The global public health landscape has been profoundly reshaped by the COVID-19 pandemic. In the face of thousands of infections, the Chinese government has implemented stringent measures to curtail the virus’s spread. These include inter-city transport restrictions, mandatory home quarantines, centralized isolation of affected individuals, and strict enforcement of social distancing protocols (1), as a consequence, this has triggered a significant transformation in people’s lifestyles. One of the pivotal changes is the 24-h movement behaviour. Extensive research has revealed that it has deteriorated notably during the pandemic (2, 3). People usually experienced longer periods of sedentary time, lower frequency of physical activity and shorter periods of sleep (4, 5). These changes may be related to mental health problems, including anxiety and depression (3, 6–9). One manifestation of these changes is the shift in teaching models from traditional face-to-face instruction to online learning, which may lead to increased periods of prolonged sitting in front of screens for both pupils and teachers, alongside a reduction in physical activity (10, 11).

In China, due to high expectations from society and parents, teachers are facing increasing pressure, the implementation of the “double reduction” policy has further increased the rate of job burnout among teachers, leading to a pessimistic outlook on their mental health (12, 13). Moreover, the dread of COVID-19, coupled with shifts in teaching methodologies, the constant worry of infecting themselves or their students, inadequate sleep, and a lack of physical activity all these factors may collectively contribute to the rising anxiety and depression among teachers (14, 15). Previous researches suggest that physical activity, sleep duration and sedentary behaviour may be effective moderators of mental health (16–19), however the impact of the epidemic has led to disruptions in 24-h patterns of behavioural activity, which may have led to negative impacts on teachers’ mental health. In the cross-sectional studies of teachers at the beginning of the epidemic in China, it was shown that teachers were approximately three times more likely to suffer from anxiety during the epidemic than they were before (20), and more than half of teachers suffer from depression. Chronic mental health problems can lead to physical health problems in teachers. These include cardiovascular disease and metabolic syndrome (21, 22). Moreover, a higher level of depression among teachers can lead to a lower level of well-being among their pupils (23). Therefore, it becomes necessary to study the mental health status of teachers in the specific context of the epidemic.

Despite the fact that up to now a number of existing studies have demonstrated the effects of the triad of sedentary time, physical activity, and sleep duration on mental health (8, 24–26), professional groups represented by teachers are often neglected. Although some studies in recent years have recognized the importance of the teacher population as a special group, for example looking at the impact of physical activity and screen time on aspects of teachers’ mental health (27), they have not considered all three 24-h behavioural factors (28, 29). Consequently, the aim of our study was to investigate whether sedentary behaviour, sleep duration and physical activity are related to teachers’ mental health during the COVID-19 pandemic and aims to provide a reference framework for enhancing the mental wellbeing of the teaching profession.

A total of 1,670 teachers were included in this study, as detailed in Table 1. The majority of participants were female (59.6%), with an average age of 37.0 ± 10.2 years. 15.3% had depressive symptoms and 80.7% suffered from anxiety symptoms. A significant portion held a bachelor’s degree (61.0%), were married (76.8%), and reported an average family economic status (81.6%). The average BMI among the teachers was 23.5 kg/m². Over half (51.5%) of the teachers reported their health status as good, while 45.7% reported average sleep quality. In terms of lifestyle-related factors 60.1% did not consume alcohol. Most participants engaged in physical activity 1–2 times a week (54.4%), and a substantial number reported sitting for 0–2 h daily (39.3%). With regards to work-related factors, the majority experienced high levels of work stress (63.8%), yet 50.5% expressed satisfaction with their jobs. The teachers worked longer hours (72.2%) and had less overtime (37.6%). The average weekly class time was 18.7 h.

As shown in Table 2, GVIF (1/2df) < 2 for all cases, which rules out collinearity among the factors in the model. In the one-way logistic regression analysis, we have found that the correlates of anxiety and depression are age, family economic status, etc. (more details in Table 1) Following adjustments to the models, as presented in Table 3, we observed that sitting for 3 to 4 h did not significantly influence anxiety or depression. However, a sitting duration of 5–6 h is statistically significant for depression. Physical activity is associated with a lower prevalence of depression, which is only significant in Model 1 and Model 3. When sleep duration exceeds 7 hours, the probability of both anxiety and depression levels are significantly reduced (p < 0.05).

