Efficacy of light therapy on nonseasonal depression among elderly adults: a systematic review and meta-analysis

Figures & data

Figure 1 Flowchart of the selection strategy and inclusion and exclusion criteria for this meta-analysis.

Notes: Database: PubMed (n=425), Cochrane Central Register of Controlled Trials (n=63), Cochrane Database of Systematic Reviews (n=5), ClinicalTrials.gov (n=17). (Phototherapy OR light therapy) AND (depress* OR mood) AND (old* OR elder* OR geriatric). Date: available until July 14, 2018.

Table 1 Summary of the characteristics of studies in the current meta-analysis

Study	Depression criteria/study population	Primary depression measure	Length of trial	Light therapy (intervention) group	N (% female) of intervention group	Mean age (SD) of intervention group	Control group	N (% female) of control group	Mean age (SD) of control group	Country	Study design
Sumaya et al, 2001^Citation15	Score of 11–20 on GDS/institutionalized older adults	GDS	5 days	Bright light 10,000 lux 30 min/day Morning	10 (60)	83.8 (9.56)	1 week (5 days) of 300 lux (placebo), or 1 week of no treatment (control)	10 (60)	83.8 (9.56)	USA	Placebo controlled, crossover
Tsai et al, 2004^Citation14	GDS ≥10; MDD or depressive disorders (DSM-IV)/inpatients	GDS	5 days	Bright light 5,000 lux 50 min/day Morning	30 (40)	75.3 (7.4)	Did not receive any treatment	30 (50)	74.6 (5.7)	Taiwan	RCT
Loving et al, 2005^Citation16	GDS score ≥11 (indicating probably major depression)/outpatients	GDS	4 weeks	Bright white light 8,500 lux, 1 hour Morning (n=13) Midday (n=15) Evening (n=13)	41 (58; total sample)	67.7 (5.45; total sample)	Red light 10 lux, 1 hour Morning (n=15) Midday (n=16) Evening (n=9)	40 (58; total sample)	67.7 (5.45; total sample)	USA	RCT
Loving et al, 2005^Citation17	GDS score of 11 (indicating probable major depression)/outpatients	GDS	4 weeks	1 hour of bright green light (1,200 lux)	17 (84.8; total sample)	67.7 (6.35; total sample)	1 hour of dim red light placebo (<10 lux)	16 (84.8; total sample)	67.7 (6.35; total sample)	USA	RCT
Paus et al, 2007^Citation63	Depression was measured by BDI/outpatients	BDI	15 days	White fluorescent light 7,500 lux, 30 minutes daily	18 (33.3)	63.6 (9.8)	White fluorescent light 950 lux, 30 minutes daily	18 (38.9)	63.4 (9.7)	Germany	RCT
Lieverse et al, 2008^Citation13	Depression was diagnosed using the Structured Clinical Interview for DSM-IV Axis I Disorders/outpatients	HAM-D	3 weeks	Bright light 1-hour early morning (pale blue, ~7,500 lux)	42 (67)	69.7 (8.5)	Placebo (dim red light, ~50 lux)	47 (64)	69.0 (6.6)	the Netherlands	RCT
Wu et al, 2015^Citation18	Depression was measured using the GDS-SF/long-term care facility	GDS-SF	4 weeks	10,000-lux light box 30 minutes in the morning	34 (47.1)	80.97 (9.84)	Routine care without light therapy	31 (38.7)	79.03 (10.06)	Taiwan	RCT
Canazei et al, 2017^Citation62	Depression was measured using ICD 10/inpatients	KUSTA	2 days	Artificial sunlight, mean vertical illuminance (at eye level), 1,700 lux, 30 minutes	21 (81.0)	70.1 (5.6; total sample)	Conventional room light, mean vertical illuminance (at eye level), 140 lux, 30 minutes	21 (81.0)	70.1 (5.6; total sample)	Austria	Crossover

Abbreviations: BDI, Beck’s Depression Inventory; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, fourth edition; GDS-SF, the short-form Geriatric Depression Scale; GDS, Geriatric Depression Scale; HAM-D, Hamilton rating scale; KUSTA, Kurz-Skala Stimmung/Aktivierung rating scale; MDD, major depressive disorder; ICD, International Classification of Diseases; RCT, randomized controlled trial.

Figure 2 Meta-analysis of studies comparing depression severity before and after light therapy in elderly adults.

Figure 3 Funnel plots for the effect sizes of light therapy.

Figure 4 Meta-regression of the effects of (A) female sex, (B) mean age, (C) treatment duration in days, and (D) light intensity in lux.

Figure 5 Subgroup meta-analyses of light colors.

Figure 6 Subgroup meta-analyses of comparators.

Figure 7 Subgroup meta-analyses of depression measures.

Abbreviations: BDI, Beck’s Depression Inventory; GDS, Geriatric Depression Scale; HAM-D, Hamilton rating scale; KUSTA, Kurz-Skala Stimmung/Aktivierung rating scale.

Figure 8 Subgroup meta-analyses of study designs.

Abbreviation: RCT, randomized controlled trial.

Figure 9 Subgroup meta-analyses of intervention lengths.

Figure 10 Subgroup meta-analyses of mean age ranges.

Figure 11 Subgroup meta-analyses of effects at 3-month follow-up after stopping light therapy.

Figure 12 Funnel plots for effect size of subgroup meta-analysis on (A) white light, (B) standard care, (C) GDS, (D) <1 week.

Abbreviation: GDS, Geriatric Depression Scale.

SumayaICRienziBMDeeganJFMossDEBright light treatment decreases depression in institutionalized older adults: a placebo-controlled crossover studyJ Gerontol A Biol Sci Med Sci2001566M356M36011382795

 PubMed Web of Science ®Google Scholar

TsaiYFWongTKJuangYYTsaiHHThe effects of light therapy on depressed eldersInt J Geriatr Psychiatry200419654554815211533

 PubMed Web of Science ®Google Scholar

LovingRTKripkeDFElliottJAKnickerbockerNCGrandnerMABright light treatment of depression for older adults [ISRCTN55452501]BMC Psychiatry200554116283925

 PubMed Web of Science ®Google Scholar

LovingRTKripkeDFKnickerbockerNCGrandnerMABright green light treatment of depression for older adults [ISRCTN69400161]BMC Psychiatry200554216283926

 PubMed Web of Science ®Google Scholar

PausSSchmitz-HübschTWüllnerUVogelAKlockgetherTAbeleMBright light therapy in Parkinson’s disease: a pilot studyMov Disord200722101495149817516492

 PubMed Web of Science ®Google Scholar

LieverseRNielenMMVeltmanDJBright light in elderly subjects with nonseasonal major depressive disorder: a double blind randomised clinical trial using early morning bright blue light comparing dim red light treatmentTrials200894818671864

 PubMed Web of Science ®Google Scholar

WuMCSungHCLeeWLSmithGDThe effects of light therapy on depression and sleep disruption in older adults in a long-term care facilityInt J Nurs Pract201521565365924750268

 PubMed Web of Science ®Google Scholar

CanazeiMPohlWBauernhoferKPsychophysiological Effects of a Single, Short, and Moderately Bright Room Light Exposure on Mildly Depressed Geriatric Inpatients: A Pilot StudyGerontology201763430831728103597

 PubMed Web of Science ®Google Scholar