Table 3 Results of parallel four-mediator model in the discovery sample (UCC).

From: Childhood maltreatment mediates the effect of the genetic background on psychosis risk in young adults

Regressions

Unstandardized B estimate (95% CI)

Standardized estimate

p value

Direct effects

   

Outcome model: PLE (R2 = 0.424)

   

 SZ-PRS

0.187 (0.035; 0.339)

0.072

0.016

 CM

0.223 (0.117; 0.328)

0.143

<0.001

 Cannabis use

2.78 (0.538; 5.02)

0.094

0.015

 Neuroticism

0.302 (0.265; 0.338)

0.541

<0.001

 Nicotine

2.57 (0.168; 4.61)

0.082

0.035

Mediator model: CM (R2 = 0.011)

   

 SZ-PRS

0.172 (0.036; 0.307)

0.103

0.013

Mediator model: cannabis use (R2 = 0.021)

   

 SZ-PRS

0.013 (0.006; 0.019)

0.144

<0.001

Mediator model: neuroticism (R2 = 0.003)

   

 SZ-PRS

0.247 (−0.120; 0.614)

0.053

0.186

Mediator model: nicotine (R2 = 0.018)

   

 SZ-PRS

0.012 (0.006; 0.018)

0.136

<0.001

Covariances

   

CM—cannabis use

0.389 (0.065; 0.713)

0.101

0.018

CM—neuroticism

80.89 (62.81; 98.97)

0.390

<0.001

CM—nicotine

0.149 (−0.183; 0.481)

0.038

0.379

Cannabis—nicotine

0.129 (0.112; 0.146)

0.628

<0.001

Cannabis—neuroticism

0.080 (−0.791; 0.951)

0.007

0.858

Neuroticism—nicotine

0.472 (−0.423; 1.37)

0.043

0.301

Indirect effects (proportion mediated)

   

CM (10.5%)

0.038 (0.003; 0.073)

0.015

0.033

Cannabis (9.6%)

0.035 (0.002; 0.068)

0.013

0.038

Neuroticism (20.7%)

0.075 (−0.036; 0.186)

0.029

0.188

Nicotine (8.0%)

0.029 (−0.002; 0.059)

0.011

0.065

Sum

0.176 (0.045; 0.308)

0.068

0.009

Total effect

0.363 (0.181; 0.545)

0.140

<0.001

  1. 95% CI 95% bias-corrected bootstrap confidence interval, PLE psychotic-like experiences, CM childhood maltreatment, SZ-PRS schizophrenia polygenic risk score at p value threshold 0.5.