Geraldine Macdonald1,2, Nuala Livingstone1, Jennifer Hanratty1, Claire McCartan1, Richard Cotmore3, Maria Cary4, Danya Glaser5, Sarah Byford4, Nicky J Welton6, Tania Bosqui1, Lucy Bowes7, Suzanne Audrey6, Gill Mezey8, Helen L Fisher4, Wendy Riches9, Rachel Churchill6. 1. Institute of Child Care Research, School of Sociology, Social Policy and Social Work, Queen's University Belfast, Belfast, UK. 2. School for Policy Studies, University of Bristol, Bristol, UK. 3. Evaluation Department, National Society for the Prevention of Cruelty to Children (NSPCC), London, UK. 4. King's Health Economics, King's College London, London, UK. 5. University College London and Great Ormond Street Hospital for Sick Children, London, UK. 6. School of Social and Community Medicine, University of Bristol, Bristol, UK. 7. Department of Experimental Psychology, University of Oxford, Oxford, UK. 8. Population Health Sciences and Education, St George's, University of London, London, UK. 9. Riches and Ullman Limited Liability Partnership, London, UK.
Abstract
BACKGROUND: Child maltreatment is a substantial social problem that affects large numbers of children and young people in the UK, resulting in a range of significant short- and long-term psychosocial problems. OBJECTIVES: To synthesise evidence of the effectiveness, cost-effectiveness and acceptability of interventions addressing the adverse consequences of child maltreatment. STUDY DESIGN: For effectiveness, we included any controlled study. Other study designs were considered for economic decision modelling. For acceptability, we included any study that asked participants for their views. PARTICIPANTS: Children and young people up to 24 years 11 months, who had experienced maltreatment before the age of 17 years 11 months. INTERVENTIONS: Any psychosocial intervention provided in any setting aiming to address the consequences of maltreatment. MAIN OUTCOME MEASURES: Psychological distress [particularly post-traumatic stress disorder (PTSD), depression and anxiety, and self-harm], behaviour, social functioning, quality of life and acceptability. METHODS: Young Persons and Professional Advisory Groups guided the project, which was conducted in accordance with Cochrane Collaboration and NHS Centre for Reviews and Dissemination guidance. Departures from the published protocol were recorded and explained. Meta-analyses and cost-effectiveness analyses of available data were undertaken where possible. RESULTS: We identified 198 effectiveness studies (including 62 randomised trials); six economic evaluations (five using trial data and one decision-analytic model); and 73 studies investigating treatment acceptability. Pooled data on cognitive-behavioural therapy (CBT) for sexual abuse suggested post-treatment reductions in PTSD [standardised mean difference (SMD) -0.44 (95% CI -4.43 to -1.53)], depression [mean difference -2.83 (95% CI -4.53 to -1.13)] and anxiety [SMD -0.23 (95% CI -0.03 to -0.42)]. No differences were observed for post-treatment sexualised behaviour, externalising behaviour, behaviour management skills of parents, or parental support to the child. Findings from attachment-focused interventions suggested improvements in secure attachment [odds ratio 0.14 (95% CI 0.03 to 0.70)] and reductions in disorganised behaviour [SMD 0.23 (95% CI 0.13 to 0.42)], but no differences in avoidant attachment or externalising behaviour. Few studies addressed the role of caregivers, or the impact of the therapist-child relationship. Economic evaluations suffered methodological limitations and provided conflicting results. As a result, decision-analytic modelling was not possible, but cost-effectiveness analysis using effectiveness data from meta-analyses was undertaken for the most promising intervention: CBT for sexual abuse. Analyses of the cost-effectiveness of CBT were limited by the lack of cost data beyond the cost of CBT itself. CONCLUSIONS: It is not possible to draw firm conclusions about which interventions are effective for children with different maltreatment profiles, which are of no benefit or are harmful, and which factors encourage people to seek therapy, accept the offer of therapy and actively engage with therapy. Little is known about the cost-effectiveness of alternative interventions. LIMITATIONS: Studies were largely conducted outside the UK. The heterogeneity of outcomes and measures seriously impacted on the ability to conduct meta-analyses. FUTURE WORK: Studies are needed that assess the effectiveness of interventions within a UK context, which address the wider effects of maltreatment, as well as specific clinical outcomes. STUDY REGISTRATION: This study is registered as PROSPERO CRD42013003889. FUNDING: The National Institute for Health Research Health Technology Assessment programme.
