Responding to children and adolescents who have been sexually abused: WHO clinical guidelines: web annex 5b: psychosocial and mental health interventions – children only: GRADE tables

1 | P a g e RESPONDING TO CHILDREN AND ADOLESCENTS WHO HAVE BEEN SEXUALLY ABUSED WHO CLINICAL GUIDELINES Web Annex 5b: Psychosocial and mental health interventions – children only: GRADE tables *Full guide: https://apps.who.int/iris/bitstream/handle/10665/259270/9789241550147-eng.pdf 2 | P a g e WHO/RHR/17.29 © World Health Organization 2017 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial- ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization. Suggested citation. Responding to children and adolescents who have been sexually abused: WHO clinical guidelines: Web Annex 5b: Psychosocial and mental health interventions – children only: GRADE tables: Geneva: World Health Organization; 2017. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/ licensing. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. Edited by Mary Stewart 3 | P a g e Web Annex 5b. Grading of Recommendations, Assessment, Development and Evaluation (GRADE) tables: psychosocial and mental health interventions for children only Recommendation 11 Question – Among children and adolescents who have or may have been exposed to sexual abuse and who are diagnosed with mental health disorders (P), do any psychosocial interventions (e.g. psychological counselling or psychotherapeutic interventions) (I), as compared to no or any other psychosocial interventions (C), improve the children or adolescents’ mental health outcomes (e.g. emotional, behavioural disorders, post-traumatic stress disorder (PTSD), depression, subjective well-being, daily functioning) and/or parent/caregiver outcomes (O)? Table 1. Study and participant characteristics Study Intervention (n at baseline) Comparison (n at baseline) Intervention provider Child mean age in years (age range of inclusion criteria) Participant gender Participant symptomology Country (income level, World Bank) Foa et al. (1) Prolonged exposure (n = 31) Supportive counselling (n = 30) Supervised, Master’s-level counsellors 15.3 (13–18) 100% adolescent girls Chronic or subthreshold PTSD United States of America (high) O’Callaghan et al. (2) Group-based TF-CBT (n = 24) Wait-list control (n = 28) Non-mental- health social workers 16.02 (12–17) 100% adolescent girls 60% of sample had PTSD Democratic Republic of the Congo (low) Jaberghaderi et al. (3) CBT with Trauma Focus (n = 5) EMDR (n = 6) Clinical psychologists Not stated (12–13) 100% adolescent girls PTSD Islamic Republic of Iran (upper middle) Trowell et al. (4), McCrone et al. (5) Group psychotherapy (n = 36) Individual psychotherapy (n = 35) Trainee psychotherapists or experienced mental health professionals 10 (6–14) 100% girls Symptoms of emotional or behavioural disturbance warranting treatment, including PTSD United Kingdom (high) King et al.a (6) Child-alone CBT with Trauma Focus (n = 12) Family CBT with trauma focus (n = 12) Wait list control (n = 12) Registered psychologists 11.4 (5–17) 69% girls 31% boys Diagnostic criteria for PSTD or short by several symptoms Australia (high) Berliner & Saunders (7) SIT (n = 48) Conventional therapy (n = 32) Master’s-level clinical social workers 8 (4–13) 71% girls 29% boys PTSD was the chart diagnosis in 81% of cases United States of America (high) 4 | P a g e Study Intervention (n at baseline) Comparison (n at baseline) Intervention provider Child mean age in years (age range of inclusion criteria) Participant gender Participant symptomology Country (income level, World Bank) Deblinger et al. (8) Deblinger et al.a (9) Child-only TF-CBT (n = 25) Mother-only TF-CBT (n = 25) Mother-and-child TF- CBT (n = 25) Community control (standard of care) (n = 25) Mental-health therapists trained in experimental CBT 9.84 (7–13) 83% girls 17% boys Presence of a total of 3 post-traumatic stress symptoms United States of America (high) Sullivan et al. (10) Individual psychotherapy (n = 35) (2 hours for 36 weeks) No treatment (n = 37) Master’s degree in counselling, clinical psychologist, visiting psychiatrist Not stated (12–16) 71% adolescent boys 29% adolescent girls Baseline symptomatology not provided United States of America (high) Notes. CBT = cognitive behavioural therapy; TF-CBT = trauma-focused CBT; EMDR = eye-movement desensitization and reprocessing therapy; SIT = stress inoculation training plus gradual exposure. a Relevant to PICO question for children only interventions (i.e. Recommendation 11) as well as PICO question for children and their caregivers (i.e. Recommendation 12 – Annex 6b). 5 | P a g e Table 2. Risk-of-bias assessment for the included studies Notes. Risk of bias: low (L), unclear (U) or high (H). Study Sequenc e generati on Allocatio n concealm ent Blinding of participants/ providers Blinding of outcome assessment Incomplete outcome data Selective reporting Other bias Overall rating King et al. (6) U U H H H U L U Sullivan et al. (10) H H U U L L H H Jaberghad eri et al. (3) U U H L H U L U Trowell et al. (4) U U H H L U L U O’Callagha n et al. (2) L U U U L H L U Berliner& Saunders (7) L L H H H U L U Deblinger et al. (8); Deblinger et al. (9) U U U U H U L U Foa et al. (1) U U H L L L L U 6 | P a g e Table 3.1. GRADE evidence profile: CBT with a trauma focus compared to wait-list control for child sexual maltreatment Patient or population – Child sexual maltreatment Intervention – CBT with a trauma focus Comparison – Wait-list control Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments 7 | P a g e Risk with wait-list controla Risk with CBT with trauma focusa Immediate post-treatment Change in PTSD (mixed gender) The mean change in PTSD was -1.47 The mean change in PTSD in the intervention group was 4.28 lower (6.23 lower to 2.33 lower) – 24 (1 randomized controlled trial – RCT)1 ⨁◯◯◯ VERY LOW2 Assessed using 24-item Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) PTSD section of the Anxiety Disorders Interview Schedule (ADIS). The total possible range for the ADIS is 0–28, where a score of ≥ 7 indicates PTSD. Calculated important differenceb shows a largec effect (SMD = -1.70). Change in PTSD (female only) The mean change in PTSD was 2.64 The mean change in PTSD in the intervention group was 25.14 lower (33.24 lower to 17.04 lower) – 52 (1 RCT)3 ⨁◯◯◯ VERY LOW2 Assessed using the self-report University of California at Los Angeles Posttraumatic Stress Disorder Reaction Index (UCLA-PTSDRI), 22 items (range 0–88), where a score of ≥ 38 indicates PTSD. Calculated important difference shows a large effect (SMD = -1.70). Change in depression (mixed gender) The mean change in depression was -1.83 The mean change in depression in the intervention group was 2.17 lower (7.24 lower to 2.9 higher) – 24 (1 RCT)1 ⨁◯◯◯ VERY LOW4,5 Assessed using the Children’s Depression Inventory (CDI), a 27-item scale (items scored on range 0–2). The total possible range for this scale is 0–54, where ≥ 19 indicates severe depression. Calculated important difference shows a small effect (SMD = -0.33). a The units in all the GRADE evidence profiles depend on the scale used to measure the outcomes and hence varies across different outcomes. See last column on comments which specifies how outcomes were assessed. b refers to the size of the intervention effect in each study. Where this size of intervention effect is expressed as SMD – it is using a summary statistic because included studies use a variety of measures or scale to measure the same outcome. therefore SMD expresses the size of the intervention effect in each study relative to variability observed in that study – i.e. difference in means. c According to Cohen’s rule of thumb, where a Standardized Mean Difference (SMD) value under 0.2 indicates no effect; a value of 0.2 to 0.49 indicates a small effect; a value of 0.5–0.79 indicates a medium effect; and a value of 0.8 and above a large effect. Hence the SMD of -1.70 is indicated as having a large effect. 