Steven A. Zilber, B.E.
Independent Lighting Consultant
Cleveland, Ohio

(1) In various forums, two related questions have been raised about indoor lighting:

  1. Are current standards of lighting responsible for poor health?
  2. Can health be maintained, or even enhanced, by new standards of lighting ?

Studies that have actually tried to address these questions are few and far between. Only now is a major effort being made to address them Ref.1. This paper reviews what has been done so far, including preliminary results on some controversial topics. The focus is on the lighting of indoor environments where people spend a significant amount of time, such as offices, schools, and nursing homes. It is important to point out that current lighting "standards" are based not on health and well-being, but on task performance and energy use Ref.2 Ref.3.

I. Light and Human Physiology

(2) Visible light is defined as electromagnetic radiation spanning the wavelengths of 380 nm (UV) to 760 nm (IR). There are two physical pathways for light to have an effect: the skin and the eyes. Currently, there appears to be no indication or evidence that skin needs visible light to remain healthy, or that the body needs skin exposure to remain healthy. However, the lack of evidence does not preclude the possibility. Photo- chemical reactions tend to involve very specific wavelengths, and it's doubtful that we've identified all such reactions which might occur in human skin. Most of the skin research has focused on the invisible UV bands.

(3) As for the eye, we know that light can produce both systemic physiologic responses and cognitive stimuli (it affects the body and the mind). For light to be perceived, it must traverse the optical pathway to the brain. This means that the eye itself must be optically clear, the retina healthy, the optic nerve intact, and the various parts of the brain responsible for light detection and vision, functional. For now, let's assume an operational visual pathway.

A. Autonomous Physiologic Effects

(4) Research has shown conclusively that light affects the human body in ways other than producing vision. The effects identified so far involve body rhythms. There are currently over 3000 references on light's affect on human chronobiology Ref.4. We'll look first at a controversial one.

B. Seasonal Affective Disorder (SAD)

(5) SAD, sometimes referred to as "the sad syndrome" or "the winter blues", was discovered by a multi-disciplinary research team at the National Institute of Mental Health (NIMH) in 1980, with the help of a researcher at Bell Laboratories Ref.5. It is listed in the DSM-IIIR as a psychiatric disorder. SAD's major symptoms are: markedly depressed mood ( >15 on Hamilton Depression), irritability, pronounced fatigue, increased need for sleep (hypersomnia), and increased consumption of carbohydrates (carbo-craving). SAD differs from common depression in the following ways:

        Common Depression                     SAD
        No evidence of Seasonality            Seasonal
        Insomnia                              Hypersomnia
        Loss of Appetite                      Carbo-Craving

(6) SAD generally manifests itself after puberty, and affects females more than males (5:1). The prevalence increases with latitude, going from about 1% in Miami (26N) to over 10% in Alaska (50N). For every person with full-blown SAD, there are another two or three who are sub-syndromal. People with lower- than- normal retinal sensitivity seem especially prone to developing SAD. Sad is triggered by the shortening of the photoperiod as winter approaches. The shortened photoperiod allows melatonin, the hormone responsible for signalling that it's time to sleep, to build up to higher-than-normal levels in the blood, which in turn produces the symptoms of SAD Ref.2 Ref.4 Ref.5. As summer approaches, the photoperiod lengthens, so melatonin levels drop and the symptoms alleviate. Phase-shifting of the sleep- wake cycle has also been implicated, as this also involves melatonin. Melatonin is only produced by the pineal gland(our third eye), and only under conditions of darkness.

(7) SAD can be treated very effectively by several methods. The most desirable one for most people is to head south for the winter: lower latitude means more daylight. The usual treatment involves some form of light therapy, from either light boxes, or dawn simulators. Light boxes generally consist of a fluorescent fixture mounted on a stand, with full-spectrum lamps and a diffuser or lens intended to produce a uniform luminous surface. The patient is instructed to look at the box for thirty to sixty minutes, usually in the morning. There is a dose response; the brighter the light, the shorter the exposure needed. The minimum level is usually 2500 lux on the cornea, while the maximum is 10,000 lux (10.76 lux: 1 foot-candle). There are very few indoor environments bright enough to produce 2500 lux of illumination. Light visors are currently under study as a more convenient system. The other lighting approach is the use of a dawn simulator, which gradually increases illumination in a bedroom before the desired wake-up time; results so far are very promising Ref.30. Some very recent research suggests that Prozac (fluoxetine) may also be effective, due to the linkage between melatonin and serotonin (both are neurotransmitters) Ref.4.

