Literature searching yielded 12,653 articles after duplicates were removed (Figure 1), which were screened by title. After further screening of abstracts (n=72), and full papers (n=51), a total of 26 studies (reported in 24 papers – note: Anderson et al. (1996) included three studies presented within the one paper [33]) were included in the review. Characteristics of included studies are summarised in Table 1. Sixteen studies employed a cross-sectional design [33-48] (n=72 to 34,129 participants). Four studies were longitudinal [33, 49-51] (n=271 to 11,676), two were controlled clinical trials [52, 53] (n=43-231), two were pilot interventions [54, 55] (n=12 to 20) and two were direct observation studies (i.e. where time in sedentary behaviour was collected for one year or one month and averaged) [33, 56] (n=79 to 140). The majority of these studies were conducted in Australia (n=8) and the US (n=8). 21 of the 26 study samples were comprised of men and women (age 18-98y) [33-38, 40-42, 45-49, 52-56], while the remaining five studies were conducted among women only (age 18-65y) [39, 43, 44, 50, 51].
Objective measures of stress were used in n=7 studies, and included salivary [37, 52, 54, 55] or hair [44, 47, 48] cortisol, blood pressure (systolic and diastolic) [52] and heart rate [52]. Stress was self-reported using measures including the Life Events Inventory (LEI, n=3) [33], the Cohen Perceived Stress Scale (PSS, n=8) [35, 38-40, 43, 48-50], the Depression, Anxiety and Stress Scale (DASS-21, n=1) [41], the Perceived Stress Questionnaire for Young Women (PSQYW, n=1) [51], the affective experience component of the Princeton Affect and Time Survey (n=1) [36], the Daily Stress Inventory (DSI, n=1) [56], the Chronic Burden Scale (CBS, n=1) [45], the Traumatic Stress Schedule (TSS, n=1) [45], job stressors (n=1) [42], the Health and work Questionnaire (n=1) [53] and the Effort Reward Imbalance Scale (ERIS, n=1) [46]. One study measured stress using self-report survey items designed specifically for that study [34].
Ten studies measured sedentary behaviour objectively (n=7 utilised accelerometers [37, 45, 48, 49, 52, 53, 56]; n=2 directly observed participants’ sitting time in a laboratory setting [54, 55]; n=1 objectively measured TV viewing time via time-lapse video recordings [33]). Self-report measures of sedentary behaviour were used in 16 studies. Types of sedentary behaviour that were self-reported included: total daily and/or weekly sitting [34, 35, 38, 40-42, 44-46, 50, 51]; TV viewing time [33, 36, 39, 41, 43, 44, 47, 50]; occupational sitting time [41]; computer use [41, 44]; transport-related sitting time [41]. Reliability and validity of all the above measures of sedentary behaviour and stress are reported in Supplemental Table 2.
Methodological quality
Table 1 presents methodological quality scores. Overall, two studies were rated as strong [50, 51], 10 studies [34, 35, 37, 41, 43, 44, 47, 48, 52, 53] were rated as moderate and 14 studies [33, 36, 38-40, 42, 45, 46, 49, 54-56] received a weak methodological quality rating. Studies received weak or moderate ratings primarily due to study design limitations (e.g. 16 of 26 studies employed a cross-sectional study design [33-48]), participants were not likely to be representative of target population (8 of 26 studies) [38-40, 42, 52, 54-56], or the studies did not employ/report reliable and/or valid measures for sedentary behaviour (11 of 26 studies) [33-36, 38, 40, 42, 51, 54] (see Supplemental Table 2).
Table 1 Characteristics of studies investigating associations between sedentary behaviour and stress (alphabetically ordered)
Authors (date) and country
|
Study design & sample
|
Sedentary behaviour indicator
|
Stress indicator
|
Association
(and direction)
|
Consistency in findings i.e. ≥75% of results in same direction
|
Methodological quality
score
|
An, Jang and Kim (2015) [34]
Republic of Korea
|
Cross-sectional
n=4,674
Age: ≥20y
58.5% women
|
Self-report:
Total daily sitting (including sitting at work, home, studying and during leisure time); Korean version of IPAQ-SV.
|
Self-report:
Stress symptoms: “Have you felt sad/desperate to the extent it disturbed daily life for more than two weeks during the past year, so much that it disturbed your daily life?”
