It is widely recognized that first year university students are exposed to various stressors when leaving the parental home and starting a new independent life (1). Generally, beginning college is associated with increased responsibilities and is coupled with a period of rising insecurities and emotional pressure (2, 3). Many students also experience a change in their overall lifestyles and social activities, including a decrease in physical activity, frequent use of alcohol, changes in diet, and increased stress levels (2). Furthermore, students often face a competitive environment and the need to create new social networks (4). These factors shape the student population and increase their susceptibility to mental health problems (2, 5).

With the onset of the COVID-19 pandemic, a potential additional stressor arose for students. In Colombia, for instance, a state of emergency was declared on March 17, 2020, after the first COVID-19 cases were detected on March 6, 2020, which resulted in university life changing immediately. Students were asked to stay away from campus, study remotely, and attend virtual classes. The measures taken to contain the spread of the virus have impacted everyone worldwide, but one of the most affected groups was probably the student population (6, 7). The closures of educational institutions and the shift to online education have caused changes in students' social life and lifestyles due to social isolation and fewer activities with peers (8, 9). Recent data show that slightly more than half of the college students experienced worsening mental health in 2020, including increased stress, anxiety, and depression (10).

Several factors have been proposed to reduce the effect of crises on health. One crucial factor discussed is social capital (11). For instance, Pitas and Ehmer (11) demonstrate that communities with high social capital respond more effectively than those with low levels of social capital. Social capital, however, has not only been shown to have beneficial effects on health during crises, but also in general. The above is consistent with what is proposed in the social determinants of health model, which explains how social, economic, and environmental factors determine the health status of individuals or populations (12, 13).

Regarding social capital, there is a growing literature showing that higher levels of social capital can benefit health (14–16). Studies that have analyzed the association between social capital and lifestyles, for instance, have shown that strengthening social networks can eradicate risky behaviors in the general population (14). In addition, higher levels of cognitive social capital have been associated with a lower risk of developing major depression (15) and with higher levels of regular, moderate, or vigorous physical activity (14–16).

Studies assessing the association between social capital and young adults' health, especially first year university students, are currently limited globally and in Colombia (17). Furthermore, most studies on mental health and social capital have been done in school, and adult populations in developed countries, and only a very few studies have evaluated this association in developing countries, such as Colombia (18–22). Additionally, most studies used cross-sectional data, which does not allow to make inferences about causality. Therefore, the present study aimed to explore the relationship between depressive symptoms and social capital and lifestyles through a repeated measure study, including two measurements 6 months apart during the first year of university studies. The COVID-19 pandemic started in the middle of our first wave of data collection.

This is the first longitudinal study assessing clinically relevant depressive symptoms and the relationship with levels of social capital and lifestyles among university students. Data collection had to be held during the COVID-19 pandemic. The main results included a clinically relevant depressive symptom rate of 42.2%, an 88% higher risk of depressive symptoms among participants with low levels of behavioral social capital, 90% higher risk in women, 300% higher risk in participants with high levels of perceived stress, and 200% higher risk in participants who do not sleep well. Regular moderate physical activity was a protective factor against depressive symptoms, sleeping well decreased the risk of suffering depressive symptoms by 14%, and smoking increased that risk by 23%.

When comparing results with the study of Backhaus et al. (17) and the prevalence rates of other countries, the rate of depressive symptoms in Colombia is not as low as in Belgium (22%) and not as high as Brazil's (86%) but similar to rate detected in Oman (41%) and Taiwan (40%).

Being a female student, having lower behavioral social capital, high levels of perceived stress, and suffering from poor sleep quality significantly increased the risk of depressive symptoms. The findings regarding social capital are in line with previous findings demonstrating the beneficial effect of social capital on mental health (17, 33–35). Individuals with high behavioral social capital levels may receive more outstanding social support from their social networks. Therefore, it is possible that experiencing less social isolation results in a lower risk of depression (35, 36). Based on some published studies, high levels of social capital have been a protective factor for mental health by preventing not only depression but also anxiety and by favoring healthy lifestyles such as regular physical activity, a balanced diet, and less substance use (21, 33, 37). The literature also highlights the protective role of social capital in reducing stress levels, a known risk factor for depression (38).

