INTRODUCTION
Mental health problems are gradually becoming a serious medical condition for adolescents. Robust evidence suggests that religiosity may be related to the severity of depressive symptoms [1]. In particular, a previous meta-analysis reported that religiosity seems to have a small positive association with psychological distress [2]. However, the results regarding the association of religious involvement with depressive symptoms were inconsistent [3,4]. Some studies revealed that religiousness was positively associated with longevity [5,6] and negatively related to depressive symptoms [7]. A meta-analysis to analyze the relationship between religious/spiritual belief and depression or depressive symptoms in 147 studies (n=98,975) reported a negative correlation between religious belief and depression symptoms [1]. Particularly, religious beliefs showed a stronger protective effect on depressive symptoms after recent severe stressful life events. Taken together, the relationship between mental health and religiosity deserves further investigation.
Coronavirus disease-2019 (COVID-19) has led to unprecedented and uncertain hazards to mental health globally. In the context of the epidemic, people’s lives are generally affected and stress is widespread, especially among teenagers. Studies based on the general population suggest that young adults may be especially vulnerable to the psychological impact of a COVID-19 outbreak. Religiosity may play a role in relieving psychological stress. Thus, the research on religiosity may provide evidence for alleviating psychologically stressful events all over the world. In particular, in China, only a small proportion of people have religious beliefs and Chinese people are more interested in material interests and fear death. When COVID-19 is popular, people are more afraid of death and have more anxiety and depression, while those with religious beliefs may have better resistance to negative emotions such as anxiety, depression, and fear. Particularly, closing schools and keeping social distance have been implemented to prevent the spread of the COVID-19 virus. Since governments closed schools and substituted teaching at home online, adolescents have been staying at home, relatively isolated from their peers. Everyday reports about the increase in COVID-19 cases and deaths have caused an increase in mental distress among adolescents. Studies in China have shown that COVID-19 might heighten adolescents’ psychological distress and psychiatric impairment during the pandemic [8-10]. However, the study of religious belief in the Chinese population during the epidemic has not been studied yet. Therefore, this study investigated the relationship between religious belief and depressive or anxiety symptoms under COVID-19, which is of great practical significance.
Studies support that religious belief, sex, stress, and family relationships can be associated directly with the symptoms of depression and anxiety [11,12]. In addition, psychological resilience and coping strategies can lower the depressive and anxiety symptoms caused by stressful circumstances [13]. Nevertheless, there is a lack of similar epidemiological studies concerning this subject among Chinese adolescents. Hence, in this study, we investigate the relationship between religious belief, infection exposure to COVID-19, confinement at home and mental status after controlling for the parent-child relationship, different types of childhood trauma, and psychological resilience for coping strategies in adolescents (grade 7, 8, and 9).
METHODS
Subjects and design
This study was carried out from March 20 to 31, 2020 in 5 Chinese cities, including Shanxi, Shandong, Henan, Fujian, and Liaoning. Students in grades 7-9 and aged 12-18 years were recruited and investigated by online survey questionnaire. Before the survey, permission from parents/teachers was obtained. Fifty teachers involved in data collection were trained online to use the “SurveyStar” network platform to ensure the accuracy and reliability of the survey. Then all trained teachers posted the link to the survey to the participants after training and instructed them to complete and submit the survey. All adolescents participated in this study voluntarily. Finally, 11,603 adolescents with valid data were recruited in this study.
This study protocol was approved by the Ethics in Human Research Committee of the Third Affiliated Hospital of Beijing University of Chinese Medicine (No.ZYSY-2019KYKTPJ-21), and each student provided online informed consent.
Assessments
A structured questionnaire was used to record the demographic data, religious beliefs, parent-child relationship, infection exposure, childhood maltreatment, and depressive and anxiety symptoms. In addition, the students were asked whether they held religious beliefs and which type of religious belief they believed. The major religiosity was Buddhism, Christianity, Islam, and others.
About the COVID-19 infection, the students were asked two questions, such as “Is any relative or friend infected with COVID-19?” and “Whether anyone in the community where you live infected with COVID-19?”. If the answers to the 2 questions were yes, the adolescents were grouped with a high exposure risk of infection.
