This study examines how online communities on Reddit can reveal connections between diabetes and other health conditions, as well as potential early signs of diabetes. We focused on three primary diabetes-related subreddits and their links to 85 other health conditions. By tracking user participation across different health-related communities, we mapped a multimorbidity network, highlighting common co-occurring conditions.

The results show strong links between diabetes and mental health conditions, including depression, anxiety, and ADHD, reinforcing the known psychological burden of diabetes. Additionally, users frequently engaged in discussions about weight loss, reproductive health, and autoimmune diseases, such as fibromyalgia, multiple sclerosis, and chronic pain, highlighting potential multimorbidity concerns.

We also investigated early signals of diabetes by analysing the temporal patterns of individual users. We found that many people discussed mental health, obesity, or pregnancy conditions before participating in diabetes-related subreddits, particularly for Type 2 diabetes. This suggests that digital behavioural data could help detect individuals at higher risk of developing diabetes, offering opportunities for earlier health interventions and support.

The opioid epidemic in the United States is an ongoing public health crisis, causing devastating harms at the individual, interpersonal, community, and societal levels. Morality-based discourse around addiction and treatment, particularly medication-assisted treatment, is pervasive in society. It informs attitudes towards harm reduction efforts, treatment approaches, and the self-image of substance users. It can shape and limit access to supportive alters. Moral foundations theory posits that moral attitudes and evaluations are based in a small number of “foundations” or dimensions.

Online social support from communities can play a key role in recovery. Prior work has examined online Reddit health communities (subreddits) organized around recovery and management of opioid use disorder, and the actors who engage in opioid and treatment-related conversations within these communities, from the perspective of moral foundations. It found that moral content is common, moral profiles within communities reflects their specific purposes and values, and that central actors are often prominent in moral discourse.

In this work, we examine central actors more closely, exploring their moral expressiveness towards alters, and their role in shaping the moral tenor of the network, its cohesiveness, and other network properties. These findings could provide insight into the effectiveness of these health communities, and the influencers within them.

This is me is the largest online counselling service for young people in Slovenia. Since its establishment in 2001 at the National Institute of Public Health, more than 57,000 questions have been answered. Safety, privacy and optimal user experience is managed by a professional editorial office, that takes care of continuous online communication between adolescents and counsellors (doctors, psychologists, teachers, social workers and other experts). Part of the daily work of the editorial office is also to assign a topic to each question received. The most common topics of questions relate to relationships and mental health (about half of all questions) and to body and physical health (about a quarter of all questions). Some topics appear together more frequently than others. Understanding the co-occurrence of issues is crucial as it provides a deeper insight into the complex challenges adolescents face. This understanding can help to develop more comprehensive and effective support strategies for addressing mental illness comorbidity and also enable the adaptation and improvement of other activities within the This is me prevention programme, ensuring safe and positive user experiences for those who seek help. Therefore, the results of the exploratory analysis applied will be presented using the social network methodology. Here, the nodes represent the topics and the links between the topics represent the extent of co-occurrence. Since the co-occurrence depends on the frequency of occurrence of the individual topics, the links are normalised using Jaccard normalization. Special attention is paid to the most severe topics (e.g., suicide, anxiety, depression, self-harm behaviour, and others) and other selected topics. The results show an increasing trend in mental health issues, with specific topics strongly influenced by age and gender. Several topics that require an interdisciplinary approach (i.e., health professionals from different fields) were identified, such as domestic violence, eating disorders, sleep disorders, sexual orientation and identity, and others. Understanding the topics co-occurrence can enhance online support platforms, positioning them as an important first source of (mental) health support, especially for adolescents, who might not otherwise take the first step toward seeking help.