Study of youth mental well-being strategies in Stockholm municipalities using participatory sessions and text mining (Preprint)

BACKGROUND Youth mental well-being is influenced by socio-economic and environmental factors of their surroundings. Promotion of mental well-being is of importance for society for supporting youth in their development towards adults with good mental health. Different municipalities of Stockholm, Sweden, have adopted strategies to promote youth well-being. However, each municipality has identified different promotion strategies. Thus, it has been hard to track them in terms of their conception, purpose and characteristics. OBJECTIVE Our aim was to use data visualizations developed using a fused dataset, as boundary objects among different stakeholders in municipalities, to identify strategies they for promoting youth mental well-being. Visualization-based boundary objects in collaborative workshops allowed us to introduce stakeholders to data and strategies from other related domains. Analysis of their interactions provided a summary of different strategies used for promoting youth mental well-being. METHODS We used a fused dataset created based on themes affecting youth well-being to develop visualizations. Workshop sessions were conducted via Microsoft Teams or as physical sessions in Swedish and lasted between 60 to 90 minutes. Interactions were recorded with consent from participants. Recordings were transcribed using Ambescript software. We employed matrix factorization with Kullback-Leibler divergence to extracted 1000 features and created 10 topic cluster with 20 top words. We used the identified words and phrases to backtrace within the transcripts and identified dialogues where they were used. We summarized these interactions of participants across all the workshops to identify factors or strategies that were discussed for youth well-being. RESULTS Personnel from the municipalities of Lidingö and Nynäshamn in Stockholm participated in the study. A total of 15 workshops were conducted, 6 with personnel from Lidingö, 6 with personnel from Nynäshamn and 3 with mixed participants. National statistics identified that Lidingö has a larger number of high-income families than Nynäshamn and is closer to the city. The summary indicated that youth well-being strategies were different in the municipalities. Promotion of sports was a common strategy while options for leisure activities differed between them. The municipalities collected different datasets to track the progress of youth in addition to the common data that is nationally mandated. CONCLUSIONS Participants considered the ability to connect and visualize data across domains in each municipality as helpful. They noted that these sessions allowed them to better situate their municipality with respect to the status presented in national statistics. Participants observed that they defined well-being differently in their respective municipalities. Source of stress on youth and their aspirations were different. They noted that there are differences in the strategy and data collected for tracking youth well-being. Maintaining sustainable collaborations across departments within each municipality was identified as a constant challenge.