Symptoms of depression, anxiety and stress in health students and impact on quality of life

Objective: to evaluate the association between quality of life and presence of symptoms of depression, anxiety, and stress in college students in the health area. Method: cross-sectional study that included 321 students from undergraduate courses in the health area. Quality of life was measured using the World Health Organization scale, abbreviated version, in the physical, psychological, social relations and environment domains, and symptoms were assessed by the depression, anxiety and stress scale. Multivariate analysis was performed using robust linear regression to evaluate the association between quality of life and symptoms. Results: a negative association was observed between the quality of life and depression symptoms in all domains, while anxiety symptoms showed a negative association in the environment domain, and stress symptoms had a negative association in the psychological domain. Symptom severity was unfavorably associated with quality of life, that is, the greater the symptom severity, the lower the mean scores in all domains. Conclusion: symptoms of depression, anxiety, and stress were prevalent and had a negative impact on students’ quality of life, especially in the presence of depressive symptoms. The decrease in scores was significantly associated with the severity of symptoms.


Introduction
Quality of life (QoL) is a concept that emerges, in contemporary times, as an important indicator of health conditions and effects of treatments and interventions, based on the individual's perception of the various aspects of their life within a comprehensive and complex field (1) .
The QoL of college students is a topic that has been increasingly discussed in health and education, considering the importance that this population assumes in the social and public health context (2) .
College students in the health field often experience significant changes when they enter university, especially when they need to move away from their family environment and migrate to other cities, facing a period of transition that can impact their health and QoL (3)(4) .
Some specificities during the academic experience may negatively impact their physical and mental health, such as the high workload in teaching activities, overload of academic tasks, and curricular internships in hospitals and primary care with stressful situations, many of them related to death and the dying process (5) .
From this perspective, the insertion in a context of high complexity during academic training may contribute to greater vulnerability in the development of mental disorders, notably stress, anxiety, depression, and suicidal ideation, which significantly affect their academic performance and overall health (6) . Moreover, there are indications that health students have a lower QoL when compared to the general population and students from other areas of knowledge and similar age groups, especially in the psychological aspect (7) .
The specificities that involve the health education process, associated with other individual factors that negatively influence the mental health of students, such as genetic load and family support network, may be related to the high prevalence of mental disorders in this population, especially depression and anxiety disorders, which can affect more than 30% of these students (8)(9)(10) . In Brazilian college students, for example, there are indications that depression is one of the main predictors of a worse QoL with a negative impact on academic performance and on their future perspectives both in the professional and personal contexts (11) .
In general, it is observed that there is a gap in the literature, especially in Brazil, regarding multicenter studies that assess the relationship between QoL and symptoms of depression, anxiety, and stress in college students in the health area, encompassing different courses, considering that most of these studies are usually limited to a single course or institution (12)(13)(14) . In this context, most investigations, both national and international, focus on assessing QoL in isolation without elucidating its relationship with aspects of mental health (11,15) .
The assessment of QoL and the characteristics associated with health students, especially those related to mental health, may point to indicators for planning interventions and timely care, seeking to avoid future unfavorable outcomes such as suicide (16) . Thus, this investigation aims to evaluate the association between QoL and the presence of symptoms of depression, anxiety, and stress in health students.

Study type
This is a cross-sectional study conducted with undergraduate health students from public higher education institutions in Minas Gerais.

Scenario in which the data collection took place
The study was conducted in three Federal Institutions part of an epidemiological, multicenter survey, carried out with FIHEs in Minas Gerais, aiming to assess the prevalence of depression, anxiety, and stress among college students (17) .

Time period
The data was collected between the months of May to December 2019.

Population and sample
To determine the sample composition, an eligible population of 5,847 students from the three educational institutions described was considered. The sample size calculation was performed taking into account the following parameters: 30% estimated prevalence for anxiety and depression disorders (18) , a margin of error of 3%, design effect 1.0, and significance level of 5%. The sample size required to meet these parameters was 306 students and was calculated using the OpenEpi ® software.
Proportional stratified sampling was used by means of random drawing with replacement, based on the list of www.eerp.usp.br/rlae 3 Freitas PHB, Meireles AL, Ribeiro IKS, Abreu MNS, Paula W, Cardoso CS enrolled students and the proportional quantity of each university. For the replacement, a quantitative of 30% of students was considered, besides the calculated sample size. Thus, 400 students were contacted.