8 | P a g e Risk with wait-list controla Risk with CBT with trauma focusa Change in anxiety (mixed gender) The mean change in anxiety was -1.59 The mean change in anxiety in the intervention group was 5.83 lower (13.35 lower to 1.69 higher) – 24 (1 RCT)1 ⨁◯◯◯ VERY LOW4,5 Assessed using the Revised Children’s Manifest Anxiety Scale (R-CMAS), a 37-item instrument (items scored yes or no) with 28 items assessing anxiety (range 0–56) and ≥ 38 indicates anxiety. Calculated important difference shows a medium effect (SMD = -0.60). Change in anxiety/depression (female only) The mean change in anxiety/depression was 0.86 The mean change in anxiety/depression in the intervention group was 24.86 lower (32.52 lower to 17.2 lower) – 52 (1 RCT)3 ⨁◯◯◯ VERY LOW2 Assessed using the African Youth Psychosocial Assessment Instrument (AYPA), a 40-item measure that assesses four locally defined domains of biopsychosocial deficits (range 0–54). Calculated important difference shows a large effect (SMD = -1.73). Change in internalizing symptoms (mixed gender) The mean change in internalizing symptoms was -5.65 The mean change in internalizing symptoms in the intervention group was 1.6 lower (7.15 lower to 3.95 higher) – 24 (1 RCT)1 ⨁◯◯◯ VERY LOW4,5 Assessed using the Children’s Behavior Checklist (CBCL-Int). T scores (range 0–100) < 60 are considered normal range, 60–63 as borderline, and scores > 63 in the clinical range. Calculated important difference shows a small effect (SMD = -0.22). Change in externalizing symptoms (mixed gender) The mean change in externalizing symptoms was -5.85 The mean change in externalizing symptoms in the intervention group was 3.77 higher (4.29 lower to 11.83 higher) - 24 (1 RCT)1 ⨁◯◯◯ VERY LOW4,5 Assessed using the Children’s Behavior Checklist (CBCL-Ext). T scores (range 0–100) < 60 are considered normal range, 60–63 as borderline, and scores > 63 in the clinical range. Calculated important difference shows a small effect (SMD = 0.36). Follow-up at 6 months 9 | P a g e Risk with wait-list controla Risk with CBT with trauma focusa Change in PTSD (mixed gender) Follow-up: mean 6 months The mean change in PTSD was -1.91 The mean change in PTSD in the intervention group was 2.75 lower (4.56 lower to 0.94 lower) - 24 (1 RCT)1 ⨁◯◯◯ VERY LOW2 Assessed using 24-item DSM-IV PTSD section of the Anxiety Disorders Interview Schedule (ADIS). The total possible range for the ADIS is 0–28, where a score of 7 or above indicates PTSD. Calculated important difference shows a large effect (SMD = -1.18). Change in depression (mixed gender) Follow-up: mean 6 months The mean change in depression was -3.5 The mean change in depression in the intervention group was 2.16 lower (7.62 lower to 3.3 higher) – 24 (1 RCT) 1 ⨁◯◯◯ VERY LOW4,5 Assessed using CDI. Calculated important difference shows a small effect (SMD = -0.31). Change in anxiety (mixed gender) Follow-up: mean 6 months The mean change in anxiety was -1.59 The mean change in anxiety in the intervention group was 8.33 lower (16.21 lower to 0.45 lower) – 24 (1 RCT) 1 ⨁◯◯◯ VERY LOW2 Assessed using R-CMAS. Calculated important difference shows a large effect (SMD = -0.82). Change in internalizing symptoms (mixed gender) Follow-up: mean 6 months The mean change in internalizing symptoms was -2.7 The mean change in internalizing symptoms in the intervention group was 6.15 lower (12.83 lower to 0.53 higher) – 24 (1 RCT)1 ⨁◯◯◯ VERY LOW 4,5 Assessed using CBCL-Int. Calculated important difference shows a medium effect (SMD = -0.71)- 10 | P a g e Risk with wait-list controla Risk with CBT with trauma focusa Change in externalizing symptoms (mixed gender) Follow-up: mean 6 months The mean change in externalizing symptoms was 0.86 The mean change in externalizing symptoms in the intervention group was 7.43 lower (15.27 lower to 0.41 higher) – 24 (1 RCT)1 ⨁◯◯◯ VERY LOW4,5 Assessed using CBCL-Ext. Calculated important difference shows a medium effect (SMD = -0.73). –Notes: *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 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 1. King et al. (6) 2. Downgraded to very low because of serious concerns for risk of bias and sample size <300 and imprecise effect sizes – see table on quality assessments 3. O’Callaghan et al.(2) 4. Downgraded to very low because of serious concerns regarding risk of bias, see table with quality assessments 5. Downgraded to very low because the sample size is <300 and effect estimate is imprecise – see table with quality assessments 11 | P a g e Table 4.1.Quality assessments: CBT with trauma focus compared to wait-list control for child sexual maltreatment Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of biasd Inconsistency Indirectness Imprecision Other considerations CBT with trauma focus Wait-list control Relative (95% CI) Absolute (95% CI) Change in PTSD (immediate post-treatment) 1 a Randomized trials Serious b Not serious Not serious Very serious b None 12 12 – MD 4.28 lower (6.23 lower to 2.33 lower) ⨁◯◯◯ VERY LOW CRITICAL Change in PTSD (immediate post-treatment) 1 c Randomized trials Serious b Not serious Not serious Very serious b None 24 28 – MD 25.14 lower (33.24 lower to 17.04 lower) ⨁◯◯◯ VERY LOW CRITICAL Change in depression d See Table 2 for risk of bias assessments. 12 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of biasd Inconsistency Indirectness Imprecision Other considerations CBT with trauma focus Wait-list control Relative (95% CI) Absolute (95% CI) 1 a Randomized trials Serious d Not serious Not serious Very serious e None 12 12 – MD 2.17 lower (7.24 lower to 2.9 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in anxiety (immediate post-treatment) 1 a Randomized trials Serious d Not serious Not serious Very serious e None 12 12 – MD 5.83 lower (13.35 lower to 1.69 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in anxiety/depression (immediate post-treatment) 1 c Randomized trials Serious b Not serious Not serious Very serious b None 24 28 – MD 24.86 lower (32.52 lower to 17.2 lower) ⨁◯◯◯ VERY LOW CRITICAL Change in internalizing symptoms (immediate post-treatment) 13 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of biasd Inconsistency Indirectness Imprecision Other considerations CBT with trauma focus Wait-list control Relative (95% CI) Absolute (95% CI) 1 a Randomized trials Serious d Not serious Not serious Very serious e None 12 12 – MD 1.6 lower (7.15 lower to 3.95 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in externalizing symptoms (immediate post-treatment) 1 a Randomized trials Serious d Not serious Not serious Very serious e None 12 12 – MD 3.77 higher (4.29 lower to 11.83 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in PTSD (follow-up: mean 6 months) 1 a Randomized trials Serious b Not