The association between prior physical fitness and depression in young adults during the COVID-19 pandemic—a cross-sectional, retrospective study

Background The COVID-19 pandemic has led to a spike in deleterious mental health. This dual-center retrospective cross-sectional study assessed the prevalence of depression in young adults during this pandemic and explored its association with various physical fitness measures. Methods This study enrolled 12,889 (80% female) young adults (mean age 20 ± 1) who performed a National Student Physical Fitness battery from December 1st, 2019, to January 20th, 2020, and completed a questionnaire including Beck’s Depression Inventory in May 2020. Independent associations between prior physical fitness and depression during the pandemic were assessed using multivariable linear and binary logistic regressions accordingly, covariates including age, dwelling location, economic level, smoking, alcohol, living status, weight change, and exercise volume during the pandemic. Sex- and baseline stress-stratified analyses were performed. Results Of the study population 13.9% of men and 15.0% of women sampled qualified for a diagnosis of depression. After multivariable adjustment, anaerobic (mean change 95% CI −3.3 [−4.8 to 1.8]) aerobic (−1.5 [−2.64 to −0.5]), explosive (−1.64 [−2.7 to −0.6]) and muscular (−1.7 [−3.0 to −0.5]) fitness were independently and inversely associated with depression for the overall population. These remained consistent after sex- and baseline stress-stratification. In binary logistic regression, the combined participants with moderate, high or excellent fitness also showed a much lower risk compared to those least fit in anaerobic (odd ratio (OR) 95% CI 0.68 [0.55–0.82]), aerobic (0.80 [0.68–0.91]), explosive (0.72 [0.61–0.82]), and muscular (0.66 [0.57–0.75]) fitness. Conclusions These findings suggest that prior physical fitness may be inversely associated with depression in young adults during a pandemic.


INTRODUCTION
Since its onset, coronavirus-2019 (COVID-19) has personified overwhelming stresses, stemming from-infections, loss of work, loss of freedoms, isolation, and death, leaving behind an aura of uncertainty amongst the world's citizens. COVID-19 has led to a spike in deleterious mental health issues (Wang et al., 2020), particularly in depression (Brooks et al., 2020). There's concern that such a surge may lead to an increased rate of suicides (Reger, Stanley & Joiner, 2020) domestic abuse, economic, and somatic health issues (Moulton, Pickup & Ismail, 2015). Hence, evidence-based research is needed to address the rise in depression across multiple populations as a result of COVID-19.
Physical fitness, particularly cardiorespiratory fitness, and exercise training have repeatedly been shown to be negatively associated with future CVD and diabetes risk (Dun et al., 2019a(Dun et al., , 2019bMyers et al., 2015;Pedersen et al., 2019) as well as having a positive relationship with mental health in multiple populations (Baumeister et al., 2017;Cho et al., 2019;Kerling et al., 2015). Studies have been published on COVID-19 and previous outbreaks and their adverse effects on mental health (Brooks et al., 2020;Torales et al., 2020). Furthermore, recent studies have alluded to an association between current physical activity and mental health during the pandemic (Lopez-Bueno et al., 2020;Stanton et al., 2020), with a commonality that decreased physical activity appears with lockdowns. However, these studies are largely limited to current physical activity and mental health. To our knowledge, none have explored the possible association between prior parameters of physical fitness and pandemic-related depression.
From the existing evidence, it may be hypothesized that physical fitness factors are inversely correlated with depression during COVID-19. Therefore, this retrospective cross-sectional study's aims are two-fold: to assess the prevalence of depression in young adults during the pandemic and to investigate the associations between a variety of different measures of physical fitness, including prior anaerobic, aerobic, explosive, muscular, flexibility, and pulmonary fitness on the prevalence of depression during the COVID-19 pandemic.

