Author
FRASER Caitlin; et al;
Intervention programs for children of parents with a mental illness: a critical review.
Journal citation/publication details
International Journal of Mental Health Promotion, 8(1), February 2006, pp.9-20.
Summary
Children whose parents are affected by mental illness are widely acknowledged to be at risk of a range of poor outcomes, and the initial searches for this systematic review identified over 500 papers dealing with interventions for this group. Twenty-six were included in the review, of which only seven were rated as methodologically strong. As a result there is very limited good quality evidence of programme effectiveness, and the authors make only general recommendations for practice.
Context
Although parents with a mental illness can, and do, carry out their parental role well, many face major challenges in this area of their lives. Past research has established that their children are at increased risk of developing their own mental health problems, as well as a range of other negative outcomes such as early parenthood and relationship violence in adulthood. Biological sources of risk to the mental health of these children are less important than social and psychological issues such as family breakdown, poverty and discrimination, and many interventions have been developed to support these families. Many, however, are not supported by research, and the aim of this review is to examine their effectiveness (or otherwise) in a systematic way.
Methods
The authors used ‘accepted' (and referenced) systematic review methods, encompassing: question formation; searching for studies; relevance appraisal; study selection; data extraction; quality assessment; and drawing conclusions.
What sources were used?
A ‘variety' of databases in health, medicine, sociology, psychology and education were searched by a professional librarian and a PhD student (both included among the authors). The reference lists of all reviews and relevant papers were checked, eight ‘particularly relevant' (but unnamed) journals were hand searched, and ‘grey literature' was sought through internet searches and contact with experts in the field.
What search terms/strategies were used?
The strategy included terms that focused on participants (family, child, teenager etc), the type of intervention (randomised controlled trial, quasi-experimental etc), a range of mental health problems (depression, schizophrenia etc) and potential outcomes (social support, quality of life, mental well-being etc). These were reviewed by a multidisciplinary group of health practitioners before the list was finalised. Although relatively little detail of sources and search strategies is given in the paper, the involvement of an information professional is a probable indicator of thoroughness.
What criteria were used to decide on which studies to include?
A standardised and referenced tool was used to select studies in which the target population was children and/or families affected by parental mental illness and subject to an intervention that had been evaluated (as opposed to merely described).
Who decided on their relevance and quality?
The initial searches delivered 532 papers. Data were extracted (using the same tool) from those deemed relevant, covering: description of study intervention and design; target population and number of participants; source of funding, intervention provider and intervention setting; geo-political context; length of intervention and follow-up; and description of authors' conclusions. A further referenced tool was used to assess study quality and provide a rating of ‘strong', ‘moderate' or ‘weak'.
How many studies were included and where were they from?
Much of the literature is descriptive, and only 26 papers met the inclusion criteria. These are summarised in Table 1 and originated from the USA (18), Australia (4), UK (3) and Israel (1). Most were undertaken in large urban areas, but in other respects (such as sample size, target populations and types of intervention) they varied markedly. Seven studies were assessed as methodologically strong, four as moderate and 15 as weak (mainly on the grounds of study design, randomisation and blinding).
How were the study findings combined?
The synthesis of findings is narrative, focusing on broad issues, with some recommendations for future interventions derived from the eleven studies deemed methodologically strong or moderate (and summarised in more detail in Table 2).
Findings of the review
Four general lessons emerged from the review. First, despite the broad range of risk and protective factors identified as important to the well-being of families affected by parental mental illness, all the studies focused on individual-level issues such as knowledge and skills. Few considered structural or social factors. Second, the studies included limited follow-up data and analysis of the cost-effectiveness of interventions. Third, there was a lack of clarity about the key needs or issues affecting the target population, with ‘enormous variety, in the outcomes measured. And finally, all but one of the studies were carried out in large urban areas, most in mental health centres, casting doubt on the transferability of the review results to different types of setting.
The methodologically strong and moderate studies provided ‘very limited evidence' of effectiveness, but none suggested that any existing programmes were harmful. There is some evidence of the effectiveness for young children of depressed mothers of baby massage and toddler-parent psychotherapy, while non-depressed adolescent children can be helped by group cognitive behavioural therapy (CBT). However, CBT was not effective in depressed adolescent offspring of depressed parents. When ‘absence of psychopathology in childhood' is measured as an outcome for those who have parents with a serious mental illness, there is no evidence for the effectiveness of education sessions, home visits from nurses, interaction coaching for mothers, counselling, CBT or brief psychodynamic psychotherapy.
There is some evidence that interventions can have positive effects on ‘surrogate markers' of child psychological well-being. For example: baby massage can lead to improved weight gain, reduced stress and better maternal-child interaction; clinician-led education sessions can improve children's understanding of parental mental illness; and home visits from nurses can produce reductions in depressive mood scores and stress among parents, and less ‘punitive parenting'. However, ‘there was no evidence that these surrogate markers resulted in improved mental health in children'.
Authors' conclusions
While finding ‘some support for interventions for children of parents with mental illness in specific circumstances, and no evidence that these programs were harmful…there is, however, insufficient evidence to reach any firm conclusions about the range of programs that might play a part in primary prevention for this target group'.
Implications for policy or practice
Despite the lack of a ‘robust body of evidence', the review does offer some broad recommendations. Practitioners should use a recognised theory in developing interventions, and link programme components to identified risk factors for their target group. Components should also ‘be selected from across the public health spectrum' so that interventions take account of broader social and structural factors, rather than being focused solely on individual-level issues. This may require a broader range of agencies to be involved in intervention development and delivery, and additional training for mental health professionals. Finally, future interventions need to be evaluated using rigorous methodological designs, validated tools, appropriate outcome measures and longer follow-up.