effects of long-term, low-level exposure on age-related macula degeneration. (Leslie Lyohs,

      a. Effects on the eyes

      • Risks related to the thermal effect of light: The risk of thermal effects is related to burns to the retina, generally resulting from short-term exposure to a very intense light. This type of danger concerns all wavelengths, from ultraviolet to infrared and the entire visible spectrum. This type of risk, usually associated with lasers, is unlikely in conventional uses of LEDs.
      • Risks related to the photochemical effects of blue light:
        • Characterisation of the risk: Evidence from human observation and experimental studies on cell cultures and various animal species has converged to demonstrate the specific toxicity of shortwave (blue) light to the retina. (...) Lesions occur on the outer retina (photoreceptors and cells of the pigment epithelium) and appear after some time has passed. (...) These interactions lead to the production of high doses of cytotoxic free radicals. The photoreactive pigments (lipofuscin) in the epithelium accumulate with age, increasing the risk of oxidative stress. Cellular death has functional consequences which are particularly significant as they concern the macular region (central vision).

          Following converging observations on experimental models, there is a strong suspicion that blue light aggravates ARMD (age-related macular degeneration). Epidemiological studies in humans have never clearly shown such effects, as a result of difficulties in evaluating the exposure and individual predisposition. (ANSES_Report 2010, p. 89: The pigment epthelium contains lipofuscin and melanin. Unlike melanin, which absorbs photons and plays a protective role, lipofuscin once excited, in particular by blue light, generates free radicals that can harm the cells themselves and neighboring cells.)

          In adults, the crystalline lens (which, as it turns yellow, partly absorbs blue radiation) and the macular pigments partially protect against this toxicity through their capacity to absorb blue light. These protective mechanisms are weaker in children (whose crystalline lenses are transparent), aphakics (with no crystalline lenses) and pseudophakics (with artificial crystalline lenses). There is also less protection available in cases of reduced macular pigment, as observed during certain macular pathologies (e.g. ARMD). (ANSES_Report 2010, p. 127: The car LED headlight may cause retinal damage in young children who find themselves on the same height as the emitting source.)

          Also CELMA (Federation of National Manufacturers Association), Optical safety of LED lighting, July 2011 gives the warning: Children are thus more sensitive to blue light hazard. Therefore, at places frequented by children particular care must be taken to ensure that lamps and luminaires are chosen and installed in such a way as to avoid people looking directly into the light source.

          The blue part of the visible radiation (410-460 nm) has a different impact on the eye, depending on the age.

          Wavelength Age < 9 years Age 10 years Age 60-70 years
          400 nm 15 % reaches the retina 15 % reaches the retina 1 % reaches the retina
          460 nm 65 % reaches the retina 60 % reaches the retina 40 % reaches the retina

          Source: SCENIHR, Health Effects of Artificial Light, March 19, 2012, p. 25-28

        • Retinal pathologies: (ANSES_Report 2010, p. 146-147)
          • Subjects with hereditary retinal lesions of the type of retinal hereditary dystrophy or of congenital optic neuropathy have an increased susceptibility to damage induced by light.
          • Patients with glaucoma and optic neuropathy may also be more sensitive because of mitochondrial damage induced by blue light on the ganglion cells (800 000 people in France) [Osborne et al., 2006, Osborne et al., 2008].
          • Subjects with accumulation of lipofuscin and alterations in pigment epithelium show an increased risk of photochemical damage induced by blue light. It is the same for ARMD patients (one million in France).
          • Patients with diabetic retinopathy have a particular sensitivity of the blue cones and may also be more sensitive to photochemical damage because of a preexisting metabolic stress (over a million in France).
        • Consumers of photosensitizing drugs: Some drugs may potentiate the phototoxic effects of light (non exhaustive list): the aminoquinolines, certain antibiotics, phenothiazines, tamoxifen, some nonsteroidal anti-inflammatory drugs, synthetic antimalarials, some antiepileptics, the hypericin, etc.). Alcohol poisoning (chronic alcoholism) is also a risk factor. (ANSES_Report 2010, p. 147)
        • Exposure to LEDs: There is currently no information about human exposure to lighting, whether for systems using LEDs or other types of light sources. (ANSES_Report 2010, p. 127:) The Working Group can only provide quantitave assessments in the case of risk to blue light, measured according to EN 62471 (Photobiological safety). Nevertheless, the Working Group notes that this standard is ill-suited to lighting using LEDs and is based on exposure limits that do not take into account the daily chronic exposure.

          Recent research:

          • October 16, 2014, UV vs. blue light: Which is more dangerous? "On the other hand, visible light does reach the retina. Blue light, sometimes referred to as high-energy visible (HEV) light, encompasses 400 nm to 500 nm. Numerous studies confirm that cumulative lifetime exposure to blue light causes photo-oxidation of retinal cells that leads to AMD. As you probably know, AMD is the leading cause of blindness for those older than age 55 in the U.S. As the Baby Boomers age, 10,000 of them will turn 65 each day from 2011 to 2029, and AMD will reach epidemic proportions. Although numbers vary on the number of new dry AMD cases per year, it is projected that there are more than 2 million new cases of dry AMD (all categories, including "pre-AMD") and more than 200,000 new cases of wet AMD each year."
          • In "", Yoshiki Kuse, Kenjiro Ogawa, Kazuhiro Tsuruma, Masamitsu Shimazawa, Hideaki Hara, Damage of photoreceptor-derived cells in culture induced by light emitting diode-derived blue light. Murine cone photoreceptor-derived cells (661 W) were exposed to blue, white, or green LED light (0.38 mW/cm2).

