The clinical effectiveness, cost-effectiveness and acceptability of community-based interventions aimed at improving or maintaining quality of life in children of parents with serious mental illness: a systematic review

Background: Serious parental mental illness poses a challenge to quality of life (QoL) in a substantial number of children and adolescents. Improving the lives of these children is a political and public health concern. Objectives: To conduct an evidence synthesis of the clinical effectiveness, cost-effectiveness and acceptability of community-based interventions for improving QoL in children of parents with serious mental illness (SMI). Data sources: Nineteen health, allied health and educational databases, searched from database inception to May 2012, and supplemented with hand searches, reference checking, searches of grey literature, dissertations, ongoing research registers, forward citation tracking and key author contact. Review methods: Inclusion criteria required >= 50% of parents to have SMI or severe depression confirmed by clinical diagnosis or baseline symptoms. Children were <= 18 years of age. Community-based interventions included any non-residential psychological/psychosocial intervention involving parents or children for the purposes of improving health or well-being. Intervention comparators were not predefined and primary outcomes were validated measures of children's QoL and emotional health. Secondary outcomes were derived from UK policy and stakeholder consultation. Data were extracted independently by two reviewers and the study quality was assessed via Cochrane criteria for randomised/non-randomised designs, Critical Appraisal Skills Programme (CASP) qualitative criteria or a standard checklist for economic evaluations. Separate syntheses were conducted for SMI and severe depression. Standardised effect size (ES) trials were pooled using random-effects modelling for which sufficient data were available. Economic data were summarised and acceptability data were synthesised via a textual narrative approach. Results: Three trials targeted mothers/the children of mothers with psychotic symptoms. Children were <= 12 years of age and no primary QoL or emotional health outcomes were reported. Insufficient secondary outcome data prevented pooling and no eligible economic evaluations were found. Twenty-six trials targeted parents/children of parents with severe depression; 18 recruited mothers of infants < 2.5 years of age. Data pooling suggested no significant short-term effect on children's emotional health [standardised ES 0.06, 95% confidence interval (CI) -0.20 to 0.33] or social function (standardised ES 0.23, 95% CI 0.00 to 0.46). Medium to large effects were observed for parents' depressive symptoms (standardised ES 0.73, 95% CI 0.51 to 0.94) and parenting behaviours (standardised ES 0.67, 95% CI 0.32 to 1.02). One non-randomised economic evaluation was found. Intervention uptake and adherence were inconsistently reported. Incomplete evidence highlighted potential barriers from child custody losses and conflicting life circumstances. Qualitative data suggesting interventions to overcome social isolation and stigma are well received by parents. Limited data suggested that children may value peer interactions and normalising activities. Limitations: Included trials were of poor or unclear quality with inadequate randomisation or allocation concealment, possible attrition biases and incomplete outcome reporting. Meaningful analysis was challenged by clinical and methodological heterogeneity and insufficient data for subgroup comparisons. Children's self-reports were lacking and evidence of effect remains biased towards parent-based interventions for severely depressed mothers of infants. Generalisability to other diagnoses, older children and children of fathers with SMI is unclear. A lack of high-quality economic data prevented economic modelling. Conclusion: Evidence for community-based interventions to enhance QoL in children of SMI parents is lacking. The capacity to recommend evidence-based approaches is limited. Rigorous development work is needed to establish feasible and acceptable child-and family-based interventions, prior to evaluating clinical effectiveness and cost-effectiveness via a randomised controlled trial (RCT). A substantial programme of pilot work is recommended to underpin the development of feasible and acceptable interventions for this population. Evaluations should incorporate validated, child-centred QoL outcome measures, high-quality cost data and nested, in-depth acceptability studies. New age-appropriate instruments that better reflect the life priorities and unique challenges faced by children of parents with SMI may need to be developed.