serious Not serious Very serious b None 12 12 – MD 2.75 lower (4.56 lower to 0.94 lower) ⨁◯◯◯ VERY LOW CRITICAL Change in depression (follow-up: mean 6 months) 14 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of biasd Inconsistency Indirectness Imprecision Other considerations CBT with trauma focus Wait-list control Relative (95% CI) Absolute (95% CI) 1 a Randomized trials Serious d Not serious Not serious Very serious e None 12 12 – MD 2.16 lower (7.62 lower to 3.3 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in anxiety (follow-up: mean 6 months) 1 a Randomized trials Serious b Not serious Not serious Very serious b None 12 12 – MD 8.33 lower (16.21 lower to 0.45 lower) ⨁◯◯◯ VERY LOW CRITICAL Change in internalizing symptoms (follow-up: mean 6 months) 1 a Randomized trials Serious d Not serious Not serious Very serious e None 12 12 – MD 6.15 lower (12.83 lower to 0.53 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in externalizing symptoms (follow-up: mean 6 months) 15 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of biasd Inconsistency Indirectness Imprecision Other considerations CBT with trauma focus Wait-list control Relative (95% CI) Absolute (95% CI) 1 a Randomized trials Serious d Not serious Not serious Very serious e None 12 12 – MD 7.43 lower (15.27 lower to 0.41 higher) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; MD: Mean difference a. King et al. (6) b. Serious concerns for risk of bias and sample size <300. c. O’Callaghan et al 2013 (2) d. Serious concerns regarding risk of bias. e. The sample size is <300 and effect estimate is imprecise. Figure 1. 1. Forest plot: CBT with trauma focus compared to wait-list control for child sexual maltreatment – immediate post treatment 16 | P a g e 17 | P a g e Figure 1.2. Forest plot : CBT with trauma focus compared to wait-list control for child sexual maltreatment – 6 months follow-up 18 | P a g e Table 3. 2. GRADE evidence profile: TF- CBT compared to community control for child sexual maltreatment Patient or population – Child sexual maltreatment Intervention – TF-CBT Comparison – Community control Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with community control Risk with TF-CBT Immediate post-treatment Change in PTSD (mixed gender) The mean change in PTSD was -3.29 The mean change in PTSD in the intervention group was 2.19 lower (3.71 lower to 0.67 lower) – 35 (1 RCT)1 ⨁◯◯◯ VERY LOW 2 Assessed using the Schedule for Affective Disorders and Schizophrenia for School-Age Children (K-SADS-E). Possible range not provided. Calculated important difference shows a large effect (SMD = -0.97), Change in depression (mixed gender) The mean change in depression was -0.14 The mean change in depression in the intervention group was 3 lower (6.25 lower to 0.25 higher) – 36 (1 RCT) 1 ⨁◯◯◯ VERY LOW 3,4 Assessed using the Children’s Depression Inventory (CDI), a 27-item scale (items scored on range 0–2). The total possible range for this scale is 0–54, where 19 or above indicates severe depression. Calculated important difference shows a medium effect (SMD = - 0.62,) Change in state anxiety (mixed gender) The mean change in state anxiety was -1.59 The mean change in state anxiety in the intervention group was 0.62 lower (2.84 lower to 1.6 higher) – 46 (1 RCT) 5 ⨁◯◯◯ VERY LOW 3,4 Assessed using the State scale (A-State) of the State/Trait Anxiety Inventory for Children (STAIC), a 20-item self-report scale used to assess situationally specific anxiety. Calculated important difference shows no effect (SMD = -0.16). Change in trait anxiety (mixed gender) The mean change in trait anxiety was -4.22 The mean change in trait anxiety in the intervention group was 1.58 lower (4.75 lower to 1.59 higher) – 46 (1 RCT) 5 ⨁◯◯◯ VERY LOW 3,4 Assessed using the Trait scale (A-Trait) of STAIC, a 20-item self-report inventory (range 0–40. There are no published cut-offs for this score. Calculated important difference shows a small effect (SMD = -0.28). 19 | P a g e Patient or population – Child sexual maltreatment Intervention – TF-CBT Comparison – Community control Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with community control Risk with TF-CBT Change in internalizing symptoms (mixed gender) The mean change in internalizing symptoms was -3.14 The mean change in internalizing symptoms in the intervention group was 2.65 lower (6.46 lower to 1.16 higher) – 45 (1 RCT) 5 ⨁◯◯◯ VERY LOW 3,4 Assessed using the Children’s Behavior Checklist (CBCL-Int). T scores (range 0–100) < 60 are considered normal range, 60–63 as borderline, and scores > 63 in the clinical range. Calculated important difference shows a small effect (SMD = -0.40) Change in externalizing symptoms (mixed gender) The mean change in externalizing symptoms was 0.0 The mean change in externalizing symptoms in the intervention group was 3.15 lower (8.94 lower to 2.64 higher) – 32 (1 RCT) 1 ⨁◯◯◯ VERY LOW 3,4 Assessed using the Children’s Behavior Checklist (CBCL-Ext). T scores (range 0–100) < 60 are considered normal range, 60–63 as borderline, and scores > ––63 are in the clinical range. Calculated important difference shows a small effect (SMD = -0.38). Follow-up at 6 months Change in PTSD (mixed gender) Follow-up: mean 6 months The mean change in PTSD was -4.22 The mean change in PTSD in the intervention group was 1.97 lower (3.59 lower to 0.35 lower) – 35 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2 Assessed using K-SADS-E. Calculated important difference shows a large effect (SMD = -0.85). Change in depression (mixed gender) Follow-up: mean 6 months The mean change in depression was -4.87 The mean change in depression in the intervention group was 1.18 lower (4.19 lower to 1.83 higher) – 36 (1 RCT) 1 ⨁◯◯◯ VERY LOW 3,4 Assessed using CDI. Calculated important difference shows a small effect (SMD = -0.26). 20 | P a g e Patient or population – Child sexual maltreatment Intervention – TF-CBT Comparison – Community control Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with community control Risk with TF-CBT Change in externalizing symptoms (mixed gender) Follow-up: mean 6 months The mean change in externalizing symptoms was -2.59 The mean change in externalizing symptoms in the intervention group was 5.66 lower (10.59 lower to 0.73 lower) – 32 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2 Assessed using CBCL-Ext––. Calculated important difference shows a large effect (SMD = -0.82). Follow-up at 12 months Change in PTSD (mixed gender) Follow-up: mean 12 months The mean change in PTSD was -4.36 The mean change in PTSD in the intervention group was 0.88 lower (2.62 lower to 0.86 higher) – 35 (1 RCT) 1 ⨁◯◯◯ VERY LOW 3,4 Assessed using K-SADS-E. Calculated important difference shows a small effect (SMD = -0.35). Change in depression (mixed gender) Follow-up: mean 12 months The mean change in depression was -5.14 The mean change in depression in the intervention group was 0.96 lower (3.91 lower to 1.99 higher) – 36 (1 RCT) 1 ⨁◯◯◯ VERY LOW 3,4 Assessed using CDI27-item ––. Calculated important difference shows a small effect (SMD = -0.21). Change in externalizing symptoms (mixed gender) Follow-up: mean 12 months The mean change in externalizing symptoms was -3.09 The mean change in externalizing symptoms in the intervention group was 5.01 lower (10.27 lower to 0.25 higher) – 32 (1 RCT) 1 ⨁◯◯◯ VERY LOW 3,4 Assessed using CBCL-Ext. ––Calculated