Social jetlag in health and behavioral research: a systematic review

Figures & data

Figure 1 Systematic review process.

Abbreviations: SJL, social jetlag; LILACS, Latin American and Caribbean Health Science Literature.

Figure 2 Risk of bias assessment.

Notes: Questions 1 and 2 (Q1 and Q2) indicate seletion bias, Q3 indicates detection bias, Q4 indicates attrition bias, Q5 indicates reporting bias, and Q6 indicates any additional bias. Crosses represent low risk of bias; dashes, high risk of bias; and interrogation marks, an inability to determine bias due to lack of information.

Table 1 Summary of findings

Study	Population	Age range/mean age (SD)	Number of males/females	Geographic location	Sample size	Design	Instrument used to define SJL	Calculation of SJL	Calculation of midpoint of sleep	Outcomes related to SJL	Study results^a
Wittmann et al 2006^18	General population	14–94	190/303	Germany (midlatitude)	501	Cross-sectional	MCTQ	MSF–MSW	Midpoint between sleep onset and sleep end	Sleep quality, use of stimulants	Higher probability of smoking
Levandovski et al 2011^29	Rural population	44.1 (13.4)	1340/2711	Southern Brazil (subtropical)	4051	Cross-sectional	MCTQ	MSF–MSW	Midpoint between sleep onset and sleep end	Depression	Higher BDIas (BDI corrected for age and sex) scores when SJL >2 hours independent of smoking status; age group 31–40 years with mild to severe BDI scores suffered from higher SJL
Roenneberg et al 2012^21	General Internet-using public	16–65	NI	Primarily central Europe (midlatitude)	64,110	Cross-sectional	MCTQ	MSF–MSW	Midpoint between sleep onset and sleep end	Obesity	Higher probability of being overweight (BMI ≥25) and weight increase in the overweight group
Johnsen et al 2013^22	General population	30–65	3053/3356	Northern Norway (arctic)	6412	Cross-sectional	MCTQ	MSFsc-MSW	Midpoint between sleep onset and sleep end	Obesity	No association between SJL, BMI ≥25 kg/m^Citation2, waist-to-hip ratio and abdominal obesity when controlled for health, lifestyle and biological factors
Kantermann et al 2013^27	Shift workers of a steel factory	42 (7.6)	77/0	Belgium (midlatitude)	77	Cross-sectional	MCTQshift	MSF–MSW	Midpoint between sleep onset and sleep end	Atherosclerotic risk	Positive correlation with heart rate, but no association with arterial stiffness measured by pulse wave velocity
Randler and Vollmer 2013^43	Undergraduate students	23.8 (3.7)	320/112	Germany (midlatitude)	432	Cross-sectional	Open questions	MSF–MSW	Midpoint between sleep onset and sleep end	Aggression	Higher physical and verbal aggression scores measured by Buss–Perry Aggression Questionnaire
Schimitt et al 2013^35	Day shift workers	18–60	NI	Southern Brazil (subtropical)	143	Cross-sectional	MCTQ	MSF–MSW	Midpoint between sleep onset and sleep end	Social rhythm and minor psychiatric symptoms	Inverse correlation with Index of Regularity of activities in bivariate, but not in multivariate analysis; no associations with SRQ-20
de Souza and Hidalgo, 2015^12	Adolescents	14.7 (1.86)	104/247	Southern Brazil (subtropical)	351	Cross-sectional	MCTQ	MSF–MSW	Midpoint between sleep onset and sleep end	Depression	No association with depressive symptoms assessed by BDI scores
Haraszti et al 2014^40	Undergraduate students	21.23 (3.12)	241/512	Hungary (midlatitude)	753	Cohort	MCTQ	MSF–MSW	Midpoint between sleep onset and sleep end	Academic performance	Poor academic grades during lecture term and better performance during exams period (no weekly fixed obligations)
Kantermann et al 2014^25	Male shift workers	43.3 (6.8)	43/0	Belgium (midlatitude)	43	Cross-sectional	MCTQshift	MSF–MSW	Midpoint between sleep onset and sleep end	Metabolic risk	No associations with blood glucose, insulin, apolipoprotein-A and -B, HDL and LDL, total cholesterol, triglycerides, minimally oxidized LDL, C-reactive protein, IL-8 and 25-OH D
Miller et al 2015^44	General population	42.89 (7.33)	197/211	Northern USA (midlatitude)	408	Cross-sectional	Actigraphy	MSF–MSW	Midpoint between sleep onset and sleep end	Mediation in the relationship between positive affect rhythms and chronotype	No significance found in correlations of acrophase and amplitude of PANAS with CSM
Rutters et al 2014^28	University students and employees	Male: 28.6 (9.4) Female: 26.9 (8.5)	67/78	The Netherlands (midlatitude)	145	Cross-sectional	Adapted MCTQ	MSF–MSW	Midpoint between sleep onset and sleep end	Adverse endocrine and cardiovascular risk profile	Subjects with ≥2 hours SJL had higher 5 hour cortisol levels, were more often physically inactive and had higher resting heart rate, compared with those who had ≤1 hour SJL; no differences in BMI, waist circumference and blood pressure
