Study on sleep maintenance in a geriatric population with Lo-LIGHT therapy lamps.

Society of Light Treatment and Biological Rhythms Abst 1999;11:39


Phillip Gehrman, Jennifer Martin, and Sonia Ancoli-Israel

University of California, San Diego Department of Psychiatry and VASDHS Department of Psychiatry. 116A, 3350 La Jolla Village Dr., CA 92161

Introduction: Older adults often complain of sleep maintenance insomnia. Individuals with this disorder fall asleep quickly, but complain of nocturnal and early morning awakening. One hypothesis is that this disorder is caused by advanced circadian rhythms. Researchers have therefore attempted to use bright full-spectrum light as a treatment (Campbell et al., 1993). Unfortunately the brightness and glare of full spectrum light made dissuade some individuals from this treatment. An alternative, which has been studied in animals, is light in the blue-green portion of the visible spectrum. An advantage of this treatment is that phase shifts may be achieved with less intense light which may be more easily tolerated. This study compared evening exposure to white full-spectrum light with light in the green portion of the spectrum, for the treatment of sleep maintenance insomnia in older adults.

Methods: As part of a larger study on alternative treatments for insomnia, complete Actillume (Ambulatory Monitoring Inc.) And sleep diary information were collected for 14 individuals (7 men, 7 women; mean age=76 y, SD=9, range=61-88). Subjects completed a 5-day baseline followed by 10 days of treatment. Actillume data were collected during baseline and the last five days of treatment. Subjects were randomized alternatively to recieve full-spectrum white light (WLTX); Apollo Light Systems, ~2500 lux) or light in the green portion of the spectrum (GLTX; Sunnex Biotechnologies, ~350 lux). On each treatment night, subjects sat within 1 m of the light apparatus from 2000-2200 in their home environment. Sleep parameters were estimated and cosinor analyses were performed using Action3 software (Ambulatory Monitoring Inc.) Based on activity data Data from baseline were compared to data from the last five days of treatment using pared t-tests for each treatment group separately. a=.05 for each test).

Results: There were no significant differences in any sleep parameters (total sleep time, etc.) in either treatment group. Inspection of mean differences suggest that there was overall improvement in both groups. Circadian rhythms parameters are summarized in the table below.

WLTX (n=6)

GLTX (n=8)




mean difference




t (df)

-.41 (5)

-.89 (5)

.74 (5)

mean difference




t (df)

-.10 (7)

-.18 (7)

3.62 (7)

Discussion: Although differences in sleep parameters were not statistically significant, the overall phase delay in activity rhythms suggests that light therapy may correct phase abnormalities in this population. Our data suggest that GLTX had a stronger effect that WLTX possibly due to greater compliance using green light treatment. Further research is needed to better understand the effects of light from different portions of the spectrum on circadian rhythms.