Key Points
Question
Can the effects of adolescent psychiatric inpatient treatment as usual for major depressive disorder be enhanced by simultaneous use of morning bright light therapy?
Findings
In this randomized clinical trial including 224 adolescents, 4-week inpatient treatment as usual augmented by bright light therapy was not superior to treatment as usual plus placebo red light. Both groups showed significant reductions in depressive symptoms, and there were no serious adverse events during the trial caused by light-related interventions.
Meaning
Bright light therapy lacked an edge over placebo in treating adolescent inpatients with depression in this study; further studies should include treatment as usual without placebo to rule out potential placebo effects.
This randomized clinical trial evaluates bright light therapy vs treatment as usual for youth with depression.
Abstract
Importance
Major depressive disorder is one of the most common mental disorders among adolescents, entailing severe, long-term psychosocial impairment and a high risk of chronicity. In view of the large number of patients requiring treatment, along with insufficient treatment responses with small effect sizes, innovative adjunctive treatment strategies are urgently needed.
Objective
To investigate whether the effect of adolescent psychiatric inpatient treatment as usual for major depressive disorder can be enhanced by simultaneous use of morning bright light therapy.
Design, Setting, and Participants
This was a double-blind, placebo-controlled randomized parallel-group trial with enrollment between March 2018 and November 2020 and follow-up completed in May 2021. The study took place among inpatients at 4 university hospitals for child and adolescent psychiatry across Germany. Of 248 eligible youth aged 12 to 18 years fulfilling ICD-10 criteria for major depressive disorder, 227 were randomized to bright light therapy (n = 116) or placebo red light (n = 111); 151 participants completed the study.
Interventions
Up to 20 sessions of either morning bright light therapy with an intensity of 10 000 lux or placebo red light (100 lux) in addition to multimodal inpatient treatment as usual over 4 weeks.
Main Outcomes and Measures
The primary outcome was the change in Beck Depression Inventory-II (BDI-II) score from baseline to posttreatment in the intention-to-treat sample.
Results
Among the 224 patients included in the intention-to-treat analyses (192 girls and 32 boys; mean [SD] age, 15.5 [1.4] years), the mean (SD) BDI-II score at baseline was 37.3 (8.7). BDI-II scores were significantly reduced after 4 weeks (postassessment) by a mean of −7.5 (95% CI, −9.0 to −6.0; Hedges g = 0.71). Bright light therapy had no impact on this change (no significant group × time effect). Loss to follow-up was 31% (n = 69) at 16 weeks and 49% (n = 110) at 28 weeks. There were 10 serious adverse events throughout the whole trial, which were not considered related to study treatment.
Conclusions and Relevance
The findings in this study did not indicate superiority of bright light therapy over placebo red light therapy in a large sample of adolescent inpatients with moderate or severe major depressive disorder. Both groups benefited equally from treatment as usual, showing relevant symptom reduction.
Trial Registration
German Clinical Trials Register: DRKS00013188
Introduction
Major depressive disorder has devastating consequences for emotional, economic, and social life and can lead to subsequent health impairments.1,2 Available data on the therapeutic efficacy of psychotherapy and antidepressant medication indicate small effect sizes and a high number of patients requiring treatment, underlining the need for complementary interventions to treatment as usual to enhance outcomes.3,4 One promising candidate that has been added to treatment as usual is morning bright light therapy, which is a particularly cost-effective, noninvasive treatment option with few adverse effects in adults. Bright light therapy involves daily exposure to bright light and is typically administered using a fluorescent light box or light glasses. It is assumed that bright light therapy activates the circadian pacemaker system in the brain by stimulating light-sensitive retinal ganglion cells.5
While it remains unclear precisely how the biomolecular mechanisms of the circadian pacemaker in the brain work, recent meta-analyses have indicated treatment effectiveness of bright light therapy for reducing depressive symptoms in adults with seasonal depression6 and nonseasonal depression.5 However, it is yet to be determined whether these findings from adults can be transferred to children and adolescents.
A small number of pilot trials have investigated the feasibility and effectiveness of bright light therapy in youth, yielding promising findings. A randomized clinical pilot trial7 including adolescent inpatients with depression compared bright light therapy (10 000 lux via light boxes as add-on treatment) with a placebo group (dim light; 100 lux) over 2 weeks, revealing significant improvements in sleep parameters and a shift toward morningness (phase advance) in the bright light therapy group. Phase advance and improved sleep quality predicted a reduction in depressive symptoms. Another study8 found evidence of a delayed effect of bright light therapy, reporting a higher overall remission rate in a bright light therapy group (46.7%) compared to dim light (25.9%) at 3-week follow-up.
A randomized clinical trial9 comparing bright light therapy only with combined bright light therapy plus 1 night wake therapy in inpatient youth with depression revealed an equal reduction of depressive symptom burden in both groups. Finally, a pilot trial10 in adolescent inpatients using light glasses with 10 000 lux delivered in the morning for 30 minutes per day for 4 weeks revealed similar efficacy with light glasses and light boxes independent of duration (2 vs 4 weeks). Based on this preliminary evidence, we assumed that morning bright light therapy applied over 4 weeks in addition to treatment as usual might yield a clinically relevant reduction in depressive symptoms in adolescent inpatients. However, large-scale randomized clinical trials to validate these effects in youth and examine the prevalence and severity of adverse effects are lacking.