important difference shows a medium effect (SMD = -0.70). Follow-up at 24 months 21 | P a g e Patient or population – Child sexual maltreatment Intervention – TF-CBT Comparison – Community control Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with community control Risk with TF-CBT Change in PTSD (mixed gender) Follow-up: mean 24 months The mean change in PTSD was -5.36 The mean change in PTSD in the intervention group was 0.5 lower (1.98 lower to 0.98 higher) – 35 (1 RCT) 1 ⨁◯◯◯ VERY LOW 3,4 Assessed using K-SADS-E. Calculated important difference shows a small effect (SMD = -0.22). Change in depression (mixed gender) Follow-up: mean 24 months The mean change in depression was -3.74 The mean change in depression in the intervention group was 0.74 lower (3.63 lower to 2.15 higher) – 36 (1 RCT) 1 ⨁◯◯◯ VERY LOW 3,4 Assessed using CDI––. Calculated important difference shows no effect (SMD = -0.17). Change in externalizing symptoms (mixed gender) Follow-up: mean 24 months The mean change in externalizing symptoms was 4.25 The mean change in externalizing symptoms in the intervention group was 10.75 lower (16.44 lower to 5.06 lower) – 32 (1 RCT) 1 ⨁◯◯◯ VERY LOW 3,4 Assessed using CBCL-Ext––. Calculated important difference shows a large effect (SMD = -1.41). Notes: *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 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 22 | P a g e 1. Deblinger et al (9) 2. Downgraded to very low because of serious concerns for risk of bias, sample size <300 and imprecise effect size – see Table 4.2. 3. Downgraded to very low because of serious concerns regarding risk of bias – see Table 4.2. 4. Downgraded to very low because also the sample size is <300 and effect estimate is imprecise – see Table 4.2. 5. Deblinger et al (8) 23 | P a g e Table 4.2.Quality Assessment: TF- CBT compared to community control for child sexual maltreatment Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations TF-CBT Community control Relative (95% CI) Absolute (95% CI) Change in PTSD (immediate post-treatment) 1 a Randomized trials Seriousb Not serious Not serious Very seriousb None 21 14 – MD 2.19 lower (3.71 lower to 0.67 lower) ⨁◯◯◯ VERY LOW CRITICAL Change in depression (immediate post-treatment) 1 a Randomized trials Seriousc Not serious Not serious Very serious d None 21 15 – MD 3 lower (6.25 lower to 0.25 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in state anxiety (immediate post-treatment) 1 e Randomized trials Seriousc Not serious Not serious Very serious d None 24 22 – MD 0.62 lower (2.84 lower to 1.6 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in trait anxiety (immediate post-treatment) 24 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations TF-CBT Community control Relative (95% CI) Absolute (95% CI) 1 e Randomized trials Seriousc Not serious Not serious Very serious d None 24 22 – MD 1.58 lower (4.75 lower to 1.59 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in internalizing symptoms (immediate post-treatment) 1 e Randomized trials Seriousc Not serious Not serious Very serious d None 24 21 – MD 2.65 lower (6.46 lower to 1.16 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in externalizing symptoms (immediate post-treatment) 1 a Randomized trials Seriousc Not serious Not serious Very serious d None 20 12 – MD 3.15 lower (8.94 lower to 2.64 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in PTSD (follow-up: mean 6 months) 25 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations TF-CBT Community control Relative (95% CI) Absolute (95% CI) 1 a Randomized trials Seriousb Not serious Not serious Very seriousb None 21 14 – MD 1.97 lower (3.59 lower to 0.35 lower) ⨁◯◯◯ VERY LOW CRITICAL Change in depression (follow-up: mean 6 months) 1 a Randomized trials Seriousc Not serious Not serious Very seriousd None 21 15 – MD 1.18 lower (4.19 lower to 1.83 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in externalizing symptoms (follow-up: mean 6 months) 1 a Randomized trials Seriousb Not serious Not serious Very seriousb None 20 12 – MD 5.66 lower (10.59 lower to 0.73 lower) ⨁◯◯◯ VERY LOW CRITICAL Change in PTSD (follow-up: mean 12 months) 26 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations TF-CBT Community control Relative (95% CI) Absolute (95% CI) 1 a Randomized trials Seriousc Not serious Not serious Very seriousd None 21 14 – MD 0.88 lower (2.62 lower to 0.86 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in depression (follow-up: mean 12 months) 1 a Randomized trials Seriousc Not serious Not serious Very seriousd None 21 15 – MD 0.96 lower (3.91 lower to 1.99 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in externalizing symptoms (follow-up: mean 12 months) 1 a Randomized trials Seriousc Not serious Not serious Very seriousd None 20 12 – MD 5.01 lower (10.27 lower to 0.25 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in PTSD (follow-up: mean 24 months) 27 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations TF-CBT Community control Relative (95% CI) Absolute (95% CI) 1 a Randomized trials Seriousc Not serious Not serious Very seriousd None 21 14 – MD 0.5 lower (1.98 lower to 0.98 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in depression (follow-up: mean 24 months) 1 a Randomized trials Seriousc Not serious Not serious Very seriousd None 21 15 – MD 0.74 lower (3.63 lower to 2.15 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in externalizing symptoms (follow-up: mean 24 months) 1 a Randomized trials Seriousc Not serious Not serious Very seriousd None 20 12 – MD 10.75 lower (16.44 lower to 5.06 lower) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; MD: Mean difference a Deblinger et al. (9). b There are serious concerns about the risk of bias and the sample size is < 300. c There are serious concerns about the risk of bias. d The sample size is < 300 and the effect estimate is imprecise. e Deblinger et al. (8). 28 | P a g e Fig. 2.1. Forest plot: TF-CBT compared to community control for child sexual maltreatment – immediate post treatment 29 | P a g e Figure 2.2. Forest plot: TF- CBT compared to community control for child sexual maltreatment – 6 month follow-up Figure 2.3. Forest plot: TF- CBT compared to community control for child sexual maltreatment – 12 month follow-up Fig. 2.4. Forest plot: TF-CBT compared to community control for child sexual maltreatment – 24 month follow-up 30 | P a g e 31 | P a g e Table 3.3. GRADE evidence profile: CBT with trauma focus compared to EMDR for child sexual maltreatment Patient or population – Child sexual maltreatment Intervention CBT with trauma focus Comparison – EMDR Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with EMDR Risk with CBT with trauma focus Immediate post-treatment Change in PTSD (child report) (female only) The mean change in PTSD (child report) was -16.0 The mean change in PTSD (child report) in the intervention group was 8.71 higher (3.49 higher to 13.93 higher) - 14 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using Child Report of Post-traumatic Symptoms (CROPS), a 26-item scale (range 0– 52), where a score of 19 or over indicates a clinical concern. Calculated important difference shows a medium effect (SMD = 1.64). PTSD Diagnosis (child report) (female only) 571 per 1000 715 per 1000 (211 to 958) OR 1.88 (0.20 to 17.27) 14 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,4 