C. Shift Work and Circadian Disorders

(8) Much of the physiology behind SAD also plays a role in our daily sleep-wake cycle. When this cycle is disrupted, we get the symptoms of jet lag: clinical sleep disorders, severe fatigue, major digestive problems, and the inability to react or concentrate. How is our sleep-wake cycle disrupted? Our bodies expect to receive a daily stimulus of light every morning: the sunrise. That the sunrise is the key stimulus makes sense when we realize that the rising and setting of the sun have for eons been the major daily environmental stimuli in the evolution of life. If we don't receive these signals, or we receive them at the wrong time, several physiologic cycles desynchronize, or are phase-shifted, causing jet-lag symptoms. So while we do have internal clocks (pacemakers), they need to be reset daily. The prime pacemaker is a pair of very small organs in the hypothalamus, called the SupraChiasmatic Nuclei (SCN). The SCN wants to 'free-run' at about a 25- hour cycle; the sunrise entrains it to a 24-hour cycle Ref.2,p144. The entrainment process is an elegant feedback system; if the 'free-run' time was less than twenty-four hours, it could not work. The signal is first received by the retina, then is detected by the pineal gland via the optic nerve; the pineal gland then signals the SCN, which in turn signals the pineal to start producing melatonin at the proper time (dusk). Therefore, if someone has major visual problems, such as blindness, the SCN may never be reset, and he or she lives in a state of perpetual disynchrony. (There is some hope: experiments are suggesting that melatonin therapy can achieve entrainment in the blind.) So, what about those with functional visual paths? Shift workers and air travelers still have problems.

(9) Research has shown that the human body needs at least three days, and sometimes as many as seven, to adapt fully to a 6-hour time shift Ref.6. So what happens if someone is working on a rotating shift schedule, especially if it is a rapidly-rotating (weekly) schedule? His body never has time to fully adjust. On the other side of the coin, if someone permanently works on the graveyard shift (midnight - 8AM), he never have a chance to experience a sunrise, so his SCN free-runs, and he lives in a state of perpetual jet lag. Why should we care if we are fortunate enough not to have to work shifts? Because there are about 5 million shiftworkers in the US, and a sizable portion of them hold critical positions in our modern society, such as EMS, police, air traffic controllers, nurses, power plant operators, etc. Ref.7. That these workers cannot perform at their full cognitive and physical capacities has been thoroughly documented Ref.7 Ref.8. What can be done to solve this problem?

(10) A couple of things can, and should be, done. First, we need to shift the sleep-wake cycle so that it is synchronized with the work schedule. It is possible to shift a person's sleep-wake cycle a full twelve hours using light therapy or melatonin Ref.2 Ref.4 Ref.6. Second, we can ensure that the work environment is lit brightly enough to suppress melatonin production, so that workers stay alert. This means lighting to over 2500 lux on the typical work surface Ref.9. Social stimulus has also been shown to be very important Ref.8 Ref.9.

D. Psychological Effects

(11) Visible light, or the lack thereof, can affect our mood through neuro-physiology. Lighting can also affect our mood based on our cognitive perceptions, and possibly unconscious perceptions. Unfortunately, there has been very little research in this area. Studies of the effects of ambient color used in indoor environments have been proposed, but rarely funded. There is anecdotal mention of the use of pink in jails to 'mellow out' prisoners, but there has been no real follow-up. This is a ripe area for study.

E. Aging's Effects on the Eye and Vision

(12) As an optical instrument, the eye is very poor. It introduces several kinds of distortion and aberration in an image before the retina receives it. Ref.2 That we can see clearly at all is really a testament to the image-processing capability of the brain. Starting from this sorry state, things get much worse as we age. The first ability to go is accommodation, the ability to focus on objects at different distances. The loss of accommodation is called presbyopia. Presbyopia usually starts at around age forty and continues until all focusing ability is lost at about age sixty, resulting in a fixed-focus system. Ref.2,p65 The mechanisms behind presbyopia are reasonably well understood (basically, the lens hardens), as are some common solutions (bifocals, reading glasses, and magnifiers).