Thus, we performed a multicenter randomized clinical trial (DeLight) comparing the efficacy of morning bright light therapy to a placebo red light condition in addition to treatment as usual in adolescent inpatients with major depressive disorder over 4 weeks. We hypothesized that added bright light therapy would be superior to placebo.
Methods
Trial Design
This double-blind, placebo-controlled randomized parallel-group trial was conducted at 4 German university hospitals for child and adolescent psychiatry. The study protocol and its amendments were approved by all 4 responsible ethics committees. The trial was registered in the German Clinical Trials Register (DRKS00013188). There were no changes to the methods after the trial commenced. The study protocol including sample size calculations and informed consent procedures has been published and is available online11 and in Supplement 1. The study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.
Randomization and Blinding
Randomization was performed using a validated computer program, in permuted blocks of variable size at a 1:1 ratio stratified by prior pharmacological antidepressant therapy (yes/no), trial site, and sex. Patients, parents, clinical evaluators, and staff on the ward were blinded regarding treatment allocation until the end of the trial. Blinding was assured by separating diagnostic assessments (performed by trained, clinically experienced research assistants) from general patient and data handling, recruitment, and randomization.
Patients
In total, 248 patients were screened for eligibility between March 26, 2018, and November 16, 2020 (center 1, n = 148; center 2, n = 29; center 3, n = 38; center 4, n = 33). Follow-up was completed in May 2021, with the last patient out by June 2021. Of those screened, 227 patients were randomized to bright light therapy (n = 116) and placebo red light (n = 111), 224 were analyzed as the intention-to-treat sample (114 in the bright light therapy group and 111 in the placebo red light group), and 152 patients (61%) completed the trial. The main reason for loss to follow-up was failure to reach patients after discharge despite repeated calls and letters. For further details on patient flow, see Figure 1.
Figure 1. Patient Flow From Screening to Follow-Up Assessment Time Points.
Bright light therapy consisted of up to 20 sessions of morning bright light therapy with an intensity of 10 000 lux. The control condition consisted of up to 20 sessions of placebo red light with an intensity of 100 lux. Both groups received multimodal inpatient treatment as usual.
Inpatients aged 12 to 18 years with a moderate to severe episode of major depressive disorder as their main diagnosis (ICD-10 codes F32.1, 32.2, F33.1, F33.2, or F92.0) and a Beck Depression Inventory-II (BDI-II) summary score greater than 19 (moderate to severe depression) were included. A Children’s Depression Rating Scale–Revised score of 45 or greater (indicating moderate to severe levels of depression) was used as an additional eligibility criterion.12 The following exclusion criteria were applied: insufficient German language skills of patients or guardians, IQ less than 70, treatment with β-blockers or high-potency neuroleptics, diseases of the retina, acute suicidality, bipolar disorder, schizophrenia, and pregnancy or lactation.
Interventions
All patients received multimodal inpatient treatment as usual comprising psychotherapy in group and individual settings; family sessions; complementary therapies such as music, art, and sports therapy; and occupational treatment. If psychopharmacological antidepressant treatment was warranted, it was provided in accordance with the attending physician’s clinical judgment (noninterventional). Changes in the treatment regimen (eg, initiation, cessation, and dose-adjustment of antidepressants) were assessed and documented weekly. In addition to treatment as usual, intervention patients received morning bright light therapy with 10 000 lux light glasses (Luminette) over 4 weeks (except weekends, amounting to 20 days of intervention). Patients in the control group received light therapy with identical glasses with dim red light (100 lux), which has no known effects on depression. Interventions took place in the morning for 30 minutes, 7.5 to 9.5 hours after the individual dim light melatonin onset estimated using the Morningness-Eveningness Questionnaire.10,13,14
Outcomes
The primary outcome was change in BDI-II15 scores from baseline to 4 weeks (0-8, no depression; 9-13, minimal depression; 14-19, mild depression; 20-28, moderate depression; 29-63, severe depression16). Secondary outcomes included change in BDI-II scores at follow-up (weeks 16 and 28), remission and response rates (remission defined as a BDI-II summary score ≤19 and a Clinical Global Impression–Improvement17 score ≤2; response defined as 50% reduction of the baseline BDI-II score), Children’s Depression Rating Scale–Revised, health-related quality of life assessed with the KIDSCREEN-10,18 the physical activity scale of the Child Health and Illness Profile–Adolescent Edition19, and sleep quality measured with the Sleep Questionnaire20 from baseline to 4 weeks and from baseline to 16 weeks and 28 weeks. Additionally, levels of the hormone melatonin were analyzed to capture the biological mechanisms associated with bright light therapy. Saliva samples were collected at 5 time points (1 per hour, individually determined) in the evening at baseline and after 4 weeks to estimate dim light melatonin onset (it was not possible to collect saliva samples at follow-up, as almost all patients had been discharged). The sampling times were determined by the German version of the Morningness-Eveningness Questionnaire13 to measure melatonin onset (see Griefahn et al21 for a detailed description). The Seasonal Pattern Assessment Questionnaire22,23,24 and therapy expectations and acceptance were included as control variables. Staff and patients reported on treatment adherence (days glasses were worn). There were no changes in the primary and secondary end points after study commencement. See detailed information on all measurement instruments in the study protocol.11
Tolerability and Safety
Frequency and severity of adverse events were monitored in a standardized manner during the intervention. Serious adverse events were coded according to Medical Dictionary for Regulatory Activities (MedDRA) version 24.0 terminology. Detailed information collected for each adverse event included a description of the event, duration, severity, relationship to trial treatment, action taken, and clinical outcome. All serious adverse events had to be documented and reported to the external safety monitoring board within 24 hours. Annual reports with all relevant safety data were provided to the board and evaluated by its members.
