Ching-Fu Yeh, Wen-Han Chang, Chen-Yun Cheng, Jiann-Horng Yeh, Ming-Hsing Chang
Background: Depressive symptoms are highly prevalent among patients with thymoma-associated myasthenia gravis (MG) and are associated with poorer clinical status and quality of life. However, their relationship with psychological counseling outcomes remains insufficiently investigated. Objectives: This study aimed to examine the association between baseline depressive tendency, baseline clinical status, and post-counseling clinical outcomes in patients with thymoma-associated MG. Methods: This single-group pre-post exploratory pilot study enrolled 17 patients with thymoma-associated MG who received eight sessions of Gestalt-oriented psychological counseling. A depressive tendency was defined as a baseline Patient Health Questionnaire-9 (PHQ-9) score ≥10 (n = 6, 35.3%). Clinical outcomes, including the Myasthenia Gravis Activities of Daily Living scale (MG-ADL), Myasthenia Gravis Quality of Life 15-item scale (MG-QoL15), mental BMI (mBMI), and PHQ-9 scores, were assessed before and after the intervention. Multivariable linear regression analyses were performed using baseline depressive tendency as the primary predictor of each post-counseling outcome. The models were sequentially adjusted for the corresponding baseline outcome score and subsequently for age and sex. Results: Following the counseling intervention, statistically significant pre-post changes were observed in MG-ADL (p = 0.044), mBMI (p = 0.018), and PHQ-9 (p = 0.002) scores. After adjustment for the corresponding baseline outcome score, age, and sex, baseline depressive tendency was associated with less favorable post-counseling mBMI (β = -3.52, robust 95% CI -6.19 to -0.86, p = 0.014) and MG-ADL scores (β = 1.70, robust 95% CI 0.51 to 2.89, p = 0.009). Conclusions: Baseline depressive tendency was associated with poorer baseline functional and disease-specific quality-of-life status and with less favorable adjusted post-counseling MG-ADL and mBMI scores. These associations met the Benjamini-Hochberg threshold but not the more conservative Benjamini-Yekutieli threshold. Given the small sample and uncontrolled pilot design, the findings should not be regarded as evidence of counseling effectiveness and require confirmation in prospective controlled studies.