Table 2.
Within-effect sizes (pretest – follow-up) and subgroup analyses for PTSD severity
| k | g | 95% CI | pa | Q | pb | I2 | pc | |
|---|---|---|---|---|---|---|---|---|
| Condition | <0.01 | |||||||
| Active treatment | 32 | 1.36 | 1.14–1.57 | <0.001 | 194.34 | <0.001 | 84.05 | |
| Active control | 11 | 0.59 | 0.14–1.04 | <0.05 | 132.38 | <0.001 | 92.45 | |
| Passive control | 1 | – | ||||||
| Treatment typed | 0.24 | |||||||
| TFT | 25 | 1.44 | 1.20–1.67 | <0.001 | 132.08 | <0.001 | 81.83 | |
| Non-TFT | 7 | 1.08 | 0.52–1.63 | <0.001 | 62.21 | <0.001 | 90.36 | |
| Population | <0.01 | |||||||
| Military | 10 | 0.93 | 0.69–1.18 | <0.001 | 32.38 | <0.001 | 72.21 | |
| Civilian | 21 | 1.58 | 1.30–1.86 | <0.001 | 102.89 | <0.001 | 80.56 | |
| Proportion female | ||||||||
| <50% | 12 | 1.04 | 0.79–1.30 | <0.001 | 47.23 | <0.001 | 76.71 | <0.001 |
| ⩾50% | 18 | 1.62 | 1.34–1.90 | <0.001 | 81.64 | <0.001 | 79.18 | |
| PTSD measure | <0.001 | |||||||
| Interview-based | 28 | 1.44 | 1.20–1.67 | <0.001 | 158.82 | <0.001 | 83.00 | |
| Self-rated | 4 | 0.80 | 0.53–1.07 | <0.001 | 5.16 | 0.16 | 41.91 | |
| Format | 0.35 | |||||||
| Group | 4 | 1.00 | 0.18–1.82 | <0.001 | 47.11 | <0.001 | 93.63 | |
| Individual | 28 | 1.41 | 1.18–1.63 | <0.001 | 146.60 | <0.001 | 81.58 | |
| Number of sessions | 0.71 | |||||||
| <10 sessions | 14 | 1.31 | 1.05–1.57 | <0.001 | 61.40 | <0.001 | 78.82 | |
| ⩾10 sessions | 18 | 1.39 | 1.04–1.75 | <0.001 | 132.88 | <0.001 | 87.21 | |
| Type of analysis1 | 0.43 | |||||||
| Intention-to-treat | 18 | 1.42 | 1.15–1.70 | <0.001 | 112.05 | <0.001 | 84.83 | |
| Completer | 14 | 1.25 | 0.94–1.57 | <0.001 | 59.30 | <0.001 | 78.08 | |
| Follow-up duration | 0.05 | |||||||
| 12 months | 26 | 1.26 | 1.03–1.49 | <0.001 | 157.48 | <0.001 | 84.13 | |
| >12 months | 6 | 1.77 | 1.33–2.21 | <0.001 | 14.84 | <0.05 | 66.32 |
Note. Analyses based on 21 studies (20 for proportion female).
p-values less than 0.05 represent statistical significance. 95% CI = 95% confidence intervals, g = Hedges’ g, k = number of comparisons, PTSD = posttraumatic stress disorder.
Additional subgroup analyses for PTSD indicated significantly larger treatment effects for intention-to-treat samples with higher compared to lower dropout rates (⩾20%; <20%, p < 0.01), while treatment effects remained unaffected by losses to follow-up (<40%; ⩾40%, p = 0.61). In completer samples, attrition rates had no impact on treatment outcomes (p = 0.56; p = 0.34).
p-value of Hedges’ g.
p-value of Q-statistics.
p-value between groups.
Analyses include active treatments only.