Physical exercise as a therapeutic approach for adults with insomnia: systematic review and meta-analysis

Corresponding author: Carolina Vicaria Rodrigues D’Aurea Avenida Albert Einstein, 627/701 - Morumbi Zip code: 05652-900 - São Paulo, SP, Brazil Phone: (55 11) 2151-1233 E-mail: fisiocarol@uol.com.br

Conflict of interest: none.

Authors statement

The final version of this manuscript has been read and approved by all authors, who have each contributed to ensure the integrity of this work.

AUTHORS’ CONTRIBUTION

Carolina Vicaria Rodrigues D’Aurea: term definition, conceptualization, methodology, formal analysis, investigation, writing, writing - review & editing, visualization, preparation, creation and/or presentation, project administration. Cristina Frange: writing - review & editing, visualization, preparation, creation and/or presentation, project administration. Dalva Poyares: term definition, writing - review & editing. Altay Alves Lino de Souza: formal analysis, writing - review & editing. Mario Lenza: term definition, methodology, formal analysis, writing - review & editing, supervision, project administration.

Studies	n	Mean age (±SD) or range	Insomnia classification	Intervention(s) and controls/duration	Outcomes
Passos et al. ⁽¹⁵⁾ Brazil	48	Inactive adults Both sexes 44.4 (±8.0)	Clinical diagnosis of insomnia (DSM-IV + ICSD-2)	4 groups: G1 = moderate-intensity aerobic exercise (50 minutes, 1 session/week) G2 = high-intensity aerobic exercise (50 minutes, 1 session/week) G3 = moderate-intensity resistance exercise (50 minutes, 1 session/week) Ctrl = no treatment	PSG: TST, SE, SOL, WASO Sleep diary: TST, SE, SOL, WASO
Reid et al. ⁽³⁶⁾ United States of America	17	Sedentary adults Both sexes 61.6 (± 4.3)	Clinical diagnosis of insomnia (ACT + 7 day sleep diary)	2 groups: G1 = moderate aerobic physical activity+ sleep hygiene (increasing from 10 to 40 minutes each week, 4 sessions/week) Ctrl = sleep hygiene 4 months	PSQI Epworth Sleepiness Scale SF-36
Afonso et al. ⁽³³⁾ Brazil	44	Postmenopausal women 50-65	Clinical diagnosis of insomnia	3 groups: G1 = yoga (1 hour, 2 sessions/week) G2 = passive stretching (1 hour, 1 session/week) Ctrl = no treatment 4 months	PSG: TST, SE, SOL, WASO ISI MENQOL
Irwin et al. ⁽³⁷⁾ United States of America	123	Older adults Both sexes 55-85	Clinical diagnosis of insomnia (DSM-IV + ICSD-2)	3 groups: G1 = CBT (2 hours, 2 sessions/week) G2 =Tai Chi (2 hours, 2 sessions/week) Ctrl = sleep seminar education + sleep hygiene (2 hours, 2 sessions/week) 4 months	PSG: TST, SE, SOL, WASO Sleep diary: TST, SE, SOL, WASO PSQI Epworth Sleepiness Scale AIS
Hartescu et al. ⁽³⁵⁾ United Kingdom	41	Inactive adults Both sexes 59.80 (±9.5)	Clinical diagnosis of insomnia (Research Diagnostic Criteria)	2 groups: G1 = moderate to vigorous physical activity (≥150 minutes, 1 session/week) Ctrl = no treatment 6 months	Epworth Sleepiness Scale ISI EQ5D-5L
D’Aurea et al. ⁽³⁴⁾ Brazil	28	Inactive adults Both sexes 30-55	Clinical diagnosis of insomnia	3 groups: G1 = resistance exercise (1 hour, 3 sessions/week) G2 = low-intensity stretching (1 hour, 3 sessions/week) Ctrl = no treatment 4 months	PSG: TST, SE, SOL, WASO ACT: TST, SE, SOL, WASO PSQI ISI SF-36