BACKGROUND:Child maltreatment is a substantial social problem that affects large numbers of children and young people in the UK, resulting in a range of significant short- and long-term psychosocial problems. OBJECTIVES: To synthesise evidence of the effectiveness, cost-effectiveness and acceptability of interventions addressing the adverse consequences of child maltreatment. STUDY DESIGN: For effectiveness, we included any controlled study. Other study designs were considered for economic decision modelling. For acceptability, we included any study that asked participants for their views. PARTICIPANTS: Children and young people up to 24 years 11 months, who had experienced maltreatment before the age of 17 years 11 months. INTERVENTIONS: Any psychosocial intervention provided in any setting aiming to address the consequences of maltreatment. MAIN OUTCOME MEASURES: Psychological distress [particularly post-traumatic stress disorder (PTSD), depression and anxiety, and self-harm], behaviour, social functioning, quality of life and acceptability. METHODS: Young Persons and Professional Advisory Groups guided the project, which was conducted in accordance with Cochrane Collaboration and NHS Centre for Reviews and Dissemination guidance. Departures from the published protocol were recorded and explained. Meta-analyses and cost-effectiveness analyses of available data were undertaken where possible. RESULTS: We identified 198 effectiveness studies (including 62 randomised trials); six economic evaluations (five using trial data and one decision-analytic model); and 73 studies investigating treatment acceptability. Pooled data on cognitive-behavioural therapy (CBT) for sexual abuse suggested post-treatment reductions in PTSD [standardised mean difference (SMD) -0.44 (95% CI -4.43 to -1.53)], depression [mean difference -2.83 (95% CI -4.53 to -1.13)] and anxiety [SMD -0.23 (95% CI -0.03 to -0.42)]. No differences were observed for post-treatment sexualised behaviour, externalising behaviour, behaviour management skills of parents, or parental support to the child. Findings from attachment-focused interventions suggested improvements in secure attachment [odds ratio 0.14 (95% CI 0.03 to 0.70)] and reductions in disorganised behaviour [SMD 0.23 (95% CI 0.13 to 0.42)], but no differences in avoidant attachment or externalising behaviour. Few studies addressed the role of caregivers, or the impact of the therapist-child relationship. Economic evaluations suffered methodological limitations and provided conflicting results. As a result, decision-analytic modelling was not possible, but cost-effectiveness analysis using effectiveness data from meta-analyses was undertaken for the most promising intervention: CBT for sexual abuse. Analyses of the cost-effectiveness of CBT were limited by the lack of cost data beyond the cost of CBT itself. CONCLUSIONS: It is not possible to draw firm conclusions about which interventions are effective for children with different maltreatment profiles, which are of no benefit or are harmful, and which factors encourage people to seek therapy, accept the offer of therapy and actively engage with therapy. Little is known about the cost-effectiveness of alternative interventions. LIMITATIONS: Studies were largely conducted outside the UK. The heterogeneity of outcomes and measures seriously impacted on the ability to conduct meta-analyses. FUTURE WORK: Studies are needed that assess the effectiveness of interventions within a UK context, which address the wider effects of maltreatment, as well as specific clinical outcomes. STUDY REGISTRATION: This study is registered as PROSPERO CRD42013003889. FUNDING: The National Institute for Health Research Health Technology Assessment programme.
Authors: Jill R McTavish; Melissa Kimber; Karen Devries; Manuela Colombini; Jennifer C D MacGregor; Nadine Wathen; Harriet L MacMillan Journal: BMJ Open Date: 2019-04-04 Impact factor: 2.692