Study design and participants
This retrospective cross-sectional study enrolled two universities (Hunan Traditional Chinese Medical College, Hunan, China, and Medical College of Jinhua Polytechnic, Zhejiang, China) selected by convenience sampling, that performed the Chinese National Student Physical Fitness Standard (CNSPFS) battery between December 1st, 2019 to January 20th, 2020 when government-issued sanctioned lockdowns and social distancing. A total of 14,059 university students who were free of chronic diseases and had completed a CNSPFS were screened. Of these, 13,013 participants (response rate of 93.2%) completed a follow-up questionnaire from May 1st to 23rd, 2020. Participants who provided poor quality questionnaires were excluded (n = 124). The criteria of poor quality were: (1) If the ID information in the CNSPFS system did not match that of the follow-up questionnaire; or (2) If the 81-question survey was completed in less than 3 min. A total of 12,889 participants were included in the study. All baseline data were extracted from the CNSPFS system; the data during the COVID-19 pandemic were collected from the survey platform (https://www.wjx.cn). Data from the two time-points were linked by identifying each participant's university student ID number and matching them accordingly. This study was evaluated and approved by the Review Board of Xiangya Hospital Central South University (approval No. 202005126). Written informed consent was documented during the baseline, and digital informed consent was given upon initiating the survey. All participants were codified and anonymized to protect the confidentiality of individual participants.

Physical fitness
Fitness measures were obtained from the completion of the CNSPFS battery, and scores were attained through the standardized scoring system that weighted each fitness indicator score by age-and sex-specific percentage. The CNSPFS battery included a 50-m sprint, (Duffield, Dawson & Goodman, 2004) an 800-m (women) and 1,000-m (men) run. Margaria, Aghemo & Piñera Limas (1975) a standing long jump (Robertson & Fleming, 1987) timed sit-ups (women) and pull-ups (men) (Rutherford & Corbin, 1994) sit and reach test and vital lung capacity respectively (Ranu, Wilde & Madden, 2011). Anaerobic, aerobic, explosive, muscular fitness, flexibility, and pulmonary fitness were assessed through these tests. The scores were classified as follows: low fitness (<60), moderate fitness (60-79), high fitness (80-89), and excellent fitness (>90). All tests were administered by trained physical education teachers following the CNSPFS standard operating procedures. The test-retest reliability across all assessments employed as an intraclass correlation coefficient (ICC) > 0.90. The details about performing CNSPFS have been described previously (Yi et al., 2019;Zhu et al., 2017).

Baseline stress and depression during the COVID-19 pandemic
Baseline stress was assessed concurrently with the CNSPFS test through a modified question as described previously (Frazier et al., 2014). Beck's Depression Inventory, second edition (BDI-II) was also administered during the follow-up questionnaire performed in May (Yang & Stewart, 2020). The BDI-II is a 21-item self-report questionnaire validated in young Chinese adults (Yang & Stewart, 2020;Zhu et al., 2018) and well correlated with a clinical diagnosis of depression (Moullec et al., 2015) with four response options for each item. The added total scores of the BDI-II can vary from 0 to 63 and are classified as: 0-13 no depression, 14-19 mild depression, 20-28 , PeerJ, DOI 10.7717/peerj.11091 3/18

Characteristics of lockdown
The first lockdown order in China was delivered on January 23rd, 2020 by the government of Wuhan, Hubei province, followed by the other provinces across China. The main requirements were: (1) all individuals were ordered to stay home or at their place of residence, except for permitted work, local shopping, or other permitted errands, or as otherwise authorized.
(2) All schools, sports facilities, entertainment, and recreational venues, personal care and beauty services, and the majority of factories and markets were closed.

