Cross-sectional post COVID-19 study of the relationship between psychological distress and obsessive-compulsive symptoms in Mexico

Introduction

In psychological studies, university students are often analyzed as a homogeneous population, yet even within one institution there can be important differences between certain subgroups of individuals such as women and men, younger versus older students, and among different areas of study. Such characteristics should be considered when assessing the prevalence of psychiatric symptoms and their possible relationship to academic achievement.^1–4 In Mexico, data on the prevalence of psychiatric symptoms in students in public schools are not typically collected, and data that do exist are not widely available. University students are a particularly relevant population for such studies, considering that psychiatric disorders often have their onset during adolescence and early adulthood.^4,5 Furthermore, college entails challenges that are intrinsic to the age and social conditionsthat will have consequences for the individual’s future development and success.⁶

Depression is closely associated with stress and anxiety symptoms, as well as maladaptive behaviors such as smoking, unhealthy diet, lack of exercise and poor sleep habits.^7,8 Further, individuals with anxiety disorders report a worse quality of life, compared to those without high levels of anxiety.⁹ Diverse factors contribute to the onset of depression and other mental health problems in university students, but the number of recent stressful life events is a prominent factor, along with with mental health and eating disorders, substance abuse, suicidal tendencies, among others.^10–14

The COVID-19 pandemic was a time of considerable stress that negatively impacted mental health in ways that are still not completely understood.¹⁵ In Mexico, mental disorders are one of the most disabling diseases with the year 2020 marked by an increase of depression and anxiety in the general population, with a prevalence of 50% for anxiety and 28% for depression.¹⁶ It is necessary to to collect data on the prevalence of anxiety and depression in diverse demographic groups, including undergraduate university students.

As in much of the world, schools and universities were closed and there was a prolonged period of home confinement, in addition to COVID-19 associated sickness and deaths, often precipitating psychological distress.¹⁷ Students’ mental health was hit particularly hard during the pandemic, with degrees of depression, anxiety, and stress above normal levels, which can augment physical and psychiatric problems, including suicidal and self-injurious behaviors.¹⁸ The full impact of pandemic-associated alterations in daily routine and social life on the mental health of university students worldwide is still being assessed. To this end, it is necessary to analyze data collected from various world regions, sociocultural contexts, and time points relative to the disruptions caused by the pandemic. There is a general scarcity of information on the prevalence of psychiatric symptoms in university students of Mexico, both during and after the COVID-19 pandemic. Moreover, while most studies worldwide have focused on possible exacerbation of psychiatric symptoms during the COVID-19 pandemic, few have posed the question of whether the reinitiation of normal activities following the pandemic may have been associated with additional symptom exacerbation. The present study aims to address this question. We predicted that the reinitiation of normal, on-site university activities after pandemic would be associated with a notable increase in symptoms of depression, anxiety and stress.

The objective of the present study was to assess the prevalence and severity of depression, anxiety, and tension/stress with obsessive-compulsive symptoms in a population of undergraduate students in a public university in central Mexico that had recently returned to on-site classes after a prolonged period of home confinement with virtual classes. Additionally, we examined the relationships between obsessive-compulsive symptoms and the constructs of depression, anxiety, and tension/stress.

Method

Study design and participants

Students from different biological/health faculties in a public university (Autonomous University of Tlaxcala, in the city of Tlaxcala, Mexico) were invited to participate in this study, by completing an in-class questionnaire, which were either administered in November of 2022 or in January of 2023. At this university, on-site classes had been suspended for the academic year 2020 – 2021 and replaced by virtual learning. On-site education resumed in the fall semester of 2021. The students that were willing to participate gave informed consent and were allowed as much time as needed to complete the brief questionnaire; most finished within around 10 minutes. The questionnaire described the purpose of the study and contained a statement of informed consent, followed by general questions on age, gender, and academic major, followed by the DASS-21 scale¹⁹ and the OCI-R.²⁰ The final sample comprised 346 (256 females and 90 males) undergraduate students belonging to different academic semesters, ranging in age from 17 to 25 years old. The academic majors of the students were within the areas of odontology, biological sciences, and psychology (psychotherapy). Being a convenience sample, the composition of each group corresponding to academic major was entirely dependent the individual’s willingness to participate in the study, and we assume that academic major had no effect on the individual`s decision to participate. No other method of randomization was applied for grouping the participants.

