n (k)* | Severity of suicidal ideation | Reliable changes (RCI)† | Treatment response (50% symptom reduction) | |||||||
b (SE) | 95% CI | P value | b (SE) | 95% CI | P value | b (SE) | 95% CI | P value | ||
Effects on severity of suicidal ideation | ||||||||||
Treatment effect at postintervention | 2037 (9) | −0.247 (0.038) | −0.322; −0.173 | <0.001 | 0.633 (0.115) | 0.408; 0.859 | <0.001 | 0.606 (0.100) | 0.410; 0.801 | <0.001 |
Treatment effect at follow-up | 891 (4) | −0.189 (0.054) | −0.296; −0.083 | 0.001 | 0.441 (0.195) | 0.056; 0.826 | 0.075 | 0.603 (0.157) | 0.295; 0.911 | <0.001 |
Moderator analyses | ||||||||||
Suicidal ideation‡ | 2037 (9) | −0.202 (0.099) | −0.396; −0.008 | 0.125 | 0.220 (0.293) | −0.355; 0.795 | 1.000 | 0.196 (0.294) | −0.380; 0.771 | 1.000 |
History of suicide attempts‡ | 1850 (6) | −0.123 (0.080) | −0.279; 0.033 | 0.370 | 0.184 (0.224) | −0.255; 0.624 | 1.000 | 0.023 (0.222) | −0.411; 0.457 | 1.000 |
Depressiveness‡ | 1980 (8) | −0.008 (0.034) | −0.074; 0.058 | 1.000 | −0.011 (0.100) | −0.207; 0.184 | 1.000 | −0.105 (0.091)§ | −0.283; 0.074 | 0.750 |
Hopelessness‡ | 1785 (5) | −0.055 (0.034) | −0.121; 0.011 | 0.303 | 0.146 (0.097) | −0.044; 0.337 | 0.395 | 0.182 (0.091) | 0.004; 0.360 | 0.135 |
Anxiety‡ | 1516 (5) | −0.069 (0.040) | −0.148; 0.009 | 0.249 | 0.151 (0.117) | −0.079; 0.380 | 0.595 | 0.152 (0.103) | −0.051; 0.355 | 0.426 |
Worrying | 1369 (4) | 0.000 (0.003) | −0.007; 0.007 | 1.000 | 0.003 (0.010) | −0.016; 0.023 | 1.000 | −0.008 (0.009) | −0.026; 0.010 | 1.000 |
Age‡ | 1907 (7) | −0.001 (0.003) | −0.007; 0.004 | 1.000 | 0.003 (0.008) | −0.013; 0.019 | 1.000 | −0.007 (0.007) | −0.021; 0.008 | 1.000 |
Female gender‡ | 2019 (9) | −0.034 (0.082) | −0.195; 0.127 | 1.000 | 0.056 (0.240) | −0.414; 0.527 | 1.000 | −0.031 (0.223) | −0.468; 0.406 | 1.000 |
Secondary education or higher‡ | 1872 (7) | 0.025 (0.140) | −0.248; 0.299 | 1.000 | −0.026 (0.425) | −0.862; 0.809 | 1.000 | −0.072 (0.422) | −0.898; 0.755 | 1.000 |
Married/living with partner‡ | 1616 (5) | −0.043 (0.096) | −0.231; 0.144 | 1.000 | 0.124 (0.268) | −0.402; 0.649 | 1.000 | −0.020 (0.263) | −0.535; 0.495 | 1.000 |
Employed‡ | 710 (6) | 0.151 (0.135) | −0.112; 0.415 | 0.782 | −0.702 (0.407) | −1.502; 0.098 | 0.256 | −0.470 (0.371) | −1.198; 0.258 | 0.616 |
Current treatment | 1829 (6) | 0.082 (0.081) | −0.076; 0.241 | 0.929 | −0.105 (0.236) | −0.569; 0.358 | 1.000 | −0.415 (0.223) | −0.852; 0.022 | 0.187 |
Alcohol use | 558 (3) | −0.022 (0.026) | −0.073; 0.029 | 1.000 | 0.038 (0.090) | −0.139; 0.214 | 1.000 | −0.012 (0.070) | −0.149; 0.124 | 1.000 |
Human support during intervention | 2037 (9) | 0.002 (0.090) | −0.173; 0.178 | 1.000 | 0.137 (0.244) | −0.342; 0.617 | 1.000 | 0.212 (0.234) | −0.246; 0.670 | 1.000 |
Treatment dose (No of modules)‡ | 2037 (9) | 0.014 (0.031) | −0.046; 0.074 | 1.000 | −0.088 (0.083) | −0.250; 0.075 | 0.868 | −0.048 (0.078) | −0.201; 0.104 | 1.000 |
Treatment dose (weeks)‡ | 2037 (9) | −0.042 (0.034) | −0.108; 0.025 | 0.661 | 0.082 (0.092) | −0.098; 0.263 | 1.000 | −0.020 (0.087) | −0.190; 0.150 | 1.000 |
Type of control group‡ | 2037 (9) | 0.182 (0.091) | 0.003; 0.360 | 0.138 | −0.578 (0.268) | −1.105; −0.051 | 0.095 | −0.694 (0.258) | −1.200; −0.188 | 0.021 |
Note: These analyses are based on imputed data. P values have been corrected for multiple testing across the three indices of suicidal ideation using the Bonferroni correction term; corrected p>1.000 have been rounded to 1.000. The CIs have not been corrected. For moderators, the treatment×moderator interaction is displayed.
*n (k): total number of participants included in the respective analyses (number of studies).
†RCI: categorised RCI per person (improvement, no change, deterioration).
‡These moderator variables were prespecified in the study protocol.
§The moderator depressiveness in the response rate model was modelled as a random effect as indicated in model comparisons; all other moderators were modelled as fixed effects.
iCBT, interventions based on cognitive–behavioural therapy; RCI, Reliable Change Index.