Statistical methods
For data of participants' demographics and characteristics, independent t-test and Chi-square test were used for assessment in mean difference between sexes of continuous and categorical variables, respectively. The primary outcomes of the present study were the change in BDI-II depression scores (continuous variable) and the presence of depression (categorical variable), defined as a BDI-II score ≥ 14 (Beck, Steer & Brown, 1996;Zhu et al., 2018) during the COVID-19 pandemic. The independent variables included prior fitness factors, stress, socioeconomic status, dwelling location, and smoking, alcohol, living status, changes in body weight, and exercise volume per week during the pandemic. The associations between each independent variable and BDI-II depression score change were assessed by univariate analyses. The independent relationships of prior physical fitness factors and BDI-II depression score change were assessed using multivariable linear regression. The associations between prior physical fitness factors and the presence of depression, defined as a BDI-II depression score ≥ 14, were evaluated via binary logistic regression in which the lowest level was set as the reference. The variables included in the regression models were demographics, the independent variables mentioned above and potentially associated with depression or those with a p value < 0.20 in the univariate analyses were included in the multivariable linear and binary logistic regression analyses.
To minimize confounding from baseline psychological status, baseline stress-stratified multivariable linear regression was performed. Analyses were carried out using SAS software, version 9.4 (SAS Institute), a two-tailed alpha level of 0.05 was considered to be statistically significant.

Demographics
Demographics of 12,889 participants are presented in Table 1. Male participants were proportionally a smaller percentage of the population. In the overall population, there were 12.2%, 60.6%, and 27.3% of participants that came from lower-income, middleincome, and upper-income families, respectively, and 35.4%, 42.4%, and 22.2% of participants lived in a rural, urban-rural junction, and urban areas, respectively. During the COVID-19 pandemic, 41% and 59% of participants lived alone or with family/friends, respectively.

Characteristics of prior physical fitness and prevalence of depression during the COVID-19 pandemic
Each of five physical fitness parameters was scored and graded as low fitness (not pass), moderate fitness, good, or excellent according to CNSPFS that was established referring to national physical fitness test data while taking age and sex into account. Across multiple physical fitness tests, the mean score ranged from 53 to 78 in men and 67 to 78 in women, there were significant differences in physical fitness score between sexes (P < 0.001 for each comparison except pulmonary fitness score). The grade distributions were evaluated by the histogram and conformed to a normal distribution; the majority of participants were graded between moderate and high fitness. Less than 25% of participants were graded as low fitness or excellent. The median BDI-II depression during the pandemic scores were 2 for men and 1 for women (P < 0.001). Men and women who presented depression made up 13.9% and 15.0% of their respective populations, no significant difference was found between sexes (P = 0.14) (More in Table 1).

Association between anaerobic fitness and depression during the pandemic
After multivariable adjustment, anaerobic fitness was independently and negatively associated with the BDI-II score. Across four grades, participants with excellent anaerobic fitness had an average BDI-II score that was −3.300 (95% CI [−4.761 to −1.838]) points lower than those with low fitness (Tables 2 and 3). This finding showed consistency in subjects with and without baseline stress (−2.689 [−4.199 to −1.178]) (Table 4). In binary (participants with depression vs without depression) logistic regression, compared to low fitness participants, participants with an excellent anaerobic fitness had less than half the risk of the presence of depression during the pandemic (OR 95% CI 0.58 [0.37-0.90]), and combined participants with moderate, high or excellent anaerobic fitness also showed a much lower risk (0.68 [0.55-0.82]) ( Fig. 1).

Association between muscular fitness and depression during the pandemic
Muscular fitness was also independently and negatively correlated with depression. Participants with excellent muscular fitness had an average of −1.713 (95% CI [−2.956 to −0.470]) lower BDI-II score than participants with low fitness (Table 3). This finding was consistent in those with and without baseline stress (−2.136 [−3.383 to −0.889]) (Table 4). In binary logistic regression, compared to those who were categorized as low fitness, participants with excellent muscular fitness had less than half risk of the presence of