Assessed using CROPS. Change in PTSD (parent report) (female only) The mean change in PTSD (parent report) was -10.86 The mean change in PTSD (parent report) in the intervention group was 0.28 lower (6.26 lower to 5.7 higher) - 14 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using Parent Report of Post-traumatic Symptoms (PROPS), a 32-item scale (range 0– 64), where a score of ≥ 16 indicates a clinical concern. Calculated important difference shows no effect (SMD = -0.05). PTSD Diagnosis (parent report) (female only) 143 per 1000 286 per 1,000 (26 to 853) OR 2.40 (0.16 to 34.93) 14 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,4 Assessed using PROPS. *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 32 | P a g e Patient or population – Child sexual maltreatment Intervention CBT with trauma focus Comparison – EMDR Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with EMDR Risk with CBT with trauma focus 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 1. Jaberghaderi et al (3) 2. Downgraded to very low due to serious concerns regarding risk of bias – see Table 4.3 below. 3. Also downgraded to very low due to the sample size is <300 and effect estimate is imprecise – see Table 4.3 below. 4. Also downgraded to very low because the sample size / number of events are <300 and effect estimate is imprecise – see Table 4.3 below. 33 | P a g e Table 4.3. Quality assessments: CBT with trauma focus compared to EMDR for child sexual maltreatment Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations CBT with trauma focus EMDR Relative (95% CI) Absolute (95% CI) Change in PTSD (child report) (immediate post-treatment) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 7 7 – MD 8.71 higher (3.49 higher to 13.93 higher) ⨁◯◯◯ VERY LOW CRITICAL PTSD Diagnosis (child report) (immediate post-treatment) 1a Randomized trials Seriousb Not serious Not serious Very seriousd None 5/7 (71.4%) 4/7 (57.1%) OR 1.88 (0.20 to 17.27) 143 more per 1000 (from 361 fewer to 387 more) ⨁◯◯◯ VERY LOW CRITICAL Change in PTSD (parent report) (immediate post-treatment) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 7 7 – MD 0.28 lower (6.26 lower to 5.7 higher) ⨁◯◯◯ VERY LOW CRITICAL PTSD diagnosis (parent report) (immediate post-treatment) 34 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations CBT with trauma focus EMDR Relative (95% CI) Absolute (95% CI) 1a Randomized trials Seriousb Not serious Not serious Very serious d None 2/7 (28.6%) 1/7 (14.3%) OR 2.40 (0.16 to 34.93) 143 more per 1000 (from 117 fewer to 711 more) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; MD: Mean difference; OR: Odds ratio a Jaberghaderi et al (3). b There are serious concerns about the risk of bias. c The sample size is < 300 and the effect estimate is imprecise. d The sample size/number of events is < 300 and the effect estimate is imprecise. 35 | P a g e Fig. 3.1. Forest plot: - CBT with trauma focus compared to EMDR for child sexual maltreatment – immediate post treatment 36 | P a g e Fig. 3.2. Forest Plot: CBT with trauma focus compared to EMDR for child sexual maltreatment – immediate post treatment (dichotomous) 37 | P a g e Table 3.4. GRADE evidence profile: stress inoculation training and gradual exposure therapy (SIT) compared to conventional therapy for child sexual maltreatment Patient or population – Child sexual maltreatment Intervention – Stress inoculation training and gradual exposure therapy Comparison – Conventional therapy Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with conventional therapy Risk with stress inoculation training and gradual exposure therapy Immediate post-treatment Change in depression (mixed gender) The mean change in depression was -3.1 The mean change in depression in the intervention group was 1.1 higher (1.72 lower to 3.92 higher) - 52 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using the Children’s Depression Inventory (CDI), a 27-item scale (items scored on range 0–2). The total possible range for this scale is 0–54, where ≥ 19 indicates severe depression. Calculated important difference shows a small effect (SMD = 0.20). Change in anxiety (mixed gender) The mean change in anxiety was -1.7 The mean change in anxiety in the intervention group was 0.5 higher (2.02 lower to 3.02 higher) - 65 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using the Revised Children’s Manifest Anxiety Scale (R-CMAS), a 37-item instrument (items scored yes or no) with 28 items assessing anxiety (range 0–56) and ≥ 38 indicates anxiety. Calculated important difference shows no effect (SMD = 0.09). Change in internalizing symptoms (mixed gender) The mean change in internalizing symptoms was -6.1 The mean change in internalizing symptoms in the intervention group was 2.4 higher (0.89 lower to 5.69 higher) - 77 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using the Children’s Behavior Checklist (CBCL-Int). T scores (range 0–100) < 60 are considered normal range, 60–63 as borderline and scores > 63 in the clinical range. Calculated important difference shows a small effect (SMD = 0.33). 38 | P a g e Patient or population – Child sexual maltreatment Intervention – Stress inoculation training and gradual exposure therapy Comparison – Conventional therapy Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with conventional therapy Risk with stress inoculation training and gradual exposure therapy Change in externalizing symptoms (mixed gender) The mean change in externalizing symptoms was -4.3 The mean change in externalizing symptoms in the intervention group was 1.9 higher (1.15 lower to 4.95 higher) – 77 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using the Children’s Behavior Checklist (CBCL-Ext). T scores (range 0–100) < 60 are considered normal range, 60–63 as borderline, and scores > 63 are in the clinical range. Calculated important difference shows a small effect (SMD = 0.27). Follow-up at 12 months Change in depression (mixed gender) Follow-up: mean 12 months The mean change in depression was -2.3 The mean change in depression in the intervention group was 1.4 lower (4.26 lower to 1.46 higher) – 52 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using CDI. Calculated important difference shows a small effect (SMD = -0.26). Change in anxiety (mixed gender) Follow-up: mean 12 months The mean change in anxiety was -2.3 The mean change in anxiety in the intervention group was 2 lower (4.67 lower to 0.67 higher) – 65 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using R-CMAS. Calculated important difference shows a small effect (SMD = -0.36). Change in internalizing symptoms (mixed gender) Follow-up: mean 12 months The mean change in internalizing symptoms was -6.3 The mean change in internalizing symptoms in the intervention group was 2.2 higher (1.14 lower to 5.54 higher) – 77 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using CBCL-Int. Calculated important difference shows a small effect (SMD = 0.30). 