Resilience was assessed by the simplified Chinese version of the Connor-Davidson Resilience Scale (CD-RISC). A higher score of CD-RISC means greater resilience [14]. The CD-RISC shows good reliability and validity in Chinese adolescents [15], and its Cronbach’s α is 0.923.
Childhood abuse was assessed with the Childhood Trauma Questionnaire (CTQ). CTQ scale includes 5 subscales consisting of 27 items: emotional neglect, physical abuse, sexual abuse, emotional abuse and physical neglect. The CTQ has been validated in Chinese adolescents with a Cronbach’s α coefficient of 0.77 [16].
Depressive symptoms of adolescents were assessed by a Chinese version of the Patient Health Questionnaire-9 (PHQ-9) [17], which consists of 9 items. In this study, the cutoff score for depression was >5. Anxiety symptoms were assessed by the Chinese version of the Generalized Anxiety Disorder-7 Scale (GAD-7) [18]. In this study, the cutoff value for anxiety was set to >5.
Statistical analyses
We used Shapiro-Wilk test to test normality for continuous variables. The continuous variables of non-normal distribution were described by using the median and interquartile ranges. Student’s t and the nonparametric Wilcoxon test were conducted to test for significant differences.
All participants were grouped into a religious group and a non-religious group, according to whether they had religious beliefs. The demographic and clinical variables of the two groups were compared using analysis of variance for continuous variables and chi-square for categorical variables. Moreover, the analysis of covariance analysis was performed to test the interaction of religious belief and potential exposure risk on outcomes. To better clarify the impact of religious belief on depressive and anxiety symptoms, we performed a logistic regression to measure the odds ratios (ORs) and 95% confidence intervals (CIs) between religious belief and outcomes. Also, we conducted a linear regression analysis to determine the potential risk and protective factors related to psychological distress in adolescents. Lastly, the logistic regression analysis was conducted to determine the key factors related to outcomes in religious adolescents. Bonferroni correction was applied to adjust for multiple tests.
All analyses were conducted in SPSS, version 20.0 (IBM Corp., Armonk, NY, USA). All p-values were 2-tailed and the significance level was <0.05.
In the statistical analysis, we also set the criteria for invalid data. 1) To test the reliability of the answers, two questions were added to the survey. “I answered all of the questions truthfully.” “I answer all the questions based on my real experiences and thoughts.” If the answer to either of the 2 questions is “no”, the data is determined to be invalid. 2) We checked the IP address of each student. If the questionnaire was repeatedly submitted from the same IP address, it is determined to be invalid.
The sample size was calculated based on the standard deviation of PHQ-9 and GAD-7 scores of depressed adolescents from previous studies [19-21], which was the primary outcome of this study. G*Power 3.1.9.2 program (Heinrich Heine University, Dusseldorf, Germany) was used to perform a sample size estimation. We concluded that the recruitment of 847 participants in the religious group and 10,222 participants in the non-religious group for a total of 11,069 participants is considered to yield a sufficient sample size for 80% power at 0.05 level of significance.
RESULTS
Of all recruited adolescents (grades 7, 8 and 9), 95.4% (11,069/11,603) were valid. Table 1 shows the demographic and psychological characteristics. The mean age was 14.3 years old. The rate of depression in adolescents was 37.5% and the rate of anxiety was 23.4%. The depressive symptoms were associated with sex, parent-child relationship, psychological resilience, and a history of child abuse and neglect assessed by CTQ (all p<0.01).
Prevalence of religious belief and its association with mental status
The rate of religious belief among adolescents was 7.7% (847/11,069). The average age of the religious group was slightly younger than that of the non-religious group (p=0.001). Adolescents with religious beliefs who had a bad parent-child relationship were less likely to live with parents and had lower psychological resilience of CD-RISC score (Bonferroni corrected all p<0.05). Moreover, adolescents with religious beliefs had higher emotional abuse, physical abuse, sexual abuse, emotional neglect, and physical neglect (Bonferroni corrected all p<0.01) (Table 1).
