Aim This study aims to identify latent subgroups of dyadic coping (DC) patterns among patients with neuromyelitis optica spectrum disorder (NMOSD) and examine factors associated with subgroup classification. Design A cross-sectional study. Methods The study was conducted from June 2023 to August 2024, enrolling 137 patients with NMOSD. Data were collected via standardized instruments: a demographic questionnaire, the Dyadic Coping Inventory (DCI), the 9-item Patient Health Questionnaire (PHQ-9) for depression screening, and the 7-item Generalized Anxiety Disorder Scale (GAD-7). Latent profile analysis (LPA) was performed using Mplus 8.11 to identify coping subgroups, with distal outcomes (PHQ-9 and GAD-7) examined via the Bolck, Croon, and Hagenaars (BCH) method, followed by multivariate logistic regression to assess subgroup associated factor. Results Analysis of the positive and negative coping dimensions revealed three distinct profiles among 137 participants: LP-HN (Low positive-High negative coping, n = 22, 16.1%), MP-MHN (Moderate positive-Mild high negative coping, n = 46, 33.6%), and BC (Balanced coping, n = 69, 50.3%). Significant intergroup differences emerged in premorbid residential status, healthcare payment method, physical activity frequency, the score of PHQ-9 and GAD-7. Multivariate analysis identified healthcare payment method as a significant associated factor. BCH analysis of distal outcomes further demonstrated that the MP-MHN group exhibited the highest levels of anxiety and depression, despite maintaining moderate positive coping. Conclusion NMOSD patients demonstrate heterogeneous DC patterns. The healthcare payment method identified as the key associated factor. The MP-MHN subgroup, with the highest psychological burden despite moderate positive coping, warrants prioritized interventions.