The finding regarding gender differences also corresponds with those reported in other studies which suggested that, due to sociocultural factors, women are more likely to express their emotions and, therefore, more readily admit when they have depressive feelings (36).

Another factor associated with clinically relevant depressive symptoms was perceived stress. We found that most of the sample (98.85%) reported having moderate to high stress levels. Of the students who reported high stress levels, 77.50% had a BDI-S > 35. Students often experience stress related to their academic careers, such as lack of time to complete academic activities, exposure to classwork, and academic overload (39). These academic stressors are a subset, among other factors that generate high stress levels in this population. First year students, in particular, are more likely to report higher stress rates due to lack of time for other activities and academic overload. Although upper-level university students continue to perceive these factors as stressors, they attenuate to some extent over time (38).

Regarding lifestyle results, we found that <10% of the students reported participating in vigorous exercise. Of the participants who only engaged in a low-intensity activity, such as walking, almost half (48.96%) reported having depressive symptoms, indicating that moderate and vigorous physical exercise can be a protective factor. These findings are consistent with those reported in the SPLASH study (17). When comparing the results of wave 1 and wave 2, we observed a decrease in physical activity among participants and a corresponding increase of 31.49% who reported sitting for more than 10 h a day. This shift from an active to a sedentary lifestyle could be explained by the impact that this pandemic has had on the routines of individuals. In general, social confinement and working and studying virtually have caused this population to remain more sedentary than it is normally (10). A meta-analysis conducted to evaluate the factors involved in the relationship between physical activity and depression showed a biological component (1). Also, physical activity decreases stress levels, and attending the gym or participating in exercise groups provides new social connections. Hence, the importance of promoting physical activity directly impacts physical and mental health and contributes to increasing the levels of social capital, which in turn is a protective factor against the development of clinically relevant depressive symptoms (35).

Analysis of the paired data showed that tobacco use was found to increase the risk of depression. A study of Korean college students showed that smoking was associated with depression and suicidal behavior, at least partly because adolescents are more vulnerable to nicotine sensitivity, which is associated with effects on cognitive function (40). In this study, a decrease in alcohol and tobacco consumption was suggested in students from wave 1 to wave 2; this could be explained because, during the confinement, establishments were closed, such as restaurants, discos, where young people were accustomed to drinking alcohol regularly (41). Also, as evidenced in this study, most students had to return to live at home with their parents (29.16%), which also explains the decrease in tobacco consumption. In this sense, the influence of context on substance use has been investigated, and contextual changes, such as “moving back home,” could represent an opportunity to facilitate cessation (42).

On the other hand, getting a good night's sleep was a protective factor for the development of depression. It has been shown that sleep, especially in young people, is fundamental for maintaining normal body functions. So young people with poor sleep will negatively impact their neurocognitive functions and emotional wellbeing, which can aggravate academic difficulties (43). Thus, it is important to promote the importance of a healthy lifestyle for its potential to prevent the development of depression (44).

The World Health Organization states that mental disorders are important causes of morbidity and mortality globally, noting that one in four people will develop a mental disorder at some point in their lives (1). It is also essential to highlight the impact that the COVID-19 pandemic has had on the general population. It altered the daily routines of all people globally, which is reflected in the observed increase in mental illnesses and unhealthy lifestyles during the pandemic (45).

We discovered other interesting findings when comparing the results of wave 1 with those of wave 2. Almost 30% of the students returned to live at home with their parents due to the pandemic. Of this group, nearly 38% reported depressive symptoms, which was lower than those who lived at a relative's home, college residence, or in a rented house. A survey conducted among university students in India reported that living at home with parents favored mental health due to the family environment and good relationships with parents (10). In contrast, students who move out of their family homes have to socialize with new people and establish a new social network, which may increase their stress levels and affect their mental health (9).