(13) Another aspect of vision that degrades is total light transmission to the retina, especially at the shorter (blue) wavelengths. The loss of transmittance can approach 66% by age 60 Ref.3,p539. The loss is caused by one or several of the following: yellowing of the lens (most probably from UV exposure), cataract formation, and the development of intraocular scattering material. Intraocular scattering is especially problematic for the elderly, because it also produces disability glare. Severe retinal problems such as macular degeneration also tend to crop up as we age.

(14) To further complicate matters, standard eyewear can actually exacerbate some problems. The most common lens is made of CR-39 plastic resin, which has a refractive index n = 1.50. At this index, only 92% of light will get to the eye, due to front and back-surface reflections. Inter-reflections can also produce significant levels of glare, especially at night. If polycarbonate is used (what the author wears) with n = 1.585, or the newer high-index resins (n = 1.59 +), the light loss is greater, and glare worse. A solution will be mentioned later.

F. Complaints and Fears

1. Eyestrain

(15) Eyestrain is the most frequently mentioned problem caused by lighting, or the lack thereof. Ref.10 Ref.11 It can be caused by one or several of the following conditions: inadequate illumination; glare; flicker; uncorrected refractive errors (need for eyeglasses); and a non-ergonomic positioning of the visual task.

(16) Eyestrain can lead to headaches and fatigue, which in turn add to stress. Let's examine glare first. Glare can be diffuse (reflected off a matte surface), specular (reflected off a glossy or mirrored surface), or direct (from a bare filament). Veiling reflections off a magazine page can cause discomfort, or even disability (the inability to perform a task; in this case, reading the printed text). Glare is never desirable, and a lot of work has gone into determining how to reduce it, especially in offices with VDTs. Unfortunately, through ignorance or over- zealousness, many offices are now too dark for general task work (a trait shared by many residences). Research has shown that limiting ceiling luminance ratios to below 5:1 helps a lot. Ref.12 Ref.13 Current preferred practice is to do this with mostly indirect lighting. Ref.13 However, indirect lighting is more costly, and less energy-efficient, than the usual fluorescent troffers, often creating a conflict with an energy code. Light that is too diffuse is disorienting, like being in a white-out snowstorm. The Lighting Research Center at Rensselaer Polytechnic Institute has stated emphatically that lighting alone cannot provide the complete answer to eliminating glare from VDTs. Ref.13 The real problem is the use of specularly-reflective screens; the answer is to make those screens non-reflective, with anti-reflective (AR) filters. Many of the newer CRTs have good- quality AR coatings applied directly to their face glass.

(17) Another approach is to use multi-layer (not linear or circular) polarized filters in the overhead troffers. This is still a topic of considerable controversy within the lighting community. Some recent follow-up studies have shown that a modest reduction in glare is possible, within certain viewing angles. Ref.14 Ref.15 The loss of ambient light in the filters can be significant, but this is still an intriguing area for research.

(18) Due to energy codes, use of high-brightness lamps is increasing, and so is glare. A 50W T5 compact fluorescent lamp can have over three times the surface brightness of a standard 40W T12 linear lamp, while halogen and metal-halide sources are far too bright to look at directly. Besides the proliferation of high-brightness sources, making matters worse is the practice of putting them in open or poorly-shielded fixtures. (A well- shielded fixture hides the lamps from direct view and does not show a reflection of the lamp until a certain -- steep -- viewing angle is reached.) But well-shielded fixtures are usually best at creating sharply-limited pools of light, so obtaining even ambient lighting is difficult without a lot of closely spaced fixtures.

2. Flicker

(19) Flicker is well known for causing eyestrain. Humans can perceive flicker at frequencies as high as 200 Hz, based on measured physiologic responses in the retina. Ref.3 Ref.16 This is why there are now standards for flicker from CRTs (70 Hz + ; it's still pretty bad). Flicker is most apparent in the peripheral areas of vision, where the rods are concentrated; however, if the frequency is low enough, the cones have sufficient time to react, and then flicker really becomes objectionable because it is cognitively noticeable as a distraction. Gas-discharge lamps such as fluorescents and metal-halides usually flicker at 120 Hz (twice the power line frequency) because their arcs are extinguished (and re-struck afterward) at every zero-crossing of current. The newer electronic ballasts can drive these lamps at 25 - 40 KHz, well beyond visual and audible perception by humans (but perhaps not so for certain pets). Incandescent lamps also flicker sinusoidally at 120 Hz, but the filament remains hot enough for it not to be noticeable by most people. However, there are some halogen lamps with integral diodes which exhibit severe flicker; they can also audibly hum or click.

