Prevalence of depression and its relationship with quality of life among university students in Macau, Hong Kong and mainland China

There is compelling evidence that depressive symptoms (depression hereafter) are common in university students and are considerably influenced by the given socioeconomic context. Being former European colonies, Macau and Hong Kong are China’s special administrative regions, with different sociocultural and economic background compared to mainland China. This study compared the prevalence of depression in university students between Macau, Hong Kong and mainland China and examined the association between depression and quality of life (QOL). The Beck Depression Inventory-II and the World Health Organization Quality of Life—Brief Version (WHOQOL-BREF) were used to measure depression and QOL, respectively. Altogether, 2,312 university students participated in this study. The overall prevalence of depression was 28.9%; 35.2% in Macau, 41.0% in Hong Kong, and 16.8% in mainland China. Compared to the “No depression” group, students with depression had significantly lower QOL scores in the physical, psychological, social and environmental domains. Factors associated with depression were different between the three study sites. Sleep disturbances and high academic pressure were positively associated with depression in all the three samples. In mainland China, male students (OR = 1.68; 95% CI: 1.10–2.56) were more likely to have depression while those who were interested in their major (OR = 0.45; 95% CI: 0.29–0.69) were less likely to have depression. In Macau, students in Grade 3 (OR = 0.56; 95% CI: 0.36–0.89) and those who were interested in their major (OR = 0.58; 95% CI: 0.42–0.81) or had optimistic perspective about their future (OR = 0.51; 95% CI: 0.36–0.73) were less likely to have depression. Nursing students (OR = 1.86; 95% CI: 1.21–2.87) and students with the average score on major subject less than 65 (OR = 3.13; 95% CI: 1.70–5.78) were more likely to have depression. In Hong Kong, students with optimistic perspective about their future (OR = 0.44; 95% CI: 0.22–0.91) were less prone to have depression. Depression is common among Chinese university students, particularly in Macau and Hong Kong. Considering the negative impact of depression on QOL, regular screening and effective treatments should be offered to this population.

Depressive symptoms (depression hereafter) are serious health challenge for university students 1,2 . University students are in transition from adolescence to adulthood, and they are confronted by considerable stress in relation to academic demands. Further, they are expected to live independently and maintain good social relationship with others, which probably increase the risk of mental health problems. A meta-analysis found that the global prevalence of depression among baccalaureate students was 30.6%, ranging from 7.9 to 84.5% 3 . The variation of prevalence figures across different areas could be due to the use of different instruments, and a variety of sociocultural and environmental factors 4 . In 2018, the number of university students in China was 38.33 million accounting for around one fifth of university students globally 5 . In order to reduce the negative health outcomes of depression and develop preventive measures, it is important to understand the epidemiology and risk factors of depression in this segment of the population. Demographic and clinical characteristics significantly associated with depression in university students include female sex [6][7][8] , family history of depression 4 , and sleep disturbances 9,10 . Good academic results reduced the risk of depression 8,11 . Students who were dissatisfied with their major subject or had learning difficulties were more likely to suffer from depression 8 .
Depressive symptoms such as poor concentration, sadness, loss of interest or pleasure, disturbed sleep or appetite, fatigue, guilty feelings or low self-worth 11 , all have negative impact on quality of life (QOL). QOL is defined as the persons' perception of their lives in the context of their culture and value systems. QOL is related to the person's goals, expectations and concerns 12 . QOL comprises the following components of overall health and well-being: physical and mental health, satisfaction with the environment and social relationships 13 .
The epidemiology of mood disorders is considerably influenced by the given socioeconomic context [14][15][16] , thus the epidemiology of depression in different countries and areas with different socioeconomic backgrounds need to be examined separately. In China, there are special administrative regions which have different sociocultural and political systems, and levels of industrialization and urbanization 17 . Being the former colonies of European countries, the two special administrative regions of China, Hong Kong and Macau, have quite different sociocultural and economic backgrounds compared to the other areas of China. Western upbringing in ways of values, mixed with elements of traditional Chinese culture could increase the risk of depression among young people 18 . To the best of our knowledge, no studies have compared the prevalence of depression in general or in students in particular between Macau, Hong Kong and mainland China. Thus the aim of this study was to examine the prevalence of depression and its relationship with demographic data and QOL in university students in Macau, Hong Kong and mainland China.

