دانلود مقاله پیش بینی افسردگی و اضطراب جمعیت چین در طول COVID-19
چکیده
1. مقدمه
2. روش ها
3. نتایج
4. بحث
منابع مالی
بیانیه مشارکت نویسنده CRediT
تضاد منافع
قدردانی
منابع
Abstract
1. Introduction
2. Methods
3. Results
4. Discussion
Funding
CRediT authorship contribution statement
Conflict of interest
Acknowledgments
References
چکیده
زمینه و هدف: با توجه به شروع استرس ناگهانی، کووید-19 تا حد زیادی بر بروز افسردگی و اضطراب تأثیر گذاشته است. با این حال، هنوز چالش‌هایی در شناسایی گروه‌های پرخطر افسردگی و اضطراب در طول COVID-19 وجود دارد. مطالعات نشان داده‌اند که چگونه تاب‌آوری و حمایت اجتماعی می‌توانند به عنوان پیش‌بینی‌کننده‌های موثر افسردگی و اضطراب به کار گرفته شوند. این مطالعه با هدف انتخاب بهترین ترکیب از متغیرها از معیارهای تاب آوری، حمایت اجتماعی و ناگویی هیجانی برای پیش بینی افسردگی و اضطراب انجام شده است.
روش‌ها: مدل تقویت شیب شدید (XGBoost1) روی مجموعه داده‌ای شامل داده‌های ۲۹۸۴۱ شرکت‌کننده که در طول همه‌گیری کووید-۱۹ جمع‌آوری شده بود، اعمال شد. تجزیه و تحلیل تمایز بر روی گروه‌های شرکت‌کننده با افسردگی (DE2)، اضطراب (AN3)، افسردگی و اضطراب همراه (DA4)، و گروه کنترل سالم (HC5) انجام شد. همه متغیرها با توجه به اهمیت آنها برای طبقه بندی انتخاب شدند. علاوه بر این، تجزیه و تحلیل با ویژگی های انتخاب شده برای تعیین بهترین ترکیب متغیر انجام شد.
یافته‌ها: میانگین دقت به‌دست‌آمده با سه کار طبقه‌بندی، DE در مقابل HC، AN در مقابل HC، و DA در مقابل HC، 0.78، 0.77 و 0.89 بود. علاوه بر این، ترکیب 19 ویژگی انتخاب شده تقریباً عملکرد یکسانی را با تمام 56 متغیر نشان داد (دقت = 0.75، 0.75، و 0.86).
نتیجه‌گیری: تاب‌آوری، حمایت اجتماعی و برخی از داده‌های جمعیت‌شناختی می‌توانند به طور دقیق DE، AN و DA را از HC متمایز کنند. نتایج را می توان برای اطلاع رسانی شیوه های غربالگری افسردگی و اضطراب مورد استفاده قرار داد. علاوه بر این، عملکرد مدل یک مقیاس محدود شامل تنها 19 ویژگی نشان می دهد که استفاده از یک مقیاس ساده امکان پذیر است.
توجه! این متن ترجمه ماشینی بوده و توسط مترجمین ای ترجمه، ترجمه نشده است.
Abstract
Background
Due to the onset of sudden stress, COVID-19 has greatly impacted the incidence of depression and anxiety. However, challenges still exist in identifying high-risk groups for depression and anxiety during COVID-19. Studies have identified how resilience and social support can be employed as effective predictors of depression and anxiety. This study aims to select the best combination of variables from measures of resilience, social support, and alexithymia for predicting depression and anxiety.
Methods
The eXtreme Gradient Boosting (XGBoost1) model was applied to a dataset including data on 29,841 participants that was collected during the COVID-19 pandemic. Discriminant analyses on groups of participants with depression (DE2), anxiety (AN3), comorbid depression and anxiety (DA4), and healthy controls (HC5), were performed. All variables were selected according to their importance for classification. Further, analyses were performed with selected features to determine the best variable combination.
Results
The mean accuracies achieved by three classification tasks, DE vs HC, AN vs HC, and DA vs HC, were 0.78, 0.77, and 0.89. Further, the combination of 19 selected features almost exhibited the same performance as all 56 variables (accuracies = 0.75, 0.75, and 0.86).
Conclusions
Resilience, social support, and some demographic data can accurately distinguish DE, AN, and DA from HC. The results can be used to inform screening practices for depression and anxiety. Additionally, the model performance of a limited scale including only 19 features indicates that using a simplified scale is feasible.
Introduction
Since its outbreak, COVID-19 rapidly became a pandemic (Wang et al., 2020a). Several factors including demographic characteristics (e.g., gender, occupation, education level, health status) and those related to COVID-19 (e.g., physical symptoms, contact history, worry level, and preventive measures) significantly impacted people's mental health, which, in some cases, further developed into psychiatric disorders (Banerjee and Rai, 2020; Minihan et al., 2020; Wang et al., 2020c; de Figueiredo et al., 2021), such as depression, anxiety, insomnia, and post-traumatic stress symptoms (Bao et al., 2020; Huang and Zhao, 2020; Luo et al., 2020; Shader, 2020; Li et al., 2022). Typically, diagnoses for depression and anxiety depend on the clinical evaluation of symptoms, as well as scales, such as the Patient Health Questionnaire-9 (PHQ-9) and Generalized Anxiety Disorder-7 (GAD-7). However, medical resource shortages during the COVID-19 pandemic made it increasingly challenging to identify these psychiatric disorders and intervene (Emanuel et al., 2020). This necessitated the development of psychiatric screening tools with minimal demand on the already limited resources of clinical staff. Although the aforementioned measures are readily accessible, they only offer short-term evaluations based on patients' subjective experiences, which may only detect the recent abnormal (last two weeks) psychological fluctuations of such patients (Garabiles et al., 2020). Therefore, it is difficult for PHQ-9 and GAD-7 to effectively describe the risk of depression or anxiety. We hope to use some indicators that can describe the risk of depression or anxiety to predict depression and anxiety, so as to quantify the probability of depression and anxiety. In addition, because there are many risk factors related to depression and anxiety, it is difficult for participants to complete if all the factors are included. It may ultimately affect the prediction results. Thus, we hope to find some stable key variables to simplify the whole process without affecting the prediction effect.
Results
3.1. Descriptive and data analysis of demographic characteristics, COVID-19 related factors, current health status, and psychological factors
The results of the descriptive analysis of the 29,841 participants (male: 10,592, female: 19,249) is presented in Table 1. The demographic data reveals that older, higher education level and divorced women are more likely to suffer from depression and anxiety (p<0.01). As for the factors related to the COVID-19 epidemic, having patients with infection (including family members, friends or colleagues) around them, having COVID-19 contact history or being infected are tend to depression and anxiety (p<0.01). In terms of general health status, the worse the general health status, the more prone to depression and anxiety (p<0.01). In psychological assessment, social support and resilience are protective factors of depression and anxiety (p<0.01). The higher the score of SSQ or CDRISC, the less likely to suffer from depression and anxiety. The results of the descriptive analysis showed all factors are related to depression and anxiety. However, it is hard to explain the impact of these variables on depression and anxiety, so it is necessary to further quantify the predictive effect of these variables on depression and anxiety with machine learning models.