Table 2 Data extraction: Outcomes
First author | Primary Outcome Measure | Other Outcome Measure(s) | Assessment measure(s) for depression/anxiety | Assessment time-points | Attrition and Adherence | Results | Limitations |
|---|---|---|---|---|---|---|---|
Danaher [45] | Depressive symptoms, acceptability and feasibility | Automatic thoughts, dyadic adjustment, parenting sense of competence, self-efficacy | EPDS (only for pre-test screening) and HRSD, PHQ-9 | Pre-test, Post-test (3 months) and follow-up (6 months), PHQ-9 during coach calls at 2 and 4 weeks additionally | All 6 sessions of the program were completed by 87 % (46/53) of participants. Posttest data were collected from 89 % of participants (47/53) with the exception of the HRSD (45/53, 85 %) and 6-month follow-up data were collected from 87 % of participants (46/53). Overall attrition was 13 % (7/53) from pretest to 6-month follow-up. Average of 5.6/6 sessions viewed. | PHQ-9 scores decreased from pretest (mean 12.6, SD 4.1) to posttest (mean 5.0, SD 4.4) and the 6-month follow-up (mean 4.2, SD 3.9) (p < 0.001) with large effects post-test (partial r = 0.77) and 6-month follow-up (partial r = 0.82). At pretest, 55 % (29/53) participants met PHQ-9 criteria for minor or major depression. At posttest, 90 % (26/29) no longer met these PHQ-9 criteria. HRSD scores also decreased from pretest (mean 16.9, SD 6.9) to posttest (mean 7.0, SD 5.6) and the 6-month follow-up (mean 6.6, SD 6.8). Changes from pretest were statistically significant (P < .001) with large effects at posttest (partial r = .75) and 6-month follow-up (partial r = .71).a | No comparator group, women were allowed to engage with other therapies (e.g., pharmacotherapy, counselling) during the trial and thus it is difficult to deduce individual effect of intervention, ‘coach’ reliant. Quasi-experimental design with small convenience sample Quality score: poor |
Kersting [40] | Prolonged grief, PTSD | General psychopathology (including depression and anxiety) | ICG, BSI | Baseline, post-treatment and 3- month and 12-month follow up | 86.1 % in the TG completed the intervention. WLC had a completion rate of 88.5 %. Dropouts were younger. | % of participants scoring > ICG-R cut-off for prolonged grief differed significantly at post-treatment (TG = 28.7 %, WLC = 47.8 %) Mean depression scores for TG were significantly decreased at post-treatment (1.19 → 0.61, t(114) = 7.98, p < 0.001) Same for anxiety (0.7 → 0.37). Depression scores continued to improve in follow-up measurements. | Heavily therapist reliant, well-educated sample, questionable relevance to perinatal depression, intensive – high level of participant engagement required. Male participants were included. Self-rating questionnaire to rate psychotherapy Quality score: intermediate |
O’Mahen [43] | Feasibility, acceptability, depressive symptoms | None | EPDS | At sign-up to the trial and 15-weeks | 18.9 % (172/910) completed the longer baseline questionnaires. The 15-week follow-up EPDS was completed by 39 % (181/462) in treatment group and 36 % in TAU (162/448). Fewer participants completed the acceptability questionnaires. | Improvement in depressive symptoms for 61.3 % (n = 111/181) of TG and 41.4 % (n = 67/162) for TAU group. When all non-respondents are counted as depressed the intervention is still favoured. 115/462 (intervention) vs 71/448 (TAU) were not depressed. | Extremely high attrition rates – follow-up EPDS was completed by less than 40 % in each group, only 1 measure of depressive symptoms, Imperfect intervention – women reported struggle ‘keeping up.’ Online Recruitment. Quality score: intermediate |
O’Mahen [44] | Depression and anxiety, attrition and adherence | Work and social impairment, social support, postnatal bonding, health service utilization. | EPDS, GAD-7 | Baseline, 17 weeks and 6 months post-treatment | 86 % (71/83) completed EPDS at post-treatment and 71 % (59/83) at 3 month follow-up. Women completed an average of 8 (SD 4.5) telephone support sessions and 5.36 (SD = 4.62) online modules | Clinically significant improvement in depression scores in 62.2 % (n = 23/37) of TG compared to 29.4 % (n = 10/34) of TAU. Odds ratio = 0.26 (95 % CI 0.10–0.71) after adjustment for baseline EPDS scores. Large Cohen’s db effect sizes for EPDS (−0.87, 95 % CI −0.42 to −1.32) and GAD-7 (−0.59, 95 % CI −1.11 to −0.07). | Online sample recruitment might give a sample that is more accepting and responsive to internet-based therapy, unable to assess the impact of telephone support vs. web-modules, only 1 follow-up assessment point, not ethnically diverse sample. Quality score: good |