Table 2.
Growth Curve Estimates for Each Model of Time and Baseline to Session 12 Changes for PTSD Symptoms and SI
| Parameter | Condition | b | 95% CI | p |
|---|---|---|---|---|
|
| ||||
| Intercept/Baseline | ||||
| PTSD | 61.12 | [51.27, 70.97] | <.001 | |
| SI | 1.14 | [0.96, 1.32] | <.001 | |
| Linear Time (Days/30) | ||||
| PTSD | −0.90 | [−1.06, −0.74] | <.001 | |
| SI | −0.01 | [−.021, −0.001] | .026 | |
| Log function of time [Ln(time+1)] | ||||
| PTSD | −1.81 | [−3.31, −2.33] | <.001 | |
| SI | −0.05 | [−0.08, −0.02] | .003 | |
| Baseline - S12 Change | Est | d | ||
| PTSD | −17.35 | −1.56 | ||
| SI | −0.33 | −0.27 | ||
Note. Initial estimates from PTSD and SI symptom growth curve models did not fully display in the Mplus output due to program limitations. We divided the temporal variable in all models by a constant (30) to change timing from daily to monthly. Dividing both variables by a constant does not change the results other than altering the coefficient metric. B = unstandardized regression coefficient; 95% CI = 95 percent confidence interval; p = p-value; Est = estimate; d = Cohen’s d effect size (0.20 small, 0.50 medium, 0.80; (Cohen, 1977); S12 = session 12; Days = days since baseline therapy session; SI = suicidal ideation; PTSD = posttraumatic stress disorder.