| Title: | Evaluating the heterogeneous treatment effects of major depression polygenic risk scores on self-harm using causal forest |
| Journal: | Journal of Mental Health |
| Published: | 13 Apr 2026 |
| Pubmed: | https://pubmed.ncbi.nlm.nih.gov/41969057/ |
| DOI: | https://doi.org/10.1080/09638237.2026.2645539 |
Publication 18019
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
Abstract
BACKGROUND: Self-harm, defined as non-fatal self-inflicted harm regardless of suicidal intent, is a critical global health issue influenced by the interaction between genetic and environmental factors.</p>
AIMS: This study investigates the causal effect of major depression polygenic risk scores (MD PRS) on self-harm in individuals of European ancestry and explores how environmental and personal factors moderate this effect.</p>
METHODS: Using data from 91,778 UK Biobank participants, we compared the top 10% with the highest MD PRS (the genetically vulnerable group) against the remainder. The causal relationship was examined using a causal forest approach to estimate average and heterogeneous treatment effects across non-genetic contributors.</p>
RESULTS: Higher MD PRS was associated with an increased risk of self-harm. Significant heterogeneity was observed: the association was stronger among individuals exposed to physical violence or partner belittlement, but weaker among older adults and those with higher income.</p>
CONCLUSIONS: These findings highlight the importance of personalized prevention. Genetic vulnerability to depression reflects a heightened sensitivity to adverse environments, suggesting that genetic risk should be considered alongside trauma and socioeconomic contexts.</p>
10 Authors
- Eunji Jeong
- Chanhui Lee
- Hyejin Kim
- Hyeona Yu
- Jakyung Lee
- Tae Hyon Ha
- Hong-Hee Won
- Woojae Myung
- Hyo Shin Kang
- Jungkyu Park
1 Application
| Application ID | Title |
|---|---|
| 33002 | Development of a computational method to predict disease risk |
Enabling scientific discoveries that improve human health