39 | P a g e Patient or population – Child sexual maltreatment Intervention – Stress inoculation training and gradual exposure therapy Comparison – Conventional therapy Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with conventional therapy Risk with stress inoculation training and gradual exposure therapy Change in externalizing symptoms (mixed gender) Follow-up: mean 12 months The mean change in externalizing symptoms was -4.3 The mean change in externalizing symptoms in the intervention group was 1.9 higher (1.15 lower to 4.95 higher) – 77 (1 RCT) 1 ⨁◯◯◯ VERY LOW2,3 Assessed using CBCL-Ext. Calculated important difference shows no effect (SMD = 0.11). Follow-up at 24 months Change in depression (mixed gender) Follow-up: mean 24 months The mean change in depression was -2.1 The mean change in depression in the intervention group was 0.9 lower (3.78 lower to 1.98 higher) – 52 (1 RCT) 1 ⨁◯◯◯ VERY LOW2,3 Assessed using CDI. Calculated important difference shows no effect (SMD = -0.16). Change in anxiety (mixed gender) Follow-up: mean 24 months The mean change in anxiety was -2.1 The mean change in anxiety in the intervention group was 2.5 lower (5.11 lower to 0.11 higher) – 65 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using R-CMAS. Calculated important difference shows a small effect (SMD = -0.46). Change in internalizing symptoms (mixed gender) Follow-up: mean 24 months The mean change in internalizing symptoms was -6.7 The mean change in internalizing symptoms in the intervention group was 1.3 higher (2.03 lower to 4.63 higher) – 77 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using CBCL-Int. Calculated important difference shows no effect (SMD = 0.18). 40 | P a g e Patient or population – Child sexual maltreatment Intervention – Stress inoculation training and gradual exposure therapy Comparison – Conventional therapy Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with conventional therapy Risk with stress inoculation training and gradual exposure therapy Change in externalizing symptoms (mixed gender) Follow-up: mean 24 months The mean change in externalizing symptoms was -3.4 The mean change in externalizing symptoms in the intervention group was 1.3 lower (4.21 lower to 1.61 higher) – 77 (1 RCT) 1 ⨁◯◯◯ VERY LOW 2,3 Assessed using CBCL-Ext. Calculated important difference shows a small effect (SMD = -0.20). *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 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 1. Berliner and Saunders (7) 2. Downgraded to very low due to serious concerns regarding risk of bias – See Table 4.4 below 3. Also downgraded to very low because the sample size is <300 and effect estimate is imprecise – see Table 4.4 below. 41 | P a g e Table 4.4. Quality assessments: stress inoculation training and gradual exposure therapy (SIT) compared to conventional therapy for child sexual maltreatment Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Stress inoculation training and gradual exposure therapy Conventional therapy Relative (95% CI) Absolute (95% CI) Change in depression (immediate post-treatment) 1 a Randomized trials Seriousb Not serious Not serious Very seriousc None 29 23 – MD 1.1 higher (1.72 lower to 3.92 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in anxiety (immediate post-treatment) 1 a Randomized trials Seriousb Not serious Not serious Very seriousc None 37 28 – MD 0.5 higher (2.02 lower to 3.02 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in internalizing symptoms (immediate post-treatment) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 46 31 – MD 2.4 higher (0.89 lower to 5.69 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in externalizing symptoms (immediate post-treatment) 42 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Stress inoculation training and gradual exposure therapy Conventional therapy Relative (95% CI) Absolute (95% CI) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 46 31 – MD 1.9 higher (1.15 lower to 4.95 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in depression (follow-up: mean 12 months) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 29 23 – MD 1.4 lower (4.26 lower to 1.46 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in anxiety (follow-up: mean 12 months) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 37 28 – MD 2 lower (4.67 lower to 0.67 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in internalizing symptoms (follow-up: mean 12 months) 43 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Stress inoculation training and gradual exposure therapy Conventional therapy Relative (95% CI) Absolute (95% CI) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 46 31 – MD 2.2 higher (1.14 lower to 5.54 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in externalizing symptoms (follow-up: mean 12 months) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 46 31 – MD 1.9 higher (1.15 lower to 4.95 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in depression (follow-up: mean 24 months) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 29 23 – MD 0.9 lower (3.78 lower to 1.98 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in anxiety (follow-up: mean 24 months) 44 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Stress inoculation training and gradual exposure therapy Conventional therapy Relative (95% CI) Absolute (95% CI) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 37 28 – MD 2.5 lower (5.11 lower to 0.11 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in internalizing symptoms (follow-up: mean 24 months) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 46 31 – MD 1.3 higher (2.03 lower to 4.63 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in externalizing symptoms (follow-up: mean 24 months) 1a Randomized trials Seriousb Not serious Not serious Very seriousc None 46 31 – MD 1.3 lower (4.21 lower to 1.61 higher) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; MD: Mean difference a Berliner & Saunders (7) b Serious concerns about the risk of bias. 45 | P a g e c The sample size is < 300 and the effect estimate is imprecise. 46 | P a g e Fig. 4.1.Forest plot: stress inoculation training and gradual exposure therapy (SIT) compared to conventional therapy for child sexual maltreatment – immediate post treatment 47 | P a g e Fig. 4.2. Forest plot: SIT compared to Conventional therapy for child sexual maltreatment – 12 months follow-up 48 | P a g e Fig. 4.3. Forest plot: SIT compared to conventional therapy for child sexual maltreatment – 24 months follow-up 49 | P a g e Table 3. 5. GRADE evidence profile: prolonged exposure therapy compared to supportive counseling for child sexual maltreatment Patient or population – Child sexual maltreatment Intervention – Prolonged exposure therapy Comparison – Supportive counselling Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with supportive counselling Risk with prolonged exposure therapy Immediate post-treatment Change in PTSD (female only) The mean change in PTSD was -13.3 The mean change in PTSD in the intervention group was 7.3 lower (10.6 lower to 4 lower) - 61 (1 RCT)1 ⨁◯◯◯ VERY LOW2 Assessed using the Child PTSD Symptom Scale-Interview (CPSS-I). Scores range from 0 to 51: 0–10, below threshold; 11–15, subclinical; 16–20, mild; 21–25, moderate; 26–30, moderately severe; 31–40, severe. Calculated important difference shows a large effect (SMD = -1.10). PTSD diagnosis (female only) 567 per 1000 223 per 1000 (84 to 471) OR 0.22 (0.07 to 0.68) 61 (1 RCT)1 ⨁◯◯◯ VERY LOW3 Assessed using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) Schedule for Affective Disorders and Schizophrenia for School-Age Children (K- SADS). Possible range not provided. Change in depression (female only) The mean change in depression was -8.4 The mean change