Selection criteria
The study population consisted of students from nine undergraduate health courses, over 18 years of age, of both genders, who were regularly enrolled in the three universities at the time of the study. Those enrolled in the following health courses at these institutions were eligible: Biomedicine, Physical Education, Nursing, Pharmacy, Physiotherapy, Medicine, Nutrition, Psychology, and Occupational Therapy. We excluded from the study students who were on exchange at the time of data collection and those who presented repeated answers throughout the questionnaire items. For each student excluded or who did not answer the questionnaire (refusal or non-response), a replacement draw was carried out within the same profile.

Data collection and instruments used
Data were collected online by means of a virtual, selfapplied and confidential questionnaire, made available on the online forms platform "Google Forms" and sent by email to each selected student. Access to this questionnaire was possible by smartphone, tablet or computer.
The participants were previously informed: a) that the average time spent to answer the questionnaire was 30 minutes; b) that it was not mandatory to answer all the questions, being free not to answer those they did The results of the individual domains are scaled in a positive direction, i.e., the higher the score, the better the QoL in the last 15 days. In the validation study for Brazil, this instrument showed satisfactory internal consistency, discriminant validity, criterion validity, concurrent validity, and test-retest reliability (19) .
The presence of symptoms of depression, anxiety, and stress was assessed through DASS-21 (Depression, Anxiety and Stress Scale), Depression, Anxiety and Stress Scale adapted and validated for Brazil, presenting adequate characteristics of reliability and validity in the adaptation and validation study (20) . It is composed of 21 affirmative sentences, subdivided into three subscales that assess, by self-report, the symptoms of anxiety, depression, and stress during the last week. Each of these subscales is composed of seven questions, and the answers are obtained according to a 4-point Likert scale (0 to 3). The scores for each subscale are obtained by adding the scores of its items and multiplying the total by two. The scores generate the following symptom severity categories: "normal", "mild/moderate", severe/ very severe".

Study variables
The study variables were: a) QoL indicators To evaluate the association between QoL and symptoms of depression, anxiety, and stress, a robust linear regression analysis was performed, considering that the variables did not have a normal distribution.
The sociodemographic, clinical, and lifestyle variables that were significant (p<0.20) in the bivariate analysis were used as an adjustment in the multivariate analysis, using the backward method. All explanatory variables were tested for multicollinearity. Missing answers were treated as missing, using the pairwise method.
In the multivariate analysis, depression, anxiety, and stress symptoms were initially included, and then the adjustment variables were included one by one in increasing order of p-value from the bivariate. www.eerp.usp.br/rlae 4 Rev. Latino-Am. Enfermagem 2023;31:e3885.
Covariates with p-value < 0.05 remained in the final model. We used the course variable as an adjustment, regardless of the p-value because of its relevance in other investigations (3,5,7,11) .

Results
Out of the 400 students who received the invitation to participate in the study, one was on an exchange program and two presented repeated answers throughout the questionnaire items (exclusions) as well as two refused to participate in the study, and 74 did not answer the questionnaire and did not provide a justification. Thus, 321 students from the health area courses participated in the study, distributed among the three institutions: UFOP (28.9%), UFTM (37.0%), and UFSJ (33.9%).
Regarding the courses that comprised the sample, the    The correlation between depression, anxiety, and stress symptoms and QoL is presented in Table 3. It is observed that all QoL domains were significantly (p-value < 0.05) and negatively (r < 0) correlated with depression, anxiety, and stress, i.e., the greater the severity of symptoms, the lower the QoL scores tended to be in all domains. The strongest correlation observed (r= -0.68) was between the psychological domain and depression.
In contrast, the weakest correlation observed (r = -0. 25) was between the social relations domain and anxiety.  The comparison of QoL scores in the WHOQOLbref scale according to the severity of symptoms of depression, anxiety and stress is presented in Table 4.

