Predicting the Effectiveness of a Mindfulness Virtual Community Intervention for University Students Targeting Symptoms of Depression, Anxiety and Stress (Preprint)

Abstract

Students’ mental health crisis has been recognized before COVID-19 pandemic and deepened during the pandemic. Mindfulness Virtual Community (MVC), an eight-week web-based mindfulness and cognitive behavioural therapy (CBT) program and featuring online videos, discussion forums, and videoconferencing, has proven to be an effective web-based program to reduce symptoms of depression, anxiety, and stress. Predicting the success of MVC before a student enrolls in the program is important to advise students’ accordingly.

Prediction of the effectiveness (i.e., success) of MVC in reducing symptoms of depression, anxiety, and stress with undergraduate students at a large Canadian university.

Machine learning models were developed to assess MVC’s effectiveness defined as success in reducing symptoms of depression, anxiety as measured using the Patient Health Questionniare-9 (PHQ9), the Beck Anxiety Inventory (BAI), and the Perceived Stress Scale (PSS), to at least the minimal clinically important difference (MCID). A dataset representing a sample of undergraduate students (n = 209) who took the MVC intervention between Fall 2017 and Fall 2018 was used. Several algorithms were trained based on the dataset to predict MVC’s effectiveness using sociodemographic and self-reported data.

Random Forest and Gradient Boosting (AUC=.89, Accuracy=.88) achieved the best performance both in terms of AUC and accuracy for predicting PHQ 9; and SVM (AUC=.91, Accuracy=.92) had best performance for predicting BAI, while random forest and several other algorithms were the best performing in predicting intervention effectiveness for PSS (AUC=.1, Accuracy=.1). The exposure to online mindfulness videos was the most important predictor for the intervention’s effectiveness for PHQ9, BAI and PSS.

The performances of the random forest models to predict MVC intervention effectiveness for depression, anxiety, and stress, are very high. These models might be useful for professionals to advise students early enough on taking the intervention or choose other alternatives. The students’ exposure to online mindfulness videos is the most important predictor for the effectiveness for the MVC intervention.