Table 3.
Binary logistic regressions classifying SB history (Model 1) & predicting SB risk (Model 2)
| Model 1: Cross-sectional (N = 131) | Model 2: Longitudinal (n = 87) | |||||
|---|---|---|---|---|---|---|
| Covariates | B (SE) | Odds Ratio (95% CI) | Wald | B (SE) | Odds Ratio (95% CI) | Wald |
| Education | −0.01 (0.08) | 1.00 (0.85–1.17) | 0.00 | 0.24 (0.10) | 1.03 (0.85–1.24) | 0.05 |
| SI Frequency (lifetime) | −0.17 (0.34) | 0.92 (0.48–1.79) | 0.08 | −0.62 (0.68) | 0.48 (0.13–1.84) | 2.04 |
| SI Frequency (past-month) | −0.15 (0.31) | 0.93 (0.51–1.70) | 0.07 | 0.33 (0.64) | 1.48 (0.42–5.18) | 0.64 |
| SI Severity (lifetime) | 2.05 (0.44) | 3.04 (1.27–7.26) | 27.26* | 0.86 (0.71) | 2.96 (0.74–11.86) | 2.99 |
| SI Severity (past-month) | 1.37 (0.44) | 2.05 (0.87 – 4.86) | 3.81 | −0.29 (0.83) | 0.67 (0.13–3.42) | 0.36 |
| SB History (lifetime) | – | – | – | 0.03 (1.46) | 0.63 (0.04–10.86) | 0.11 |
| SB History (past-month) | – | – | – | 0.07 (1.20) | 2.11 (0.20–22.07) | 0.30 |
| Accuracy | −1.03 (0.02) | 0.94 (0.91 – 0.98) | 5.52** | 0.13 (0.01) | 0.99 (0.96–1.02) | 0.38 |
| False Alarms | −0.36 (0.02) | 0.98 (0.95–1.02) | 1.60 | −0.18 (0.04) | 0.93 (0.87–1.00) | 3.73* |
| Misses | −0.26 (0.04) | 0.98 (0.91–1.05) | 0.98 | 0.40 (0.04) | 1.04 (0.96–1.14) | 0.95 |
| NRB | −0.25 (0.03) | 0.98 (0.92–1.04) | 0.51 | 0.24 (0.03) | 1.03 (0.96–1.09) | 0.45 |
| Step 1: Childhood maltreatment | ||||||
| CTQ Total | 0.40 (0.01) | 1.02 (1.00–1.05) | 1.33 | 0.33 (0.02) | 1.01 (0.97–1.05) | 0.20 |
| Model fit: χ2(117) = 91.48, p < 0.001, McFadden R2 = 0.52 | χ2(74) = 57.62, p < 0.001, McFadden R2 = 0.48 | |||||
| Step 2: Emotional response inhibition | ||||||
| PEAT | 0.20 (0.01) | 1.01 (0.96–1.06) | 0.14 | 0.15 (0.06) | 1.08 (0.96–1.21) | 2.57 |
| NEAT * | 1.04 (0.02) | 1.05 (1.00–1.10) | 4.05* | 0.23 (0.04) | 1.09 (1.01–1.17) | 5.02* |
| Full model fit: χ2(115) = 96.19, p < 0.001, McFadden R2 = 0.55 | χ2(72) = 64.60, p < 0.01, McFadden R2 = 0.54 | |||||
Note. 95% CI = 95% Confidence interval (odds ratio scale); SI = Suicidal Ideation; NRB = Negative Response Bias, i.e., the proportion of negative judgments during no-signal trials, regardless of stimulus valence; PEAT = Positive Emotional Action Termination, i.e., percentage of positive false alarms to positive stimuli presented during stop-signal trials on the Emotional Stop-Signal Task. All predictor variables measured at baseline; standardized parameter estimates with robust standard errors are shown at entry step;
*
p < .05;
**
p < .01;
***
p < .001.