Statistical Analysis
All randomized participants were included in the intention-to-treat population used for further analysis. Given a deviation from the study protocol due to 39% loss to follow-up, the main analysis was performed as a mixed model with fixed effects for group (bright light therapy vs placebo red light), week (4, 8, and 12), previous antidepressant therapy (yes or no), sex (male or female), and the interaction term of group and week. Remission and response data were analyzed using logistic regression models. A 2-sided significance level of α = .05 was assumed for P values, and 95% CIs were calculated. A sensitivity analysis on the completer population (those completing at least 16 sessions) was conducted (eAppendix in Supplement 2). For the analysis of secondary parameters, analogous statistical models were used. For melatonin analyses, dim light melatonin onset was determined only when a melatonin concentration increase could be observed, using the hockey stick method.25 Patients were analyzed if pretreatment and posttreatment dim light melatonin onset could be identified. Finally, correlations were performed to check whether change in depressive symptoms was associated with number of therapy sessions. All analyses were done with SAS version 9.4 (SAS Institute).
Results
Patients
Among the 224 patients included in the intention-to-treat analyses (192 girls and 32 boys; mean [SD] age, 15.5 [1.4] years), the mean (SD) BDI-II score at baseline was 37.3 (8.7). Table 1 presents further baseline characteristics of the total sample and both study groups. Based on BDI-II scores, 36 patients (16%) (14 [12%] in bright light therapy and 22 [20%] in placebo red light) were categorized as moderately ill and 188 (84%) (100 [88%] in bright light therapy and 88 [80%] in placebo red light) as severely ill.
Table 1. Sample Characteristics at Baseline.
| Characteristic | Mean (SD) | ||
|---|---|---|---|
| Bright light therapy (n = 114) | Placebo red light therapy (n = 110) | Total (N = 224) | |
| Sex, No. (%) | |||
| Female | 97 (85) | 95 (86) | 192 (86) |
| Male | 17 (15) | 15 (14) | 32 (14) |
| Age, y | 15.5 (1.6) | 15.7 (1.4) | 15.6 (1.5) |
| Weight, kg | 65.0 (17.8) | 64.3 (14.9) | 64.7 (16.4) |
| Height, cm | 166.9 (8.2) | 166.3 (7.2) | 166.6 (7.7) |
| Region of origin, No. (%)a | |||
| European | 108 (95) | 105 (95) | 213 (95) |
| Otherb | 6 (5) | 5 (5) | 11 (5) |
| Siblings, No. | 2.1 (1.8) | 1.8 (1.5) | 1.9 (1.7) |
| Lives with both parents, No. (%) | |||
| Yes | 42 (37) | 35 (32) | 77 (34) |
| No | 72 (63) | 75 (68) | 147 (66) |
| Comorbidity | 0.6 (1.3) | 0.7 (1.4) | 0.7 (1.3) |
| BDI II score | 37.8 (8.9) | 36.9 (8.5) | 37.3 (8.7) |
| CDRS-R score | 64.9 (9.6) | 65.6 (11.0) | 65.3 (10.3) |
| CGI severity score | 5.2 (0.6) | 5.2 (0.6) | 5.2 (0.6) |
| KIDSCREEN-10 score | 26.3 (4.9) | 26.4 (5.1) | 26.3 (5.0) |
| CHIP-AE Physical Activity score | 10.3 (3.4) | 11.5 (3.9) | 10.9 (3.7) |
| SPAQ score | 9.7 (5.4) | 9.8 (5.3) | 9.8 (5.4) |
| MEQ score | 34.2 (8.6) | 35.3 (9.3) | 34.8 (8.9) |
| SF-B/R general sleep score | 2.4 (0.8) | 2.3 (0.8) | 2.3 (0.8) |
| Antidepressant medication, No. (%)c | 22 (19.3) | 17 (15.5) | 39 (17.4) |
Abbreviations: BDI-II, Beck Depression-Inventory-II; CDRS-R, Children’s Depression Rating Scale-Revised; CGI, Clinical Global Impression; CHIP-AE, Child Health and Illness Profile–Adolescent Edition; SPAQ, Seasonal Pattern Assessment Questionnaire; MEQ, Morningness-Eveningness Questionnaire; SF-BR, Sleep Questionnaire (Schlaffragebogen-B/Revised).
Region of origin was collected via self-report and included to assess the representativeness of the data.
