The trusted source for
healthcare information and
Shining a Bright Light on Depression
Abstract & Commentary
By Allan J. Wilke, MD, MA, Chair, Department of Integrative Medicine, Ross University School of Medicine, Commonwealth of Dominica Dr. Wilke reports no financial relationship to this field of study.
Synopsis: Elderly patients with depression respond favorably to bright light treatment.
Source: Lieverse R, et al. Bright light treatment in elderly patients with nonseasonal major depressive disorder: A randomized placebo-controlled trial. Arch Gen Psychiatry 2011;68:61-70.
We are approaching the 30th anniversary of the first article published about the effect of bright light treatment (BLT) on affective and sleep disorders.1 A Cochrane review from 2002 supported the use of BLT for sleep disorders in the elderly.2 Another review from 2004 concluded that BLT was modestly effective in nonseasonal depression.3 A systematic review in 2008 concluded that BLT was "an excellent candidate for inclusion ... as adjuvant therapy to antidepressant medication."4 BLT is well established as treatment for seasonal affective disorder (SAD).5
This double-blind, placebo-controlled randomized clinical trial, conducted in the Netherlands, hypothesized that BLT use in elderly patients with nonseasonal major depressive disorder (MDD) would result in symptomatic relief, improved sleep, and a change in hormonal levels associated with depression and sleep (cortisol and melatonin). The participants were recruited between 2003 and 2007. The inclusion criteria were age ≥ 60 years and MDD. They were screened initially with the 15-item Geriatric Depression Scale.6 Those scoring ≥ 5 (the standard cutoff for depression) were interviewed, using the Structured Clinical Interview Guide for DSM-IV Axis I Disorders7 to establish the diagnosis of MDD. The severity of depression was measured with the Hamilton Depression Rating Scale (HAM-D). The 89 subjects were randomized to receive one of two light boxes: one emitting 750 lux light through a pale blue filter or one emitting 50 lux light through a blood-red filter. The study was blinded to subjects by telling them that the study was comparing the effects of blue and red light on depression. (Dim red light has been shown to have no effect.)
Subjects averaged 69 years of age and 65% were female. The two groups were very similar, except for three variables. The Mini-Mental State Examination score in the BLT group was slightly lower than in the placebo group (27.6 vs 28.5 out of 30), the baseline average HAM-D score in the BLT group was higher (18.6 vs 16.2 out of 38), and more subjects in the BLT group had previously received psychotherapy (31 vs 21). Subjects were exposed to light for 60 minutes in the early morning for 3 weeks. At baseline (T0), the end of treatment (T1), and 3 weeks after treatment (T2), several assessments were carried out: urinary free cortisol (UFC), saliva cortisol levels, saliva melatonin levels, autographic estimates of sleep, and light sensor estimates of BLT exposure. The salivary cortisol samples were collected after awakening and just before bedtime, but the melatonin samples were collected only at bedtime. The primary outcome measure was the change in the HAM-D score from T0 to T1. Secondary outcomes included: change in HAM-D scores from T0 to T2, response rates from T0 to T1 and T0 to T2 (defined as ≥ 50% decrease in HAM-D score), and hormone levels. The melatonin samples were obtained four times hourly before bedtime and results were reported as the slope of the rise in levels.
No subjects died (including suicide) during the study, and there were no hospitalizations. Analysis was by intention-to-treat. During the treatment phase (T0 to T1), the average HAM-D score in the BTL group fell to 10.1 and in the placebo group to 10.4. The difference (8.5 vs 5.8) was statistically significant. More subjects in the BLT group responded (50% vs 41%), but this did not reach statistical significance. However, at T2, 53% in the BLT group vs 34% in the placebo group responded, which was significant (number-needed-to-treat = 5). At T2, the average HAM-D score in the BLT group was 8.6, a 10.0-point drop. The placebo group decreased by 5.4, to 10.6. Both of these scores are in the "mild depression" range.
At T0, both groups had the same average bedtime and get-up time, and total sleep time, sleep efficiency, and sleep latency were the same. No change to bedtime was noted during or after treatment, but the BLT group decreased significantly the amount of time in bed compared to the placebo group, and sleep efficiency increased.
The average UFC level decreased in the BLT group and increased in the placebo group during the treatment phase (T0 to T1), but this did not reach statistical significance. However, by T2, it had reached statistical significance, and in the BLT group, the average UFC was in the normal range. Similar results were found for average evening saliva cortisol levels, but the morning decrease did not reach statistical significance. The slope of the average saliva melatonin level increased dramatically in the BLT group, but decreased in the placebo group, during treatment.
The adverse effects profiles of both groups were the same, but the placebo group noted more fatigue and daytime sleepiness.
The good news, take-away message from this study is that BLT works, and works as well as antidepressants in the treatment of MDD in the elderly. This is important, because pharmacotherapy of depression in the elderly is littered with land mines. Tricyclic antidepressants have anticholinergic side effects, and selective serotonin reuptake inhibitors, while better tolerated than TCAs, interact with many drugs also taken by the elderly.
There are several areas of weakness in this study. First is the small population. The investigators had calculated that they would need 63 patients in each arm. They recruited two-thirds of that. Recruitment was halted after 4.5 years for reasons of "limited resources." Secondly, the groups were not evenly matched, despite randomization. The subjects in the BLT group had an average score at the boundary of "moderate depression" and "severe depression."8 The placebo group was in the "moderate depression" mid-range. Finally, the study ended after 6 weeks and the subjects were still "mildly depressed." A much longer study would be welcome.
I did not include the actual numerical results of the cortisol and melatonin levels, mainly because these are intermediate results and not patient-oriented. Cortisol levels are higher in depression, but there is some question about the ability of saliva cortisol levels to distinguish depressed patients from controls.9
Rant alert! BLT requires bright lights, and these are not cheap. I checked Amazon.com for a Philips BriteLITE 6, which emits 1000 lux. It was $280, plus shipping, and Medicare does not cover it. Counter-intuitively, Medicare will pay for antidepressant drugs, and, should your patient need hospitalization after suffering an adverse drug effect, it will pay for that, too. Before ordering a bright light box for your patient, I recommend checking with a local medical supply store to see if it can offer your patient a better deal. In the long run, you may be doing your patient a favor.
1. Lewy AJ, et al. Bright artificial light treatment of a manic-depressive patient with a seasonal mood cycle. Am J Psychiatry 1982;139:1496-1498.
2. Montgomery P, Dennis J. Bright light therapy for sleep problems in adults aged 60+. Cochrane Database Syst Rev 2002;(2):CD003403.
3. Tuunainen A, et al. Light therapy for non-seasonal depression. Cochrane Database Syst Rev 2004;(2):CD004050.
4. Even C, et al. Efficacy of light therapy in nonseasonal depression: A systematic review. J Affect Disord 2008;108:11-23.
5. Partonen T, Lönnqvist J. Seasonal affective disorder. Lancet 1998;352:1369-1374.
6. Geriatric Depression Scale (short form). Available at: www.chcr.brown.edu/GDS_SHORT_FORM.PDF. Accessed Feb. 6, 2011.
7. First MB, et al. Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV). Washington, DC: American Psychiatric Press, Inc.; 1996.
8. Hamilton Depression Rating Scale (HAM-D). Available at: www.assessmentpsychology.com/HAM-D-scoring.pdf. Accessed Feb. 7, 2011.
9. Knorr U, et al. Salivary cortisol in depressed patients versus control persons: A systematic review and meta-analysis. Psychoneuroendocrinology 2010; 35:1275-1286.