Summary of findings for the main comparison. Psychotherapy as an adjunct to usual care compared with usual care alone for treatment‐resistant depression in adults ‐ short‐term effects.
| Psychotherapy as an adjunct to usual care compared with usual care alone for treatment‐resistant depression in adults | ||||||
| Patient or population: adults with treatment‐resistant depression Setting: primary or secondary care Intervention: psychotherapy with usual care Comparison: usual care alone | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | № of participants (studies) | Quality of the evidence (GRADE) | Comments | |
| Risk with usual care alone | Risk with psychotherapy as an adjunct to usual care | |||||
| Self‐reported depressive symptoms short term (up to 6 months) ‐ BDI (BDI) | Mean depressive symptoms short term (up to 6 months) ‐ BDI was 21.1 | MD 4.07 lower (7.07 lower to 1.07 lower) | MD ‐4.07 (‐7.01 to ‐1.07) | 575 (5 RCTs) | ⊕⊕⊕⊝ MODERATEa,b |
One large and 4 small studies comprising mainly women. Third‐wave cognitive/behavioural therapies given (individual CBT in 3 studies, group DBT in 1, and individual IPT in 1) |
| Self‐reported depressive symptoms short term (up to 6 months) ‐ SMD (BDI & PHQ9) | Mean depressive symptoms short term (up to 6 months) ‐ BDI was 21.1, and PHQ9 was 14.79 | SMD 0.4 SD lower (0.65 lower to 0.14 lower) | SMD ‐0.4 (‐0.65 to ‐0.14) | 635 (6 RCTs) | ⊕⊕⊕⊝ MODERATEa,b |
All 6 studies combined |
| Observer‐rated depressive symptoms short term (up to 6 months) ‐ PHQ‐9 | Mean depressive symptoms short term (up to 6 months) ‐ PHQ‐9 was 14.8 | MD 4.66 lower (8.72 lower to 0.59 lower) | MD ‐4.66 (‐8.72 to ‐0.59) | 482 (2 RCTs) | ⊕⊕⊕⊝ MODERATEa |
One large study from UK and one relatively small one from Canada |
| Observer‐rated depressive symptoms short term (up to 6 months) ‐ HAMD | Mean depressive symptoms short term (up to 6 months) ‐ HAMD was 14.76 | MD 3.28 lower (5.71 lower to 0.85 lower) | MD ‐3.28 (‐5.71 to ‐0.85) | 193 (4 RCTs) | ⊕⊕⊝⊝
LOW c,d |
Although blinded outcome assessment, 4 small studies each using a different type of psychotherapy: group DBT; ISTDP; CBT; IPT |
| Dropout short term (up to 6 months) | Study population | RR 0.85 (0.58 to 1.24) | 698 (6 RCTs) | ⊕⊕⊕⊕ HIGH | Objective outcome; data reported in all studies; Although a proxy for acceptability, it suggests that intervention may be as acceptable as usual care | |
| 149 per 1000 (14%) | 126 per 1000 (12.6%) (86 to 184) | |||||
| Response (50% reduction in depressive symptoms from baseline) short term (up to 6 months) | Study population | RR 1.80 (1.20 to 2.69) | 556 (4 RCTs) | ⊕⊕⊝⊝ LOW a,e |
‐ | |
| 264 per 1000 | 476 per 1000 (317 to 711) | |||||
| Remission (< 7 on HAMD or < 10 on BDI) short term (up to 6 months) | Study population | RR 1.92 (1.46 to 2.52) | 635 (6 RCTs) | ⊕⊕⊕⊝ MODERATEa,b |
One large and 5 small studies comprising mainly women. Third‐wave cognitive/behavioural therapies given (individual CBT in 3 studies; individual IPT, ISTDP, and group DBT in 1 study each) | |
| 166 per 1000 | 319 per 1000 (243 to 419) | |||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OR: odds ratio; RCT: randomised controlled trial; RR: risk ratio; SMD: standardised mean difference. | ||||||
| GRADE Working Group grades of evidence. High quality: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate quality: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low quality: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low quality: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. | ||||||
aOutcome assessment not blind.
bAllocation concealment unclear for one of the two smaller studies.
cRisk of bias due to incomplete outcome data in two of the studies.
dStudies are small. Effects not in the same direction for IPT study (n = 30).
eReporting bias likely as less frequently reported than remission or mean scores.