3. Noise

(20) As mentioned above, flicker and noise often accompany each other. Here the lighting system is to blame , not the light itself. All lamp ballasts and transformers produce some noise at harmonics of the line frequency. Noise produces stress in humans Ref.3. Currently, of all the types of ballasts and transformers for all the different lamp technologies, only some fluorescent ballasts are "rated" for noise; there is no certification requirement -- a scandalous situation. The best rating, Class A, allows as much as 24dBA of noise, which is quite noticeable in otherwise quiet places like libraries and rural homes. The best electronic ballasts are virtually silent, but many are not, and when you consider that multiple ballasts are usually used in a room, the noise adds up. The ballasts for cold-cathode (neon), high-intensity discharge (HID: metal-halide and sodium), and low-voltage (<30V) halogen lamps tend to be the noisiest, and the higher the wattage, the louder the noise. These almost always need to be mounted in a remote location.

4. Fears

(21) This section discusses issues of some controversy, some of which are being aggressively researched for health risks. Non- technical readers are strongly advised to acquire a basic understanding of risk assessment, and are directed to Reference 17 as an excellent place to start.

5. UV and Skin Cancer Risk from Lighting

(22) There have been many claims for and against skin exposure to UV. The CIE defines three bands of UV, listed in order of increasing energy:

UV-C will quickly kill most micro-organisms, and is used for that purpose. Human skin uses UV-B to produce vitamin D3, which is needed for maintenance of proper calcium:phosphorus balance. Most milk is fortified with vitamin D3, which seems to have obviated the need for skin exposure (assuming a person drinks fortified milk) Ref.2,p143. UV-B is strongly suspected as the cause of lens yellowing and cataract formation, besides carcinoma and melanoma Ref.2. Benefits and hazards of UV-A have not been thoroughly researched. UV exposure can have systemic effects on the immune system Ref.2,p141. On the other hand, there may still be a minimum daily requirement for UV exposure. There is evidence that certain wavelengths of UV-A and blue light can repair damage to DNA caused by UV-B Ref.2,p152. Also, the greater transparency to UV of the eyes of children may play a role in their physical development [George C. Brainard, Ph.D. (Chair, IES Photobiology Committee), quoted in 18].

(23) Many types of lamps produce UV: fluorescent, metal-halide, quartz halogen. The first two technologies rely on using the UV-C line at 253.7 nm produced by the plasma arc through mercury vapor. In fluorescent lamps, most of the UV is converted to visible light by the phosphors coating the lamp, and almost all of the rest is absorbed by the glass bulb wall, so very little UV is actually emitted. Metal-halide lamps tend to produce more UV, since they rarely have phosphor coatings designed for UV conversion (for the few which are coated), and they have much more intense arcs. If their outer bulb breaks, significant amounts of UV can be emitted, so the lamps are almost always shielded behind a cover glass, or they are designed to self- extinguish, or they have a special third layer for security. Because of the quartz bulb, quartz halogen lamps can transmit a large percentage of the small amount of UV they produce (the amount can be calculated from the black-body equations). Most halogen lamps use a cover glass for protection from hot (250C) glass if the bulb pops; this cover glass absorbs almost all of the UV. A bare bulb of 100W or more could produce erythema (sunburn) given enough constant exposure; 500W and higher would be more effective. Radiation dose for a point-source radiator (a filament) equals the source intensity times exposure duration divided by the square of the distance between the source and the exposee (inverse-square law). So, the higher the wattage or number of lamps, the longer the duration of exposure, and the closer one is to the lamps, the more likely any risk is increased. I have received many questions from clients concerned about UV from their 20W desk lamps or 500W torchieres; my position has been to tell them that they have more to worry about from the burn or fire risk. Artwork, on the other hand, is definitely at risk of degradation, and should be lit with filtered sources. It is really quite easy to filter out UV. Last year an epidemiological study was published by a Canadian research team which claimed to show a correlation between cumulative exposure to fluorescent lighting and increased risk of melanoma on the order of two to three times, principally to adult males Ref.19. Assuming the paper is correct, a two- to three-fold increase of a very rare disorder still leaves us with a very rare disorder: it is probably not a significant risk to society. And it's important to remember that correlation does not equal causation. To the authors credit, they do raise the possibility of masking from several factors known to increase melanoma risk. Ref.19,p752 A common problem with this type of study is that we do not have good quality data on the actual exposure of the sample population to light and UV; knowledge of the dose helps tremendously. Meanwhile, there exist a few lamp manufacturers who claim all kinds of health benefits for exposure to the enhanced levels of UV produced by their lamps. This matter should be pursued vigorously by the research community, and then put to bed.