Our findings suggest that the mental health of university students may have been substantially affected during the COVID-19 pandemic.

Our study shows significant findings on the mental health of Universidad de Los Andes students and their health behaviors. It also shows the value of mental health protective factors, such as social capital. It is necessary to provide opportunities that favor social interaction and promote the creation of interest groups among students, thus strengthening their support networks and levels of social capital. In turn, encourage the practice of sports or activities that promote a healthy lifestyle and help with stress management, positively impacting the mental health of these young people.

In addition, the COVID-19 pandemic has impacted the lives of young adults across the globe. Before the pandemic, university students comprised a population considered particularly vulnerable to mental health problems (36). The COVID-19 pandemic has brought unprecedented stress to students. Our findings corroborate recent studies that high levels of social capital have contributed to some populations being less affected by the measures used to cope with this pandemic (10, 11, 46).

Universidad de los Andes has some general guidelines to promote student wellness. It has a variety of services available, such as counseling, advising, academic support, and leisure time strategies focused on positively impacting the health and wellbeing of students. In this institution, a culture of wellness is promoted throughout the campus. The directors and professors are trained in mental health issues not to diagnose and prescribe but to operate under a wellness mentality in every interaction and detect risks that can be referred to on time. Increasing the availability of places that favor socialization and the practice of sports or other activities that promote a healthy lifestyle will remain fundamental to mitigate the impact of the transition to university life, in this case, aggravated by the current pandemic in our country and other university settings.

Furthermore, we need additional studies that evaluate the social capital in Latin American contexts to help determine the causes of poor health outcomes. We also need studies that contribute to developing and implementing interventions that strengthen social capital in different communities and strategies to promote healthy lifestyles in this population.

This study has some limitations that must be acknowledged. First, the study and data collection fell right into the pandemic, and therefore, different recruitment approaches had to be applied. While the first wave was conducted in person during January and February 2020 and directly on the university campus, the second wave, which took place in August 2020, had to be entirely online. This has resulted in a much lower response rate than expected. Therefore, the results of longitudinal changes have to be interpreted with caution and take into account the potential loss of statistical power in light of the valuable information collected during an unprecedented global public health crisis such as the COVID-19 pandemic. Second, self-reported measures were used concerning questions investigating lifestyles and depressive symptoms, and we cannot exclude recall bias.

Despite these limitations, the present study has important strengths. It should be emphasized that our study is not a cross-sectional study, but uses a two-wave panel design that allows us to make early inferences about the directionality of the effects of social capital on mental health. Furthermore, this is the first study assessing the presence of clinically relevant depressive symptoms and the relationship with levels of social capital, perceived stress, and lifestyles among university students both before and during the COVID-19 pandemic. Therefore, the findings offer important information about the value of social capital and the importance of maintaining lifestyles during times of crisis.

Our study showed relevant findings on the role of social capital in preventing depressive symptoms and promoting healthy lifestyles in this population. This study found that the proportion of first semester university students with clinically depressive symptoms was already elevated before the COVID-19 pandemic and may have increased even further during the pandemic, particularly among those with lower levels of social capital. Thus, students' health has to be prioritized and put at the top of public health agendas, and the implementation of measures strengthening social capital should be considered.

The data set of this study is available upon request to the corresponding author.

The Ethics Committee of the Universidad de los Andes approved this study (act number: 201909242). Written informed assent or consent was obtained, depending on whether participants were adolescents or young adults. Participants were guaranteed confidential treatment of their data and were given the possibility to withdraw from the study at any time. When participants reported having relevant depressive symptoms or suicidal ideas, they received psychological support from the university health services.

LS, PS, and AR collected data at Universidad de los Andes (Colombia). LS, PS, AR, JG, and IB conducted the statistical analyses. All authors participated in revising the results and the manuscript and approved the final version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.