|
Total sitting = +
|
Sitting = +
|
Moderate
|
Anderson, et al. (1996) [33]
US
|
Study A
Cross sectional
n=491
Age: 18-88y
54.8% women
|
Self-report:
Total weekly TV viewing.
|
Self-report:
51-item Life Events Inventory.
|
TV = 0
|
TV = 0
|
Weak
|
Anderson, et al. (1996) [33]
US
|
Study B
Longitudinal
n=651
Age: men=34.8y, women= 32.8y
50.5% women
|
Self-report:
Weekly TV viewing (calculated from two 10-day TV viewing diaries, recorded one month apart).
|
Self-report:
51-item Life Events Inventory.
|
TV (women) = +
TV (men) = 0
|
TV = +/0
|
Weak
|
Anderson, et al. (1996) [33]
US
|
Study C
Direct observation and longitudinal (survey)
n=140
Age: mid 20s to late 30s
50.7% women
|
Objective:
Time-lapse video of TV and TV viewing areas recorded for 10 continuous days to ascertain when participant was looking at the TV.
|
Self-report:
51-item Life Events Inventory.
|
TV (men) = +
TV (women) = 0
|
TV = +/0
|
Weak
|
Ashdown-Franks, et al. (2018) [35]
China, Ghana, India, Mexico, Russia and
South Africa
|
Cross-sectional
n=34,129
Age: ≥50y (mean±SD = 62.4y±16.0)
55.0% women
|
Self-report:
Total mins/day spent sitting/ reclining.
|
Self-report:
Two items of the Cohen Perceived Stress Scale.
|
Total sitting (adjusted) = +
Total sitting (50-64y) = +
Total sitting (≥65y) = +
|
Sitting = +
|
Moderate
|
Depp, et al. (2010) [36]
US
|
Cross-sectional
n=3,982
Age: 15-98y (mean±SD = 51.4±18.0)
61.0% women
|
Self-report:
15 minute ‘episodes’ engaged in TV viewing, as defined in the American Time-Use Survey lexicon.
|
Self-report:
Affective experience:
‘Feeling stressed’ (from the Princeton Affect and Time Survey).
|
TV = –
|
TV = –
|
Weak
|
Diaz, et al. (2018) [56]
US
|
Direct observation
n=79
Age: mean±SD = 31.9y±9.5
14.1% women
|
Objective: Accelerometry. Total sitting time (min/day), and mean sitting bout duration (min/bout).
|
Self-report:
Participants recorded momentary stress (prompted randomly three times/day), and end-of-day (prompted in evening, once/day). Prompts based on Daily Stress Inventory (including work, argument, traffic, deadlines, bills, running late, or other).
|
Total sitting time:
Work = 0
Argument = +
Traffic jam = 0
Deadlines = 0
Bills = 0
Running late = –
Other stress = 0
|
Sitting = 0
|
Weak
|
Ellingson, et al. (2018) [49]
US
|
Longitudinal
n=271
Age: mean±SD = 27.8y±3.7
49.0% women
|
Objective: Accelerometry. Total sedentary time (hrs/day); and low (<10.5 hr), medium (10.5–12 hr), and high (>12 hr) sub-groups. Sedentary bout duration (< 30 min, ≥30 min).
|
Self-report:
Ten-item Cohen Perceived
Stress Scale.
|
Baseline:
Total sitting = 0
Stratified by sitting sub-group = 0
Sitting bout duration <30 min = 0
Sitting bout duration ≥30 min = 0
Total sitting Δ over time = +
(as sedentary behaviour increased over time, stress increased)
Total sitting Δ over time (stratified by baseline sedentary behaviour sub-group):
Low = +
Medium = +
High = +
(as sedentary behaviour increased over time, stress increased)
Sitting bout duration Δ over time:
<30 min = 0
≥30 min = +
|
Sitting = +/0
|
Weak
|
Endrighi, Steptoe and Hamer (2016) [52]
UK
|
Intervention
n=43
Age mean±SD: 23.86y±4.71 (men), 25.73y±0.13 (women)
44.2% women
|
Objective: Accelerometry. Change in total sedentary time (min/day) between treatment (sedentary) and control (usual behaviour) conditions.