Other regions of origin included Africa, Asia, and other, consolidated owing to small numbers.
Including selective serotonin reuptake inhibitors (n = 34), serotonin-norepinephrine reuptake inhibitors (n = 2), and other antidepressants (n = 3).
Treatment Adherence
The mean (SD; median; range) number of therapy sessions was 15.9 (5.4; 18; 1-24). A total of 110 patients (49%) received the target of 20 light therapy sessions. The mean (SD) participation duration was 156 (81.4) days and did not differ significantly between treatment groups (bright light therapy: 149 [81] days vs placebo red light, 163 [81]; P = .22). Loss to follow-up was 31% (n = 69) at 16 weeks and 49% (n = 110) at 28 weeks.
Primary Outcome
Changes in depression symptoms are shown in Figure 2. BDI-II scores changed from baseline to week 4 (mean change, −7.5; 95% CI, −9.0 to −6.0; Hedges g = 0.71), with substantial clinical effect sizes in both groups (bright light therapy: mean change: −6.6; 95% CI, −8.3 to −4.9; Hedges g = 0.74 and placebo red light: mean change, −8.6; 95% CI, −10.8 to −6.4; Hedges g = 0.78). Treatment differences adjusted for covariates from the primary analysis were not statistically significant (mean change, 2.3; 95% CI, −0.8 to 5.3; P = .15).
Figure 2. Change in Depressive Symptoms Over the Study Period.
BDI-II indicates Beck Depression Inventory-II; BLT, bright light therapy (10 000 lux); RED, control condition with red placebo light (100 lux).
Secondary Outcomes
During follow-up, larger effect sizes for the reductions in BDI-II scores were observed in both groups (bright light therapy: mean change, −11.0; 95% CI, −13.0 to −8.9; Hedges g = 1.23 at week 16 and mean change, −12.2; 95% CI, −14.6 to −9.7; Hedges g = 1.37 at week 28 vs placebo red light: mean change, −12.4; 95% CI, −14.3 to −10.5; Hedges g = 1.46 at week 16 and mean change, −11.2; 95% CI, −13.4 to −9.0; Hedges g = 1.32 at week 28). Treatment differences from baseline to follow-up did not differ significantly between groups (week 16 Δ, 2.16; 95% CI −1.88 to 6.19; P = .29 vs week 28 Δ, 0.43; 95% CI, −3.93 to 4.78; P = .85). The results of the sensitivity analysis on the completer population supported these findings (eAppendix in Supplement 2). Post hoc correlation analyses revealed that session number correlated with the reduction in BDI-II scores (total sample: r, −0.21; P = .003; bright light therapy: P = .04; placebo red light: P = .40) but not with response (r, 0.10; P = .18) or remission (r, 0.077; P = .28).
Some patients could not be included in the remission and response analysis, mostly due to study discontinuation. Groups did not differ regarding number of remitted patients or regarding response to treatment. Remission and response rates were low at postassessment and increased up to one-third of patients during follow-up (Table 2).
Table 2. Response and Remission Rates at Posttreatment Assessment (4 Weeks) and Follow-Up Period (Weeks 16 and 28).
| Outcome | No. | No. (%) | Odds ratio (95% CI) | P value | |
|---|---|---|---|---|---|
| Bright light therapy | Placebo red light therapy | ||||
| Remissiona | |||||
| 4 wk | 196 | 11 (11) | 15 (15) | 0.68 (0.30-1.57) | .37 |
| 16 wk | 151 | 22 (29) | 27 (36) | 0.72 (0.37-1.44) | .36 |
| 28 wk | 109 | 16 (31) | 16 (28) | 1.20 (0.53-2.74) | .67 |
| Responseb | |||||
| 4 wk | 196 | 14 (14) | 19 (19) | 0.69 (0.32-1.46) | .33 |
| 16 wk | 151 | 25 (32) | 27 (36) | 0.86 (0.44-1.67) | .65 |
| 28 wk | 109 | 19 (36) | 18 (30) | 1.34 (0.61-2.93) | .47 |
Remission was defined as a Beck Depression-Inventory-II summary score ≤19 and a Clinical Global Impression score ≤2.
Response was defined as a 50% reduction of the baseline Beck Depression-Inventory-II score.
No differences between the groups emerged for medication status (baseline: bright light therapy, 21 of 114 [19.3%] vs placebo red light, 17 of 110 [15.5%], newly prescribed during intervention: bright light therapy, 33 of 114 [28.9%] vs placebo red light, 30 of 110 [27.3%]). In total, 54 of 114 patients in the bright light therapy group (47.4%) and 47 of 110 in the placebo red light group (42.7%) received antidepressant medication. Significant improvements emerged in Children’s Depression Rating Scale–Revised and Clinical Global Impression–Improvement scores over time but did not differ significantly between the groups. No significant changes in self-reported sleep quality, psychosocial functioning, or physical activity were observed during the study (Table 3). Treatment expectation and acceptance were comparable between groups, showing good evaluations—179 of 222 individuals (81%) expected a positive effect with treatment and 177 of 195 (91%) reported they might recommend the treatment to a friend with similar problems (Table 3). Regarding melatonin, no group differences were apparent at pretreatment or posttreatment assessments.