Instruments

The Depression Anxiety Stress Scale (DASS) is a self-report measure of anxiety, depression and stress. The original DASS scale was based on a dimensional conception of the psychological disorder.²¹ In the present study, we applied the Spanish version that has been validated in a Hispanic sample.¹⁹ The abbreviated version DASS-21, made up of 21 items, has the advantage of being a short self-report instrument that is easy to administer and answer, is in the public domain, and possesses adequate reliability and validity.²² DASS-21 contains three subscales with 7 items each; each subscale was designed to respectively assess the constructs of depression, anxiety and tension/stress. DASS-21 has been used to measure psychological symptoms and distress and has been used for screening and evaluation.²³ The final scores can be compared to established cut-off levels to assign categories of normal, mild, moderate, severe, and extremely severe. This scale has been frequently used with university students and has been correlated with other variables like addiction to internet and smartphones and the impact of the COVID-19 pandemic.^24,25

The Obsessive-Compulsive Inventory-Revised (OCI-R) evaluates the severity of obsessive-compulsive symptoms in clinical and non-clinical individuals.^26–28 The OCI-R has been administered worldwide and has been validated for different cultures and languages. In the present study, we applied the Spanish version of the OCI-R.²⁰ The OCI-R comprises six subscales: washing, checking, ordering, obsessing, hoarding, and neutralizing, each one with three items, increasing the sensitivity and specificity of this test. OCI-R has been used to assess symptoms in different groups of university students to assess the prevalence of obsessive-compulsive symptoms and their correlates.^27,29

Data Analysis

Data was analyzed using SPSS version 17.0. The Chi2 test was performed to compare symptom severity levels of the DASS-21 scale (from normal to extremely severe) between genders, academic majors, and age groups. Pair-wise comparisons were done using the Fisher test. ANOVA or T-tests were applied to compare DASS-21 and OCI-R total and subscale scores between genders, academic majors, and age groups. Where appropriate, post-hoc Tukey tests were applied. Statistical significance was assumed where p<0.05. In order to explore the relationships between the OCI-R subscales and the DASS-21 constructs of depression, anxiety and tension/stress, we applied univariate General Linear Model (GLM) analysis, the dependent variables being DASS-21 depression, anxiety, or tension/stress scores, the OCI-R subscale scores as covariates, and sex, age group and academic major as fixed factors. Tests of skewness and kurtosis indicated that all variables showed a normal distribution (skewness values ranged from -0.013 to 1; kurtosis values ranged from -1.0 to 0.3). Shapiro Wilk test confirmed normal distribution of residuals (p>0.05).

Bioethical considerations

Individuals agreed to participate in the research project with signed informed consent. This study was done in accordance with the Helsinki Declaration of Human Studies and was approved by the Ethical Committee of Faculty of Odontology, Autonomous University of Tlaxcala, Mexico.

Results

Prevalence of depression, anxiety, and tension/stress (DASS-21 scale)

We first categorized individuals within the sample as showing normal, mild, moderate, severe, and extremely severe depression, anxiety, or stress symptoms, based on their DASS-21 scores.²¹ With respect to the entire sample (n=346), 55.6% of all students showed moderate, severe, or very severe symptoms of depression. Women and men did not differ significantly in distribution among severity categories. With respect to anxiety, 76.19% of the sample were categorized with moderate, severe, or extremely severe symptoms, and women and men differed significantly in this regard (Chi-square=22.5, df=4, p<0.05). Fisher post-hoc comparisons indicated differences between women and men in the mild group (p<0.05) and in the extremely severe (p<0.05), with fewer women reporting mild symptoms compared to men and more women reporting extremely severe symptoms. With respect to tension/stress, 63.3% of the sample fell into moderate, severe, or extremely severe categories. Again, men and women differed significantly (Chi-square=10.87, df=4). Fisher post-hoc comparisons indicated that a lower proportion of women than men showed moderate symptoms (p<0.05), while severe stress was more frequent in women than in men (p<0.05; Table 1).