Discussion
Symptoms of depression, anxiety, and stress are interconnected phenomena that move between negative affect, emotional discomfort, and physiological changes (6,20) . In this investigation, it was observed that approximately 50% of health students presented mild to  Depressive symptoms are considered the psychological predictor most often associated with low QoL observed in undergraduate health students, especially among nursing and medical students in most domains of the WHOQOLbref (2,9,11,21) . This finding was also demonstrated in this gender, and economic difficulties (8,16,21,23) .
The decrease in QoL observed in health students, especially in the psychological domain, is closely related to the high prevalence of depression symptoms in this population, according to some evidence (8)(9)14,16) . These students experience a higher prevalence of depressive symptoms, especially severe symptoms, when compared to other areas of knowledge, considering the specificities of their training process, such as the challenges and responsibilities that involve caring for human life, including the process of death and dying (7,24) . This high burden of depressive symptoms can cause a negative impact on QoL, with complex risk factors related to individual characteristics (personality traits, for example) and the academic environment (25) . Consistent with the findings of this study, it is estimated that more than 30% of students in this area, both nationally and internationally, present moderate to severe symptoms of depression (9)(10) , which may vary according to different geographic regions and age groups, showing an inverse relationship with QoL (23) .
This scenario is increasingly alarming within higher education institutions and public health, considering the negative impact of these symptoms on the physical and mental health of students with unfavorable outcomes, especially a significant decrease in academic performance, disruption of social interaction, suicidal ideation, and increased risk of suicide (6,14) , which may be reflected in QoL and health services delivery by these future professionals (2,9,15,22) . It is believed that the prevalence of these symptoms is at least partially related to the learning environment, considering the overload of teaching activities, pressure for high performance, contact with challenging situations in practice fields, distance from family members, and low self-esteem (8,14) . Paradoxically, meta-analysis studies suggested that the learning environment, by itself, does not present itself as the preponderant aspect for the high burden of depressive symptoms in this population, but influences the training process, contributing to several negative impacts on the health and QoL of these students in the medium and long term (11,18) .
A high prevalence of anxiety symptoms among students in this study was found, showing mild and This finding can be corroborated by a recent metaanalysis estimating that about one in three medical students presented anxiety symptoms with an overall prevalence of 33.8%, showing a higher prevalence than in the general population, which may substantially impact QoL (26) . Another investigation, which included students from health courses, showed that anxiety symptoms were the most prevalent, affecting 74.6% of students with important participation of stressors related to the academic environment in their development (6) . In Brazil, the prevalence of anxiety symptoms in health students is also above 50%, contributing to the worsening of QoL (24,27) . Although this condition is as debilitating and common as depression, it is still little discussed and diagnosed among college students (27)(28)(29) .
It was observed that anxiety symptoms were negatively associated with QoL in the environment domain. The environment domain involves issues related to financial resources, safety and security, physical environment, and recreation and leisure opportunities (19) .  (15,17,22) .  (12,(24)(25) .
The presence of stress symptoms was negatively associated with QoL in the psychological domain and a high prevalence of moderate and severe symptoms was evidenced. Stress can be defined as a non-specific state of excessive excitement or tension, resulting from the inefficiency or exhaustion of coping strategies, having a complex and close relationship with depression and anxiety (30) . Students who experience high and persistent levels of stress are more susceptible to decreased academic performance, changes in general health status, and the development of major depressive symptoms (4,25,31) . Thus, as demonstrated by this investigation, high levels of stress are associated with worse QoL with a greater impact on the psychological domain, causing a significant decrease in scores according to the severity of symptoms (6,(32)(33) .
This condition has been shown to be common among students of undergraduate courses in the health field.
A prevalence of stress of 42.3% of mild and moderate symptoms and 23.3% of severe and very severe symptoms was observed, possibly impacting the perception of QoL.
Studies carried out with health students both nationally and internationally point out that stress is considered the most prevalent mental health problem among these students, affecting more than half of them (6,13,18) . Besides, there are indications that the prevalence of moderate and severe stress symptoms is higher among health students when compared to other areas of knowledge (32) .
In this study, some sociodemographic aspects, especially female gender and economic class, were used as adjustment variables in the association between symptoms of depression, anxiety and stress, and QoL. These sociodemographic aspects demonstrate a possible influence on the worsening of mental health and QoL in undergraduate health course students. Observational (12,22,28,34) , meta-analysis (8,11) , and longitudinal (23,25) studies conducted with health students indicated that female students showed a significant decrease in QoL scores in the physical and psychological domains when compared to males. This finding is possibly due to the high frequency of mental disorders in female students (18,24,27) . The stresses and difficulties related to economic aspects (income and housing conditions, for example) can also affect the mental health and QoL of these students (11,21,28) . In this context, in the present investigation, even adjusting for socioeconomic, demographic, and physical health characteristics, the symptoms of anxiety, depression, and stress were negatively associated with a worse QoL in all dimensions.
This study has some limitations that deserve discussion. The first one is related to the cross-sectional