|
Objective:
Systolic and diastolic blood pressure, heart rate, salivary cortisol
|
Total sitting (systolic and diastolic blood pressure) = 0
Total sitting (heart rate) = 0
Total sitting (salivary cortisol) = 0
|
Sitting = 0
|
Moderate
|
Gilson, et al. (2017) [54]
Australia
|
Pilot intervention
n=20
Age: mean±SD = 37.9±11.6
55.0% women
|
Objective:
Observed three 1.5 hour work periods per treatment group: 1) usual chair and desk use, 2) sit–stand desk, and 3) treadmill desk.
|
Objective:
Salivary cortisol.
|
Occupational sitting (usual chair and desk use) = +
|
Sitting = +
|
Weak
|
Gubelmann, et al. (2018) [37]
Switzerland
|
Cross-sectional
n=1,948
Age: 45-86y
55.0% women
|
Objective:
Accelerometry. Mean time (%) spent sitting (all days). Mean sitting stratified into tertiles, classified as ‘high sedentary behaviour’ if in the highest tertile, and as ‘low sedentary behaviour’ if in the remaining tertiles.
|
Objective:
Salivary cortisol measured at T1 (waking), T2 (30 min after T1), T3 (11:00 am) and T4 (8:00 pm). Mean cortisol and diurnal cortisol slope (Steeper decline tends to be associated with more favourable health outcomes [63]; T4-T1 cortisol divided by number of hours separating T1-T4).
|
Low vs high mean sitting:
Mean cortisol:
Adjusted = 0
Awakening cortisol:
Adjusted = 0
Diurnal cortisol slope:
Adjusted = +
|
Sitting = 0/+
|
Moderate
|
Jackson, et al. (2019) [47]
UK
|
Cross-sectional
n=3,555
Age: ≥50y (mean±SD = 68.34±7.86)
66.6% women
|
Self-report:
Mean daily hours TV viewing (combined weekdays and weekend days); < 2 h/day; 2 to<4 h/day; 4 to<6 h/day; ≥6 h/day.
|
Objective:
Hair cortisol.
|
TV (adjusted) = 0
|
TV = 0
|
Moderate
|
Lee and Kim (2018) [38]
Republic of Korea
|
Cross-sectional
n=244
Age: University students (age not reported)
80.0% women
|
Self-report:
Mean hours/day spent engaged in activities that do not increase energy expenditure above resting, i.e. ~1.0-1.5 METs (total, week and weekend days).
|
Self-report:
Ten-item Cohen Perceived
Stress Scale.
|
Total sitting (adjusted) = +
Week day sitting (adjusted) = +
Weekend day sitting (adjusted) = 0
|
Sitting = +/0
|
Weak
|
Mouchacca, Abbott and Ball (2013) [50]
Australia
|
Longitudinal
n=1,382
Age: 18-46y (mean±SD = 35.7±7.7)
100.0% women
|
Self-report (T1 and T2):
Total weekly hours spent sitting (IPAQ-L) and total weekly TV viewing.
|
Self-report (T1):
Four-item Perceived Stress Scale.
|
TV (baseline) = 0
TV (at follow-up) = +
Total weekly sitting (baseline) = 0
Total weekly sitting (follow-up) = 0
|
TV = 0/+
Sitting = 0
* Overall SB = 0
|
Strong
|
Pavić and Rijavec (2013) [39]
Croatia
|
Cross-sectional
n=216
Age: 18-45y (mean±SD = 26.51±7.63)
100.0% women
|
Self-report:
Total weekly TV viewing.
|
Self-report:
10-item version of the Cohen Perceived Stress Scale and sub-scales ‘Negative emotions’ and ‘Lack of control’.
|
TV (overall stress) = +
TV (‘Negative emotions’) = +
TV (‘Lack of control’) = +
|
TV = +
|
Weak
|
Pelletier, Lytle, and Laska (2016) [40]
US
|
Cross-sectional
n=441
Age: <21y (50.6%), ≥21 (49.4%)
67.6% women
|
Self-report:
Total daily sitting and reclining (WHO Global Physical Activity Questionnaire).
|
Self-report:
Four-item Cohen Perceived Stress Scale.