            Results: "Blue LED light

            • damaged most severely compared to white and green LED light
            • exposure increased ROS [reactive oxygen species] generation compared to white and green LED lights
            • altered the protein expression level
            • induced the aggregation of short-wavelength opsins (S-opsin), resulting in severe cell damage.

            Age-related macular degeneration (AMD), a retinal degenerative disease, affects more than 30% of the people at or over 75 years of age. The pathogenesis of AMD usually advances with retinal photic injury caused by excessive light exposure and consequent oxidative stress. The retina contains much chromophores which can lead to the photochemical damage when excited at the each wavelength light, and age-related decrease of antioxidants such as superoxide dismutase (SOD) and increase of ROS following light exposure can progress to the pathology of AMD. The loss of vision is the major symptom of retinal diseases such as AMD.

            Although the effect of blue LED light in altering the circadian rhythm has been reported, the retinal photoreceptor cell damage induced by blue LED light is not fully understood.

            Overall, the LED light induced cell damage was wavelength-, but not energy-dependent and may cause more severe retinal photoreceptor cell damage than the other LED light."

          • Chamorro, E., Bonnin-Arias, C., Pérez-Carrasco, M. J., de Luna, J. M., Vázquez, D. and Sánchez-Ramos, C. (2013), Effects of Light-emitting Diode Radiations on Human Retinal Pigment Epithelial Cells In Vitro. Photochemistry and Photobiology, 89: 468–473. doi: 10.1111/j.1751-1097.2012.01237.x: Light-emitting diodes (LEDs) are the basic lighting components in screens of PCs, phones and TV sets; hence it is so important to know the implications of LED radiations on the human visual system. The aim of this study was to investigate the effect of LEDs radiations on human retinal pigment epithelial cells (HRPEpiC). They were exposed to three light–darkness (12 h/12 h) cycles, using blue-468 nm, green-525 nm, red-616 nm and white light. It is shown that LED radiations decrease 75–99% cellular viability, and increase 66–89% cellular apoptosis. They also increase ROS [Reactive oxygen species] production and DNA damage. Fluorescence intensity of apoptosis was 3.7% in nonirradiated cells and 88.8%, 86.1%, 83.9% and 65.5% in cells exposed to white, blue, green or red light, respectively.

            An investigator at Madrid’s Complutense University, Dr Celia Sánchez Ramos, says the retina, a highly sensitive tissue covering the eye, never regenerates itself once it has become damaged.'This problem is going to get worse, because humans are living longer and children are using electronic devices from a young age, particularly for schoolwork,' Sánchez-Ramos told. (...) Humans have their eyes open for around 6,000 hours a year, and most of this time they are exposed to artificial light, for which reason Dr Sánchez Ramos says the best way to prevent damage is to "close your eyes often to soften the impact". (Source: Complutense University)

          • 30 March 2017: Could light from LED screens cause irreversible eye damage? "Retinas of animals exposed to unprotected LED screen showed a 23% retinal cell death. In addition, all factors involved in the process of apoptosis (programmed cell death) were inhibited or neutralized by the use of the Reticare protector. (...) Reticare eye filters should be used on the screen, so that the emitted light can be the healthier as possible. Periodic breaks should also be performed using the 20/20/20 rule (ie, every 20 minutes rest 20 seconds looking at 20 feet). The study carried out by the Complutense University of Madrid (Spain) - UCM - is the first to evaluate retinal damage using tablets with LED screens available in the market with living animals."
          • L. Udovičić, F. Mainusch, M. Janßen, D. Nowack, G. Ott, Photobiolobigische Sicherheit von Licht emittierenden Dioden (LED), Bundesanstalt für Arbeitsschutz und Arbeitsmedizin, 2013. Almost all LEDs which exceeded the emission limit value of the Exempt Group were white or blue LEDs (except for one green LED). During a deliberate long-term close-distance look into a white or blue LED, the exposure limit value for photochemical retinal hazard can be exceeded in only 10 seconds. The sum of single exposures at certain workplaces (such as the LED industry, the installation of lighting systems or the theater and stage lighting) can exceed this time rapidly.
          • Mei-Lin Peng, Cheng-Yu Tsai, Chung-Liang Chien, John Ching-Jen Hsiao, Shuan-Yu Huang, Ching-Ju Lee, Hsiang-Yin Lin, Yang-Cheng Wen, Kuang-Wen Tseng. The Influence of Low-powered Family LED Lighting on Eyes in Mice Experimental Model. Life Science Journal. 2012;9(1):477-482. This study set up the LED-irradiated animal model to test if any potential risk on the retina of eyes.

            a. The white house LED exhibited (in this study) a power-peak of at 450 nm. These data indicate that white house LED lamps might have additional risk of degenerative factors on photoreceptor cells of retinal tissues. b. The findings of the present study demonstrate that the thicknesss of ONL (outer nuclear layer) decreases by long blue light exposure over time. c. Removing the lens by cataract surgery increases the amount of light exposure that approaches the retina. The light from white house LED has a peak at 450 nm. Accordingly, the longer blue light from the LED lamps was not filtered through the yellow intraocular lens and might was hazardous to the photoreceptor cells of retina.