We then compared mean depression, anxiety and stress subscale scores between men and women. Men and women differed significantly with respect to mean anxiety [t (344) =-3.96, p<0.05)] and stress [t (344) =-2.66, p<0.05; Table 2]. When mean symptom scores were compared between age groups, we found that the younger age group (17 – 20 years old) showed more severe anxiety and stress symptoms compared to older students (21 – 25, and over 25 years old). Mean depression scores were also numerically higher in the youngest age group, but this difference did not reach statistical significance (Table 3). With regards to academic major, students in the biological sciences showed higher depression scores than did odontology students (Table 4).

We searched the PubMed database for studies in which the DASS-21 scale was applied to university students, and that reported results in the form of proportions of subjects having normal, mild, moderate, severe, and extremely severe symptoms, according to the established cutoff points.²¹These studies were further categorized as having data collected before the pandemic (6 studies, data collected prior to 2019), during the pandemic (15 studies, data collected during the year 2020), and after the pandemic (9 studies; data collected during and after 2021, when normal activities were largely resumed). Means of the reported proportions were calculated and are shown in Table 5 (see also Supplementary Material 2). Although this analysis suggested an increase in depression associated with the pandemic and post-pandemic epochs, and an increase in anxiety severity associated particularly with the post-pandemic epoch (Table 5, and Supplementary Material 2), an ANOVA comparing these three epochs (pre-pandemic, pandemic, post-pandemic) did not reveal statistically significant differences across these three epochs for any of the depression, anxiety or stress severity categories (data not shown). Symptom severity in the present sample was, in general, consistent with that reported in other post-pandemic studies world-wide (Table 5, Supplementary Material 2).

Prevalence of obsessive-compulsive symptoms (OCI-R scale)

Obsessive-compulsive symptoms also differed significantly between women and men. Specifically, women had significantly greater ordering symptoms, and there was a trend toward greater checking symptoms in women compared to men (Table 6). Younger students had more severe OCI-R total scores, as well as significantly higher obsessing scores, compared to those students 21 – 25 years old (Table 7). When the different academic majors were considered, psychotherapy students showed lower OCI-R total scores, as well as lower washing, obsessing, ordering, checking, and neutralizing scores, compared to biological sciences and/or odontology (Table 8; Supplementary Material 1).

Relationship between OCI-R subscales and DASS-21 depression, anxiety, and tension/stress

We then explored the relationship between OCI-R subscale scores and DASS-21 depression by applying a univariate general linear model (GLM, SPSS software). Sex, age group and academic major were entered as fixed factors, and DASS-21 anxiety and stress were entered as covariates. The GLM was significant (F(13)=37.4, p<0.05, partial Eta2=0.58). Obsessing subscale was a significant predictor of DASS-21 depression (F(1)=9.9, p<0.05, partial Eta2=0.029), while washing, ordering and hoarding showed trend-level significance. (p<0.1) We then constructed a second GLM in which we replaced the obsessing subscale with the three individual items of the obsessing subscale. These items are Item 6: “I find it difficult to control my own thoughts”; Item 12: “I am upset by unpleasant thoughts that come into my mind against my will”, and Item 18: “I frequently get nasty thoughts and have difficulty getting rid of them”. Thus, these three items respectively reflect perceived lack of control over thoughts, negative emotional appraisal of thought, and frequency and persistence of intrusive thought. In the present sample, these three items were only moderately correlated with each other (Pearson correlation coefficients in the range of 0.58 – 0.68). This model was significant (F(15)=32.6 p<0.05, partial Eta2=0.60). Items 6 and 18 of the obsessing subscale showed trend-level significance (p=0.07 and 0.06, respectively; partial Eta2=0.01 for both), while item 12 showed no association (p=0.68; partial Eta2 <0.001). DASS-21 anxiety (F(1)=21.4, p<0.05, partial Eta2=0.06) and stress (F(1)=40.2, p<0.05, partial Eta2=0.11) were also significant predictor variables.