DISCUSSION
This study is the first to examine the relationship between physical fitness across multiple domains and the presence of depressive symptoms during the COVID-19 pandemic. Our data showed that 13.9% of young men and 15.0% of women had BDI scores that qualified them for the diagnosis of depression, which is unexpectedly similar to the level of depression reported previously in older adult populations (16.5%) (Wang et al., 2020).
Findings that are particularly implicative as both depression and low physical fitness are risk factors for future CVD and diabetes (Moulton, Pickup & Ismail, 2015;Myers et al., 2015;Pedersen et al., 2019). This study further explored the association between previous levels of explosive, anaerobic, flexibility, and pulmonary domains of fitness, with depression. Our data demonstrate that apart from flexibility and pulmonary fitness, all of these fitness parameters were independently inversely associated with depression in young adults free from chronic diseases. Depression and depressive symptom severity vary in individuals, and at different periods, at times, it may impact functional capabilities, whose effects are deleterious on individuals' physical and psychological wellbeing. While the etiology of depression remains largely unknown, the onset of depression may come in various forms, ranging from physiological factors, psychological factors, and those coming from an individual's environment. Both chronic and acute stresses are believed to play an integral part in the development of depression. Chronic stress, encompassing long-term negative environmental circumstances, such as financial difficulties; conflictual relationships with family, friends, or romantic partners (Hammen et al., 2009), acute stresses coming from episodes of stress such as those environmental or from personal loss. The COVID-19 pandemic and lockdowns have led to a surge in negative emotions and increased depression (Choi, Hui & Wan, 2020;Lei et al., 2020).
Recent research has shown that anaerobic training to be inversely associated with depression severity independent of aerobic activity (Cangin et al., 2018). Our research showed for the first time that anaerobic fitness was the greatest predictor of a lower BDI-II score among the fitness parameters. After stratification of baseline stress, increasing levels of fitness were significantly correlated with lower BDI-II scores in a near dose-response fashion, with those with the highest fitness being less likely to be depressed (Table 4). Furthermore, in conjunction with previous research (Harvey et al., 2018;Kerling et al., 2015) our findings showed that aerobic fitness was independently associated with lower depression scores. A number of studies have found that aerobic exercise is an effective therapy for moderate forms of depression while also showing it may be equally effective compared to other traditional methods of psychotherapy. The association of aerobic fitness with lower levels of depression is not limited to young adults as longitudinal studies have similar results in reducing depression in middle-aged and older individuals and younger adults (Jaworska et al., 2019). Muscular fitness was measured through two strength tests to best evaluate for muscular strength and endurance, given the variability of muscular fitness across sexes. Muscular fitness and its inverse association with depressive symptoms have been previously investigated (Marques et al., 2020). This association was also shown in this study after multivariable analysis, showing that participants in the highest subset of muscular fitness demonstrated a BDI-II score lower than participants with low fitness scores. These findings are consistent with those found during pre-pandemic conditions, where increased handgrip strength as well as comprehensive muscular fitness appears to act as a buffer against depressive symptoms and are associated with improvements in depression (Krogh et al., 2009;Suija et al., 2013). Previous large-scale observational studies suggested that muscular fitness may even have greater effects than those associated with aerobic fitness in the reduction of depression (Bennie et al., 2019). Our findings extend those of others, in that we found muscular fitness to be associated with lower depression scores during the COVID-19 pandemic. Moreover, Explosive fitness, utilizing the phosphocreatine system (Baker, McCormick & Robergs, 2010), was found to be independently and inversely associated with depressive symptoms, with the highest level of explosive fitness being associated with a lower BDI-II score. Furthermore, individuals in the highest explosive fitness level had less than half the associated risk of depression than individuals in the lowest levels. This is a particularly interesting finding as it appears to suggest that different types of fitness than those reported elsewhere (Ren et al., 2020) may have an effect of protection from depressive symptoms. Future prospective cohort or intervention studies are required to better elucidate the relationships between explosive fitness and the risk of depressive symptoms. Physical fitness and its inverse relationship with depression could be brought about through increased physical activity and exercise. Accumulating evidence suggests that exercise could be associated with primary monoamines, whereby higher amounts of exercise have a positive impact on neurotransmitter systems that regulate primary monoamines, dopamine, noradrenaline, and serotonin (Dishman, 1997;Poulton & Muir, 2005). As well as having a positive effect on physiological states, regular exercise and higher fitness could impact psychological states. Mental resilience may be improved through exercise and attainability of higher fitness (Childs & De Wit, 2014) as persons who feel or believe themselves to be healthier may be less inclined to fear the COVID-19 pandemic or at least feel more resilient against it. This has been demonstrated as those with higher mental resilience were less likely to be depressed (Kirby et al., 2017) as participation in regular exercise can usually distract individuals from noxious stimuli, thereby improving depression. Furthermore, since exercise is often extrinsic, increased self-efficacy and self-esteem, garnered from higher muscle mass and fitness could lead to improved mental health (Blumenthal et al., 2007).
While this study provides certain insights into the associations through which higher physical fitness levels in several parameters of fitness is related to a decreased risk of depressive symptoms, the causation between fitness parameters and depressive symptoms could not be established due to the cross-sectional study design. Associations between fitness measures and depressive symptoms in young adults may be bidirectional. Studies have indicated that previous depression could be a factor in the cessation of exercise, and the development of a sedentary lifestyle since those with depression may have lower self-worth, and confidence, increased self-criticism, and unwarranted guilt (Faulkner, Carson & Stone, 2014;Ren et al., 2020). Future prospective cohort or intervention studies are required to better elucidate the relationships between fitness, particularly anaerobic and explosive fitness, and the risk of depressive symptoms.
Furthermore, the sample was selected through convenience sampling and therefore may be subject to a degree of information bias, may not be representative of the whole population, and variations within individuals exist. As is with all cross-sectional studies, our data demonstrate an association between physical fitness and prevalence of rather than incidence of depression during the COVID-19 pandemic, although baseline stress status was included as a covariate in all multivariable analyses, and baseline stress-stratified multivariable analysis was performed, with consistent findings. The present study only comprised university-educated Chinese young adults, which potentially limits the generalizability of the findings. However, our findings were consistent among sexes, and with multivariable adjustment of age, baseline stress, geographic attribute, socioeconomic level, and smoking, alcohol, living status, changes in weight, and exercise volume during the pandemic. In addition, all findings were consistent in both multivariable linear and binary logistic regressions. It seems plausible that the biologic effects of many factors would be qualitatively similar in other populations.