The rate of depressive symptoms was different between the adolescents with religious beliefs and non-religious beliefs (43.8% vs. 37.0%). The rate of anxiety also was significantly different between the adolescents with religious beliefs and non-religious beliefs (30.8% vs. 22.8%). Logistic regression analysis found that religious belief was a risk factor for depressive symptoms (p=0.001) and anxiety symptoms (p<0.001) (Tables 2 and 3). Further analysis found that compared to those non-religious adolescents, religious adolescents had a higher severity of depressive symptoms (F=4.8, p<0.001) and anxiety symptoms (F=4.9, p<0.001). After controlling for age, sex, parent-child relationship, resilience, and child abuse experience, the differences between the two groups remained significant.
Associations between infection exposure to COVID-19, religious belief and mental status
All adolescents were divided into two groups, according to whether the adolescent’s relatives were infected or there were infected people in the community. We found significant differences in depressive symptoms (F=24.6, p<0.001) and anxiety symptoms (F=18.8, p<0.001) between the two groups, after controlling for the confounding variables. We further analyze the interaction effects of infection exposure and religious beliefs on depressive symptoms and anxiety symptoms. The results showed that there was no significant interactive effect of infection exposure and religious belief on depressive symptoms (F=0.10, p=0.762) and anxiety symptoms (F=0.458, p=0.458).
Factors associated with depressive symptoms and anxiety symptoms
Multiple regression analysis and logistic regression were carried out to identify demographic and psychological characteristics that were related to depressive symptoms. In this model, the religious belief, age, sex, marital status of parents, parent-child relationship, living with parents or not, child abuse, psychological resilience, and infection exposure were used as dependent variables and depressive symptoms were added as independent variables. We found that religious belief, older age, girl, infection exposure, poor parent-child relationship, child abuse, and low psychological resilience score were associated with depressive symptoms (p<0.05, R2=0.41).
Multiple regression and logistic regression analyses were performed to determine the key risk and protective factors for adolescents’ anxiety symptoms. We found that religious belief, girls, older age, infection exposure, poor parent-child relationship, the low score of resilience, and child abuse experiences were associated with depressive symptoms (p<0.01, R2=0.34). The coefficients of the variables are shown in Table 3.
Behavioral protective and risk factors in adolescents with religious believers
Multivariable logistic regression analysis revealed that in 847 adolescents with religious belief, greater resilience was protective in preventing depressive (OR: 0.60; 95% CI 0.47-0.75; p<0.001) and anxiety symptoms (OR: 0.51; 95% CI 0.39-0.65; p<0.001). Adolescents reporting emotional abuse were more likely to develop depressive (OR: 1.39; 95% CI 1.30-1.50; p<0.001) and anxiety symptoms (OR: 1.37; 95% CI 1.27-1.47; p<0.001) (Table 4). Another risk factor was relevant to depressive symptoms alone including the poor parent-child relationship (OR: 1.52; 95% CI 1.04-2.21; p=0.029). Adolescents reporting emotional neglect were only more likely to develop anxiety symptoms (OR: 1.07; 95% CI 1.02-1.12; p=0.002), not depressive symptoms (Table 4).
DISCUSSION
Our study has three major findings. 1) Adolescents with religious beliefs had more severe symptoms of anxiety and depression. 2) Even though infection exposure risk was a strong risk factor for anxiety and depression during confinement, we found no significant interaction effect of infection exposure risk and religious belief on both anxiety symptoms and depressive symptoms. 3) High psychological resilience was protective in depressive and anxiety symptoms, while poor parent-child relationship and child trauma experience were risk factors in religious adolescents.
Our findings of more severe depressive and anxiety symptoms in adolescents are in line with previous studies [22,23]. However, there is a great inconsistency in their associations with different ages and sex, cultural backgrounds and different religious orientations [1,2,5,6,24,25]. Recent studies have shown two opposite epidemiological forces of religion, which lead to opposite directions in the association between religiosity and mental health [26]. Some adolescents maintain religious beliefs intrinsically [27]. Kendler et al. [28] conducted a genetic epidemiological study of adult female twins with an average age of 30.1 years (n=2,163) and found that broad heritability accounted for 29% of the variance in religious beliefs. In addition, adolescents with depressive symptoms may seek and maintain religious beliefs. However, religious adolescents may be treated differently from those non-religious adolescents in China, which may cause negative emotions and anxiety.
