Abstract
BACKGROUND: Childhood adversity is an established risk factor for depression, yet critical gaps remain in understanding its nuanced impact on emerging adults in academic settings, especially in conjunction with stressful life events and how these associations vary across the severity spectrum of depression. METHODS: We employed probability proportional to size (PPS) sampling to recruit 1,470 undergraduates from a Chinese university. Partial proportional odds models (PPOM) were fitted to analyze data, as Brant tests indicated a violation of the proportional odds assumption specifically for life events (χ²=7.35, p = 0.03). A likelihood ratio test confirmed that the PPOM provided a significantly better fit than the standard proportional odds model (χ²(7) = 28.97, p < 0.001). RESULTS: Compared to students without childhood adversities, those with moderate and high adversities had 1.34-fold (aOR = 1.34, 95% CI: 1.02–1.75) and 1.84-fold (aOR = 1.84, 95% CI: 1.22–2.77) higher odds of depressive symptoms, respectively. Positive childhood experiences showed protective association, with 32% (aOR = 0.68, 95% CI: 0.51–0.91) and 52% (aOR = 0.48, 95% CI: 0.34–0.68) lower odds of depressive symptoms for moderate and high levels. Life events exhibited severity-dependent associations: the aORs increased progressively from mild (2.61, 95% CI: 2.02–3.37) to moderate (4.60, 95% CI: 2.91–7.27) to severe depressive symptoms (7.67, 95% CI: 2.85–20.62). Post-hoc pairwise Wald tests confirmed that these category-specific associations were statistically distinct across adjacent severity thresholds (e.g., mild vs. moderate: χ²(1) = 8.77, p = 0.003). CONCLUSIONS: Our findings reveal a severity-dependent stress-reactivity pattern consistent with stress sensitization principles, where the association between recent life events and depression becomes disproporionately stronger among students with more severe symptoms. The results highlight the importance of considering both childhood adversity and current stressors in campus mental health interventions, with particular attention to students exhibiting moderate-to-severe symptoms. The PPOM approach proved valuable for detecting these severity-specific associations that operationalize a key manifestation of sensitization, which conventional analyses would obscure.