6. Electromagnetic Fields (EMF)

(24) This is one of the most controversial topics to hit the public health scene in quite a while. EMF is being blamed for all sorts of things. People are afraid of their electric blankets, cellular phones, CRTs, and microwave ovens, not to forget power lines. An epidemiological study from Sweden purported to show an increased risk of childhood leukemia (another very rare disease) of two to three times from exposure to power lines Ref.20. And then there was the 60 Minutes segment on police radar seeming to cause rare cancers. An interesting dialogue on EMF can be found in reference Ref.21. EMF is non- ionizing radiation. Its effects vary with its field strength, frequency, and exposure time. In current usage, it can mean AC or DC electric fields, magnetic fields, or any combination of the above. Because of the popularity of the issue, lots of research is currently being conducted by organizations like the IEEE, EPRI, etc. The Energy Act of 1992 included a $65M allotment for a five-year study, to be co-funded by industry. The official position of the IEEE so far is one of extreme skepticism, which at this stage is probably a wise stance. However, we know that an applied electric current can promote tissue generation and healing, while magnetic fields have been shown to aid bone repair, so it is feasible that EMF might affect us, but whether it is for better or worse, we just do not yet know.

(25) So let's relate the above to indoor lighting. The Lawrence Berkeley Laboratory did a study on this very question, where they actually measured the fields produced by typical fixtures in typical settings; they were barely measurable above ambient levels Ref.22. The contribution of electric lighting to EMF is minuscule.

7. Mercury Exposure

(26) Mercury (Hg), which is a toxic heavy metal, is used in every fluorescent and metal-halide lamp, and much cold-cathode lighting. Without the mercury, the technologies do not work. The mercury is contained within the sealed lamp, and is not released while the lamp is intact. If a lamp breaks, almost all of it will amalgamate with the phosphors on the bulb wall, rather than dashing off into the environment. The amount in a typical fluorescent tube is 40 mg, and a concerted effort is under way to cut that amount in half (for mercury is expensive). Mercury is mainly an issue for the lamp disposal and recycling operations.

II. Lighting Measurements: Are We Measuring the Right Things, in the Right Way?

A. Scotopic Vs. Photopic Vision--- (Rods vs. Cones; Night Vision vs. Day Vision)

(27) In the last three years, research has shown that many of the eye's responses to light are determined by the rods, not the cones Ref.23. Particularly important was the revelation that pupil < diameter is controlled by the rods. The higher the proportion of short wavelengths (blue end of spectrum), the smaller the pupil diameter. A smaller pupil diameter improves visual acuity (the ability to see detail), and depth-of-field (the ability to see clearly and simultaneously things at different distances). This situation could account for some of the perceived and measured benefits of full-spectrum lighting (a topic to be dealt with later), besides reducing eyestrain. The rods have their peak response at 509nm (blue-green), therefore pupil size is optimized at higher color temperatures, 5000 K and higher. The possibility that fewer lamps of high color temperature could replace more lamps of lower color temperature to achieve the same task performance has excited some members of the energy-efficiency crowd, but more research is needed.

(28) The research into rod function also raises the question as to whether we are measuring illuminance based on the correct eye response curve? A movement is growing for use of the scotopic (dark-adapted) curve, with its peak response at 509nm, over the photopic (light-adapted) curve, with a peak response of 555 nm (yellow-green). Much of the data behind these response curves came from research done back in the 1920s and 1930s. Needless to say, technology and methodology are vastly superior today. Unfortunately, it will nevertheless be a few years yet before more accurate data becomes available.