          • more studies about LED lighting can be found on
        • Attitude of U.S. Department of Energy (DOE): As could be expected, DOE minimizes the dangers of LEDs and does not consider recent research findings in this area. See True Colors, October 2014.
          • "Just because there may be a distinct blue peak in their SPD [spectral power distribution] - in contrast with some other light sources, such as incandescent or daylight - LEDs do not necessarily have greater potential to cause retinal, material, or photobiological harm.
          • As with material and optical safety, it is sometimes argued that LED sources have a greater potential to affect the circadian system, which may have undesirable consequences if it occurs at the wrong time for the individual. As with the other risks, concerns often arise from the short-wavelength peak of a blue-pump LED package, which leads to the perception that LEDs emit more blue light. The situation may appear especially alarming if relative SPDs are shown.

            However, there are two important things to consider. First, the overall sensitivity of the human circadian system is still being rigorously debated. It is known that the photic input to the nonvisual system comes not just from melanopsin-containing ipRGCs (intrinsically photosensitive retinal ganglion cells), but also from rods and cones, the photoreceptors typically associated with visual function. Further, nonvisual photosensitivity is potentially mediated by a person's state of adaptation, the time of day, and the quantity of light. Thus, modeling circadian stimulation with a simple spectral weighting function is generally insufficient."

        • SCHEER, Preliminary Opinion on Potential risks to human health of Light Emitting Diodes (LEDs), July 2017.
          • p. 77: "Porphyrias are rare diseases. Prevalence and incidence figures vary substantially between type of porphyria and countries. The absorption spectrum of the porphyrins present in patients with photosensitive porphyrias overlaps the emission spectra of LED lighting sources. The SCHEER could not find evidence for increases in the incidence of porphyrias and photodermatoses since the publication of the Opinion on artificial light (SCENIR, 2012). Theoretically, the incidence of the chemical/drug-induced types of porphyrias and induction and aggravation of any of the photodermatoses may increase with increased light exposure in general. Although it seems possible to elicit certain visible light-induced photosensitivity disorders with LED lighting sources, it must be kept in mind that these diseases are rare."
      • Photobiological safety standards:

        The EN 62471 standard is unsuited to lighting systems using LEDs.

        1. Maximum exposure limits unsuitable for repeated exposure to blue light: In the light of current knowledge, the maximum exposure limits in force do not allow evaluation of daily chronic exposure limits to blue light. The classification of lamps by these values does not take account of the long-term risk resulting from accumulated exposure. This means that repeated and prolonged exposure could induce an accumulated risk potentially greater than that assessed using the maximum exposure limits. (ANSES_Opinion 2010, p. 7) (ANSES_Report 2010, p. 122)
        2. Ambiguity in measurement distances: For the most common lighting lamps, the NF EN 62 471 standard requires the risk group to be evaluated at the distance at which they produce a brightness of 500 lx. For other types of lamps, the risk group must be determined for the worst observation case, i.e. a distance of 200 mm.

          The risk group for any lighting system using LEDs can be determined using either of these measurement protocols, leading to very different classifications (evaluation at 500 lx always gives a lower value than evaluation at 200 mm). There is therefore ambiguity concerning the distance at which these measurements should be taken.

        3. Failure to take into account population groups sensitive to blue light: To assess the risk related to blue light, the NF EN 62 471 standard recommends using the phototoxicity curve for blue light suggested by the ICNIRP (International Commission for Non-Ionising Radiation Protection). This curve is only suitable for adults. The standard includes no specific recommendations for population groups whose natural mechanisms for filtering blue light are diminished (children, aphakics and pseudophakics), or who are more sensitive to blue light as a result of retinal diseases. In fact, the ICNIRP gives a different phototoxicity curve for blue light for aphakics. The current standard does not take account of the situation of population groups sensitive to blue light.

        It seems that certain LEDs that are very widely used in lighting, signage and guide lights fall into Risk Group 2, whereas all other light sources currently on sale to the public fall into either Risk Group 0 or 1.