in depression in the intervention group was 2.8 lower (5.35 lower to 0.25 lower) - 61 (1 RCT)1 ⨁◯◯◯ VERY LOW2 Assessed using the Children’s Depression Inventory (CDI), with scores from 0–54, where 0–13 indicates no depressive disorder; 14–19, possible depressive disorder; and 20–54, depressive disorder. Calculated important difference shows a medium effect (SMD = - 0.54). Follow-up at 12 months 50 | P a g e Patient or population – Child sexual maltreatment Intervention – Prolonged exposure therapy Comparison – Supportive counselling Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with supportive counselling Risk with prolonged exposure therapy Change in PTSD (female only) follow-up: mean 12 months The mean change in PTSD was -14.3 The mean change in PTSD in the intervention group was 5.7 lower (8.77 lower to 2.63 lower) - 61 (1 RCT)1 ⨁◯◯◯ VERY LOW2 Assessed using CPSS-I. Calculated important difference shows a large effect (SMD = -0.92). PTSD diagnosis (female only) Follow-up: mean 12 months 467 per 1000 95 per 1000 (26 to 300) OR 0.12 (0.03 to 0.49) 61 (1 RCT)1 ⨁◯◯◯ VERY LOW3 Assessed using K-SADS. Change in depression (female only) Follow-up: mean 12 months The mean change in depression was -6.7 The mean change in depression in the intervention group was 4.7 lower (7.57 lower to 1.83 lower) - 61 (1 RCT)1 ⨁◯◯◯ VERY LOW2 Assessed using CDI. Calculated important difference shows a large effect (SMD = -0.81). *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 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 1 Foa, McLean, Capaldi & Rosenfield (1). 2 Downgraded to very low because of serious concerns about the risk of bias – see Table 4.5 below. 3 Downgraded to very low because of serious concerns about the risk of bias, and because the sample size/number of events is <300 – see Table 4.5 below. Table 4.5. Quality assessment: prolonged exposure therapy compared to supportive counselling for child sexual maltreatment 51 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Prolonged exposure therapy Supportive counselling Relative (95% CI) Absolute (95% CI) Change in PTSD (immediate post-treatment) 1a Randomized trials Seriousb Not serious Not serious Very seriousb None 31 30 – MD 7.3 lower (10.6 lower to 4 lower) ⨁◯◯◯ VERY LOW CRITICAL PTSD diagnosis (immediate post-treatment) 1a Randomized trials Seriousc Not serious Not serious Very seriousc None 7/31 (22.6%) 17/30 (56.7%) OR 0.22 (0.07 to 0.68) 34 fewer per 100 (from 10 fewer to 48 fewer) ⨁◯◯◯ VERY LOW CRITICAL Change in depression (immediate post-treatment) 1a Randomized trials Seriousb Not serious Not serious Very seriousb None 31 30 – MD 2.8 lower (5.35 lower to 0.25 lower) ⨁◯◯◯ VERY LOW CRITICAL Change in PTSD (follow-up: mean 12 months) 52 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Prolonged exposure therapy Supportive counselling Relative (95% CI) Absolute (95% CI) 1a Randomized trials Seriousb Not serious Not serious Very seriousb None 31 30 – MD 5.7 lower (8.77 lower to 2.63 lower) ⨁◯◯◯ VERY LOW CRITICAL PTSD Diagnosis (follow-up: mean 12 months) 1a Randomized trials Seriousc Not serious Not serious Very seriousc None 3/31 (9.7%) 14/30 (46.7%) OR 0.12 (0.03 to 0.49) 37 fewer per 100 (from 17 fewer to 44 fewer) ⨁◯◯◯ VERY LOW CRITICAL Change in depression (follow-up: mean 12 months) 1a Randomized trials Seriousb Not serious Not serious Very seriousb None 31 30 – MD 4.7 lower (7.57 lower to 1.83 lower) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; MD: Mean difference; OR: Odds ratio a Foa, McLean, Capaldi & Rosenfield (1). b Serious concerns about the risk of bias and a sample size < 300. c Serious concerns about the risk of bias and a sample size/number of events < 300. 53 | P a g e Fig. 5.1. Forest plot: prolonged exposure therapy (EXP) compared to supportive counselling for child sexual maltreatment – immediate post treatment Fig. 5.2. Forest plot: prolonged EXP compared to supportive counselling for child sexual maltreatment – 12- month follow-up 54 | P a g e Fig. 5.3. Forest plot: prolonged exposure therapy compared to supportive counselling for child sexual maltreatment – dichotomous 55 | P a g e Table 3.6. GRADE evidence profile: individual psychotherapy compared to no treatment for child sexual maltreatment Patient or population – Child sexual maltreatment Intervention – Individual psychotherapy Comparison – No treatment Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with No treatment Risk with Individual Psychotherapy Immediate post-treatment Internalizing symptoms (female only) The mean internalizing symptoms was 67.431 The mean internalizing symptoms in the intervention group was 9.43 lower (15.85 lower to 3.01 lower)1 – 21 (1 RCT) 1,2 ⨁◯◯◯ VERY LOW 3 Assessed using the Children’s Behavior Checklist (CBCL-Int). T scores (range 0–100) < 60 are considered normal range, 60–63 as borderline, and scores > 63 in the clinical range. Calculated important difference shows a large effect (SMD = -1.27). Internalizing symptoms (male only) The mean internalizing symptoms was 69.71 The mean internalizing symptoms in the intervention group was 9.6 lower (13.47 lower to 5.73 lower)1 – 51 (1 RCT) 1,2 ⨁◯◯◯ VERY LOW 3 Assessed using CBCL-Int. Calculated important difference shows a large effect (SMD = -1.44). Externalizing symptoms (female only) The mean externalizing symptoms was 64.141 The mean externalizing symptoms in the intervention group was 7.14 lower (12.05 lower to 2.23 lower) 1 – 21 (1 RCT) 1,2 ⨁◯◯◯ VERY LOW 3 Assessed using the Children’s Behavior Checklist (CBCL-Ext). T scores (range 0–100) < 60 are considered normal range, 60–63 as borderline, and scores > 63 in the clinical range. Calculated important difference shows a large effect (SMD = -1.25). 56 | P a g e Patient or population – Child sexual maltreatment Intervention – Individual psychotherapy Comparison – No treatment Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with No treatment Risk with Individual Psychotherapy Externalizing symptoms (male only) The mean externalizing symptoms was 70.031 The mean externalizing symptoms in the intervention group was 15.22 lower (20.26 lower to 10.18 lower)1 – 51 (1 RCT) 1,2 ⨁◯◯◯ VERY LOW3 Assessed using CBCL-Ext. Calculated important difference shows a large effect (SMD = -1.79). 1 Data at post-treatment/end-point only. Baseline data not available for all participants. 2 Sullivan et al (10). 3 Downgraded to very low because of serious concerns about the risk of bias and a sample size < 300 – see Table 4.6 below. 57 | P a g e Table 4.6. Quality assessment: individual psychotherapy compared to no treatment for child sexual maltreatment Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Individual psychotherapy No treatment Relative (95% CI) Absolute (95% CI) Internalizing symptoms (immediate post-treatment) 1a Randomized trials Seriousb Not serious Not serious Very seriousb None 14c 7c – MD 9.43 lower (15.85 lower to 3.01 lower) ⨁◯◯◯ VERY LOW CRITICAL Internalizing symptoms (immediate post-treatment) 1a Randomized trials Seriousb Not serious Not serious Very seriousb None 21c 30 c – MD 9.6 lower (13.47 lower to 5.73 lower) ⨁◯◯◯ VERY LOW CRITICAL Externalizing symptoms (immediate post-treatment) 1a Randomized trials Seriousb Not serious Not serious Very seriousb None 14c 7 c – MD 7.14 lower (12.05 lower to 2.23 lower) ⨁◯◯◯ VERY LOW CRITICAL Externalizing symptoms (immediate post-treatment) 58 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Individual psychotherapy No treatment Relative (95% CI) Absolute (95% CI) 1a Randomized trials Seriousb Not serious Not serious Very seriousb None 21 c 30c – MD 15.22 lower (20.26 lower to 10.18 lower) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; MD: Mean difference a Sullivan et al (10) b Serious concerns about the risk of bias and a sample size < 300. c. Data at post-treatment/end-point only. Baseline data not available for all participants. 