Results
Altogether 2,523 students were invited to participate in this study; 2,312 students completed the study yielding a response rate of 91.6%. The overall prevalence of depression was 28 Table 1 shows the basic demographic and clinical characteristics of the whole sample according to study sites. Table 2 presents the comparison between the 'Depression' and 'No-depression' groups in terms of demographic and clinical characteristics and QOL. After controlling for the potential confounders, students without depression had higher QOL in physical (F (11,2312) = 315.7, p < 0.001), psychological(F (11,2312) = 474.6, p < 0.001), social relationship (F (11,2312) = 167.3, p < 0.001), and environmental (F (11,2312) = 165.8, p < 0.001) domains. Table 3 shows the independent correlates of depression by study sites. Factors associated with depression were different between the three study sites. Sleep disturbances and high academic pressure were positively associated with depression in all three samples. In mainland China, male students (OR = 1.68; 95% CI: 1.10-2.56) were more likely to have depression while those who were interested in their major (OR = 0.45; 95% CI: 0.29-0.69) were less likely to have depression. In Macau, students in Grade 3 (OR = 0.56; 95% CI: 0.36-0.89) and those who were interested in their major (OR = 0.58; 95% CI: 0.42-0.81) or had optimistic perspective about their future (OR = 0.51; 95% CI: 0.36-0.73) were less likely to have depression. Nursing students (OR = 1.86; 95% CI: 1.21-2.87) and students with the average score on major subject less than 65 (OR = 3.13; 95% CI: 1.70-5.78) were more likely to have depression. In Hong Kong, students with optimistic perspective about their future (OR = 0.44; 95% CI: 0.22-0.91) were less prone to have depression.