Borisenkov et al 2015^32	High school students	14.8 (2.6)	1517/1918	Northern European Russia (subarctic and arctic)	3435	Cross-sectional	MCTQ	MSF–MSW	Midpoint between sleep onset and sleep end	Winter seasonal affective disorder	Higher winter scores of SPAQ in women
Díaz-Morales and Escribano 2015^39	High school students	14.1 (1.48)	371/425	Spain (midlatitude)	796	Cross-sectional	School Sleep Habits Survey	MSF–MSW	Midpoint between bedtime and rising time	Cognitive performance and academic achievement	Lower general cognitive ability; lower scores on Thurstone’s primary mental abilities (except for verbal fluency); lower GPA. Greater effects on women
Jankowski 2015^34	Undergraduate students	22.2 (1.9)	16/101	Poland (midlatitude)	117	Cohort	Open questions	MSF–MSW	Midpoint between sleep onset and sleep end	Psychological well-being	Positive association with nervous mood, assessed by TA subscale of UMACL
Lin and Yi 2015^37	Junior-high school students	Wave 2: 14.3 (0.48) Wave 3: 15.3 (0.47) Wave 4: 16.3 (0.47)	1368/1315	Taiwan (subtropical)	2683	Cohort	Open questions	Weekend bedtime–weekday bedtime	SJL calculation did not use midpoint of sleep	Conduct problems	Poor academic performance and defiant attitude assessed by open questions
Parsons et al 2015^23	General population	38	NI	New Zealand (midlatitude)	815	Cross-sectional	MCTQ	MSF–MSW	Midpoint between sleep onset and sleep end	Obesity and metabolic disorder	Higher BMI and fat mass; higher probability of being obese and meet the criteria for metabolic syndrome
Tavernier et al 2015^42	Undergraduate students	19.01 (0.9)	268/674	Southwestern Canada (midlatitude)	942	Cohort	Adapted MCTQ	MSF–MSW	Midpoint between bedtime and wake time	Academic adjustment and substance use	No predictive effect on substance use or academic adjustment assessed by a subscale of the SACQ; greater substance use was a significant predictor of SJL
Wong et al 2015^26	General population	42.7 (7.4)	210/237	Northern USA (midlatitude)	447	Cross-sectional	Actigraphy	MSF–MSW	Midpoint between sleep onset and sleep end	Cardiometabolic risk	Higher triglycerides, fasting insulin, insulin resistance estimated by the HOMA-IR, waist circumference and BMI; lower HDL
Choi et al 2016^45	Epileptic and healthy control participants	Epilepsy: 33.0 (8.24) Control: 33.6 (8.80)	155/135	South Korea (midlatitude)	290	Cross-sectional	Sleep diary	MSF–MSW	Midpoint between sleep onset and sleep end	Prevalence in epileptic compared to healthy control subjects and among epilepsy subtypes	Larger SJL in healthy controls and in general epilepsy compared to focal epilepsy
Díaz-Morales et al 2016^36	High school students	13.95 (1.69)	690/716	Spain (midlatitude)	1406	Cross-sectional	Open questions	MSF–MSW	Midpoint between bedtime and rising time	Anxiety	No correlation found with the STAI
Kolomeichuk et al 2016^38	Schoolchildren and college students	10–18	808/858	Northern European Russia (subarctic)	1666	Cross-sectional	MCTQ	MSFsc-MSW	Midpoint between sleep onset and sleep end	Academic achievement	No influence in academic achievement (assessed by an open question) in a linear regression model together with average sleep duration, chronotype, age and sex
Polugrudov et al 2016^24	General population	22 (2)	27/35	Northern Asian Russia (subarctic)	62	Cross-sectional	MCTQ	MSF–MSW	Midpoint between sleep onset and sleep end	Circadian system state, depression, anxiety, anger, obesity and cognitive function	Higher incidence of depression measured by BDI, higher wrist temperature at night and lower during the day, lower wrist temperature rhythm amplitude and a tendency for increase in CAR; no association with BMI, waist circumference, Spielberger Anxiety Inventory and Clinical Anger Scale
Sheaves et al 2016^30	Undergraduate students	21 (20–23)^c	612/780	Southern UK (midlatitude)	1403	Cross-sectional	MCTQ	MSF–MSW	Midpoint between sleep onset and sleep end	Hallucinations, paranoia, anxiety, depression, mania and hypomania	No associations found for the following outcomes: hallucinations or paranoia assessed by subscales from the SPEQ; depression or anxiety, assessed by two subscales from the DASS-21; mania or hypomania assessed by the Mood Disorder Questionnaire
Silva et al 2016^46	Undergraduate students	21.6 (3.9)	92/112	Central Brazil (tropical)	204	Cross-sectional	Open questions	MSF–MSW	Midpoint between sleep onset and sleep end	Inadequate diet	Negative association with servings per day of beans assessed by the FFQ
Yong et al 2016^41	Day and shift workers (midlatitude)	42 (10)	1831/643	Germany	2474	Cross-sectional	Modified version of MCTQ	MSF–MSW^b	Midpoint between sleep onset and sleep end	Work ability	Poor WAI when SJL was associated with short sleep duration

