Lower Urinary Tract Symptoms in Mental Illness: A Topic Modeling Approach to Online Mental Health Communities.

Abstract

This study aimed to provide foundational data to enhance integrated interventions by gaining an in-depth understanding of the perceptions of patients experiencing both mental illness and lower urinary tract symptoms (LUTS), which carry a high risk of comorbidity and potential mutual exacerbation. Data were collected from a large online mental health community in South Korea, active among individuals with mental illness and their families (as of October 21, 2024: 113,060 members and 368,352 posts). Posts containing the keywords 'pee' or 'urine' (a total of 986 posts) were analyzed including their titles, content, and categories. Analyses included word cloud, latent Dirichlet allocation topic modeling, category frequency analysis, and qualitative analysis. The findings indicate that among individuals with mental illness, LUTS are perceived as side effects of psychiatric medications and regarded as inevitable. Many patients attempted self-regulation or discontinuation of medication instead of seeking urological treatment, but these attempts often led to failure. The study revealed that participants used the community to explore the relationship between LUTS and mental illness. LUTS had negative impacts on family and social life, and urinary incontinence exacerbated emotional distress such as self-blame and despair. Patients with mental illness experiencing LUTS often engage in self-regulation or discontinuation of medication. It underscores the need for accurate information and warnings about risks. As these patients endure discomfort and frustration in daily life, symptom deterioration is likely and necessitates proactive urological intervention to improve their quality of life. This study enhances understanding of the impact of co-occurring mental illness and LUTS and highlights the need for proper information and guidance. However, it has limitations including reliance on self-reported data and limited sample representativeness. Future research can address these issues by linking medical diagnoses with objective data and expanding the sample size.