        Sensitive or highly exposed population groups: (ANSES_Opinion 2010) Three population groups have been identified as being either especially sensitive to the risk or highly exposed to blue light:

        • children (because of the transparency of their crystalline lens) and both aphakics (with no crystalline lens) and pseudophakics (with artificial crystalline lenses) who consequently either cannot or can only insufficiently filter short wavelengths (especially blue light); (ANSES_report 2010, p. 73-74:) In young people (before 10 years), the lens passes virtually all blue light (80%), especially waves between 430 and 440 nm, the most dangerous to the retina (peak absorption at 360 nm). With age, the lens becomes yellow and absorbs shorter wavelengths. This change in age-dependent transmission protects the retina from the blue light and reduces scotopic vision (vision at night) significantly (about 33% at 50 years compared to 5 years). At the age of 50, protection against UVA, UVB and blue light grew 80%.
        • (ANSES_Opinion 2010) population groups which are already light-sensitive: patients suffering from certain eye and skin conditions and patients taking treatments one of whose side-effects is to increase photosensitivity, etc., for whom blue light can be an aggravating factor for their condition;
        • population groups highly exposed to LEDs (certain categories of workers: those installing lighting systems, theatre and film industry professionals, etc.) which are subjected to high-intensity lighting, and are therefore susceptible to exposure to large quantities of blue light.


      The arrival of LEDs on the lighting market for the general public is an unprecedented development: it is the first time that sources classified in Risk Group 2 have become accessible to the general public, for use in the home and, most importantly, with no indication of the risk. (Source: ANSES_Opinion 2010, p. 8) We can therefore conclude that a definition of a method of measurement and classification of risk specific to LEDs is needed. (ANSES_Report 2010, p. 142)

    • Risks related to glare: Because the emission surfaces of LEDs are highly concentrated point sources, the luminance of each individual source can be at least 1000 times higher than the luminance from traditional lighting sources. The level of direct radiation from this type of source greatly exceeds the level of visual discomfort.
    • Risk of deregulating the biological clock and pupil contraction: In humans, the biological clock and pupil contraction are regulated by wavelengths close to 480 nm which suppress the production of melatonin (a hormone participating in the regulation of the biological clock and therefore the circadian cycle). The spectrum produced by LEDs differs fundamentally from that of natural light, with a very low proportion near 480 nm. This could expose subjects to a risk of deregulation of their biological clocks and, in consequence, of their circadian rhythms. These risks are exacerbated by high-temperature colours (cold white and blue), which are frequently found in LED-based lighting systems.

      Deregulation of the biological clock can affect the metabolism, the thymus (depression, mood swings), the waking/sleeping rhythm, etc.

      • Light pollution is often considered to be of concern mostly to professional and amateur astronomers, but it has wider-reaching effects than this. It disrupts nocturnal animal behavior, including mammals, birds, amphibians, fish, and insects. It also disrupts the circadian rhythms of humans, animals, and plants; and it has even been implicated in the global obesity epidemic - light pollution may be making us fat. (Source: LEDs Magazine, October 2015)
      • Study after study has linked working the night shift and exposure to light at night to several types of cancer (breast, prostate), diabetes, heart disease, and obesity. It's not exactly clear why nighttime light exposure seems to be so bad for us. But we do know that exposure to light suppresses the secretion of melatonin, a hormone that influences circadian rhythms, and there's some experimental evidence (it's very preliminary) that lower melatonin levels might explain the association with cancer. (Harvard Health Letter, May 2012)
      • Abraham Haim of the University of Haifa in Israel considers white LEDs a form of "light pollution." "What is called 'friendly' environmental illumination is unfriendly," said Haim, who is a chronobiologist, a scientist who studies biological rhythms and cycles in animals. He has conducted studies showing that blue light can disrupt circadian-related hormones in nocturnal animals such as voles, moles and rats. (Do White LEDs Disrupt our Biological Clocks?) Melatonin is a compound that adjusts our biological clock and is known for its anti-oxidant and anti-cancer properties. The fact that "white" artificial light (which is actually blue light on the spectrum, emitted at wavelengths of between 440-500 nanometers) suppresses the production of melatonin in the brain's pineal gland is already known. Also known is the fact that suppressing the production of melatonin, which is responsible, among other things, for the regulation of our biological clock, causes behaviour disruptions and health problems. (Melatonin disruption by 'eco lighting' a rising health threat, 12 September 2012)
      • An other excellent study about the health effects of LEDs is "Joan E. Roberts Ph.D., Light and Dark and Human Health: The first sentences of the article are: Humans evolved under both a light and dark night cycle. Therefore, it is not surprising that modifying natural, cyclical daylight and dark exposure would lead to severe health risks. The issue can be found in "Environmental Impact of Light Pollution and its Abatement, Special Report of the Journal of The Royal Astronomical Society of Canada, design 21 LED Lighting Systems" and can be downloaded here (7 MB).

      Comments from SCHEER, Preliminary Opinion on Potential risks to human health of Light Emitting Diodes (LEDs), July 2017. "Generation of the circadian rhythm The biological clocks consists of multiple 'clocks': 1) the central clock in the brain (the suprachiasmatic nucleus or SCN) and 2) peripheral clocks in almost all organs including heart, liver and kidneys. The peripheral clocks are regulated by the central clock (Dibner, Schibler et al. 2010).

      Function of circadian rhythms

      Circadian rhythms most likely evolved to adapt and respond optimally to daily environmental cycles. It enables anticipation to expected events and ensures that bodily processes occur in a temporal and synchronized fashion at the most optimal timing related to the environment. A simplified example: eating when food is present and subsequently optimize metabolism processes after eating. The bodily processes regulated in a circadian fashion are widespread and linked. Ranging from behaviour (sleep/wake cycles), cognition (attention, concentration), the immune system and repair mechanisms, to numerous physiological processes including endocrine functioning, metabolism, cardiovascular functioning etc.

