Associations between early antidepressant exposure, affective states, and subjective sleep quality among euthymic depressive-onset bipolar patients: a cross-sectional parallel mediation analysis

Bipolar Disorder (BD) is often misdiagnosed as major depressive disorder, leading to prolonged antidepressant use during the early stages. Antidepressant treatment has shown significant effects on affective states and sleep. However, the relationship between early antidepressant use, affective states, and sleep quality has yet to be explored in the BD population. A cross-sectional, single-blind study was conducted in Tianjin Anding Hospital between 1st September 2020 and 30th May 2023. BD patients were categorized into antidepressant treatment (AT) and non-antidepressant treatment (NT) groups based on whether they received early-stage systematic antidepressant treatment. Affective states and subjective sleep quality were measured using the Positive and Negative Affect Schedule (PANAS) and Pittsburgh Sleep Quality Index (PSQI), respectively. Spearman’s correlation examined associations among antidepressant use, affective states, and sleep; mediation effects of affective states were analyzed using a bootstrap method. A total of 124 participants were included. Descriptive statistics showed that the AT group had higher PSQI scores than the NT group (Z = − 5.414, p < 0.001). Preliminary correlation analysis revealed positive correlations between systematic antidepressant use and both scores of PSQI (r = 0.488, p < 0.001) and negative affect (NA) (r = 0.216, p = 0.016). Mediation analysis showed positive correlations between early antidepressant use and scores of PSQI (β = 0.8035, p < 0.001) and NA (β = 0.3804, p = 0.0297), while positive affect scores were negatively correlated with PSQI scores (β = − 0.1968, p = 0.0229). The total effect of antidepressant use on sleep quality was significant (95% bootstrap CI = 1.2940, 3.0595). The indirect effect of early antidepressant use on sleep quality through affective states was not significant (95% bootstrap CI = − 0.2190, 0.3602), but the direct effect was significant (95% bootstrap CI = 1.2467, 3.0220). Early systematic antidepressant use in the cohort is associated with poor subjective sleep quality and higher levels of NA, even in the euthymic phase, emphasizing the need for cautious early use of antidepressants in treating BD and the importance of early diagnosis.