CONCLUSION
Multiple physical fitness parameters are inversely associated with the prevalence of depression in young adults during the COVID-19 pandemic. These associations were independent of other potential confounders, such as sex, age, baseline stress, dwelling location, socio-economic level, and smoking, alcohol, living status, weight changes, and exercise volume during the pandemic. These findings along with previous research suggest that that techniques and lifestyle management that lead to improved comprehensive fitness including a wider range of muscle groups and energy systems may be considered as a possible approach to help prevent and/or reverse depression during the pandemic.
Randal J. Thomas performed the experiments, authored or reviewed drafts of the paper, and approved the final draft. Thomas P. Olson performed the experiments, authored or reviewed drafts of the paper, and approved the final draft. Jie Liu analyzed the data, prepared figures and/or tables, coordinated the study; conduct and collect data, and approved the final draft. Yuchen Dong analyzed the data, prepared figures and/or tables, coordinated the study; conduct and collect data, and approved the final draft. Suixin Liu conceived and designed the experiments, prepared figures and/or tables, authored or reviewed drafts of the paper, and approved the final draft.

Human Ethics
The following information was supplied relating to ethical approvals (i.e., approving body and any reference numbers): The study was approved by Ethics Committee of Xiangya Hospital of Central South University (approval no. 202005126) and written informed consent was documented. All participants were codified and anonymized to protect the confidentiality of individual participants.

Data Availability
The following information was supplied regarding data availability: Raw data, including the data of anonymized participants fitness scores, anthropometric measurements and psychological status, are available as a Supplemental File.