      Consequences of disturbance of the circadian rhythm by light

      (...) negative health effects of optical radiation from LEDs, specifically, have not been investigated. It is expected that these effects are not LED-specific; they apply to exposure to light during the evening that influences the circadian system in general. The effects may, however, be enhanced for LEDs compared to traditional light sources at similar illumination levels, due to the particular spectral emission pattern of certain types of LEDs. In addition, it is important to note that direct causal relations of the use of LEDs or other artificial light sources during the evening on health have not been investigated. Indications are obtained from association studies, circumstantial evidence and hypothesized effects based on studies investigating other types of circadian disturbance. An important consequence of the circadian disturbance due to light during the evening is its effect on sleep. As described in more detail above, the studies by Cajochen et al. and Chang et al. indicate that use of certain types of LEDs, similar to other artificial light sources, can result in reduced sleepiness (Cajochen, Frey et al. 2011, Chang, Aeschbach 25 et al. 2014) and increased latency to sleep (Chang, Aeschbach et al. 2014), possibly causing shorter sleep duration and poorer sleep quality. It is important to note that, regardless of the cause (i.e. being artificial light or other factors), reduced sleep duration and quality is associated with poorer cognitive performance, fatigue, altered mood and increased health and safety risks (Christoffersson, Vagesjo et al. 2014, Engle-Friedman 2014, Burke, Scheer et al. 2015, Cedernaes, Schioth et al. 2015). Furthermore, additional light during the evening has been hypothesized to phase delay circadian rhythms. Delay in the circadian rhythm can result in 'social jetlag'. This refers to the phenomenon that the circadian rhythm is delayed but the social environment requires behavioural patterns to remain at the earlier phase (Wittmann, Dinich et al. 2006). In other words, a person still has to get up early in the morning to go to work/school. This can cause several important bodily processes to occur 'out of sync' with the biological clock, such as food consumption. This desynchronization external and internal stimuli might be underlying some of the health effects related to disturbances of the circadian system. Social jetlag has mainly been associated with risk factors for cardio-metabolic diseases (Parsons, Moffitt et al. 2015, Wong, Hasler et al. 2015). Furthermore, evening light exposure might enhance delayed sleep-wake phase disorder (DSWPD) in sensitive persons. This disorder is characterized by late sleep and wake times and poorer sleep quality (Joo, Abbott et al. 2017, Magee, Marbas et al. 2016).In summary, disturbances of the circadian rhythm can result in negative consequences on sleep, cognitive performance and, in the long term, on metabolic risk factors. Since no experimental studies have been performed with chronic exposure (multiple years) to artificial light during the evening, it is currently unknown if the disturbance of the circadian rhythm remains, increases or reduces after chronic exposure to light during the evening.

    Furthermore, the pupil contraction reflex is induced in strong light by these same wavelengths. It could be reduced under LED lighting, which could lead to stronger light falling on the retina and an increase in the risks associated with blue light. (ANSES_Opinion, 2010, p. 9-10)

    LED streetlights: a new danger! (Melatonin disruption by 'eco lighting' a rising health threat, 12 September 2012)

    • High-pressure sodium (HPS) bulbs, which off orange-yellow light and are often used for street and road lighting, suppressed melatonin the least.
    • the metal halide bulb, which gives off a white light and is used for stadium lighting, among other uses, suppresses melatonin at a rate more than 3 times greater than the HPS bulb.
    • Although the LED is considered to be more 'eco friendly', the researchers predict that "The current migration from the now widely used sodium lamps to white lamps will increase melatonin suppression in humans and animals". This example is striking. The city of Davis, CA, for example, was obliged to replace newly-installed 4800K street lighting with 2700K luminaires at a cost of $350,000 following residents' complaints about "prison-white" lighting. (Source: LEDs Magazine, October 2015)

    Conclusions (Melatonin disruption by 'eco lighting' a rising health threat, 12 September 2012):

    • It is necessary to understand that artificial light creates "light pollution" that ought to be addressed in the realms of regulation and legislation.
    • In addition they suggest limiting the use of "white" light to those instances where it is absolutely necessary.
    • Finally, manufacturers should be compelled to state clearly on their packaging what wavelengths are produced by each bulb. If wavelength indeed influences melatonin production, this is health information that needs to be brought to the public's attention, so consumers can make informed choices about whether to buy this lighting or not. "Just as there are regulations and standards for 'classic' pollutants, there should also be regulations and rules for pollution stemming from artificial light at night," says Prof. Abraham Haim.
  • Risk related to flicker in the light emitted by LEDs:
    • As a consequence of the manner in which they are powered electronically, the light emitted by LEDs may be subject to rapid fluctuation of great amplitude. This fluctuation, combined with the fact that LEDs have very low remanence, is usually imperceptible to human vision. In situations involving movement or in confined spaces with periodic variations in contrast, it can be responsible for stroboscopic effects. Although such stroboscopic effects have never been studied in depth, they can have a direct impact on health (epileptic seizures for subjects at risk), visual performance and safety. A recent publication showed that LEDs can produce fluctuations in light at frequencies known to produce effects on health (from 3 to 60 Hz for visible fluctuations and from 120 à 150 Hz for non-visible fluctuations). (See IEEE, A review of the Literature on Light Flicker) Only fluctuations in light at frequencies lower than 100 Hz are visible. A veil (see graphs!) is partially lifted in the DOE fact sheet of March 2013. More research is needed.
    • One of the most common complaints associated with LEDs is still flicker. Flicker may only seem to be a problem associated with dimming LEDs (specifically because of the eye's sensitivity to fluctuations in low light levels), but recent research has shown there may be health effects associated with flicker even when not directly visible.