Discussion
To the best of our knowledge, this was the first study that compared the prevalence of depression in university students between Macau, Hong Kong and mainland China. The prevalence of depression was 28.9% in the whole sample; Hong Kong had the highest prevalence, followed by Macau and mainland China. Post-hoc comparisons with Bonferroni correction found that university students in mainland China were less likely to have depression compared to those in Macau (16.8% vs. 35.2%; χ 2 = 82.0, p < 0.001) and Hong Kong (16.8% vs. 41.0%; χ 2 = 95.2, p < 0.001). There was no significant difference in the prevalence of depression between Macau and Hong Kong at the significance level of 0.017 using Bonferroni corrections. The difference between Hong Kong and mainland China was similar to a previous study (43.9% vs. 24.8%) 19 , while the figure in mainland China was similar to findings of a meta-analysis of 39 studies in Chinese university students (23.8%, 95% CI: 19.9-28.5%) 7 .
Socioeconomic and cultural factors are closely associated with the occurrence and subjective experience of depression 20 . Although the socioeconomic and cultural context characterizing the three sites were not measured Scientific RepoRtS | (2020) 10:15798 | https://doi.org/10.1038/s41598-020-72458-w www.nature.com/scientificreports/ due to the lack of standardized instruments in China, we hypothesized that the difference of depression prevalence between the three sites could be partly explained by the interaction and clash between Western lifestyle and traditional Chinese values and culture resulting in the higher rate of depression among Macau and Hong Kong students. Students in Macau and Hong Kong are affected by Western culture and lifestyle although the two cities are Chinese societies. The gap between Western and Chinese traditions may lead to conflicting cultural idioms thereby increasing the risk of depression 19 . Cultural values affect the way people perceive and present with psychological distress, help-seeking behavior and the stigma of mental illness 21,22 . For example, self-control, conformity, and discipline are emphasized in Chinese culture; subthreshold depression, or depressive symptoms are thought to be acceptable to persons growing up with Western values, but in China many people tend to deny depression or only express it with somatic signs and symptoms, which results in lower prevalence of depression 21 . In addition, the higher rate of depression in Macau and Hong Kong could be related to personal characteristics. A comparison between Hong Kong, Shanghai and Beijing students found that Hong Kong students' self-esteem was significantly lower than those from Shanghai and Beijing, which was associated with higher rate of depression in Hong Kong 23 . Moreover, concealing psychological or psychiatric disturbances among students in mainland China could also result in the lower rate 24 . Factors associated with depression were different between the three study sites apart from sleep disturbances and high academic pressure that were both positively associated with depression in each site, similar to previous findings 25,26 . A meta-analysis concluded that people with insomnia were more likely to develop depression 27 . Sleep disturbances and depression share risk factors and biological basis; in addition, sleep disturbances are frequent symptoms of depression 8 . Consistent with previous studies 28,29 , students reporting higher academic pressure were more likely to have depression in this study. Ongoing stress could lead to loneliness, fear, helplessness, uselessness, anger, and guilt among other negative feelings and experiences, particularly in individuals with poor coping mechanisms 30 , who are more liable to psychological problems including depression 31,32 . Furthermore, people with low self-esteem 33 usually have higher level of stress, which is also a contributor to depression.
In this study, an optimistic perspective of the future and greater interest in one's major subject were negatively associated with depression, which echoes the findings in a study involving Turkish university students 34 . Positive www.nature.com/scientificreports/ coping styles and attitude reduce the risk of anxiety and depression 35 . Similar to previous findings 8,36,37 , students with average scores of major subject of less than 65 points were more prone to present with depression. Students with poor academic performance cannot meet their parents' or teachers' expectations, therefore they experience higher stress, anxiety and low self-esteem, paving the way towards depression 38,39 . Conversely, students with psychiatric disorders, such as depression, are more likely to experience difficulties in their studies 40 . Compared to other grades, third grade Macau students were less likely to rate themselves depressed, probably because by then they could adapt better to university life while they are not yet looking for jobs and face uncertainty 41 . Unlike in mainland China and Hong Kong, in Macau people graduating elsewhere cannot get a work permit thus Macau students have less pressure related to job-hunting, which may also lower the risk of depression.
Findings regarding gender differences with respect to the prevalence of depression among college students have been contradictory 42,43 . In this study, male students had a higher prevalence of depression in mainland China, but not in Macau and Hong Kong. This may be related to gender differences in seeking psychological help in China, where male students are less likely to utilize mental health services 44 . Inconsistent with results of previous studies 45 , no association was found between family history of mental disorders and depression in either of the sites. Data on family history of mental disorders were collected by self-report in this study, which method is less reliable due to fear of stigma and discrimination. Nursing students were more prone to have depression (OR = 1.86; 95% CI: 1.21-2.87) in Macau, but not in mainland China. We hypothesized that this may be related www.nature.com/scientificreports/ to the relatively low social status of nursing by the young people. This assumption is supported by the increasingly diminishing rate of applications to nursing schools in Macau. As expected, depression was positively associated with poor QOL in all domains. Depressed students experience low mood, reduced or loss of interest and pleasure, feelings of guilt or low self-worth, disturbed sleep, low energy, poor concentration, impaired functions, all of which significantly affect their QOL 40 .
The strengths of this study are the large sample size, the inclusion of Macau, Hong Kong and mainland China, and the use of standardized instruments. However, the results should be interpreted with caution due to its methodological limitations. First, this was a cross-sectional study, thus, the causal relationships between depression and other variables could not be investigated. Second, due to logistical reasons, university students were recruited only from Beijing and Jilin province in mainland China. The sample size in Hong Kong was relatively small and only one university in this city participated in the study. The relatively small number of participating universities also limits the generalizability of the findings to all university students in China. In addition, the relatively high percentage of nursing students in the study sample is likely to have caused selection bias. Third, due to logistical reasons, interest in major subject and academic pressure were assessed only by two standardized questions. More sophisticated measures, preferably standardized scales should be used in future studies.
In conclusion, depression was common in Chinese university students, particularly in Macau and Hong Kong. Considering the negative impact of depression on their QOL, regular screening for depressive symptoms should be conducted and effective treatments should be offered to this vulnerable segment of the population.