|
Total daily sitting/reclining = 0
|
Sitting = 0
|
Weak
|
Peterman, et al. (2019) [53]
Australia
|
Intervention
n=231
Age: 18-65y (mean±SD = 45.6±9.4)
68.0% women
|
Objective:
Accelerometry. Change in mean min spent sitting/8-hour workday between treatment (reduced workplace sitting) and control (usual working conditions).
|
Self-report:
Single stress item from the Health and Work Questionnaire (HWQ)
|
Occupational sitting = 0
|
Sitting = 0
|
Moderate
|
Rebar, et al. (2016) [41]
Australia
|
Cross-sectional
n=1,104
Age: mean=58y
55.0% women
|
Self-report:
Daily sitting in the following contexts: leisure, occupation, computer use, TV, and transport; and overall sitting time (10-item Workforce Sitting Questionnaire).
|
Self-report:
Depression, Anxiety, and Stress Scale (DASS-21).
|
Transport sitting = +
Leisure = 0
Occupational sitting = 0
Overall sitting time = 0
Computer = 0
TV = 0
|
Sitting = 0
Computer = 0
TV = 0
* Overall SB = 0
|
Moderate
|
Ryde, et al. (2019) [48]
UK
|
Cross-sectional
n=77
Age: mean±SD = 40.8±9.7
78.0% women
|
Objective:
Accelerometry (mean min/day)
|
Objective:
Hair cortisol.
Self-report:
Ten-item Cohen Perceived
Stress Scale.
|
Occupational sitting (objective stress, adjusted) = 0
Occupational sitting (self-report stress, adjusted) = 0
|
Sitting = 0
|
Moderate
|
Sonnentag and Jelden (2009) [42]
Germany
|
Cross-sectional
n=78
Age: mean±SD = 43.8y±7.7
14.1% women
|
Self-report:
Overall time daily spent engaged in ‘low effort’ activities (e.g. watching TV, reading newspaper, doing nothing).
|
Self-report:
Job stressors: ‘time pressure’, ‘role ambiguity’ and ‘situational constraints’ (from shortened job stressor scales).
|
Total sitting (‘situational constraints’) = +
Total sitting (‘time pressure’) = 0
Total sitting (‘role ambiguity’) = 0
|
Sitting = +/0
|
Weak
|
Sperlich, et al. (2018) [55]
Germany
|
Pilot intervention
n=12
Age: mean±SD = 22.0y±2.0
58.0% women
|
Objective:
Researchers observed participants completing a control or treatment routine.
Treatment routine included:
T0: Resting lying down for 30 min
T1: Consumed breakfast
T2: Sitting for one hour
T3: Six min HIIT session
T4-T7: Sitting for two hours (T4 = 30-, T5 = 60-, T6 = 90-, and T7 = 120-min after HIIT session).
Control routine included:
T0 and T1 described above, followed by 186 min of sitting.
|
Objective:
Salivary cortisol. Samples collected at T0, T2, T3, T4, T5, T6 and T7.
|
Compared to baseline (T0) measurement (results presented for control group only):
T2:
Sitting = –
T3:
Sitting = –
T4:
Sitting = –
T5:
Sitting = –
T6:
Sitting = –
T7:
Sitting = –
|
Sitting = –
|
Weak
|
Teychenne, Ball and Salmon (2012) [43]
Australia
|
Cross-sectional
n=1,554
Age: 18-65y (mean±SD = 42.0y±12.78)
100.0% women
|
Self-report:
Total weekly TV viewing.
|
Self-report:
Four-item Cohen Perceived
Stress Scale.
|
TV = 0
|
TV = 0
|
Moderate
|
Teychenne, et al. (2018) [44]
Australia
|
Cross-sectional
n=72
Age:18-46y (mean±SD = 43.5y±7.1)
100.0% women
|
Self-report:
Weekly hours engaged in TV viewing, computer use and overall sitting time.
|
Objective:
Hair cortisol.
|
TV = 0
Computer = 0
Sitting = 0
|
TV = 0
Computer = 0
Sitting = 0
* Overall SB = 0
|
Moderate
|
Uijtdewilligen, et al. (2014) [51]
Australia
|
Longitudinal
n=11,676
Age (mean±SD): 2000 – 24.6y±1.5; 2003 – 27.6y±1.5; 2006 – 30.6y±1.5; 2009 – 33.7y±1.5
100.0% women
|
Self-report:
Total daily hours sitting on weekdays, and on weekend days.
|
Self-report:
Perceived Stress Questionnaire for Young Women.
|
Weekday sitting (multivariate) = +
Weekend day sitting (multivariate) = +
|
Sitting = +
|
Strong
|
Vasquez, et al. (2016) [45]
US
|
Cross-sectional
n=4,244
Age: 18-74y
62.0% women
|
Objective:
Accelerometry. Mean min/day.