(29) An even larger issue is, should we even be measuring illuminance (the amount of light impinging on a surface), as is most current practice; or instead, should we be measuring luminance (the amount of light emitted or reflected from a surface) and contrast (the difference in luminance between two objects)? Historically, the technology to measure illuminance has been much easier to implement. That is still true today, and is reflected by the large price differential for instruments: a good luminance meter costs at least five times what an equally good illuminance meter costs, on average ($2500 vs. $500). Also, the Illuminating Engineering Society has formalized illuminance measurements with very extensive tables of recommended light levels for many diverse tasks and environments Ref.2, providing what some consider to be a 'cookbook' approach to lighting design. However, while the eye itself relies on being illuminated, what's doing the illuminating are luminous objects. A fair proportion of current lighting research now uses luminance and contrast measurements, and as mentioned previously in the section on glare (above), the recommended practice for office lighting depends on contrast measurement.

B. Are We Meeting the Needs of Enough People?

(30) As has been related, aging has definite effects on the eye and vision. Now the population of elderly is increasing. What might be bright enough for a healthy thirty-year old is not close to being adequate for an aged individual with impaired vision. Ref.2 Ref.3 Ref.27 Ref.28

C. Quality of Lighting

(31) As we developed lighting and other building technologies, we gained control of our immediate environment, but in many cases, we also became isolated from the natural environment. There are some major differences between the outdoors lit environment and the typical office lighting. First, the outdoors is more brightly lit -- much more. The illuminance from the sky alone (sun blocked) can exceed 10,000 lux; with direct sunlight added, illuminance can exceed 100,000 lux. Ref.2 Ref.24 Compare this to a "brightly lit" office at 1000 lux (93 foot-candles). Second, the light from the sky is very uniform and very diffuse; it is almost like being inside a giant fluorescent tube. Third, the amount of light varies smoothly over the course of the day. Fourth, the sky is blue (color temp > 6000K); not many offices have blue walls and ceilings (for good reasons...). Fifth, the color temperature changes smoothly throughout the course of the day, rarely dropping below 5000K (at noon), then increasing to over 10,000K at dusk. Ref.2 Lastly, there is significantly more UV present, at all wavelengths.

(32) Ponder what happens if we make the indoor lit environment more similar to the natural daylit environment. One could expect that any negative effects of indoor lighting would be mitigated, as would any positive ones (reduced heat, glare, and UV). It is interesting to note that the recommendations for lighting levels in the United States have gotten steadily higher over the last thirty years Ref.3,p 528, at least until the energy codes were passed. Now, what often happens is that rather than improving the efficiency of indoor lighting, which is expensive, even with subsidies, less lighting is used, resulting in inadequately-lit, cave-like environments. Designers, building owners, and tenants must ensure that this does not happen.

D. Spectral Distribution of Light: Full-Spectrum Lighting and High-Color-Temp Lighting

1. Full-Spectrum Lighting

(33) Full-spectrum lighting has been receiving rapidly growing attention, especially from pet owners. Our first difficulty is that there is no precise or sanctioned definition of "full- spectrum." Common usage implies similarity to the spectral power distribution (SPD) produced by the sun, which is very uniform across the visible spectrum Ref.2Ref.24, compared to gas-discharge sources like fluorescents, which have several large spikes. Ref.2 Most of the discussion refers to fluorescent lamps with a color rendering index (CRI) of 91 out of 100, which ostensibly means that they render colors more accurately than lamps of lower CRI. Common examples are DuroTest's Vita-Lite and GE's Chroma 50, among others. Their SPD's are smoother and less spiky than standard lamps, with a better balance between reds and blues. They are frequently used in emergency rooms, pet stores, and artists' studios. To achieve their high CRI, they need to use phosphors which are typically 30% less efficient than standard lamps, so either 30% more lamps are needed, or 30% lower light levels have to be endured. Of course, they cost more than standard lamps also. A recent study claimed that students showed significantly better performance, and were less likely to be absent due to illness, because of full-spectrum lighting in the classroom Ref.18.