      The cause of flicker arises from one or more of three possible sources: instability of the driver's output, instability of the control system's input, or interference from an outside source such as power-line noise. Specifying a high-quality, digitally-controlled dimmer with superior filtering capability is the best way to eliminate many of these problems. (Source: LEDs Magazine, October 2015)

  • b. Effects on the skin

    Skin and optical radiation

    By convention, the visible radiation ranges from 380 to 800 nm. The blue part of the visible radiation (410-460 nm) bears an energy close to the ultraviolet radiation and is able to induce, as UVA, oxidative stress. Therefore, the penetration depth and the molecular nature of the tissues passed determine the nature, site and the consequences of agression. Example: the epidermis, in constant renewal, absorbs 90% of the ultraviolet radiation B and let penetrate deeper 50% of the radiation UVA and almost all of the visible radiation. (ANSES_Report 2010, p. 75-76; 78)

    SCENIHR [SCENIHR, 2009] estimated that about 250 000 people (0.05% of the population) the number of hyper-sensitive in Europe. Effects on the skin can not so far be clearly established. However, recent publications involving carcinomas and exposure to visible light do not exclude the risk of increased skin cancer in people exposed to prolonged LED emitting radiation between 380 and 500 nm [Zastrow et al ., 2009]. (ANSES_Report 2010, p. 127)

    Skin diseases that can be triggered or aggravated by bright artificial radiation emitted by the LEDs are from diverse origin and represent a significant fraction of the population. Because the realization of high-power LEDs and their various uses is expected, we can estimate that patients with some "solar" diseases will be exposed and could potentially see their pathologies triggered and / or aggravated.

  • The report SCENIHR, Health Effects of Artificial Light, March 19, 2012 came to the following conclusions:
    • "Relevant data are lacking regarding the effects of specific lighting technologies on medical conditions. The most important areas where knowledge gaps have to be filled in order to be able to draw firm conclusions are outlined" on p. 12 and p. 83-84.
    • (p. 4-5) Potential health impacts on the general public caused by artificial light. In general, the probability is low that artificial lighting for visibility purposes induces acute pathologic conditions, since expected exposure levels are much lower than those at which effects normally occur, and are also much lower than typical daylight exposures.
    • Certain lamp types (including also incandescent light bulbs) may emit low level UV radiation. According to a worst case scenario the highest measured UV emissions from lamps used in offices and schools, but not the very low emissions lamps used for household lighting, could add to the number of squamous cell carcinomas in the EU population.
    • Aggravation of the symptoms of pathological conditions. The blue or UV components of light tend to be more effective than red components in aggravating skin disease symptoms related to pre-existing conditions such as lupus erythematosus, chronic actinic dermatitis and solar urticaria. UV and/or blue light could also possibly aggravate the systemic form of lupus erythematosus. It is recommended that all patients with retinal dystrophy should be protected from light by wearing special protective eyeware that filters the shorter and intermediate wavelengths.
    • Short-term UV effects from artificial lighting on healthy people are thought to be negligible.
    • (p. 67) With photofrin, photosensitivity might be expected to occur with CFL and LED sources to a greater extent than that currently seen with incandescent lighting. This is due to a combination of greater sensitivity of porphyrins to blue light (soret band), coupled with an enhanced blue light emission of these sources.
    • The previous SCENIHR opinion (SCENIHR 2008) stated that a number of patients are exceptionally sensitive to UV/blue light exposure. The number of EU citizens with light-associated skin disorders that would be affected by exposures from CFLs was estimated in the report to be around 250,000. Clearly, the risk for this group of patients is not limited to CFL, but includes all light sources with significant UV/blue light emissions. The lack of proper data precludes any improvement of the estimate of the size of the affected group. (p. 11)
    • It may be advisable to make sufficient information on the emitted spectrum for individual lamp models available to the healthcare professionals and the patients to allow them to choose their lighting solutions optimally. (p. 83)


