Social support, distress, stress, anxiety, and depression as predictors of suicidal thoughts among selected university students in Bangladesh

This study examines the association of perceived social support with suicidal thoughts among a young adult sample of university students and estimates the degree to which perceived stress, distress, anxiety, and depression may explain the association. A cross-sectional survey was conducted from June to September 2022 among 642 students, selected using the probability proportional to size procedure, at a large university in Bangladesh. We used the Multidimensional Scale of Perceived Social Support (MSPSS) to measure support. The Impact of Events Scale-Revised (IES-R), the Perceived Stress Scale (PSS-10), the Generalized Anxiety Disorders (GAD-7) scale, and the Patient Health Questionnaire (PHQ-9) were used to assess distress, stress, anxiety, and depression. The association between social support and suicidal thoughts was examined using multivariable logistic regression models. Mediation analyses were carried out using the Karlson, Holm, and Breen (KHB) method. We found that perceived social support was associated with lower odds of suicidal thoughts after controlling for other variables (adjusted odds ratio = 0.94; 95% confidence interval = 0.92-0.96). Mediation analysis showed that, after adjusting for potential confounders, the mediating effects among the total effect of perceived social support on suicidal thoughts were 56.1%, 20.8%, 22.5%, 38.8%, and 50.9% for all mental disorders together and four specific adverse mental health outcomes, i.e., perceived distress, stress, anxiety, and depression, respectively. This study demonstrates that perceived social support reduces the likelihood of suicidal thoughts among university students, and mental disorders have a partial mediating effect on the association. Suicide prevention and mental health promotion efforts among university students should consider encouraging students to build and strengthen a strong social support network.