Evidence Table 4Seasonal Affective Disorder

STUDY:Authors: Lam et al.108, Michalek et al.109
Year: 2006, 2007
Country: Canada
FUNDING:Canadian Institute of Health Research (CIHR) & CIHR/Wyeth post-doc fellowship award (Michalak)
DESIGN:Study design: RCT
Setting: multi-centre
Sample size: 96
Drug:Light therapyFluoxetine
Dose:10 000 lux20mg/d
Duration:8 weeks8 weeks
Sample size:
INCLUSION:Out-patients aged 18–65 years
DSM-IV criteria for major depressive episodes with a seasonal pattern
>20 on HAMD-17 or >14 on HAMD-17 if >23 on HAMD-24
  1. pregnant or lactating women or could become pregnant
  2. serious suicidal risk
  3. DSM-IV diagnoses of organic mental disorders, substance use disorders, including alcohol, active within the last year, schizophrenia, paranoid or delusional disorders, other psychotic disorders, bipolar I disorder, panic disorder or generalized anxiety disorder not concurrent with major depressive episodes;
  4. serious unstable medical illnesses;
  5. retinal disease that precluded the use of bright light;
  6. history of severe allergies and/or multiple drug adverse reactions;
  7. current use of certain other psychotropic drugs (inc lithium, L-tryptophan, St John’s wort or melatonin)
  8. current use of beta blocking drugs;
  9. use of antidepressants or mood-altering medications within 7 days of baseline;
  10. previous use of fluoxetine or light therapy;
  11. formal psychotherapy started within 3 months of baseline or initiated during the study period;
  12. shift work or southbound travel during the protocol.
POPULATION CHARACTERISTICS:Groups similar at baseline: yes (previous antidepressant therapy 45.8% vs. 33.3%)
Mean age: 42.3, 44.6 Gender (female %): 66.7%
Ethnicity: Canadian
Other population characteristics: NR
OUTCOME ASSESSMENT:Primary Outcome Measures: HAMD-24 clinical response= ≥50% reduction from baseline, clinical remission= response + score≤8, Patient perception of Quality of Life (Q-LES-Q, SF-20)
Secondary Outcome Measures: CGI, BDI-II
Timing of assessments: 1, 2, 4, 8 weeks
  • Significant effect of time, but no significant difference between light therapy and fluoxetine
  • Clinical response rate: both 67%
  • Clinical remission rate: light 50% vs. fluoxetine 54% p=0.84
  • CGI improvement rating: 1.90 vs. 1.92
  • Much/very much improved CGI: both 73%
  • No difference in sub-group “severely depressed” (HAMD-24≥30): response 70% vs. 73% remission 48% vs. 50%
  • improvements in Q-LES-Q: light 20.56 vs. fluoxetine 21.77 (not sig)
  • improvements in SF-20: light 7.82 vs. fluoxetine 9.38 (not sig)
  • improvements in depression were significantly associated with improvements in QoL
Post randomization exclusions: No
Loss to follow-up differential high: No
ATTRITION:Light therapyFluoxetine 20mg/d
Loss to follow-up:16%16%
Withdrawals due to adverse events:2%4%
Withdrawals due to lack of efficacy:NRNR
ADVERSE EVENTS:Light therapy vs. fluoxetine
At least one AE: 77% vs. 75%
Agitation 0% vs. 12.5% p<0.05
Sleep disturbance 2.1% vs. 29.2% p<0.01
Palpitations 0% vs. 10.4% p<0.05

Occurred more often in light therapy than fluoxetine group (though reported as not significant):
Headache 16.7% vs. 10.4%
Feeling faint 6.3% vs. 0
STUDY:Authors: Moscovitch et al 110
Year: 2004
Country: Multinational (Canada and Europe)
FUNDING:Pfizer International
DESIGN:Study design: RCT
Setting: multi-centre
Sample size: 187
Dose:Flexible dose 50–200mg/dn/a
Duration:8 weeks8 weeks
Sample size:9394
INCLUSION:Outpatients, older than 18,
DSM-IIIR criteria for major depression, depressive disorder NOS, bipolar disorder depressed, or bipolar disorder NOS with a seasonal pattern.
12 on HAMD, plus 10 on supplementary items for SAD evaluation, 22 on 29-item HAMD, SIGH-SAD
less than 25% improvement during washout
enrolled during winter
EXCLUSION:Very serious suicide risk, history of alcoholism, drug abuse, poor motivation or intellectual problems
OTHER MEDICATIONS/INTERVENTIONS:Any necessary for other medical conditions, not psychoactive
POPULATION CHARACTERISTICS:Groups similar at baseline: yes
Mean age: 39.6±11.6, 40.0±11.2
Gender (female %): 77.5%
Ethnicity: Austria, Canada, Finland, France, UK
Other population characteristics: NR
OUTCOME ASSESSMENT:Primary Outcome Measures: HAMD-29, HAMD-21, HAMD-17, HAMD item 1, CGI-S, HAMA, HAD-D, HAD-A
Secondary Outcome Measures: not specified
Timing of assessments: 1, 2, 4, 6, 8 weeks
  • Sertraline was better than placebo at endpoint (ITT population) for all of the above efficacy measures: HAMD-29 -17.90 vs. −13.39 p=0.019, HAMD-21 −10.63 vs. −7.51 p=0.016, HAMD-17 −9.36 vs. −6.87 p=0.033, CGI-S −1.60 vs. −1.06 p=0.018, HAMA −8.99 vs. −6.52 p=0.024, HAD-D −5.04 vs. −2.87 p=0.005, HAD-A −4.00 vs. −2.16 p=0.006.
  • Significantly more patients in the sertraline group received a CGI-I rating of one or two (eg: a CGI-I response) at endpoint than placebo (62.4% vs. 46.2% p=0.04)
  • There were no substantial differences in sleep factors (Leeds sleep evaluation)
  • The mean final dose of sertraline was 111.3±44.9 mg
Post randomization exclusions: 1
Loss to follow-up differential high: No
Loss to follow-up:NRNR
Withdrawals due to adverse events:10.8%4.3%
Withdrawals due to lack of efficacy:3.2%14.9%
ADVERSE EVENTS:Sertraline vs. placebo (%)
Treatment related AEs 81.7% vs. 50.0% p=0.001
Nausea 35.5% vs. 8.5% p=0.001
Insomnia 24.7% vs. 10.6% p= 0.01
Diarrhea 19.4% vs. 5.3% p= 0.004
Dry mouth 12.9% vs. 2.1% p=0.005
Ejaculation * 14.3% vs. 4.8 p=0.31
Abdominal pain 9.5% vs. 4.3% p=0.15
Sustained erection * 9.5% vs. 0 % p=0.15
Tremor 7.5% vs. 2.1% p=0.09
Vomiting 6.5% vs. 1.1% p=0.01
Anorexia 6.5% vs. 1.1% p= 0.053
Anxiety 4.3% vs. 1.1% p=0.17

From: Evidence Tables

Cover of Drug Class Review: Second-Generation Antidepressants
Drug Class Review: Second-Generation Antidepressants: Final Update 5 Report [Internet].
Gartlehner G, Hansen RA, Reichenpfader U, et al.
Portland (OR): Oregon Health & Science University; 2011 Mar.
Copyright © 2011 Oregon Health & Science University.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.