2. Full-Spectrum Vs. High-Color-Temp

(34) Due to the potential market, the major lamp manufactures carefully review any research claiming health and behavior improvements under full-spectrum lamps (though I don't believe any of them have bothered actually to fund this type of research), and so far have not found any meaningful results. However, recent product introductions show a rapidly growing interest in high-color-temp lamps, with a nominal standard color temp of 5000K. Japanese researchers have shown that illumination from high-color-temp (>= 5000K) fluorescent lamps produces changes in physiology, generally for the better. Ref.25 Ref.26 Remember that high color temps also improve pupil size.


A. General Recommendations

(35) Indoor lighting should not fall below 500 lux as an absolute minimum; 1000 lux is better. Daylighting should be used as much as possible, both for its superior quality, and its energy-saving potential. Good daylighting is difficult to implement, however; light shelves are becoming an increasingly popular architectural approach. High (>9ft) ceilings also help a lot. Use lamps with a color temperature of 5000 - 6000 K, full- spectrum optional. If there is concern about UV exposure, buy filters. Eyeglass wearers (especially the elderly) should have untinted lenses with anti-reflective (AR) coatings.

B. Lighting for the Vision Impaired and Elderly

(36) Use enough illumination. Persons with macular degeneration perform much better with illumination exceeding 5000 lux. Ref.27 Make all lighting transitions from one space to another as gradually as possible; the elderly take longer to adapt. Ref.28 If someone moves from a brightly-lit space into a darker corridor, he or she will not adapt in time, but will be effectively blinded. This might be why so many falls occur in hallways and stairwells. Taking extra care to avoid sources of glare will help to reduce the disability glare caused by intra-ocular scattering. In a few cases, people with very cloudy eye conditions can improve their visual clarity and contrast sensitivity by wearing "blue-blocking" lenses or filters. Ref.29

C. Night Shift Lighting

(37) Campaign against rotating shifts, and avoid the night shift. Give night-shift workers as much light as possible, preferably above 2500 lux, while being careful not to cause glare. Promote the use of sleep-phase shifting, so workers can be in sync with their environment. References 2,4,6, and 9 tell how to do this.


  • Ref.1: The Lighting Rersearch Institute (LRI), the research arm of the IES, is starting a research center on this very topic atJefferson Medical College, Philadelphia PA. Contact George C. Brainard, PhD @ FAX:(215)955-5515.

  • Ref.2: Rea, Mark S., ed. IES Lighting Handbook, 8th Ed. Illuminating Engr. Soc. of N. Amer. (IESNA) 1993.

  • Ref.3: Sanders, Mark S. and E.J. McCormick. Human Factors in Engineering and Design, 7th Ed., New York: McGraw-Hill 1993.

  • Ref.4: Kripke, Dan F., Sonia Ancoli-Israel, Jeffrey A. Elliott, and Harry Klemfuss. "Light Treatment and Biological Rhythms: An Indexed Bibliography." 1st Ed. (Mar. 1993). Available on paper or disk from: Society for Light Treatment and Biol. Rhythms, PO Box 478, Wilsonville OR 97070 (503)694-2404

  • Ref.5: Rosenthal, Norman E., D.A. Sack, J.C. Gillin, Al J. Lewy, F.K. Goodwin, Y. Davenport, P.S. Mueller, D.A. Newsome, and Tom A. Wehr. "Seasonal Affective Disorder: A Description of the Syndrome and Preliminary Findings with Light Therapy." Archives Gen. Psychiatry 41 (Jan. 1984): 72-80.

  • Ref.6: Several papers by Czeisler, Charles A. (Harvard Med Sch., Brigham and Womens' Hosp.) Referenced in [2,4,9].

  • Ref.7: U.S. Office of Technology Assessment. Biological Rhythms: Implications for the Worker. OTA-BA-463, Sep. 1991.

  • Ref.8: Special Issue: Night and Shiftwork, Ergonomics 36 (Jan. - Mar. 1993).

  • Ref.9: Monk, Timothy H. and Simon Folkard. Making Shiftwork Tolerable. London: Taylor and Francis, 1992.

  • Ref.10: 1989 Steelcase Office Environment Index, Louis Harris and Assocs.

  • Ref.11: Hedge, Alan. "Lighting the Computerized Office." Presentation given to Human Factors Society 1989. (Summary available from Peerless Lighting).