59 | P a g e Fig. 6.1. Forest plot: individual psychotherapy compared to no treatment for child sexual maltreatment – 12 month follow-up 60 | P a g e Table 3.7. GRADE evidence profile: individual psychotherapy compared to group psychotherapy for child sexual maltreatment Patient or population – Child sexual maltreatment Intervention – Individual psychotherapy Comparison – Group psychotherapy Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with group psychotherapy Risk with individual psychotherapy Follow–up 6 months Change in PTSD (re– experiencing) (female only) Follow–up: mean 6 months The mean change in PTSD (re– experiencing) was – 0.5 The mean change in PTSD (re– experiencing) in the intervention group was 1.32 lower (2.6 lower to 0.04 lower) – 56 (1 RCT)1 ⨁◯◯◯ VERY LOW2 Assessed using the 19-item Orvaschel’s PTSD scale (extension of the Schedule for Affective Disorders and Schizophrenia for School-Age Children – K-SADS), with re-experiencing of trauma as a subdomain and a 3-point coding system for summation of the relevant items. Calculated important difference shows a medium effect (SMD = -0.53). Change in PTSD (avoidance) (female only) Follow–up: mean 6 months The mean change in PTSD (avoidance) was –1.5 The mean change in PTSD (avoidance) in the intervention group was 1.1 lower (2.39 lower to 0.19 higher) – 56 (1 RCT)1 ⨁◯◯◯ VERY LOW3,4 Assessed using Orvaschel’s PTSD scale. Calculated important difference shows a large effect (SMD = -0.44) FOLLOW–UP 24 MONTHS Change in PTSD (re– experiencing) (female only) Follow–up: mean 24 months The mean change in PTSD (re– experiencing) was – 0.22 The mean change in PTSD (re– experiencing) in the intervention group was 1.74 lower (2.97 lower to 0.51 lower) – 49 (1 RCT)1 ⨁◯◯◯ VERY LOW2 Assessed using Orvaschel’s PTSD scale. Calculated important difference shows a medium effect (SMD = -0.79). 61 | P a g e Patient or population – Child sexual maltreatment Intervention – Individual psychotherapy Comparison – Group psychotherapy Setting – Any Outcomes Anticipated absolute effects (95% CI) Relative effect (95% CI) No. of participants (studies) Quality of the evidence (GRADE) Comments Risk with group psychotherapy Risk with individual psychotherapy Change in PTSD (avoidance) (female only) Follow-up: mean 24 months The mean change in PTSD (avoidance) was -1.5 The mean change in PTSD (avoidance) in the intervention group was 0.6 lower (1.75 lower to 0.55 higher) - 49 (1 RCT)1 ⨁◯◯◯ VERY LOW3,4 Assessed using Orvaschel’s PTSD scale. Calculated important difference shows a small effect (SMD = -0.28). *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 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 1 Trowell et al. (4). 2 Downgraded to very low because of serious concerns about the risk of bias and a sample size < 300 – see Table 4.7 below. 3 Downgraded to very low because of serious concerns about the risk of bias – see Table 4.7 below. 4 Downgraded to very low because the sample size is < 300 and the effect estimate is imprecise – see Table 4.7 below. 62 | P a g e Table 4.7. Quality assessment: individual psychotherapy compared to group psychotherapy for child sexual maltreatment Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Individual psychotherapy Group psychotherapy Relative (95% CI) Absolute (95% CI) Change in PTSD (re-experiencing) (follow-up: mean 6 months) 1a Randomized trials Seriousb Not serious Not serious Very serious None 28 28 – MD 1.32 lower (2.6 lower to 0.04 lower) ⨁◯◯◯ VERY LOW CRITICAL Change in PTSD (avoidance) (follow-up: mean 6 months) 1a Randomized trials Seriousc Not serious Not serious Very seriousd None 28 28 – MD 1.1 lower (2.39 lower to 0.19 higher) ⨁◯◯◯ VERY LOW CRITICAL Change in PTSD (re-experiencing) (follow-up: mean 24 months) 1a Randomized trials Seriousb Not serious Not serious Very seriousb None 27 22 – MD 1.74 lower (2.97 lower to 0.51 lower) ⨁◯◯◯ VERY LOW CRITICAL Change in PTSD (avoidance) (follow-up: mean 24 months) 63 | P a g e Quality assessment No. of patients Effect Quality Importance No. of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Individual psychotherapy Group psychotherapy Relative (95% CI) Absolute (95% CI) 1a Randomized trials Seriousc Not serious Not serious Very seriousd None 27 22 – MD 0.6 lower (1.75 lower to 0.55 higher) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; MD: Mean difference a Trowell et al. (4). b Serious concerns about the risk of bias and a sample size < 300. c Serious concerns about the risk of bias. d The sample size is < 300 and the effect estimate is imprecise. 64 | P a g e Fig. 7.1. Forest plot: individual psychotherapy compared to group psychotherapy for child sexual maltreatment – 6- month follow-up Fig. 7.2. Forest plot: individual psychotherapy compared to group psychotherapy for child sexual maltreatment – 24- month follow-up 65 | P a g e References 1. Foa EB, McLean CP, Capaldi S, Rosenfield D. Prolonged exposure vs supportive counseling for sexual abuse-related PTSD in adolescent girls: a randomized clinical trial. JAMA. 2013;310(24):2650–7. doi: 10.1001/jama.2013.282829. 2. O'Callaghan P, McMullen J, Shannon C, Rafferty H, Black A. A randomized controlled trial of trauma-focused cognitive behavioral therapy for sexually exploited, war-affected Congolese girls. J Am Acad Child Adolesc Psychiatr. 2013;52(4):359–69. 3. Jaberghaderi N, Greenwald R, Rubin A, Zand SO, Dolatabadi S. A comparison of CBT and EMDR for sexually-abused Iranian girls. Clin Psychol Psychother. 2004;11(5):358–68. 4. Trowell J, Kolvin I, Weeramanthri T, Sadowski H, Berelowitz M, Glaser D et al. Psychotherapy for sexually abused girls: psychopathological outcome findings and patterns of change. Br J Psychiatry. 2002;180:234–47. 5. McCrone P, Weeramanthri T, Knapp M, Rushton A, Trowell J, Miles G et al. Cost-effectiveness of individual versus group psychotherapy for sexually abused girls. Child Adolesc Ment Health. 2005;10(1):26–31. 6. King NJ, Tonge BJ, Mullen P, Myerson N, Heyne D, Rollings S et al. Treating sexually abused children with posttraumatic stress symptoms: a randomized clinical trial. J Am Acad Child Adolesc Psychiatry. 2000;39(11):1347–55. 7. Berliner L, Saunders BE. Treating fear and anxiety in sexually abused children: results of a controlled 2-year follow-up study. Child Maltreat. 1996;1(4):294– 309. 8. Deblinger E, Lippmann J, Steer R. Sexually abused children suffering posttraumatic stress symptoms: initial treatment outcome findings. Child Maltreat. 1996;1(4):310–21. 9. Deblinger E, Steer RA, Lippmann J. Two-year follow-up study of cognitive behavioral therapy for sexually abused children suffering post-traumatic stress symptoms. Child Abuse Negl. 1999;23(12):1371–8. 10. Sullivan PM, Scanlan JM, Brookhouser PE, Schulte LE, Knutson JF. The effects of psychotherapy on behavior problems of sexually abused deaf children. Child Abuse Negl. 1992;16(2):297–307.