Effects of physical exercise in insomnia and control patients
Patient or population: adults with insomnia Intervention: physical exercise Comparison: Control Group (no activity)
Outcomes	Illustrative comparative risks* (95%CI)		Number of participants (studies)	Quality of evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control Group	Physical Exercise
PSG - TST; PSG is the gold standard for analysis of physiological variables affecting sleep. Follow-up: 4 months	Mean TST in Control Groups was 341.3 minutes	SMD 0.09 lower (0.45 lower to 0.28 higher)	121 participants (3 studies)	⊕⊕⊝⊝ ^1,2,3 Low	Primary outcome, difference was not statistically significant
PSG - SOL Follow-up: 4 months	Mean SOL in Control Groups was 21.06 minutes	MD 0.8 higher (6.13 lower to 7.73 higher)	145 participants (4 studies)	⊕⊝⊝⊝ ^2,3,4 Very low	Primary outcome, difference was not statistically significant
PSG - SE Follow-up: 4 months	Mean SE in Control Groups was 79.02%	MD 0.9 lower (4.52 lower to 2.73 higher)	145 participants (4 studies)	⊕⊕⊝⊝ ^2,4 Low	Primary outcome, difference was not statistically significant
PSG - WASO Follow-up: 4 months	Mean WASO in Control Groups was 66.03 minutes	MD 0.79 higher (13.95 lower to 15.52 higher)	145 participants (4 studies)	⊕⊝⊝⊝ ^2,3,4 Very low	Primary outcome, difference was not statistically significant
Sleep questionnaire – Insomnia Insomnia Severity Index (0 to 28 points, 0 is best). Follow-up: 4 months	Mean score in sleep questionnaire - insomnia in Control Groups was 9.36	SMD 0.72 lower (1.1 lower to 0.34 lower)	121 participants (3 studies)	⊕⊕⊕⊕ ^1,2 High	Primary outcome, difference was statistically significant
Sleep questionnaire - Pittsburgh Sleep Questionnaire Index (0 to 21 points, >5 indicates good sleep quality). Follow-up: 4 months	Mean score in sleep questionnaire - PSQI in Control Groups was 6.83	MD 3.17 lower (4.23 lower to 2.12 lower)	108 participants (3 studies)	⊕⊝⊝⊝ ^1,2,3,5 Very low	Primary outcome, difference was statistically significant
Sleep diary - TST Follow-up: 4 months	Mean TST according to sleep diary in Control Groups was 334.25 minutes	MD 47.72 higher (46.92 higher to 48.52 higher)	97 participants (2 studies)	⊕⊕⊕⊝ ^1,2,5 Moderate	Secondary outcome, difference was statistically significant

Outcomes	Number of studies	Risk of bias	Indirect evidence	Inaccuracy	Inconsistency	Number of participants	Relative effect (95%CI)	Quality of evidence (GRADE)	Clinical outcome, significance
TST minutes (PSG)	3	Severe ¹	Severe ²	Severe ³	Not severe	121	SMD: 0.95 (-0.46-2.36)	^⨁◯◯◯ Very low	Primary outcome, difference was not statistically significant
SOL minutes (PSG)	4	Very severe ⁴	Severe ²	Severe ³	Not severe	145	MD: 0.8 (-6.13-7.73)	^⨁⨁◯◯ Low	Primary outcome, difference was not statistically significant
SE % (PSG)	4	Very severe ⁴	Severe ²	Not severe	Not severe	145	MD: -0.9 (-4.52-2.73)	^⨁◯◯◯ Very low	Primary outcome, difference was not statistically significant
WASO minutes (PSG)	4	Very severe ⁴	Severe ²	severe ³	Not severe	145	MD: - 0.79 (-13.95-15.52)	^⨁⨁◯◯ Low	Primary outcome, difference was not statistically significant
Severity of insomnia	3	Severe ¹	Severe ²	Not severe	Not severe	121	SMD: -0.72 (-1.1, -0.34)	^⨁⨁⨁⨁ High	Primary outcome, difference was statistically significant
Sleep quality (PSQI)	3	Severe ¹	Severe ²	Severe ³	Very severe ⁵	108	MD: -3.17 (-4.23, -2.12)	^⨁◯◯◯ Very low	Primary outcome, difference was statistically significant

[1] Corresponding author: Carolina Vicaria Rodrigues D’Aurea Avenida Albert Einstein, 627/701 - Morumbi Zip code: 05652-900 - São Paulo, SP, Brazil Phone: (55 11) 2151-1233 E-mail: fisiocarol@uol.com.br

Brasil

Brasil

Physical exercise as a therapeutic approach for adults with insomnia: systematic review and meta-analysis

ABSTRACT

Objective

Methods

Results

Conclusion