  • Ref.12: Florence, Noel. "The Making of RP-24." LD+A (Lighting Design + Application) (Jan. 1992): 30-33.

  • Ref.13: Rea, Mark S. "Technics: Solving the Problem of VDT Reflections." Progressive Architecture (Oct. 1991): 35-40.

  • Ref.14: Jones, Bill F. "Polarized Light: Effects on Vision, Energy Use, and Cost Reduction." LD+A (Jan. 1992): 34-36.

  • Ref.15: Bernecker, Craig A., Leslie A. Beebe, and Jianou Shi. "Blackwell: A Look Back." LD+A (Aug. 1993): 62-68. (historical evaluation of polarized lighting).

  • Ref.16: Berman, Sam M., D.S. Greenhouse, I.L. Bailey, R.D. Clear, and T.W. Raasch. "Human Electroretinogram Responses to Video Displays, Fluorescent Lighting, and Other High Frequency Sources." Optometry and Vision Science (Aug. 1991): 645-662.

  • Ref.17: Lewis H.W. Technological Risk. New York: W.W. Norton, 1990. Also: Huff, Darrell. How to Lie with Statistics, New York: W.W. Norton, 1954.

  • Ref.18: Blumenthal, Robin G. "New York Schools Consider Installing Full-Spectrum Lights to Help Students." Wall St. Journal, Eastern Ed., 31 Dec. 1992.

  • Ref.19: Walter, Stephen D., Loraine D. Marrett, Harry S. Shannon, Lynn From, and Clyde Hertzman. "The Association of Cutaneous Malignant Melanoma and Fluorescent Light Exposure." Am. J. Epidemiology 135, 1 April 1992, 749-762.

  • Ref.20: Feychting, M. and A. Ahlbom. "Magnetic Fields and Cancer in People Residing Near Swedish High-Voltage Power Lines." IMM Rapport 6, Karolinska Institutt, Stockholm, 1992.

  • Ref.21: "The EMF Uproar." Electronic Design 8 Jul. 1993, 14 and 16 Aug. 1993, 190-196.

  • Ref.22: Walitsky, Paul R. "Health Effects of Lighting Equipment: Fact or Fancy." (white paper from Philips Lighting). (May 1993).

  • Ref.23: Berman, Sam M., G. Fein, D.L. Jewett, G. Saika, and F. Ashford. "Spectral Determinants of Steady-State Pupil Size with Full Field of View." J. IES 21 (Summer '92): 3-13.

  • Ref.24: Robbins, Claude L. Daylighting: Design and Analysis. (New York: VNR 1986).

  • Ref.25: Kobayashi, H. and M. Sato. "Physiological Responses to Illuminance and Color Temperature of Lighting." Ann. Physiol. Anthropology 11 (Jan. 1992): 45-49.

  • Ref.26: Mukae K. and M. Sato. "The Effect of Color Temperature of Lighting Sources on the Autonomic Nervous Functions." Ann. Physiol. Anthropology 11 (Sep. 1992): 533-538.

  • Ref.27: Eldred, K.B. "Optimal Illumination for Reading in Patients with Age-Related Maculopathy." Optometry and Vision Science 69 (Jan. 1992): 46-50.

  • Ref.28: Kolanowski, A.M. "The Clinical Importance of Environmental Lighting to the Elderly." J. Gerontol. Nursing 18 (Jan. 1992): 10-14.

  • Ref.29: Zigman, S. "Light Filters to Improve Vision." Optometry and Vision Science 69 (Apr. 1992): 325-328.

  • Ref.30: Avery, David, M.A. Bolte, S.R. Dager, L.G. Wilson, M. Weyer, G.B. Cox, and D.L. Dunner. "Dawn Simulation Treatment of Winter Depression: A Controlled Study." Am. J. Psychiatry 150 (Jan. 1993): 113-117. Also: unpublished papers by Michael Terman (Columbia-Presbyterian Hosp) referenced in [2,4].

    Copyright 1993 Steven A. Zilber, B.E.

    Material appearing in Architronic may be distributed freel y by electronic or any other means, providing that any such distribution is without charge (unless for purposes of cost recovery by interlibrary loan services) and that Architronic is acknowledged as the source. However, no article may be reprinted in any publication without the explicit written permission of the author(s). This statement must accompany all distributions of Architronic, whether complete or partial.