Key points are not available for this paper at this time.
Kazakhstan has one of the highest suicide rates globally, yet little is known about how suicide-related content is structured and expressed on VKontakte, the country’s widely used social network. This study aimed to qualitatively analyze VKontakte communities associated with suicide, depression, and self-harm, with a focus on naming conventions, thematic characteristics, and potential indicators relevant for digital prevention strategies. A qualitative content analysis was conducted on 50 public VKontakte communities selected from a larger dataset of 2353 communities collected between December 2021 and March 2025. Communities were included if suicide- or self-harm-related references appeared in their names, descriptions, posts, or visual elements and if they had at least one subscriber with a probable connection to Kazakhstan. Textual and visual content was examined manually at the community level. Six naming typologies were identified: explicitly suicidal, self-harm-focused, depressive, ironic, supportive, and non-related. Community content ranged from direct expressions of suicidal ideation to aestheticized or romanticized representations of pain and death. Some communities contained material that encouraged or normalized self-harm with minimal moderation, while others combined supportive interactions with potentially harmful content. Overall, VKontakte communities linked to users from Kazakhstan represent a heterogeneous digital environment in which supportive and risk-related elements may coexist. These findings highlight challenges for automated detection and suggest that patterns of engagement with specific community types may serve as descriptive indicators for future ethically guided research.
Bex et al. (Tue,) studied this question.