key: cord-0992800-bqcj23bo
authors: Feng, Jie; Lau, Patrick Wing Chung; Shi, Lei; Huang, Wendy Yajun
title: Movement behaviors and posttraumatic stress disorder during the COVID-19 pandemic: A retrospective study of Chinese university students
date: 2022-05-25
journal: J Exerc Sci Fit
DOI: 10.1016/j.jesf.2022.05.002
sha: 371fa60e08c43435b518a8b31ffc7540f34d24b8
doc_id: 992800
cord_uid: bqcj23bo

Background/objective This study aimed to examine the associations between physical activity (PA), sedentary behavior, sleep and posttraumatic stress disorder (PTSD) among undergraduate students during the coronavirus disease 2019 (COVID-19) pandemic in China. Methods A total of 3178 university students responded to an online questionnaire between December 2020 and January 2021. Participants self-reported the time they spent on PA, screen time and sleep after (over the past seven days) and during the outbreak peak (from January to March 2020). Their sleep quality was measured using the Chinese version of the Pittsburgh Sleep Quality Index. The Chinese version of the Posttraumatic Stress Disorder Checklist - Civilian Version was used to measure PTSD. Logistic regressions and generalized linear mixed models were conducted. Results The final analysis included data from 2070 university students (20.2 ± 1.3 years old, 37.0% males). The prevalence of PTSD was 7.1%. Better sleep quality both during and after the outbreak peak, and longer sleep duration after the outbreak peak were associated with a lower odds ratio of having PTSD and lower re-experiencing, avoidance and hyperarousal scores. Higher total PA levels during the outbreak peak were associated with a higher odds ratio of having PTSD and higher levels of re-experiencing and avoidance. Conclusion Sleep quality and duration were negatively associated with PTSD among university students during the COVID-19 pandemic. The associations between PA, screen time and PTSD require further examination. Future interventions to enhance mental health could consider targeting university students’ sleep hygiene.

Widespread outbreaks of the coronavirus disease 2019 (COVID-19) pandemic have adversely 14 affected people's mental health. A recent systematic review found a relatively high prevalence 15 of symptoms of anxiety (6.3% to 50.9%), depression (14.6% to 48.3%), stress (8.1% to 81.9%) 16 and posttraumatic stress disorder (PTSD; 7.0% to 53.8%) among the general population during associated with mental health. For instance, sleep problems such as insomnia have been found 1 to predict depression, 16 while shorter sleep duration and lower sleep quality have been 2 associated with higher levels of PTSD among adults. 14,17 A systematic review found that 3 individuals with PTSD had lower levels of PA, 18 whereas improving PA was effective at 4 reducing PTSD symptoms. 19 As for sedentary behavior, a longer time spent sitting has been 5 associated with higher levels of PTSD. 17 After the COVID-19 pandemic outbreak, studies 6 found that university students' levels of PA declined, 20 whereas their sedentary behavior and 7 sleep duration increased. 20, 21 However, the correlations between these movement behaviors and 8 PTSD among university students during the COVID-19 pandemic remain unknown; most 9 studies have focused on influencing factors such as socio-demographic information and 10 exposures to the pandemic. 14,22 In addition, measurements at different stages of the pandemic 11 (e.g., during and after the outbreak peak) are needed to elucidate the longitudinal relationships 12 between movement behaviors and PTSD. This retrospective study therefore explored the 13 associations between movement behaviors during and after the outbreak peak and PTSD 14 among university students. We hypothesized that higher levels of PA, less screen time (ST), 15 longer sleep duration, and higher sleep quality both during (Time 0) and after the outbreak peak 16 (Time 1) were associated with lower risks of having PTSD and lower levels of re-experiencing, 17 avoidance, and hyperarousal (Time 1). schedules had returned to normal. Students were asked to recall their experiences and feelings since the outbreak of the pandemic 4 and to respond 'yes' or 'no' to each of nine statements: (1) people infected in their community;

