Algorithms to predict anxiety among university students | NDT


Introduction

Mental disorder poses a significant challenge which has become one of crucial public health problems among university students. According to epidemiological studies, 25.2–44.0% of college students suffered from anxiety symptoms,1,2 and 15.6–42.0% of college students had depression symptoms.1,3 University students were in an essential stage of transiting from adolescence to adulthood. At this stage, they were vulnerable to anxiety and depression possibly because of environmental changes, academic pressures, or obstacles to their life goals.4 Notably, anxiety and depression are of great harm, which are detrimental to their lives and academic performances and may become the major reasons for school dropout and even suicide.5 According the recent study, approximately 24% of college students were identified as having suicide risk.6 Therefore, it is extremely warranted to develop strategies so as to early identify anxiety and depression.

The identification of anxiety or depression can be realized with the help of risk factors. Studies have shown that gender, body image, years of study, and academic performances were associated with anxiety,7,8 while gender, exercise frequency,3 concern over mistakes,3 poor diet,9 sleep problems,9,10 alcohol abuse,11 and daytime sleepiness3 were significantly associated with depression. Furthermore, some prediction models have been proposed to predict anxiety and depression.12–14 However, those models were created based on varying populations, including patients with diabetes after limb amputation,12 chronic obstructive pulmonary diseases,13 and depression among undergraduate student under Coronavirus Disease 2019 (COVID-19).14 Technically, these prediction models may not be capable of being used to predict anxiety and depression especially among college students. Thus, it is of great necessity for us to develop prediction models among particular university students.

Therefore, in this study, we aim to propose algorithms to predict anxiety and depression among university students so as to identify those who are at a high risk of mental health disorders.

Materials and Methods

Samples

We included and analyzed 881 university students from eight colleges in China in November 2020. The eight universities included Xiamen University of Technology, Chengdu Sport University, Xiamen University, JiMei University, Fujian Agriculture and Forestry University, Southwest Jiaotong University, Shandong Sport University, and Chongqing Normal University. We performed a survey online and students voluntarily responded and filled in the survey based on their real conditions. Student’s basic information, lifestyles, sport habits, comorbidities, and mental health conditions were collected in the survey. A database was constructed based on the survey. Students were excluded in the analysis if he or she was reluctant to take part in the survey, had an age of less than 18 years old, or previously diagnosed with anxiety or depression in the hospital. This study was approved by the Academic Committee and Ethics Board of the Xiamen University of Technology (No. 2020.01). Formal consents were obtained from all participants and their personal information was all anonymized. This study was conducted in line with the Declaration of Helsinki.

Evaluation of Anxiety and Depression

Anxiety and depression were measured using generalized anxiety disorder 7 (GAD-7)15,16 and patient health questionnaire 9 (PHQ-9),17 respectively. The GAD-7 ranged from 0 to 21. A score of 5 to 9 was considered as mild anxiety, 10 to 14 was moderate anxiety, and 15 or above was severe anxiety. PHQ-9, ranging from 0 to 27, had the same classification to the GAD-7: participants who had a score of 4 or less was regarded as no depression, 5 to 9 was mild depression, 10 to 14 was moderate depression, and 15 or above was severe depression. GAD-7 and PHQ-9 were widely used measurement to assess anxiety and depression among university students. The reliabilities and validities of GAD-7 and PHQ-9 were good with the corresponding Cronbach alpha of 0.915 and 0.892, respectively.

Potential Risk Factors

In terms of previous studies and data availabilities, we identified 25 potential risk factors which were used to assess their ability to predict anxiety or depression scores according to the multiple linear regression model. These risk factors included sex (female vs male), age (˂20 years vs ≥20 years), grade levels (first year vs second…



News Read More: Algorithms to predict anxiety among university students | NDT

Get real time updates directly on you device, subscribe now.

Subscribe
Notify of
guest
0 Comments
Inline Feedbacks
View all comments

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More

Get more stuff like this
in your inbox

Subscribe to our mailing list and get interesting stuff and updates to your email inbox.

Thank you for subscribing.

Something went wrong.