Background: Depression frequently occurs in the elderly. Its cause is largely unknown, but several studies point to disturbances of biological rhythmicity. In both normal aging, and depression, the functioning of the suprachiasmatic nucleus (SCN) is impaired, as evidenced by an increased prevalence of day-night rhythm perturbations, such as sleeping disorders. Moreover, the inhibitory SCN neurons on the hypothalamus-pituitary adrenocortical axis (HPA-axis) have decreased activity and HPA-activity is enhanced, when compared to non-depressed elderly. Using bright light therapy (BLT) the SCN can be stimulated. In addition, the beneficial effects of BLT on seasonal depression are well accepted. BLT is a potentially safe, nonexpensive and well accepted treatment option. But the current literature on BLT for depression is inconclusive. Methods/Design: This study aims to show whether BLT can reduce non-seasonal major depression in elderly patients. Randomized double blind placebo controlled trial in 126 subjects of 60 years and older with a diagnosis of major depressive disorder (MDD, DSM-IV/SCID-I). Subjects are recruited through referrals of psychiatric outpatient clinics and from case finding from databases of general practitioners and old-people homes in the Amsterdam region. After inclusion subjects are randomly allocated to the active (bright blue light) vs. placebo (dim red light) condition using two Philips Bright Light Energy boxes type HF 3304 per subject, from which the light bulbs have been covered with bright blue- or dim red light- permitting filters. Patients will be stratified by use of antidepressants. Prior to treatment a one-week period without light treatment will be used. At three time points several endocrinological, psychophysiological, psychometrically, neuropsychological measures are performed: just before the start of light therapy, after completion of three weeks therapy period, and three weeks thereafter. Discussion: If BLT reduces nonseasonal depression in elderly patients, then additional lightning may easily be implemented in the homes of patients to serve as add-on treatment to antidepressants or as a stand-alone treatment in elderly depressed patients. In addition, if our data support the role of a dysfunctional biological clock in depressed elderly subjects, such a finding may guide further development of novel chronobiological oriented treatment strategies. © 2008 Lieverse et al; licensee BioMed Central Ltd.