These same models were repeated with DASS-21 anxiety as the dependent variable. Gender, age group and academic major were entered as fixed factors, DASS-21 depression and stress were entered as covariates. The GLM was significant (F(13)=63.7, p<0.05, partial Eta2=0.72). Obsessing subscale was a significant predictor of DASS-21 anxiety (F(1)=13.8, p<0.05, partial Eta2=0.04). We then constructed a second GLM in which we replaced the obsessing subscale with the three individual items of the obsessing subscale. This model was significant (F(15)=55.2, p<0.05, partial Eta2=0.72). Obsessing item 18 showed trend-level significance (p=0.06, partial Eta2=0.012). Female gender (F(1)=14.1, p<0.05, partial Eta2=0.04), DASS-21 depression (F(1)=21.4, p<0.05, partial Eta2=0.06) and stress (F(1)=111.7, p<0.05, partial Eta2=0.25) were also significant predictor variables.

Finally, a GLM was constructed with DASS-21 stress as the dependent variable. Gender, age group and academic major were entered as fixed factors, DASS-21 depression and anxiety were entered as covariates. The GLM was significant (F(13)=61.7, p<0.05, partial Eta2=0.71). The obsessing subscale was not a significant predictor of DASS-21 stress, while the ordering subscale was significant (F(1)=10.6, p<0.05, partial Eta2=0.031). When the ordering subscale was replaced with its individual items, only item 9 was a significant predictor (F(1)=5.4, p<0.05, partial Eta2=0.016). Item 9 is “I get upset if others change the way I have arranged things”, and Items 3 and 15 are “I get upset if objects are not arranged properly”, and “I need things to be arranged in a particular way”, respectively. These items showed low correlations with each other (Pearson’s r = 0.26 – 0.49). DASS-21 anxiety (F(1)=110, ´ p<0.05, partial Eta2=0.25) and depression (F(1)=39.1, p<0.05, partial Eta2=0.11) were also significant predictor variables.

Discussion

The present study revealed notable levels of anxiety, depression, and stress in this Mexican university undergraduate student population. There was a sex difference with respect to symptom severity, with women in general reporting more severe symptoms, and younger students showing more severe symptoms than older ones. There were also differences in symptom severity among the different academic majors: students of biological sciences showed higher depression scores than did students of psychotherapy and odontology. Similar trends were observed for OC symptoms: women showed more severe ordering symptoms than men, psychotherapy majors showed lower overall OC scores, and younger students reported more severe symptoms than older ones. Obsessing subscale scores were significantly associated with depression and anxiety while tension/stress was associated with ordering.

Prevalence and severity of depression, anxiety, tension/stress, and obsessive-compulsive symptoms

According to the severity categories that have been established for the DASS-21, approximately 22%, 13%, and 20% of the present sample reported moderate, severe, and extremely severe depression respectively, while 21%, 10%, and 45% reported moderate, severe, and extremely severe anxiety respectively, and 24%, 24%, and 16% reported moderate, severe, and extremely severe tension/stress, respectively. Overall, these prevalences resembled those reported in pre-pandemic studies of university students of Saudi Arabia,³⁰ Pakistan,²⁴ Turkey,³¹ India,³² Malaysia,¹⁴ China,³³ and Lebanon.³⁴ However, compared to these and other pre-pandemic studies, our sample comprised greater proportions of students that presented severe and extremely severe symptoms, particularly with respect to anxiety and depression. This finding is consistent with many post-pandemic studies of university students, which in general have reported increased severe and extremely severe anxiety and depression, compared to those carried out prior to the pandemic. Thus, if one compares data from pre-pandemic studies (data collected prior to December of 2019) to those in which data were collected post-pandemic (from January of 2021 onward), one observes almost a doubling in the proportion of individuals within the extremely severe anxiety category (See Table 5, and Supplementary Material2). Notably, the timing of post-pandemic studies generally corresponded to the reinitiation of on-site classes; such a drastic change from remote learning under pandemic restrictions to quasi-normalcy could be particularly conducive to anxiety symptoms, as recently suggested by Bhakat et al.³⁵