Table 3. Secondary and Exploratory Outcomes According to Treatment Conditiona.
| Outcome | Mean (SE) | P value | df | t or χ2 | |
|---|---|---|---|---|---|
| Bright light therapy | Placebo red light therapy | ||||
| CDRS-Rb | |||||
| Baseline | 64.9 (0.90) | 65.6 (1.05) | NA | NA | NA |
| 4 wk | 53.3 (1.31) | 52.9 (1.28) | .82 | 200 | 0.23 |
| 16 wk | 48.3 (1.57) | 48.9 (1.58) | .77 | 200 | −0.30 |
| 28 wk | 46.0 (1.89) | 47.2 (1.81) | .63 | 200 | −0.44 |
| CGI-improvementc | |||||
| Baseline (CGI severity) | 5.2 (0.06) | 5.2 (0.06) | NA | NA | NA |
| 4 wk | 3.2 (0.11) | 3.1 (0.11) | .46 | 200 | 0.74 |
| 16 wk | 2.7 (0.12) | 2.7 (0.12) | .85 | 200 | 0.19 |
| 28 wk | 2.8 (0.14) | 2.6 (0.14) | .39 | 200 | 0.86 |
| SF-BR sleep qualityd | |||||
| Baseline | 2.8 (0.05) | 2.8 (0.06) | NA | NA | NA |
| 4 wk | 2.9 (0.06) | 2.9 (0.06) | .49 | 185 | −0.69 |
| 16 wk | 3.0 (0.07) | 3.1 (0.07) | .14 | 185 | −1.49 |
| 28 wk | 3.0 (0.08) | 3.1 (0.08) | .46 | 185 | −0.74 |
| SF-BR feeling of being recoverede | |||||
| Baseline | 2.0 (0.06) | 2.0 (0.06) | NA | NA | NA |
| 4 wk | 2.2 (0.08) | 2.3 (0.08) | .33 | 153 | −0.98 |
| 16 wk | 2.6 (0.14) | 2.5 (0.14) | .62 | 153 | 0.50 |
| 28 wk | 2.9 (0.18) | 2.9 (0.17) | .96 | 153 | −0.05 |
| KIDSCREEN-10f | |||||
| Baseline | 26.3 (0.46) | 26.4 (0.49) | NA | NA | NA |
| 4 wk | 29.4 (0.67) | 29.7 (0.69) | .76 | 172 | −0.30 |
| 16 wk | 30.4 (0.89) | 30.5 (0.92) | .89 | 172 | −0.14 |
| 28 wk | 30.2 (1.08) | 31.4 (1.07) | .42 | 172 | −0.81 |
| CHIP-AE Physical Activityg | |||||
| Baseline | 10.4 (0.32) | 11.5 (0.37) | NA | NA | NA |
| 4 wk | 11.3 (0.31) | 11.5 (0.31) | .63 | 198 | −0.48 |
| 16 wk | 10.8 (0.41) | 11.5 (0.41) | .17 | 198 | −1.37 |
| 28 wk | 11.4 (0.48) | 10.9 (0.44) | .47 | 198 | 0.73 |
| Melatoninh | |||||
| Baseline | 77.6 (7.53) | 62.0 (4.81) | NA | NA | NA |
| 4 wk | 80.3 (5.88) | 66.1 (4.81) | .26 | 80.44 | 1.14 |
| Therapy expectancy (depression), No. (%)i | |||||
| Not correct | 5 (4) | 2 (2) | .62 | 3 | 1.77 |
| Rather not | 18 (16) | 18 (16) | |||
| Rather | 56 (50) | 52 (47) | |||
| Correct | 33 (29) | 38 (35) | |||
| Therapy acceptance, No. (%)j | |||||
| No | 2 (2) | 2 (2) | .97 | 3 | 0.2537 |
| Rather not | 8 (8) | 6 (6) | |||
| Possibly | 53 (53) | 53 (55) | |||
| Yes | 36 (36) | 35 (36) | |||
| Antidepressant medication initiation during treatment, No. (%)k | 33 (29) | 30 (28) | .78 | 1 | 0.0777 |
Abbreviations: CDRS-R, Children’s Depression Rating Scale–Revised; CGI, Clinical Global Impression; CHIP-AE, Child Health and Illness Profile–Adolescent Edition; SF-BR, Sleep Questionnaire (Schlaffragebogen-B/Revised).
Total of 224 patients were analyzed in intention-to-treat analysis; 152 completed the trial.
Scores of 35 to 40 indicate mild depression; scores higher than 44 indicate moderate to severe depression.
The CGI consists of 7 items rated from 1 “very much improved” to 7 “very much worse.”
Higher scores indicate better sleep.
Measures the general health-related quality of life on a 10-item scale. Higher scores indicate higher quality of life.
To measure physical activity, the respective subscale with 5 items of the CHIP-AE (measuring health and quality of life in adolescents) was applied. Higher scores indicate more physical activity.
A total of 5 hourly samples were taken.
Assessed with 2 items asking whether participants expected the light therapy to change their depressive symptoms and sleep.