Previous studies by Brooks et al. [29] identified the fear of infection as a primary stressor during quarantine, and a study also showed that quarantine during the outbreak of severe acute respiratory syndrome was correlated with depressive symptoms. Therefore, we also analyzed the impact of potential COVID-19 exposure in the community or relatives during home confinement on the mental status of adolescents. We found that after home confinement for two months due to the amnestic, those knowing that their neighbors or relatives were infected with COVID-19 had a significantly higher rate of depressive symptoms and anxiety symptoms than those without this risk. Contrary to our expectation, we found no significant interaction effect of religious belief×infection risk on psychiatric symptoms of adolescents. Namely, we found that the exposure risk does not outweigh the risk of religious belief in increasing those adolescents’ anxiety and depressive symptoms. Our finding is inconsistent with a previous study [30], which reported that religious beliefs were more protective in individuals exposed to severe recent stressful life events. Nonetheless, their result was limited by a small sample size and borderline significance, especially since the participants were specifically catholic or protestant Caucasians who were the offspring of depressed parents in the negative life events exposed cohort. Several reasons may explain why there was no further negative psychological impact of infection exposure and home confinement on adolescents with religious belief in this study compared with those without belief. First, only 2.2% of all adolescents (2.8% in religious adolescents and 2.2% in non-religious) reported infection in their neighborhood or relatives, therefore, most religious adolescents had no fear of being infected. In the provinces we investigated, the incidence of COVID-19 is relatively low, which might lead to low levels of anxiety and depressive symptoms. Second, since all the schools were shut down and all the adolescents were at home, there was no stigma. In addition, exemptions from multiple school examinations and strict rules and regulations may help reduce anxiety and depression during home confinement [31,32].
We further analyzed the risk factors of depressive and anxiety symptoms in adolescents with religious beliefs and showed that childhood emotional abuse, physical abuse, and poor parent-child relationship as stressors stimulated more severe psychiatric symptoms in religious adolescents. This finding is in line with previous studies on adolescents from China, which show that with an increase in emotional or physical abuse scores, the risk of suicidal attempts increases [33]. Moreover, strong associations are also observed between childhood emotional abuse and major depressive disorders [34], and having poor family relationships leads to anxiety and depression in adolescents [35]. Family support plays an important role in the mental health of those adolescents with religious beliefs. The poor parent-child relationship is regarded as one of the critical vulnerabilities and risk factors related to psychological disorders in religious adolescents. On the contrary, we found that resilience was a protective factor for the symptoms of depression and anxiety in religious adolescents. Studies have shown that psychological resilience moderates the effects of adverse circumstances and childhood abuse among adolescents [13,36]. In summary, our results suggest that adolescents with religious beliefs should be regularly monitored by parents and education workers, especially those who have poor parent-child relationships and emotional or physical abuse.
Our study has several limitations. First, this is an online survey rather than a survey using paper and pencil questionnaires. Although the online survey is much faster and easier, it may have biases in the assessment of the rate of religiosity, depression, and anxiety due to the sampling problems. Second, a methodological limitation was that we used a very short 2-item interview to assess religious beliefs. We did not comprehensively measure other aspects of religion such as religious participation, spiritual help-seeking, religious coping, or satisfaction with religious affiliation. If we used more comprehensive measures, we would have detected more relationships between religious belief and mental health. Third, in this study, all the data were self-reports. The self-reports from the adolescents might have limited the accuracy of some data such as parent-child relationship assessments. In particular, the levels of depressive and anxiety symptoms may not be in accordance with the assessment of experienced psychiatrists.
In summary, this study found that compared to those without religious beliefs, adolescents with belief showed greater severity of anxiety and depressive symptoms. Exposure risk to COVID-19 increases the severity of symptoms of depression and anxiety, however, those with religious belief do not have greater severity of psychiatric symptoms after home confinement and exposure risk to COVID-19 than those without religious belief and exposure risk. Moreover, emotional abuse, resilience scores and poor parent-child relationships were associated with psychiatric symptoms in adolescents with religious beliefs. Further research using more sensitive scales of religious beliefs, and both linear and non-linear statistical methods is needed to replicate the findings.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