    • Nevertheless, this lack of firm data has not hindered the Commission to draw such conclusions.
    • A characterisctic feature in the reasoning of the SCENIHR (and the EC) is that they are only looking for proofs of harm. They found that some lamps are harmful to certain photosensitive patients. SCENIHR and the EC then propose to these groups to use only double envelope bulbs or to wear special protective eyeware (SCENIHR 2012, p. 69). This means that the dangerous lamps remain on the market! The rights of the consumer are as a consequence of it severely damaged. The EC has the opinion that if a lamp is not harmful to the majority of the consumers, this lamp has to remain on the market, even if the lamp is harmful to certain photosentive consumers. The consumers wants a certain quality of light and this quality is not attained with CFLs or LEDs. If consumers feel discomfortable when using CFLs, they must have the right to buy incandescent bulbs because the costumer is always right.
    • Many conclusions of SCENIHR are in clear contradiction with their own findings.
    • If single envelope CFL classified as RG1 may be hazardous to a photosensitive patients if used closer than 20 cm to the skin, these may be hazardous to all people. All depends on the number of hours that the lamp is used.
    • That SCENIHR could not improve the estimate of the size of the affected group is a great shortcoming that should involve a delay of all firm conclusions.
    • The damage inflicted to the consumers is in flagrant contradiction with article 15 of the Directives 2005/32/EC and 2009/125/EC and with the Commission Regulation (EC) No 244/2009.
    • A short comparison between incandescent light bulbs, CFLs, halogen lamps and LEDs can be found here.

    5. Because LED light bulbs emit more blue light, they attract more (dangerous) insects

    It is known that kerosine and other indoor pollutants are dangerous in Africa due to their household air pollution. That's why several campaigns are going on to promote the use of solar power for domestic use. But there is an unexpected side effect: flying insects. "Many different types of insects are attracted to light sources. This is either in search of a mate or after taking a meal. Many insects attracted to lights may be harmless, but key species are known vectors of disease that affect both humans and animals. These include:

    • domestic flies, which carry the bacteria that cause blinding trachoma;
    • mosquitoes, which carry the parasites that cause malaria, filariasis, dengue; and
    • sandflies, which carry the parasites that cause leishmaniasis.
    Ironically, it is low-energy LED light bulbs that are the most attractive due to the fact that they emit more blue light. This is such a major issue that attempts are underway to tune LEDs so that they attract a smaller number of flying insects." (Source: "")

    6. Depression, a common effect of CFLs and LEDs

    After decades of intense research, the CRI of the CFLs and LEDs is only about 80 while incandescent light bulbs and halogen lamps have a CRI of 100. It is known that the spectrum of incandescent light bulb is most similar to the spectrum of the setting sun. Consumers are deprived from this light source and become depressed, also as a consequence of the poor light, the long warm-up times, the bad dimming quality of LEDs and CFLs, etc. One has to investigate what is the social cost of the treatment of these additional depressions. I guess the EU has no fund available to do this research. It is time to stop this torture of the consumers and let the incandescent light bulbs again available on the market!

    More information: Light at Night: The Latest Science

    7. Harmful effects of halogen lamps

    8. Discussion concerning the answer given by the European Commission to my letter

    My letter of 31 August 2012 Answer dated 18 October 2012 My comment
    Study of Tatsiana Mironava, Michael Hadjiargyrou, Marcia Simon, Miriam H. Rafailovich, The Effects of UV Emission from CFL Exposure on Human Dermal Fibroblasts and Keratinocytes in Vitro, Photochemistry and Photobiology, June 2012. Our results indicate that commercial CFL bulbs (chosen at random) emit UV radiation which can induce damage to various types of skin cells. Since 2009, CFLs coming on the EU market have to comply with stringent UV radiation limits that make the product safe, which is checked by market surveillance authorities of the member states. In particular, CFLs are not allowed to emit radiation in the UV-C range.

    Commission Regulation 244/2009 imposes the same limit on the total UV radiation of CFLs as the one that already exists for halogen lamps under the Low Voltage Directive (2006/95/EC). In addition it forbids UVC radiation (< 0.01 mW/klm is the scientific way of expressing a quantity that is too small for measuring). These requirements have been applicable since 1 September 2009 for any new compact fluorescent lamp placed on the EU market. For further details, please consult the Professional FAQs on the regulation phasing out conventional incandescent bulbs:

    • It depends on the authorities of the member states to control a possible UV radiation from CFLs. As long as there is no stringent control on a sufficient high percentage of the market, CFLs which emit radiation in the UV-C range will still be found on the European market.
    • SCENIHR found that even incandescent bulbs can have measurable UV emissions. Does it mean that a certain level of UV radiation is allowed for both incandescent bulbs and CFL? The Stony Brook University team found that incandescent bulbs did not damage the skin.
    • I am asking if the CFLs on the market of the U.S. are so different from the CFLs on the European market.
    The team found that after an exposure of only 5 hours at a typical working distance of 35cm of a desk lamp the TLV (Threshold Limit Value) was exceeded. The UVC emission was larger than ambient sunlight on a mountain.