A recent study of Mexican university students reported that those individuals within the “extremely severe” category of anxiety symptoms were markedly more likely to have a chronic illness, live with a chronically ill person, have themselves been infected with COVID-19 or had contact with infected person, or have experienced the death of someone close to them by COVID-19.³⁶ These same factors were also associated with extremely severe depression and stress. Interestingly, a clear characteristic of those within the “extremely severe” anxiety category was that they were much more likely to perceive that public compliance with COVID-19 restrictions was low.³⁶ This latter finding suggests that perceived lack of controllability with regards to protecting oneself from COVID-19 infection and may have been an important factor that promoted extreme anxiety in these students.

Several studies have identified factors associated with increased DASS-21 scores of depression, anxiety, and stress in university students during and since the pandemic. These factors include younger age,³⁷ being a sexual minority,³⁸ financial worries,^39,40 lack of access to accurate information about COVID-19,⁴¹ and lack of satisfaction with the educational experience during the pandemic.^36,42 Other factors included having a pre-existing chronic disease,^43,44 the illness or death of a close family member or friend,^42,45 excessive internet use and media attention to pandemic-related themes,^40,43,46 the loss of one’s daily routine,⁴⁴ poor sleep^37,42,46 and lack of physical exercise.^{36,39,45,47,48} Finally, pre-existing psychological and emotional factors were also important determinants of the severity of pandemic depression, anxiety and stress; these factors include suffering from a pre-existing mental disorder or family history of depression,^45,49 dysfunctional coping mechanisms,^50,51 loneliness,^37,42,50 lack of talking about pandemic fears and problems,³⁷ fear of infection^42,50 and lack of family and/or social support.^43,44,46 In future pandemics, the effects of these factors might be mitigated by government (local, state, federal) and university policy, as well as by preventative programs aimed at physical and mental health.

In the present study, women showed higher mean anxiety and stress scores than did men, while they did not differ from men with respect to depression. Likewise, a greater proportion of women than men fell within the severe and extremely severe categories of anxiety and stress. This difference between the sexes has been consistently reported in studies that have applied the DASS-21 scale.^14,30,31,37 In the present study, women also showed significantly higher scores on the ordering dimension of the OCI-R and trended toward higher scores on the checking dimension, while their total OCI-R scores did not differ significantly from men. This same sex difference was also observed in Smári⁵² in which checking and ordering scores were significantly elevated in women compared to men.

There was a clear effect of the students’ age on DASS-21 scores: younger students (17-20 years old) had significantly greater anxiety compared to older students, and this general trend was also observed for tension/stress. In several different countries, younger university students and/or those in the initial years of undergraduate study reportedly have more severe depression, anxiety, and tension/stress.^31,37,53 In the present study, younger students also showed more severe OC symptoms, particularly in the obsessing subscale. It is likely that young adults just entering university require some time to adapt to the academic and social conditions that university studies entail, and difficulties in dealing with the transition to university education might affect their mental health and academic performance.

Although the present OCI-R scores were similar to those of a pre-pandemic Turkish study,⁵⁴ in general they were slightly higher than in the sample of university students in which the OCI-R was originally validated,²⁷ as well as compared to many other studies published prior to the pandemic.^20,52,55 Much of this increase in OCI-R scores is likely due to stress provoked by the COVID-19 pandemic and the measures put in place to mitigate its spread. Khosravani et al.⁵⁶ reported that COVID-19 related stress accounted for pandemic-related increases in OC symptoms across symptom dimensions, in a sample of patients with OCD. Similarly, Fontanelle et al.⁵⁷ found that OC symptoms in a general population sample during the pandemic were predicted by female gender, the number of COVID-19 related stressful events, as well as by prior OC symptoms related to fear of harm and symmetry.