Self-report: Global Physical Activity Questionnaire
|
Self-report:
Chronic stress (8-item Chronic Burden scale) and traumatic stress (10-item Traumatic Stress Schedule).
|
Objective sitting:
Chronic stress:
Model 3 = +
Traumatic stress:
Model 3 = +
Self-report sitting/reclining:
Chronic stress (age and field-centre adjusted) = +
Traumatic stress (age and field-centre adjusted) = +
|
Sitting = +
|
Weak
|
Wang, et al. (2018) [46]
Australia
|
Cross-sectional
n=1,481
Age: 31-41y (mean±SD: men, 36.8±2.5; women, 36.5±2.6)
58.0% women
|
Self-report:
Mean week day and weekend day sitting time (min/day; IPAQ-L).
|
Self-report:
Effort Reward Imbalance 17-item scale.
|
Self-report sitting
Week day sitting:
Model 3 (men) = 0
Model 3 (women) = 0
Weekend day sitting:
Model 3 (men) = +
Model 3 (women) = 0
|
Sitting = 0
|
Weak
|
TV – television; + = increasing sedentary behaviour is associated with increased stress, – = increasing sedentary behaviour is associated with decreased stress, 0 = no association.
*Overall SB = composite direction/score for sedentary behaviour, whereby 75% of within-study results reflect an overall direction
Results are presented as a whole sample (i.e. combined findings) first, then for specific sedentary behaviours (i.e. overall sitting time, TV viewing, computer use), and finally for objectively versus subjectively assessed sedentary behaviour and stress.
Combined findings
Unadjusted results are reported here first, i.e. not adjusting for publication bias, nor for multiple modelling within studies. Overall, across the 26 studies, there were 78 models that assessed the association between sedentary behaviour (any indicator) and stress. This resulted in 29 positive associations (37%; i.e. higher sedentary behaviour associated with higher levels of stress); eight inverse associations (10%; i.e. higher sedentary behaviour associated with lower levels of stress) and 41 null results (53%; i.e. no association between sedentary behaviour and stress).
Figure 2 provides a graphical representation (harvest plots) of the overall evidence. Specifically, of 26 studies reviewed, six (n=1 strong quality [51]; n=2 moderate-quality [34, 35]; n=3 methodologically weak-quality [39, 45, 54]) found a positive association between time spent in any sedentary behaviour and stress (i.e. increased time spent in sedentary behaviour was associated with increased stress). Five of these studies used self-report measures of stress [34, 35, 39, 45, 51], while one study [54] examined salivary cortisol. Eleven studies showed ‘mixed’ findings between sedentary behaviour and stress. Firstly, both positive and null associations were found among nine studies (n=1 strong- [50]; n=2 moderate- [37, 41]; n=6 weak-quality [33, 38, 42, 46, 49]). Stress was self-reported in eight of these studies [33, 38, 41, 42, 46, 49, 50], while one utilised objective measures (salivary cortisol) [37]).
Both positive and inverse associations between sedentary behaviour and self-reported stress were found in one weak-quality study [56], with the direction of these associations dependent on the source of stress assessed (e.g. total sitting was positively associated with ‘argument-related’ stress, but negatively associated with ‘running late’ stress) [56]. Two weak-quality studies which examined self-reported stress [36] and salivary cortisol [55] respectively, found inverse associations between time spent in sedentary behaviour and stress, whereby increased time in sedentary behaviour was associated with lower levels of stress. Eight remaining studies (n=6 moderate- [43, 44, 47, 48, 52, 53]; n=2 weak-quality [33, 40]) found no association between sedentary behaviour and stress. Stress was self-reported in four of those studies [33, 40, 43, 53]; three used objective measures of stress (hair cortisol [44, 47]; systolic and diastolic blood pressure, heart rate, and salivary cortisol [52]); whilst one study used both a self-report and objective (hair cortisol) measure [48].