(2) lived in the worst-hit areas; (3) knew someone who died of the infection; (4) neighbors 6 infected; (5) friends infected; (6) relatives infected; (7) exposed to stressful mass media 7 information; (8) family members infected and (9) felt extreme fear of being infected. The score 8 of COVID-19 exposure (ranging from 0 to 9) was calculated as the sum of 'yes' responses. 14 9 Movement behaviors after the outbreak peak (Time 1) 10 PA during the past seven days was measured using the Chinese version of the International up time the subsequent day − bedtime). One question was asked to measure sleep quality: 'Over 2 the past month, how would you rate your sleep quality?' Four options from '0, very good' to 3 '3, very bad' were provided. Unreasonable responses were considered invalid and excluded in 4 the final analysis (e.g., sleep duration was shorter than 1 hour or longer than 20 hours, actual 5 sleep duration was longer than time in bed). 29 6 Movement behaviors during the outbreak peak (Time 0) 7 To measure movement behaviors during the outbreak peak, students responded to the following 8 questions: 'During the outbreak peak of the COVID-19 pandemic (from January to March 9 2020), how much time on average did you spend on (1) all PA intensities (e.g., walking, running, 10 playing ball, doing housework); (2) ST (e.g., watching TV, using computers/tablets/mobile 11 phones); (3) actual sleep duration per day?' To assess sleep quality, the same question as above 12 was used, based on the PSQI. 28 The same criteria of screening movement behaviors after the 13 outbreak peak were also applied. 26, 27, 29 14 Post-traumatic stress disorder (PTSD) 15 PTSD was measured by the Chinese version of the Posttraumatic Stress Disorder Checklist -16 Civilian Version (PCL-C). 30 The PCL-C consisted of 17 items, covering three components: re-17 experiencing (five items), avoidance (seven items) and hyperarousal (five items). Five options 18 ranging from '1, not at all' to '5, extremely' were provided for each item, with a higher score 19 indicating a more severe symptom. Responses from '3, moderately' to '5, extremely' were 20 defined as symptomatic for each item. If a respondent was symptomatic for at least one re-21 J o u r n a l P r e -p r o o f 1 old, 37.0% male) provided complete and valid data and were included in the final data analysis.

No differences were found between the included and excluded participants in demographic 3 factors, except that the BMI of the included students was lower than that of the excluded 4 students (21.9 ± 5.0 vs. 22.3 ± 5.7, p < 0.05). Among the included sample, 43.9% of the 5 participants were living in Guangdong province during the outbreak peak, and the others' Table 1 . Overall, 7.1% of the students were 10 diagnosed as having PTSD.

For COVID-19 exposures, 29.5% of the students reported that they felt extreme fear of being 12 infected, 7.3% were exposed to stressful mass media information, 5.3% experienced people 13 being infected in their community and fewer than 5.0% had other COVID-19 exposures.

14 Self-reported total PA was 6.9 ± 5.2 MET-hours/day at Time 1, and 1.0 ± 1.1 hours/day during Time 0 were as follows: ST (5.7 ± 2.7 vs. 6.4 ± 2.9 hours/day, p < 0.01) and sleep duration (6.9 17 ± 1.2 vs. 7.4 ± 1.9 hours/day, p < 0.01). For sleep quality, 33.1% and 50.6% of the students 18 rated their sleep quality as very good after and during the outbreak peak, respectively.

The associations between potential predictors and the risk of having PTSD are shown in Table   20 2. Having a longer sleep duration after the outbreak peak was associated with a lower risk of quality both after (OR = 1.60; 95% CI: 1.21, 2.11) and during the outbreak peak (OR = 2.29; 1 95% CI: 1.76, 2.99) was associated with a higher OR of being diagnosed with PTSD. However, 2 having more PA during the outbreak peak (OR = 1.27; 95% CI: 1.11, 1.45) was also associated 3 with a higher risk of having PTSD.