Measured with 1 item: “Would you recommend the light glasses to a friend with similar symptoms?”
Including selective serotonin reuptake inhibitors (n = 54 cases), serotonin-norepinephrine reuptake inhibitors (n = 3), and other antidepressants (n = 6).
Tolerability and Safety
Headache and dizziness were the most frequently reported adverse events (Table 3). These were classified as potentially related to bright light therapy and have been evaluated as typical for bright light therapy in adult samples.26 Ten serious adverse events were reported (Table 3). No lethal serious adverse events occurred. With the exception of 1 unknown outcome, all serious adverse events improved (n = 2) or recovered (n = 7). None of the serious adverse events were considered as related to study treatment.
Discussion
This randomized clinical trial examined the efficacy of bright light therapy at an intensity of 10 000 lux compared with a placebo condition (placebo red light, 100 lux) in a large sample of adolescent inpatients with moderate to severe major depressive disorder. The groups did not differ in the primary or secondary outcome parameters. Sex, medication status, study site, and seasonality did not affect the outcomes. Both groups showed a significant improvement in depression, with large stable effect sizes during follow-up. Remission and response rates and psychosocial improvements were similar in both groups. Health-related quality of life, self-rated sleep quality, and levels of physical activity and evening melatonin release did not change significantly over the course of treatment and follow-up. No intervention-induced serious adverse events occurred in either group. The intervention was rated as helpful and would be recommended by approximately 80% of participants. Possible explanations for the nonsignificant findings include the influence of the inpatient setting, the severity of disease, possible placebo effects, and the use of light glasses, which have rarely been used in young people.
The highly structured therapeutic inpatient setting, in which different interventions were combined, and the often additional use of antidepressants might have led to the significant improvements in depressive symptoms in both groups. Patients in a study by Bogen and colleagues7 were medication naive, whereas in the present study, almost half of the study sample took antidepressants. It may be that individuals with less severe depression respond better to adjunctive treatments, such as bright light therapy. These considerations correspond to recent meta-analytic evidence from studies of bright light therapy in adults,5 which indicated better effects in outpatient settings (SMD, 0.45) compared to inpatient settings (SMD, 0.27).
Participants’ severity of depression may also have influenced the overall results. Compared to the study by Bogen et al7 demonstrating initial evidence of an additional positive effect with bright light therapy for depressive symptoms in the inpatient setting, a greater proportion of patients in the present study had more severe depressive symptoms. Future studies should examine the response to bright light therapy in outpatients with low depressive symptom burden and low comorbidity.
Moreover, a higher placebo effect has been observed in adolescents with depression compared to other disorders and to adults with depression27; in youth, placebo response rates have been estimated at approximately 30%.28 Given the good acceptance (80% recommendation rate) and positive treatment expectation in both groups (also about 80%), it cannot be ruled out that any sort of light-based treatment ritual as an adjunctive intervention has a placebo effect.
Another important issue is the potential effect of the control condition. Previous studies using placebo red light as a control condition have found superiority of bright light therapy over placebo red light,29,30 but we cannot completely rule out the possibility that placebo red light itself had therapeutic effects and minimized the differences between groups.
The use of light glasses may have led to worse outcomes than the light boxes used in previous studies,7 although previous research has demonstrated that light glasses are as equally effective as light boxes.10 Moreover, for ease of integration into inpatient routine, the light glasses were handed out by nurses every morning and worn independently. Thus, they may not have been worn continuously or for as long as requested, which might have adversely impacted the results. It may be that adherence to therapy and patient motivation play a greater role regarding light glasses compared to more supervised treatments with light boxes. This needs to be clarified in further studies.
The lack of changes in melatonin and sleep could reflect patients’ still clinically relevant symptom burden. It seems plausible, and is consistent with previous findings,31 that depression that is still clinically relevant is associated with persistent disturbances in melatonin secretion. On the other hand, the present results regarding melatonin saliva samples are limited because many data sets were missing, and the results were consequently based on a small sample. Other studies7,32 have demonstrated changes in melatonin secretion before and after bright light therapy, which may be due to the less severely ill patient population.
Strengths and Limitations
To our knowledge, this study is the largest investigation of the efficacy of bright light therapy in youth to date, surpassing participant numbers in adult studies. The sound methodology in terms of double-blinding, the inclusion of assessments from different sources (clinicians and self-report), and the use of objective and biological variables to assess treatment outcomes minimize the risk of bias and strengthen the robustness of the results. However, the evaluation of the salivary melatonin samples was limited due to numerous dropouts. Also, without a third treatment group receiving no additional measures beyond treatment as usual, it is not possible to disentangle the specific effects of bright light therapy, any sort of light-based treatment rituals, and inpatient treatment alone.33 The proportion of participants from non-European regions (5%) was lower than in the study centers’ catchment areas (approximately 20%). The interpretation of the outcomes may therefore not be fully generalizable.