Best-evidence synthesis: The best-evidence synthesis (adjusting for publication bias and multiple modelling within studies) resulted in the following: Excluding weak studies (since ≥2 moderate/strong quality studies exist), three (25%) (n=1 strong- [51]; n=2 moderate-quality [34, 35]) studies found a positive association between time spent in any sedentary behaviour and stress. One moderate-quality study (9%) [37] showed ‘mixed’ (i.e. positive and null) findings and eight studies (66%) (n=1 strong- [50]; n=7 moderate- [41, 43, 44, 47, 48, 52, 53] found predominately no association between sedentary behaviour and stress. Based on the best-evidence synthesis, there was insufficient evidence for an overall relationship between time spent in sedentary behaviour and stress.
Overall sitting time
Unadjusted results showed that across all studies, there were 60 models that assessed the association between sitting time and stress. This resulted in 23 positive associations (38%); seven inverse associations (12%) and 30 null results (50%).
Specifically, a total of 19 studies (n= 2 strong-quality [50, 51]; n=8 moderate-quality [34, 35, 37, 41, 44, 48, 52, 53]; n=9 weak-quality [38, 40, 42, 45, 46, 49, 54-56]) reported on associations between sitting time and stress. Five studies (n=1 strong-quality [51]; n=2 moderate-quality [34, 35]; n=2 weak-quality [45, 54]) found positive associations between sitting time and stress, i.e. increased sitting time was associated with increased stress. Of those, four employed self-reported measures of stress and non-domain-specific sitting measures, whilst one [54] utilised an objective measure of stress (i.e. salivary cortisol) and a self-reported domain-specific measure of sitting (i.e. occupational sitting) .
One weak-quality study showed an inverse association between sitting time and objectively measured stress (i.e. salivary cortisol) [55]. Four studies (n=1 moderate- [37]; n=3 weak-quality [38, 42, 49]) showed mixed findings. That is, they all included both positive and null results. For example, Lee & Kim (2018) showed a positive association between total and weekday sitting time and stress, but no association between weekend sitting time and stress [38]. Gubelmann et al (2018) showed a positive association between sitting and diurnal cortisol slope, but no association with mean or awakening cortisol [37]. Nine studies (n=1 strong- [50]; n=5 moderate- [41, 44, 48, 52, 53]; n=3 weak-quality [40, 46, 56]) showed predominately no association between sitting time and stress. Of note, however, are the findings from Rebar et al. (2016) [41]. That study compared sitting in three domains (transport-related, leisure-time, occupational) as well as overall sitting. Transport-related sitting time was positively associated with stress, whilst all other models (three of the four [i.e. 75%]) showed no association (and hence the study was labelled as a null overall finding).
Best-evidence synthesis: Based on the inconsistent findings across studies (i.e. among the 10 moderate-strong quality studies included, three showed positive associations (30%), one showed mixed (10%), six showed null results (60%)), the best-evidence synthesis resulted in insufficient evidence for a relationship between overall sitting time and stress.
TV viewing
Unadjusted results showed that across all studies, there were 15 models that assessed the association between TV viewing and stress. This resulted in six positive associations (40%); one inverse associations (7%) and eight null associations (53%).
Specifically, ten studies (n=1 strong-quality [50]; n=4 moderate-quality [41, 43, 44, 47]; n=5 weak-quality [33, 36, 39]) investigated associations between TV viewing time and stress. One weak-quality study found a positive association between TV viewing and self-reported stress (i.e. increased TV viewing was associated with greater stress) [39]. Conversely, one weak-quality study found an inverse relationship between TV viewing and self-reported stress (i.e. increased TV viewing was associated with reduced stress) [36]. Mixed results, in this case positive and null associations, were found in two weak- [33] and one strong- [50] quality study. For example, one longitudinal study found no cross-sectional association between TV viewing and perceived stress at baseline, but stress at baseline predicted increased TV viewing at follow-up [50]. Five (n=1 weak-quality [33] and n=4 moderate-quality [41, 43, 44, 47]) studies found no association between TV viewing and stress.
Best-evidence synthesis: Among the five moderate-strong quality studies that examined TV viewing, four (80%) showed null associations and therefore based on the best-evidence synthesis there was strong evidence for no association between TV viewing and stress.