Table 4 shows that the prevalence of depression almost doubles when sitting for more than 6 h compared to 5 to 6 h. When the frequency of physical activity was 5–6 times/week, the probability of experiencing anxiety increases by nearly twofold compared to when 1–2 times/week. Furthermore, when sleep duration >7 h and sedentary time >6 h, the prevalence of anxiety and depression rises significantly.

Based on our prior findings, we found that 4 h of sitting was the node where the results were significant or not, so we used 4 h as the basis for grouping. As Table 5 illustrates, when the sedentary time was ≤4 h, sleeping for more than 7 h is associated with a lower probability of anxiety and depression. Conversely, when the sedentary time >4 h, sleeping >7 h notably decreased the likelihood of depression in only Model 1 (OR = 0.43, CI = 0.25–0.72) and Model 3 (OR = 0.43, CI = 0.26–0.71).

The outbreak of COVID-19 has led to significant changes in people’s learning and lifestyles. Due to the lockdown measures taken by the government, many schools have had to adopt online teaching models (35). Although the present study did not collect specific data on the duration of isolation or the intensity of remote teaching, the outbreak of COVID-19 likely served as a significant contextual factor. During this period, the shift to online teaching models and government lockdown measures presented teachers with unique challenges, such as the need to rapidly learn new techniques and balance work-life boundaries (36–38). These contextual factors may have contributed to the increased work pressure and mental health issues observed in our results, consistent with broader trends reported during the pandemic. Our cross-sectional study delved into exploring the intricate interplay between sedentary behaviour, physical activity, and sleep duration on the psychological well-being of teachers in China during the pandemic context. The study findings underscored the significant associations among these three factors and the prevalent mental health issues of anxiety and depression among teachers.

Our research indicates that prolonged periods of sitting are associated with a higher probability of anxiety and depression, a finding that aligns with prior researches (39–41). The subgroup analyses in Table 4 show that when sleep duration was ≤7 h, more sedentary time was associated with an increased probability of depression, and when sleep duration was >7 h, sedentary time >6 h was associated with an increased probability of both anxiety and depression. In Table 5, when sedentary time is ≤4 h, adequate sleep time is associated with a lower prevalence of anxiety and depression, whereas when sedentary time is >4 h, the results are only significant for depression in some models and lead to a higher probability of anxiety. This suggest that adequate sleep may not be able to reverse the negative effects of sedentary time. Therefore, one of the important measures to improve teachers’ mental health is to reduce their sedentary time.

Sedentary behaviour includes sitting for various purposes, including work and electronic screen-based activities and so on. Because of the inventive and control measures taken by the Government during the pandemic, some schools having to close and thus switch to a home-based online teaching model (14), this has led to a significant increase in the amount of time teachers spend on the screen, which further increases sedentary time. Therefore, in the special context of the pandemic, one way to reduce teachers’ sedentary behaviour is by reducing screen time. Screen-based activities lead to increased arousal in the central nervous system and less social interaction and a withdrawal from relationships, which can lead to an increased risk of anxiety and depression (42–44). In a study of Brazilian teachers, they were more likely to report a very low quality of life and to feel sad and anxious on a regular basis if they used computers/tablets for more than 5 hours a day (27). Relevant departments can set time limits for online classes, meetings, to avoid unlimited screen time and teachers can be encouraged to use monitoring software on their own devices, such as time reminder timers (27), thereby reducing the sedentary time. In addition, a German study on children found that the use of standing desks in the classroom was effective in reducing sedentary time (45), schools and educational institutions can also use standing desks or liftable desks and chairs in the office to help teachers reduce sedentary behaviour.

Sedentary behaviour, physical activity and sleep duration are integral components of the 24-h movement behaviour framework. Given that physical activity in this study was assessed as weekly frequency, these behaviours represent distinct yet interrelated dimensions of lifestyle. An increase in sedentary behaviour often reflects a sedentary lifestyle pattern, which may be associated with lower physical activity frequency or reduced sleep duration (46, 47). In our study, almost 80% of the teachers were not physically active or were only physically active 1–2 times/week. According to the results of univariate analysis, physical activity was significantly associated with depression, and although only in models 1 and 3 was the effect on depression statistically significant after modeling adjustments, we still found a decrease in the probability of depression as the frequency of weekly physical activity increased. Our findings emphasize that teachers should be actively encouraged to be physically active. Physical activities can trigger various biochemical changes in the brain, including the release of endorphins and the transmission of monoamine neurotransmitters, thereby improving mental health (48).