Relationship between obsessive compulsive symptoms and depression, anxiety and tension/stress

Depression and OCD share certain “transdiagnostic emotional vulnerabilities” with respect to distress tolerance, anxiety sensitivity, and hedonic responsivenes.⁵⁸ In the present study, we observed a selective positive association between the OCI-R obsessing subscale, and DASS-21 depression and anxiety. The OCI-R obsessing subscale comprises three items that are respectively related to: 1) the perceived lack of control over intrusive thoughts, 2) negative emotional appraisal of intrusive thoughts, and 3) frequency and persistence of intrusive thoughts. This subscale is selectively associated with low distress tolerance⁵⁹; conversely, lower distress tolerance predicted greater obsessions in individuals that experienced a greater number of negative daily life events. Low distress tolerance is suggested to be associated with internalizing symptoms (negative self-evaluation, rumination) in females and externalizing symptoms (substance use disorder, delinquency) in males.⁵⁹ Although we did not assess the construct of distress tolerance in the present sample, we did observe that the obsessing subscale item that reflects the frequency and persistence of intrusive thoughts (Item #18) was selectively associated with depression and anxiety. In future studies it will be important to assess distress tolerance, obsessing and depression/anxiety in the same population, to clarify the relationships between these constructs; specifically, one might predict that low distress tolerance might have a unique relationship with frequency and persistence of intrusive thoughts, and particularly in women.

The present study also revealed relationships between DASS-21 tension/stress and the OCI-R ordering subscale. This finding is consistent with several studies that have reported significant associations between stressful or traumatic life events and the onset of symmetry/ordering and checking OC symptoms, particularly in women.^60,61 Indeed, a recent study⁵⁹ reported a specific association between stressful experiences and OC symmetry symptoms.

Limitations

Since this is a cross-sectional study of students at a single university in Mexico, our conclusions may not fully extrapolate to other contexts. Likewise, we do not have comparable pre-pandemic data on depression, anxiety, stress, or obsessive-compulsive symptoms in this student population, and therefore we cannot unequivocally determine the effects of the COVID-19 pandemic on these symptoms. Nevertheless, our data are consistent with those reported in many other recently published post-pandemic studies of university students. Taken as a whole, this body of literature indicates an important world-wide post-pandemic increase in anxiety, depression and stress. Studies in progress include gathering more detailed socioeconomic information and information on the occurrence of pandemic-related stressful events.

Conclusions

Drawing on the experience and findings from across the past 4 years, some recommendations can be made with regards to caring for the mental health of university students during future pandemics. First and foremost, a holistic approach to mental health education should be emphasized at all educational levels – in normal times as well as during a pandemic. Such educational programs should include information on the important relationship between mental health and basic lifestyle habits, including sleep, physical exercise, diet, and internet use. Second, strategies for remote education should be improved to minimize feelings and perceptions of social isolation. Third, it must be considered that the transition from pandemic living to “normalcy” (resuming on-site education) may be a particularly delicate period with regards to anxiety and depression, and specific educational plans (e.g., mental health educational workshops; e.g., Marazziti et al.⁶²) should be prepared in order to meet this need. Finally, the present study and others have identified several clear vulnerability factors for mental health difficulties of university students: younger age, being female or a sexual minority, previous mental health problems and unhealthy lifestyle habits, maladaptive coping mechanisms, excessive internet use, and lack of family and social support. Inasmuch as possible, such factors should be considered when planning and delivering mental health education and care during a pandemic.

Funding

No funding was received for this work.

Informed consent statement

The project was approved by the Odontology Faculty in July, 2023 (Approval code #153) before administering the questionnaire, the general objective of the study was explained to the subjects, and they were given the option to decline to participate. Each participant in the study signed a statement of informed consent. The data were gathered, and the database was constructed and managed with strict confidentiality and anonymity; no information was collected that could be used to identify any individual subject.

Conflict of interest

The authors declare that they have no conflict of interest.