Computer use
Unadjusted results showed that across all studies, there were two models that assessed the association between computer use and stress, both of which found null associations (100%). Specifically, two moderate-quality studies examined the association between computer use [41, 44] and stress. No association was found between computer use and self-reported stress [41], or computer use and objectively measured stress (hair cortisol) [44] in these studies (100%). Based on the best-evidence synthesis there was strong evidence for no association between computer use and stress.
Objectively measured versus self-reported stress
Unadjusted results showed that across all studies, there were 18 models that assessed the association between sedentary behaviour and objectively measured stress. This resulted in two positive associations (11%); six inverse associations (33%) and ten null associations (56%).
Specifically, of the seven studies that used an objective measure of stress (n=2 weak- [54, 55] and n=5 moderate-quality [37, 44, 47, 48, 52]), two reported positive [54] or mixed (in this case positive and null) associations between sedentary behaviour and stress [37] (weak- and moderate quality studies, respectively). Four moderate-quality studies found null associations [44, 47, 48, 52], while one weak-quality study showed inverse associations between sedentary behaviour and stress [55]. Based on the best-evidence synthesis (i.e. four of five moderate-quality studies showed null association (80%), there was strong evidence for no association between any type of sedentary behaviour and objectively measured stress.
Unadjusted results showed that across all studies, there were 67 models that assessed the association between sedentary behaviour and self-reported stress. This resulted in 32 positive associations (48%); eight inverse associations (12%) and 27 null results (40%).
Specifically, of the 20 studies that utilised self-report measures of stress, five studies (n=1 strong-quality [51]; n=2 moderate-quality [34, 35]; n=2 weak-quality [39, 45]) showed positive associations, while one weak-quality study showed inverse associations between sedentary behaviour and stress [36]. Five weak-quality studies showed ‘mixed’ (i.e. positive and null) associations between sedentary behaviour and stress [33, 38, 42, 49]). Nine other studies (n=1 strong [50]; n=4 moderate- [41, 43, 48, 53]; and n=4 weak-quality [33, 40, 46, 56]) found no associations. Based on the best-evidence synthesis (i.e. three of eight (38%) moderate-strong quality studies showed positive associations, five (62%) showed null associations), there was insufficient evidence of a relationship between sedentary behaviour and self-reported stress.
Objectively measured versus self-reported sedentary behaviour
Unadjusted results showed that across all studies, there were 39 models that assessed the association between objectively measured sedentary behaviour (i.e. device, direct observation) and stress. This resulted in 13 positive associations (33%); seven inverse associations (18 %) and 19 null results (49%).
Specifically, ten studies utilised objective measures of sedentary behaviour. Of those, two weak-quality studies showed positive associations [45, 54]. Three studies showed ‘mixed’ (i.e. positive and null) associations between objectively measured sedentary behaviour and stress (n=2 weak- [33, 49] and n=1 moderate-quality [37] studies) One weak-quality study found an inverse association [55]; while four studies (n=3 moderate- [48, 52, 53]; n=1 weak-quality [56] found no associations. Based on the best-evidence synthesis (i.e. of the four moderate-quality studies, one (25%) showed mixed, the other three (75%) showed null associations), there was strong evidence for no relationship between objectively measured sedentary behaviour and stress
Unadjusted results showed that across all studies, there were 39 models that assessed the association between self-reported sedentary behaviour and stress. This resulted in 16 positive associations (41%); one inverse association (3%) and 22 null results (56%).
Specifically, studies that used self-report measures of sedentary behaviour (n=16) yielded the following results. Four studies (n=1 strong- [51]; n=2 moderate- [34, 35]; n=1 weak-quality [39]) found a positive association; four found ‘mixed’ associations (in this case, positive and null; n=1 strong- [50]; n=3 weak-quality [33, 38, 42]; one weak-quality study found an inverse association [36] and seven studies (n=4 moderate-quality [41, 43, 44, 47]; n=3 weak- [33, 40, 46]) found no association between self-reported sedentary behaviour and stress. Based on the best-evidence synthesis (i.e. of the eight moderate-strong quality studies, three (37%) showed positive, one (13%) showed mixed, four (50%) showed null associations) there was insufficient evidence for a positive relationship between self-reported sedentary behaviour and stress.