Skip to main content
. 2021 Jul 20;11:391. doi: 10.1038/s41398-021-01498-1

Table 3.

Univariate and multivariate associations of adversity during sensitive periods and mental health symptoms with the M1/M2 phenotype.

M1/M2 phenotype
β t-test p-value
Univariate associations (Mental health)
 PTSD symptom severity (PSSI), df = 40 0.180 1.157 0.254
 Depressive symptoms (CESD-R), df = 40 0.074 0.470 0.641
Multivariate model 1 (Adversity only, df = 39)
 Early life adversity 0.297 2.049 0.047*
 Adversity in pregnancy 0.291 2.008 0.052†
Multivariate model 2 (Adversity and mental health, both)
 Early life adversity 0.330 2.171 0.036*
 Adversity in pregnancy 0.312 1.916 0.063†
 PTSD symptom severity (PSSI) 0.123 0.554 0.583
 Depressive symptoms (CESD-R) −0.224 −1.042 0.304

Univariate associations show that neither mental health factor is significantly associated with the phenotype in regression analyses (equivalent here to Pearson correlations). Multivariate models 1 and 2 demonstrate the relative independent associations of each sensitive period from one another and also from mental health symptoms, taking into account the small sample size may limit statistical power. Correlation of PSSI and CESD-R: r = 0.731, p < 0.001; Adversity in Early Life and Pregnancy are not significantly correlated with one another. The results of Multivariate Model 2 do not substantially change when only including either PSSI or CESD-R rather than both.

PSSI PTSD symptom scale interview, total score, CESD-R Center for epidemiologic studies depression scale, total score.

*p ≤ 0.05, †p ≤ 0.10.