Conclusion
Despite the lack of significant group differences regarding depressive symptoms, this study provides insights into the treatment of severely depressed adolescents. The significant reduction in depressive symptoms observed in both the bright light therapy and placebo red light groups and the fact that more than 60% of patients were classified as remitted or responsive underscore the effectiveness of the intensive inpatient treatment setting. However, an additional effect of bright light therapy was not demonstrated. To further our understanding and improve treatment outcomes, future research should examine the efficacy of bright light therapy in less intensive treatment contexts and target patients with varying severity of depression and comorbidities.
Trial protocol
eAppendix
Data sharing statement
References
- 1.Rao U, Chen LA. Characteristics, correlates, and outcomes of childhood and adolescent depressive disorders. Dialogues Clin Neurosci. 2009;11(1):45-62. doi: 10.31887/DCNS.2009.11.1/urao [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Shorey S, Ng ED, Wong CHJ. Global prevalence of depression and elevated depressive symptoms among adolescents: a systematic review and meta-analysis. Br J Clin Psychol. 2022;61(2):287-305. doi: 10.1111/bjc.12333 [DOI] [PubMed] [Google Scholar]
- 3.Cuijpers P, Karyotaki E, Ciharova M, et al. The effects of psychological treatments of depression in children and adolescents on response, reliable change, and deterioration: a systematic review and meta-analysis. Eur Child Adolesc Psychiatry. 2023;32(1):177-192. doi: 10.1007/s00787-021-01884-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Aust S, Brakemeier EL, Spies J, et al. Efficacy of augmentation of cognitive behavioral therapy with transcranial direct current stimulation for depression: A randomized clinical trial. JAMA Psychiatry. 2022;79(6):528-537. doi: 10.1001/jamapsychiatry.2022.0696 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Tao L, Jiang R, Zhang K, et al. Light therapy in non-seasonal depression: an update meta-analysis. Psychiatry Res. 2020;291:113247. doi: 10.1016/j.psychres.2020.113247 [DOI] [PubMed] [Google Scholar]
- 6.Pjrek E, Friedrich ME, Cambioli L, et al. The efficacy of light therapy in the treatment of seasonal affective disorder: a meta-analysis of randomized controlled trials. Psychother Psychosom. 2020;89(1):17-24. doi: 10.1159/000502891 [DOI] [PubMed] [Google Scholar]
- 7.Bogen S, Legenbauer T, Gest S, Holtmann M. Lighting the mood of depressed youth: feasibility and efficacy of a 2 week-placebo controlled bright light treatment for juvenile inpatients. J Affect Disord. 2016;190:450-456. doi: 10.1016/j.jad.2015.09.026 [DOI] [PubMed] [Google Scholar]
- 8.Martiny K, Lunde M, Undén M, Dam H, Bech P. Adjunctive bright light in non-seasonal major depression: results from clinician-rated depression scales. Acta Psychiatr Scand. 2005;112(2):117-125. doi: 10.1111/j.1600-0447.2005.00574.x [DOI] [PubMed] [Google Scholar]
- 9.Gest S, Legenbauer T, Bogen S, Schulz C, Pniewski B, Holtmann M. Chronotherapeutics: an alternative treatment of juvenile depression. Med Hypotheses. 2014;82(3):346-349. doi: 10.1016/j.mehy.2014.01.002 [DOI] [PubMed] [Google Scholar]
- 10.Kirschbaum-Lesch I, Gest S, Legenbauer T, Holtmann M. Feasibility and efficacy of bright light therapy in depressed adolescent inpatients. Z Kinder Jugendpsychiatr Psychother. 2018;46(5):423-429. doi: 10.1024/1422-4917/a000603 [DOI] [PubMed] [Google Scholar]
- 11.Holtmann M, Mokros L, Kirschbaum-Lesch I, et al. Adolescent depression: study protocol for a randomized, controlled, double-blind multicenter parallel group trial of bright light therapy in a naturalistic inpatient setting (DeLight). Trials. 2018;19(1):568. doi: 10.1186/s13063-018-2949-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Plener PL, Grieb J, Spröber N, et al. Convergence of Children’s Depression Rating Scale-Revised scores and clinical diagnosis in rating adolescent depressive symptomatology. Ment Illn. 2012;4(1):e7. doi: 10.4081/mi.2012.e7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Horne JA, Ostberg O. A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms. Int J Chronobiol. 1976;4(2):97-110. [PubMed] [Google Scholar]
- 14.Wirz-Justice A, Benedetti F, Terman M. Chronotherapeutics for Affective Disorders: A Clinician’s Manual for Light and Wake Therapy. 2nd ed. S. Karger; 2013. doi: 10.1159/isbn.978-3-318-02091-5 [DOI] [PubMed] [Google Scholar]
- 15.Hautzinger M, Keller F, Kühner C. Beck Depressions-Inventar Revision—Manual. Harcourt Test Services GmbH; 2006. [Google Scholar]
