Table 2.
Comparison of high-quality (HQ) and lower-quality trials (Other) when compared to passive control conditions at posttreatment
| k | g | 95% CI (PI) |
I2 | NNT | p | |||
|---|---|---|---|---|---|---|---|---|
| High quality (HQ) | HQ | 12 | 0.87*** | 0.61–1.14 (0.03–1.72) |
77.05*** | 2.16 | 0.233 | |
| Other | 65 | 1.14*** | 0.96–1.33 (−0.19 to 2.47) |
81.34*** | 1.72 | |||
| CAPS only | HQ | 8 | 0.78*** | 0.48–1.07 (0.03–1.52) |
70.75** | 2.40 | 0.098 | |
| Other | 32 | 1.21*** | 0.96–1.45 (−0.05 to 2.47) |
80.44*** | 1.65 | |||
| Type of treatment | TF-CBT | HQ | 8 | 0.83*** | 0.54–1.13 (0.08–1.59) |
71.47** | 2.25 | 0.119 |
| Other | 39 | 1.33*** | 1.05–1.61 (−0.28 to 2.94) |
85.77*** | 1.53 | |||
| EMDR | HQ | n.a. (k = 2) | n.a. | |||||
| Other | 5 | 1.16*** | 0.70–1.61 (0.29–2.02) |
52.97 | 1.70 | |||
| OAC | HQ | n.a. (k = 2) | n.a. | |||||
| Other | 19 | 0.78 | 0.60–0.97 (0.24–1.32) |
42.93** | 2.38 | |||
| Treatment format | Individual | HQ | 12 | 0.87*** | 0.61–1.14 (0.03–1.72) |
77.05*** | 2.16 | 0.128 |
| Other | 51 | 1.24*** | 1.02–1.47 (−0.21 to 2.70) |
83.27*** | 1.61 | |||
| Group | HQ | n.a. (k = 0) | n.a. | |||||
| Other | 14 | 0.84*** | 0.56–1.11 (−0.03 to 1.70) |
65.97*** | 2.24 | |||
| k | Intercept | b | I2 | p | ||||
| Number of sessions as a continuous variable | HQ | 12 | 0.79 | 0.01 | 79.40*** | 0.928 | ||
| Other | 65 | 1.36 | −0.02 | 81.41*** | 0.209 | |||
CAPS, clinician-administered PTSD scale; k, number of trials included in the analysis for the given comparison; n.a., not applicable [i.e. number of trials too small (k < 4) to conduct analysis]; OAC, other active conditions; PCC, passive control conditions; PI, prediction interval; TF-CBT, trauma-focused cognitive behavior therapy. p values refer to the comparison of high quality v. other studies and to the significance level of b. For outlier adjustments see Appendix E.
*p < 0.05; **p < 0.01 ***p < 0.001.