Notes:

a Only outcomes related to SJL are described.

b In the case of shift workers, midsleep on day shifts, night shifts and free days were considered.

c Median age (IQR).

Abbreviations: 25-OH D, 25-hydroxyvitamin D; BDI, Beck Depression Inventory; BMI, body mass index; CAR, cortisol awakening response; CSM, composite scale of morningness; DASS-21, Depression, Anxiety and Stress Scales; FFQ, Food Frequency Questionnaire; GPA, grade point average; HDL, high-density lipoprotein; HOMA-IR, homeostatic model assessment of insulin resistance; IL-8, interleukin-8; IQR, interquartile range; LDL, low-density lipoprotein; MCTQ, Munich ChronoType Questionnaire; MSF, midpoint of sleep on free days; MSFsc, midpoint of sleep on free days corrected for sleep debt; MSW, midpoint of sleep on work days; NI, not informed; PANAS, positive affect negative affect schedule form; SACQ, Student Adaptation to College Questionnaire; SD, standard deviation; SJL, social jetlag; SPAQ, Seasonal Pattern Assessment Questionnaire; SPEQ, Specific Psychotic Experiences Questionnaire; SRQ-20, Self-Report Questionnaire-20; STAI, state–trait anxiety inventory; TA, tense arousal; UMACL, University of Wales Institute of Science and Technology mood adjective checklist; WAI, work ability index.

Figure 3 Distribution of studies across latitude zones.

Zhang J, Kaasik, K, Blackburn MR, Lee CC. Constant darkness is a circadian metabolic signal in mammals. Nature. 2006;439(7074):340–343.

 PubMed Web of Science ®Google Scholar