    The importance of this study is that the authors could point out the cause of the significant and seriously damaging effects of high UV emission from CFLs. (Scientific evidence)

    Distances of 2.5cm and 7.5cm cannot be considered realistic reproductions of normal usage, such as in an office setting, and even the worst performing CFL showed less UV-A radiation than the threshold limit values (TLV) would have allowed for a 35 cm distance. (...) there is no scientific evidence that this study shows new insights, which have not been evaluated by SCENIHR in its opinion on heath effects of artificial light. The answer seems to minimize the results of the study. The team found that after an exposure of only 5 hours at a typical working distance of 35cm of a desk lamp the TLV was exceeded. The answer states that this study shows no new insights while later on, they recognize that UV light can escape through the cracks in the phosphor coating.
    With this new study, everyone is warned to avoid using CFLs at close distance. While only single envelope CLF bulbs were used in this research, one may expect that double envelope CFLs will leak less UV light. However this has still to be demonstrated and extreme caution must be taken until double envelope bulbs are shown to be safe for all European citizens. We doubt if that day will ever come as CFLs are simply bad technology. Consumers who would be still worried can use other alternative technologies (halogens or LEDs), or CFLs with a second lamp envelope that filters any UV light coming through cracks in the phosphor coating of the incorporated tubes. However, according to Spectrum Alliance it is clear that its members have tried those double-envelope bulbs and that, although they are an improvement for some people, they still induce similar symptoms in most of those affected. (Source)
    It is unjustifiable and disproportionate that conventional incandescent lamps are being banned without knowing the consequences. A large number of people have suffered and still suffer from defects or damage to their skin as a consequence of CFLs. The government and Commissioner who ordered this ban take the full responsibility for this suffering. They made two mistakes: they did not warn citizens about the side effects in advance and they did not have a mechanism to compensate the victims for the injuries they suffered. The time will come when they have to compensate the victims for this damage. Other fluorescent lamps (neon tubes) use the same technology as CFLs, they have been around for decades and are widely used in offices and also in many households. We do not see a reason to single out CFLs. Are the leaking CFLs (through cracks) not a sufficient evidence?
    The Stony Brook University team found that incandescent bulbs did not damage the skin. SCENIHR recently reviewed research covering lamps on the EU market (opinion on health effects of artificial light) and found that even incandescent bulbs can have measurable UV emissions. According to the study of Rachel S. Klein, Robert M. Sayre, John C. Dowdy, Victoria P. Werth, The risk of ultraviolet radiation exposure from indoor lamps in lupus erythematosus, the results reported in the literature are conflicting. Chignell et al recently demonstrated that a 60 Watt incandescent bulb will begin to emit UV at 375 nm. (a safe zone past 340 nm) However, another study indicates that the emission spectra of incandescent bulbs begin as low as 280 nm, which would be considered a risk to photosensitive patients. The study concludes that the level of irradiance is quite low.
    SCENIHR estimated that about 250,000 people in the EU have a light-associated skin disorder that would be affected by exposures from CFLs. The true number of people affected could be about 50 times higher or about 12 million people. No answer was given by the EC. -
    It should be an elementary principle and also good practice to provide consumers with comparisons of the spectral output of other types of lamps and the spectral output of an incandescent lamp on the lamp packaging. What SCENIHR advises for people with light-sensitive health conditions must also be available for everyone: (p. 83) In view of the large number of patients affected by photosensitive diseases it may be advisable to make sufficient information on the emitted spectrum for individual lamp models available to the healthcare professionals and their patients to allow them to choose their lighting solutions optimally. Concerning your remarks about the aggravation of symptoms of people suffering from light sensitivity we would like to point out that using commonly available compact fluorescent lamps with a second lamp envelope can both solve the problem of light-sensitive patients and prevent overexposure of the general public even in extreme situations. However, the envelope slightly lowers (about 10%) the efficacy of the compact fluorescent lamp, meaning more lamps using more power will be needed for the same light output. Transparent or translucid luminaires that fully cover up the bare lamps have the same effect as a second lamp envelope. Also alternative technologies can be chosen by consumers, such as improved incandescent bulbs (with halogen technology) that have identical light spectrum to conventional incandescent bulbs. This is not an easy solution for consumers. Other qualities are evenly important such as the spectrum of light, the dimming possibility, the warm-up time, etc. The rights of the consumers are infringed if they saddled with the cost of the 10% lower efficacy.
    This information does not appear on the packaging at present. Consumers are confused by the overwhelming variety of other types of lamp available. Buying a CFL is about the same as taking part in a lottery but with the odds staked against the consumer. People do not know how long it will take for the bulb to warm up, if the intensity of the light and the life of the CFL will match what is stated on the packaging and how dim the light will become as the lamp ages. [labeling of the lamps]

    Commission Regulation (EC) No 244/2009 states at paragraph 14: The ecodesign requirements should not affect functionality from the user’s perspective and should not negatively affect health, safety or the environment. This intention is not being achieved.

    Consumers cannot wait until 2014 for their justified demands to be met. For the sake of the many victims of the ban on incandescent lighting, the European Commission must acknowledge that this measure was ill-considered, premature and without public discussion. It is also essential that this mistake is not repeated with directional lighting.

    Regarding the labeling of the lamps which indicates the energy level, we would like to inform you that lamps have had to display an A-G scale energy label on their packaging since 1998 (Commission Directive 98/11/EC). It is planned to reexamine the scale taking into account the phase-out of many inefficient lamps and the recent appearance of more efficient lamps, and also to extend the scope of the label to the so-far excluded reflector lamps and low voltage lamps, probably in 2012. This is not an answer to my question.

    The discussion of other items of the answer of the EC can be found here.

  • Last update 15 August 2017