However, we found in Table 4 that physical activity at 5–6 times/week was associated with increased anxiety when sleep duration was ≤7 h, although physical activity at 1–2 times/week and 3–4 times/week was not significantly associated with anxiety, we observed that the likelihood of anxiety increased as the frequency of weekly physical activity increased. Moreover, the anxiety-inducing effect of physical activity at 5–6 times/week was reduced when sleep duration was >7 h. In Table 5, we found that physical activity at 5–6 times/week was not significantly associated with a reduction in anxiety and depression when sedentary time was > 4 h and there is a slight increase in its probability compared to 1–2 times/week. Previous studies have shown that more exercise is not always better (49, 50). These findings may suggest the need to focus on body conditioning and choosing the right frequency when performing physical activities. Therefore, it is imperative that the focus is shifted toward interventions for teachers’ integrated behavioural patterns, as opposed to merely increasing the frequency of physical activity to cope with teachers’ negative emotions. Of course, we also need to consider the uniqueness of the teaching profession and make adjustments based on actual situations.

Our study showed that greater than 7 h of sleep resulted in lower probability of anxiety and depression, same results as a study in Ningbo, China (51), based on two recent Chinese cross-sectional studies, an excessive amount of sleep was also discovered to heighten the probability of experiencing anxiety and depression (52, 53). Therefore, further research is needed to determine how many hours of sleep are critical for the development of teachers’ anxiety and depression.

This study focuses on the overlooked issue of teachers’ mental health during the epidemic. It is the first to examine the impact of three factors of 24-h movement behavioural and the interrelationships between them on teachers’ mental health in China. It also indicates that physical activity must be maintained at a certain frequency to effectively improve mental health. However, several limitations need to be noticed. Firstly, cross-sectional study design has its own limitations. There may be a reverse causality between the results. Secondly, the classification of physical activity intensity relied on a verbal questionnaire, as opposed to the more robust International Physical Activity Questionnaire or other commonly used standardized questionnaires, which may lead to less accurate acquisition of physical activity frequency, thus potentially contributing to fewer statistically significant results regarding the impact of physical activity on mental health. In addition, physical activity is measured through frequency, this may lead to confusion between intensity and frequency. Despite the same frequency of physical activity per week, there may be significant differences in duration. Meanwhile, investigating physical activity, sedentary time, and sleep duration through specific questions may introduce self-reported recall bias in the subjects, affecting the accuracy of the results. Thirdly, the categorization of anxiety and depression levels in this study led to a substantial disparity between the number of individuals classified as normal and those identified as experiencing non-normal levels, thereby introducing a degree of bias into the results. Fourthly, the limitations imposed by the database’s own data volume and the restricted nature of the data collected within it, for example, the units of measurement for physical activity and sedentary time are not unified, which limits the generalisability of the research. Future research may employ longitudinal cohort studies to investigate the causal relationships among these three factors and their impact on teachers’ mental wellbeing, while expanding the study population to encompass a broader geographical scope. Although our study categorised sedentary time, sleep duration and physical activity, the classification criteria and methodology were not rigorous due to database limitations. Future research can adopt more scientific measurements and classifications when studying sedentary behaviour, physical activity, and sleep duration.

This study examined an association between sedentary behaviour, sleep duration, physical activity and mental health among teachers. The results indicates that extended periods of sedentary time are closely linked to anxiety and depression among teachers, sufficient sleep significantly reduces the likelihood of experiencing these mental health issues and appropriate physical activity may help to reduce the likelihood of depression. However, high frequency of physical activity in the context of sedentary and sleep deprivation is associated with an increase in the incidence of anxiety. Therefore, teachers may need to reduce sedentary behaviour, increase sleep duration, and maintain regular physical activity to ensure their mental health and comprehensive behavioural interventions for the prevention of negative emotions among teachers are important. Our study may be able to provide data to support the development of precise behavioural guidelines for groups of teachers. Given the cross-sectional design of this study, all findings only reveal statistical associations and cannot infer causal relationships and future longitudinal studies are warranted to clarify the directionality of these associations and disentangle potential reverse-causal pathways.