- 16.Besier T, Goldbeck L, Keller F. Psychometrische Gütekriterien des Beck-Depressions-Inventars II (BDI-II) bei jugendpsychiatrischen Patienten. Psychother Psychosom Med Psychol. 2008;58(2):63-68. doi: 10.1055/s-2007-986195 [DOI] [PubMed] [Google Scholar]
- 17.Guy W. Clinical Global Impressions (CGI) Scale. APA PsycTests; 2000. [Google Scholar]
- 18.Ravens-Sieberer U, Auquier P, Erhart M, et al. ; European KIDSCREEN Group . The KIDSCREEN-27 quality of life measure for children and adolescents: psychometric results from a cross-cultural survey in 13 European countries. Qual Life Res. 2007;16(8):1347-1356. doi: 10.1007/s11136-007-9240-2 [DOI] [PubMed] [Google Scholar]
- 19.Starfield B, Riley AW, Green BF, et al. The adolescent child health and illness profile. a population-based measure of health. Med Care. 1995;33(5):553-566. doi: 10.1097/00005650-199505000-00008 [DOI] [PubMed] [Google Scholar]
- 20.Görtelmeyer R. Manual zum SF-A/R und SF-B/R - Schlaffragebogen A und B - Revidierte Fassung. Hogrefe; 2011. [Google Scholar]
- 21.Griefahn B, Kunemund C, Brode P, Mehnert P. Zur Validitat der deutschen Übersetzung des Morningness-Eveningness-Questionnaires von Horne und Ostberg. The validity of a german version of the morningness-eveningness-questionnaire developed by Horne and Ostberg. Somnologie (Berl). 2001;5(2):71-80. doi: 10.1046/j.1439-054X.2001.01149.x [DOI] [Google Scholar]
- 22.Kasper S, Wehr TA, Bartko JJ, Gaist PA, Rosenthal NE. Epidemiological findings of seasonal changes in mood and behavior. a telephone survey of Montgomery County, Maryland. Arch Gen Psychiatry. 1989;46(9):823-833. doi: 10.1001/archpsyc.1989.01810090065010 [DOI] [PubMed] [Google Scholar]
- 23.Magnusson A, Friis S, Opjordsmoen S. Internal consistency of the Seasonal Pattern Assessment Questionnaire (SPAQ). J Affect Disord. 1997;42(2-3):113-116. doi: 10.1016/S0165-0327(96)00104-8 [DOI] [PubMed] [Google Scholar]
- 24.Steinhausen HC, Gundelfinger R, Winkler Metzke C. Prevalence of self-reported seasonal affective disorders and the validity of the seasonal pattern assessment questionnaire in young adults findings from a Swiss community study. J Affect Disord. 2009;115(3):347-354. doi: 10.1016/j.jad.2008.09.016 [DOI] [PubMed] [Google Scholar]
- 25.Danilenko KV, Verevkin EG, Antyufeev VS, Wirz-Justice A, Cajochen C. The hockey-stick method to estimate evening dim light melatonin onset (DLMO) in humans. Chronobiol Int. 2014;31(3):349-355. doi: 10.3109/07420528.2013.855226 [DOI] [PubMed] [Google Scholar]
- 26.Kirschbaum-Lesch I, Holtmann M, Legenbauer T. Chronotherapy for Adolescent Major Depression. Neurotechnology and Brain Stimulation in Pediatric Psychiatric and Neurodevelopmental Disorders. Academic Press; 2019:313-334. [Google Scholar]
- 27.Feeney A, Hock RS, Fava M, Hernández Ortiz JM, Iovieno N, Papakostas GI. Antidepressants in children and adolescents with major depressive disorder and the influence of placebo response: a meta-analysis. J Affect Disord. 2022;305:55-64. doi: 10.1016/j.jad.2022.02.074 [DOI] [PubMed] [Google Scholar]
- 28.Rutherford BR, Wager TD, Roose SP. Expectancy and the treatment of depression: a review of experimental methodology and effects on patient outcome. Curr Psychiatry Rev. 2010;6(1):1-10. doi: 10.2174/157340010790596571 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Sit DK, McGowan J, Wiltrout C, et al. Adjunctive bright light therapy for bipolar depression: a randomized double-blind placebo-controlled trial. Am J Psychiatry. 2018;175(2):131-139. doi: 10.1176/appi.ajp.2017.16101200 [DOI] [PubMed] [Google Scholar]
- 30.Wirz-Justice A, Bader A, Frisch U, et al. A randomized, double-blind, placebo-controlled study of light therapy for antepartum depression. J Clin Psychiatry. 2011;72(7):986-993. doi: 10.4088/JCP.10m06188blu [DOI] [PubMed] [Google Scholar]
- 31.Sundberg I, Ramklint M, Stridsberg M, Papadopoulos FC, Ekselius L, Cunningham JL. Salivary melatonin in relation to depressive symptom severity in young adults. PLoS One. 2016;11(4):e0152814. doi: 10.1371/journal.pone.0152814 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Niederhofer H, von Klitzing K. Therapy augmentation by bright light treatment for non-seasonal depression of adolescents. Clin Neuropsychiatry. 2011;8(3):225-227. [Google Scholar]
- 33.Locher C, Koechlin H, Zion SR, et al. Efficacy and safety of selective serotonin reuptake inhibitors, serotonin-norepinephrine reuptake inhibitors, and placebo for common psychiatric disorders among children and adolescents: a systematic review and meta-analysis. JAMA Psychiatry. 2017;74(10):1011-1020. doi: 10.1001/jamapsychiatry.2017.2432 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Trial protocol
eAppendix
Data sharing statement