Formal Aspects of Light Therapy
As noted earlier in this chapter, light therapy appears to be mediated via the eyes rather than the skin. It is important for clinicians to inform patients of this observation, as many patients will otherwise not appreciate the importance of ensuring that their eyes are open during treatment sessions and are at the appropriate distance from the light source. In addition, it is important to emphasize that there is no evidence that tanning salons, where the eyes are generally covered and the subject’s skin is exposed to light, are useful for SAD. In addition, the light sources in tanning salons are by definition high in UV rays, which can be harmful to both the eyes and the skin. Because UV light is not necessary for an antidepressant effect (Lam et al. 1991; Oren et al. 1991), use of light devices that emit UV light is contraindicated.

Although the initial light treatment studies were performed with full-spectrum lights (i.e., lights that mimic in their wavelength range the spectrum of sunlight), more recent studies have shown ordinary fluorescent lights to have significant antidepressant effects (Oren and Rosenthal 1992; Rosenthal 1993). In addition, there is no evidence that full-spectrum light is superior to ordinary fluorescent light (Lam et al. 1992). Patients often ask whether it would be helpful for them to replace all the fluorescent lights in their home or workplace with full-spectrum lights, indicating a lack of understanding that it is the intensity of light rather than its spectrum that appears to be of critical importance in achieving an antidepressant effect.

Initially, white light of 2,500-lux intensity was tested as the active treatment modality against light sources of lower intensity (Oren and Rosenthal 1992; M. Terman and Terman 1992) and was found to be superior. The 2,500-lux intensity was chosen because it had previously been found to be capable of suppressing nocturnal melatonin secretion (Lewy et al. 1980). The original 2,500-lux light boxes were placed vertically on a table or work surface 3 feet away from the user’s eyes. Currently sold light boxes are slanted toward the user’s face in such a way as to bring 10,000 lux to the eyes. The chief advantage of the new boxes is that antidepressant effects can be obtained from shorter treatment sessions (e.g., 30 minutes rather than 2 hours) in some patients (J. S. Terman et al. 1990), suggesting a reciprocal relationship between duration and intensity of effective light treatment.

The importance of the time of day when light therapy is administered has been one of the most controversial issues in the field. Although a number of studies have found no difference between morning and evening light therapy (Jacobsen et al. 1987; Wirz-Justice et al. 1993), research studies have more commonly shown morning light therapy to be superior to evening light therapy for SAD (Lewy et al. 1998; M. Terman et al. 1989b). Given the controversy surrounding the importance of time of day of administration of light therapy, a treating therapist may be forgiven for some confusion as to how best to proceed. In our view, to achieve the quickest resolution of depression, the patient should be advised to start light therapy within 10 minutes of awakening. After a week of morning light therapy, which should be long enough for the depression to show signs of improvement, the patient may be advised to try light therapy at whatever time of day might be convenient for him for her. If the beneficial effect wears off, return to the early morning time is indicated.

Novel Treatment Devices
Another innovative way of administering light therapy is by means of electronic devices that can be connected to light sources in the bedroom in such a way as to simulate a summer dawn (M. Terman et al. 1989a). In other words, the electronic device can cause the light source to turn on gradually at a predetermined time and increase to its maximum intensity over a preprogrammed period. A few controlled studies have shown “active” dawn simulation to be superior to “control” dawn simulation (Avery et al. 1993, 1994) even though the subject is asleep while receiving this treatment and the final light intensity incident on the eyes (approximately 250 lux) is far lower than that reported to be necessary for a therapeutic response to regular light therapy. Presumably, the well-documented and universally experienced increased sensitivity of the eyes to early morning light accounts for the efficacy of the lower intensities of light delivered by this type of treatment (Knoerchen and Hildebrandt 1976). Dawn simulation can both improve the patient’s mood and help the patient wake up more easily in the morning. The currently available dawn simulator is likely best used as a supplement to regular light therapy, as its efficacy alone is not as high as that of light boxes alone (Lingjaerde et al. 1998). Two studies suggest that negative-ion generators may also have antidepressant effects in SAD (M. Terman and Terman 1995; M. Terman et al. 1998). Further exploration of this possibility is indicated.