The associations between movement behaviors and the three components of PTSD are 5 presented in Table 3 . Having better sleep quality during and after the outbreak peak and having 6 a longer sleep duration after the outbreak peak were associated with lower re-experiencing, 7 avoidance and hyperarousal scores. Higher PA during the outbreak peak was associated with 8 higher scores of re-experiencing and avoidance.

To the best of our knowledge, this is the first study to examine the associations between 12 movement behaviors during and after the outbreak peak of the COVID-19 pandemic and PTSD 13 among Chinese university students. We found that higher sleep quality during and after the 14 outbreak peak and a longer sleep duration after the outbreak peak were associated with a lower 15 risk of having PTSD, as well as lower levels of re-experiencing, avoidance and hyperarousal.

Individuals with higher total PA during the outbreak peak were more likely to be diagnosed as 17 having PTSD and have higher levels of re-experiencing and avoidance approximately 10-12 18 months later.

The prevalence of PTSD (7.1%) observed in this study was higher than that reported in 20 February 2020 among university students in China (2.7%), 14 but lower than another study conducted in April 2020 (16.3%). 15 This variation could be explained by different durations of 1 exposure to the pandemic: a longer duration of exposure to traumatic stressors was associated 2 with increased PTSD symptoms. 33 In the developing stage of the COVID-19 pandemic, 3 therefore, people's mental health became worse. During the data collection period of this study, 4 however, control of the pandemic had almost been achieved in Mainland China; people's PTSD 5 symptoms may thus have been ameliorated. In addition, it's worth noting that sampling 6 methodology, characteristics of participants (e.g., sex ratio), measurements of PTSD used, and 7 the residential areas of participants during the pandemic may contribute to the discrepancy 8 between studies.

The findings that both sleep quality and duration were associated with PTSD were consistent 10 with our hypothesis. A cohort study conducted among Chinese adolescents found that those 11 who had worse sleep quality 12 months after an earthquake were more likely to have PTSD at 12 both 12 months and 24 months after the earthquake. 34 Similar to sleep quality, longer sleep 13 duration after the outbreak peak was associated with a lower OR of having PTSD and lower 14 levels of re-experiencing, avoidance and hyperarousal. These findings are supported by 15 existing evidence. Goodwin et al. found that short sleep duration was associated with a high 16 likelihood of having PTSD. 35 The relationship between sleep duration and PTSD also existed 17 during the COVID-19 pandemic: a study conducted among Chinese university students found 18 that longer sleep duration was associated with lower PTSD scores. 14 It is worth noting that the 19 relationship between sleep and mental health may be bi-directional. 16, 36 Nevertheless, our 20 finding further demonstrated that sleep during the peak of the pandemic outbreak was still 21 associated with PTSD among university students 9-12 months later when their daily routines J o u r n a l P r e -p r o o f Re-experiencing (score) 7.0 ± 3.2 Avoidance (score) 9.6 ± 4.3

Hyperarousal (score) 7.1 ± 3.2

Having post-traumatic stress disorder 147 (7.1%) a measured by the International Physical Activity Questionnaire-Short Form (IPAQ; 7-day recall) between December 2020 and January 2021; total physical activity includes walking, moderate physical activity, and vigorous physical activity; b retrospective recall between January and March 2020, total physical activity was measured by one question, including all intensities of physical activity. c range of the subscale score: re-experiencing, 5.0-25.0; avoidance, 7.0-35.0; hyperarousal, 5.0-25.0. Having PTSD was defined as being symptomatic of at least one item on re-experiencing, three items on avoidance, and two items on hyperarousal.

Abbreviations: BMI, body mass index; COVID-19, coronavirus disease 2019; MET: metabolic equivalent of task.

J o u r n a l P r e -p r o o f 

Abbreviations: OR, odds ratio; 95% CI, 95% confidence interval; MET: metabolic equivalent of task. Model 1: adjusted for age, sex, body mass index (BMI), and exposure to coronavirus disease 2019

Abbreviations: 95% CI, 95% confidence interval; MET: metabolic equivalent of task

Model 1: adjusted for age, sex, body mass index (BMI), and exposure to coronavirus disease 2019 (COVID-19) score