"If it was known that a certain spectrum of light was effective, while another was not, and these could be administered without the patient being aware which he was receiving, then the ineffective spectrum would be the ideal placebo for light therapy."[1]

There have been many attempts to find a suitable placebo for LIGHT THERAPY, dim white light, dim red light, yellow light, green light etc. Unfortunately, what appeared to be a placebo in one study, in another study seems to work as well or better than the supposed treatment.

Dim white light (<400 lux) has been used in several LIGHT BOX studies to act as a placebo when testing bright white light (>2500lux). Dim yellow, red and blue have also been used. [2] In most cases, these studies used either full spectrum or cool white fluorescent light sources.

A study comparing Green light and Red light[3] found that "The antidepressant effect of GREEN light was superior to that of RED light and (the green light) will provide a treatment effect similar to that seen with white light in earlier studies" although they further suggest the failure of red light as an antidepressant in this study should not be viewed as proving that it is ineffective, but that it was only ineffective in this paradigm."

Another study[4] hypothesized "the action spectrum for phototherapy may lie within the green waveband of light due to: 1) its ability to affect melatonin secretion; 2) the sensitivity of photoreceptors to green light; 3) its ability to penetrate to the dermal capillary beds" and then submitted; "Post-hoc Neuman-Keuls tests revealed that full spectrum phototherapy produced a significant change between pre- and post treatment HRSD scores while green spectrum phototherapy did not" and "...these results suggest that green spectrum light may not be an effective treatment for SAD during the winter.

"A study of DAWN SIMULATORS[5] comparing 30 minute dawn simulation peaking at 0.2 lux and 2 hour simulation peaking at 250 lux concludes that "This study indicates that dawn simulation is an effective treatment for winter depression." "Since the bright light therapy seems to work through the eyes, how can the efficacy of dawn simulation be explained in subjects whose eyes are closed? The eyelids are translucent to light transmitting about IO% (or 25 lux) in the red end of the visible spectrum (above 700 nm), with the transmittance declining to to I%-2% in the green and blue end (below 600 nm). In the visible spectrum, the spectral distribution of the incandescent tights used in the present study shows a predominance of red over blue and green."

Along comes a multicenter study[6] of 46 patients using a head mounted light visor with intensities of bright (7000 lux) and dim (400 lux) in which the results of the dim light were better than the bright light stating, "In fact, we found the pooled dim treatment response rate (65%) to be superior to the pooled bright treatment response rate (23%)..."

Another multicenter head mounted light visor study[7] involving 105 patients (the largest to that date) compared 60 lux, 600 lux and 3500 lux with the assumption that, based on previous studies, the dim (60 lux) light would act as placebo. There was no statistically significant difference in the response to the three intensities.

There had been a consensus that there was a "time-intensity" relationship of 5000 to light therapy. e.g.- 1/2hr @ 10,000 lux; 2hrs @ 2,500 lux but the low intensity results suggest that this would only apply when using a "light box."

To further confound the issue, a three center study[8] involving 55 patients compared a 600 lux white with a 30 lux red head mounted visor with the assumption that, based on previous experience, the red would act as an effective placebo. SIGH-SAD scores declined 37% with the white visor and 43% with the red. Remission rates were "white 36%, red 50%". In the discussion the comment was made "This study demonstrates that a DIM RED LIGHT VISOR designed to be ineffective, was as effective in reversing symptoms of SAD as a brighter white light visor" and, "THE DIM RED VISOR WAS ABOUT AS EFFECTIVE AS ANY INTENSITY WHITE VISOR USED IN PREVIOUS TRIALS."

Most studies involving red light have used dim red (<400 lux) as a placebo and no studies had compared dim red to bright red light. In 1991-92 a study[9] of dim (86 lux) versus bright (1432 lux) RED light from a head mounted unit was done again presuming the dim red would be a placebo. There was no significant difference in the number of responders between subjects treated with the dim (68%) versus bright (67%) nor was there a difference in percent change in SIGH-SAD scores. The findings suggest that dim and bright red light are similarly effective in treating SAD and that, when delivered using LED light from a light visor have equivalent efficacy as compared with previously studied light units.

With the exception of the Dawn Simulators, these "dim light" treatments used HEAD MOUNTED units. One of the theories suggested was that the light source was close to the eyes and bore a fixed relationship to them so much lower light intensities were adequate to achieve results. This does not explain the results reported by the dawn simulator studies - eg: the patient may sleep in various positions during the night. In the noted dawn simulation study, no other light source was to be allowed to 'dilute' the light from the test equipment as could easily be the case with light boxes.