Methods
Study setting and participants. This study was conducted in two nursing colleges (Kiang Wu Nursing College of Macau and Beijing Union Nursing University) and three universities with a number of faculties (University of Macau, Jilin University and Hong Kong ShueYan University) between June and December 2016. In an earlier survey 19 , the prevalence of depression in Hong Kong university students was 43.9%. Given a significance level of 0.05 (two-sided test) and a power of 0.8, at least 436 students were to be recruited in each study site according to the NCSS-PASS program. The University of Macau is the only public university in Macau with around 6,000 undergraduates in eight residential colleges. The Kiang Wu Nursing College of Macau had around 500 nursing students. Jilin University is located in Jilin province with approximately 40,000 undergraduates. Beijing Union Nursing University has around 1,000 undergraduates. Hong Kong Shue Yan University has about 5,000 undergraduates. According to the number of students, grades and residential colleges in participating universities, 2 residential colleges in University of Macau; 4 grades in Kiang Wu Nursing College of Macau; 4 grades in Hong Kong Shue Yan University, 4 grades in Jilin University and 5 grades in Beijing Union Nursing University were randomly selected using computer-generated random numbers. The inclusion criteria were: (1) students aged 18 years and above; (2) fluency in Chinese language (Cantonese or Mandarin); and (3) Willingness to provide written consent and complete the assessment. There were no exclusion criteria. All students in the selected residential colleges and grades were invited to take part in the study on a voluntary basis. Written Table 3. Multiple logistic regression analysis of the correlates of depression by study sites. Bold values are p < 0.05. Controlled variables are demographic and clinical variables that significantly associated with depression in univariate analysis and related factors reported association with depression in previous studies (gender, major subject, and family history of psychiatric disorders). CI confidence interval, OR odds ratio. www.nature.com/scientificreports/ informed consent was required prior to participation. Students who agreed to participate were given the selfreported questionnaire and were asked to return the completed questionnaires within one week. Participants were assured about anonymity and confidentiality. The study protocol was approved by the Ethics Committees of participating universities.

Assessment instruments and evaluation. A paper-and-pen version data collection sheet was designed
to gather demographic and other information, including gender, age, university, grade, major, level of interest in the major subject, academic pressure, academic performance, and perspective on future.
The presence and severity of depression were measured with the validated Chinese version of the Beck Depression Inventory (BDI-II) 46,47 . The BDI-II consists of 21 items, each ranging from 0 ("symptom not present") to 3 ("symptom strongly present"). In this study, a total BDI score of 14 or higher indicated depression 47,48 . Sleep disturbances in the previous month were examined with the following standardized questions 49,50 : difficulty initiating sleep (DIS) ("Have you had difficulty falling sleep in the last month?"); difficulty maintaining sleep (DMS) ("Have you had difficulty maintaining sleep or waking up often in the last month?"); early morning awakening (EMA) ("Have you woken up in the middle of the night or early morning in the last month?"). The answers were recorded as "no", "sometimes" and "often. " The answer of "often" to any of the three questions indicated sleep disturbance 48,51 . Quality of life was evaluated with the World Health Organization Quality of Life-Brief Version (WHOQOL-BREF) 12,19 , which is the most widely used QOL measure in Chinese populations with satisfactory reliability and validity 52,53 . It consists of 26 items covering four domains: physical health, psychological health, and social relationships and environmental factors. A higher score indicates a better QOL.

Statistical analyses.
Data analyses were performed with the SPSS Version 21.0 (Chicago, Ill. USA) for Windows. Normal distribution of continuous variables was measured with Kolmogorov-Smirnov test. Comparisons between students with and without depression in term of socio-demographic and clinical characteristics were made with the independent samples t-test and chi-square test, as appropriate. Multiple chi square tests with Bonferroni correction were conducted to compare the prevalence of depression between Macau, Hong Kong and mainland China. QOL was compared between the 'Depression' and 'No-depression' groups with analysis of covariance (ANCOVA) after controlling for the potentially confounding effects caused by variables that significantly differed between the two groups in univariate analyses. Multivariate logistic regression analysis using the "Enter" method was performed to determine the independent relationships between depression and the sociodemographic and clinical characteristics. Depression was the dependent variable, while demographic and clinical variables that significantly associated with depression in univariate analysis were the independent variables. In addition, commonly reported associated factors of depression among university students, such as gender 42 , major subject 54 , family history of mental disorders 55 were entered as the independent variables. Human ethics. The study was conducted in accordance with the guidelines of the Declaration of Helsinki.
The protocol of the survey has been approved by the Clinical Research Ethics Committee of the University of Macao (Approval number: BSEREAPP002-FHS).