A Harman's single factor test was used to examine the presence of common method bias. The results revealed that there were 15 factors with eigenvalues greater than 1, and the first factor explained 29.18% of the variance, which is below the critical threshold of 40%. This indicates that the likelihood of common method bias in this study is relatively low.
Descriptive Statistics and Correlation Analysis
This study conducted clinical assessments on 129 patients with obsessive-compulsive disorder (OCD). Of these, 16 patients were excluded due to having Hamilton Depression Rating Scale (HAMD) scores greater than 24, and 12 were excluded due to Hamilton Anxiety Rating Scale (HAMA) scores exceeding 24. Additionally, four patients were excluded because their assessment results were incomplete. Ultimately, 107 OCD patients were included in the study: 59 males and 48 females, with an average age of 27.61 ± 9.59 years. The average HAMA score was 13.59 ± 3.28, and the average HAMD score was 12.65 ± 4.13. OCD patients' scores on the Scale of Suicidal Ideation and Suicidal Orientation Screening Scale (SIOSS) were consistently below 4, indicating good data reliability. Detailed scores for each variable are provided in Table 1. Correlation analysis showed that suicidal ideation in OCD patients was significantly positively correlated with obsessive-compulsive symptoms, sleep quality, and childhood trauma; there was also a significant positive correlation between obsessive-compulsive symptoms and experiences of childhood trauma, as well as between childhood trauma and sleep disorders.
Testing for Mediation Effects
After standardizing all continuous variables, the mediating role of sleep quality between obsessive-compulsive symptoms and suicidal ideation was tested. The results indicated that obsessive-compulsive symptoms significantly positively affected sleep quality (β=0.31, t=10.67, p<0.001), and both sleep disorders (β=0.20, t=7.86, p<0.001) and obsessive-compulsive symptoms (β=0.39, t=14.65, p<0.001) significantly predicted the risk of suicide. Further analysis using the bias-corrected bootstrap method to test the significance of the mediation effect revealed a significant mediation, with a mediation effect value of 0.08 and a 95% CI of [0.22, 0.28].
The relationship between obsessive-compulsive symptoms and suicide risk: Testing for moderated mediation effects.
First, the impact of obsessive-compulsive symptoms on suicide risk and the direct effect were examined to see if they were moderated by childhood trauma. The results indicated that obsessive-compulsive symptoms significantly positively predicted suicide risk (β=0.28, t=5.98, p<0.001). However, the interaction term between obsessive-compulsive symptoms and childhood trauma did not significantly predict suicide risk (β=0.16, t=8.18, p>0.05).
Next, a moderated mediation model was established to test whether the mediation model of sleep quality on suicide risk was moderated by childhood trauma. The results showed that obsessive-compulsive symptoms significantly positively affected sleep quality (β=0.21, t=10.52, p<0.001), and the interaction term between obsessive-compulsive symptoms and childhood trauma significantly positively predicted suicide risk (β=0.15, t=7.47, p<0.001). Sleep quality significantly positively predicted suicide risk (β=0.14, t=8.34, p<0.001), and the interaction term between sleep quality and childhood trauma also significantly positively predicted suicide risk (β=0.05, t=3.71, p<0.001). The model estimates confirmed the mediating role of sleep quality between obsessive-compulsive symptoms and suicide risk, and both the first half and the second half of this mediation effect were moderated by childhood trauma.(Table 2)
To further understand the nature of the moderating effect of childhood trauma, childhood trauma was divided into high and low groups based on one standard deviation above and below the mean, and simple effects analysis diagrams were drawn (Figure 2 and Figure 3). The results found that for OCD patients with low childhood trauma, as obsessive-compulsive symptoms increased, there was no significant increase in their sleep disorder scores (β=0.03, t=0.66, p>0.05); for OCD patients with high childhood trauma, as obsessive-compulsive symptoms increased, their sleep disorder scores significantly rose (β=0.26, t=5.87, p<0.001). This means that the impact of obsessive-compulsive symptoms on sleep quality intensifies with an increase in childhood trauma, suggesting that childhood trauma enhances the positive predictive effect of obsessive-compulsive symptoms on sleep quality. As sleep disorders increased, for OCD patients with low childhood trauma, their suicide risk showed a significant decreasing trend (β=0.09, t=2.46, p<0.05); for OCD patients with high childhood trauma, as their sleep disorder scores increased, their suicide risk showed a stronger upward trend (β=0.22, t=5.87, p<0.001). This means that the predictive effect of sleep disorders on suicide risk strengthens with an increase in the level of childhood trauma.