Some other condition common to the head mounted unit and dawn simulators should be examined.

With the exception of the LED light[9], all of the successful DIM LIGHT studies noted used incandescent light sources instead of fluorescent lights. By examining the spectral power output of daylight, full spectrum & cool white fluorescents, and incandescent light sources, we find that only the incandescent light matches or exceeds the RED from natural sunlight. This RED output is common not only to the head mounted units and the dawn simulators but also to the LED (660 nm) in the dim and bright red study.[9]

Is the high intensity (2500-10000 lux) requirement of the light boxes due to their low red output?

If it is the green in the lights that is producing the results, why does the full spectrum fluorescent (close to natural daylight in green output) not produce results at a lower intensity and shorter time? Incandescents are lower in green output.

Why does 500 lux red light from a light box appear to be a good placebo? Could it be that light from a light box is diluted by ambient light? Is it possible that, when telling the patient about the red light, there is an almost imperceptible change in the researchers attitude, perhaps a very slight lowering or slowing down of the voice that is subconsciously noted by the patient and lowers their expectations of the effects of the red light?

One light therapy manufacturer has introduced a device that has three colours of lights, Green, Yellow and Red. Their information is that the Green is for relaxation and the Yellow for concentration but, "RED should be used when a person feels lethargic or tired. It can help increase physical energy etc....". It sounds exactly what SAD patients need. Since the company is quite reputable, we have to assume that there are some studies etc. to authenticate the declaration!!!

Considerable study and research is yet to be done and no area of timing, duration, intensity or wavelength should be relegated to any inferior relevance in this research.


1. Jonathan Stewart, M.D. - PLACEBOS IN EVALUATING LIGHT THERAPY FOR SEASONAL AFFECTIVE DISORDER. Psychopharmacology Bulletin Vol 26 No 4 p 525.
2. Norman E. Rosenthal, David A. Sack, Robert G. Skwerer, Frederick M. Jacobsen, Thomas A. Wehr - PHOTOTHERAPY FOR SEASONAL AFFECTIVE DISORDER. Journal of Biological Rhythm, Vol 3, No 2, 1988, pp. 102-120.
3. Dan A. Oren, M.D., George C. Brainard, Ph.D., Scott H. Johnston, B.A, Jean R. Joseph-Vanderpool, M.D., Elizabeth Sorek, R.N., and Norman E. Rosenthal, M.D. - TREATMENT OF SEASONAL AFFECTIVE DISORDER WITH GREEN AND RED LIGHT. Am. J. Psychiatry 1991; 148: 509-511.
5. David H. Avery, M.D., Mary Ann Bolte, M.D., Stephen R. Dager, M.D., Lawrence G. Wilson, M.D., Matthew Weyer, Gary B. Cox, Ph.D., and David L. Dunner, M.D. - DAWN SIMULATION TREATMENT OF WINTER DEPRESSION: A CONTROLLED STUDY. Am. J. Psychiatry 1993, 150: 113-117.
6. D.E. Moul, M.D., C.J. Hellekson, M.D., D.A. Oren, M.D., A. Frank Ph.D., G. Brainard, Ph.D., M.G. Murray, B.A., T.A. Wehr, M.D., and N.E Rosenthal, M.D. - TREATING SAD WITH A LIGHT VISOR: A MULTICENTER STUDY. SLTBR Absracts 1990, p 15.
7. Russel T. Joffe, Douglas E. Moul, Raymond W. Lam, Anthony J. Levitt, Martin H. Teicher, Breck Lebegue, Dan A. Oren, Alan Buchanan, Carol A Glod, Megan G. Murray, Joanne Brown, and Paul Schwartz. - LIGHT VISOR TREATMENT FOR SEASONAL AFFECTIVE DISORDER: A MULTICENTER STUDY - Psychiatry Research, 46: 29-39.
8. Martin H. Teicher, M.D., Ph.D., Carol A. Glod, R.N., M.S.C.S., Dan A. Oren, M.D., Chris Luetke, Paul Schwartz, M.D., Charlotte Brown, M.A and Norman E. Rosenthal, M.D. - The Phototherapy Light Visor: There is More to it Than Meets the Eye. SLTBR Absracts 4, 1992, p. 20.
9. Anthony J. Levitt, M.D., Russel T. Joffe, M.D., Eleanor King, R.N. - DIM VERSUS BRIGHT RED L[GHT IN SEASONAL AFFECTIVE DISORDER. Can. Psychiatric Assoc. 42nd Annual Meeting presentation; p. 75.