Notes on Preventive Interventions for COPMI: Systematic Review (Puchol-Martínez, Vallina-Férnandez, Santed-Germán, 2023)

Introduction

COPMI (children of parents with mental illness) represent a high-risk group for developing mental illness (MI) or behavioral problems. The background evidence notes wide variability in prevalence estimates and risk across studies. Broadly cited figures indicate COPMI prevalence ranging from about 11% to 23%, depending on sample and setting, with lifetime risk of MI elevated relative to peers (odds ratio approximately 2.4 in some meta-analytic syntheses). In particular, meta-analytic work on severe mental illness (SMI) in parents (psychotic or severe mood disorders) suggests children have roughly a one-third chance of developing SMI after age 20, compared with controls (about 32% risk). Several mechanisms are proposed to explain transmission risk, including genetic vulnerability, prenatal exposures, family dynamics, parenting quality, stress reactivity, and attachment patterns. Protective factors are also highlighted, including coping strategies (emotional awareness, problem-solving), social support, and adaptive family processes. Positive parenting programs and family-based interventions show promise for buffering risk, with evidence that universal and targeted preventive interventions can promote well-being and prevent mental disorders across development. The present review focuses on preventive psychotherapeutic interventions for COPMI, aiming to evaluate their efficacy in reducing internalizing/externalizing symptoms, resilience, coping, quality of life, family relationships, and MI risk in children aged 4–18 years who do not have a diagnosed MI themselves.

Key concepts and definitions

• COPMI: Children living with a parent who has a diagnosed MI. Protective mechanisms include coping strategies, social support, psychoeducation, and family-focused interventions.

• Preventive interventions: Psychotherapeutic approaches delivered by health/social professionals using psychological principles to help children and families understand, reduce, or manage problems related to parental MI and to bolster protective factors.

• Outcome domains: Internalizing symptoms (e.g., anxiety, depression), externalizing symptoms (e.g., conduct problems), prosocial behavior, coping style, resilience, quality of life, family relationships, and MI literacy (knowledge about MI).

• Effect size conventions: Cohen’s d used for continuous outcomes; odds ratio (OR) and hazard ratio (HR) for risk outcomes; RCI (Reliable Change Index) and various correlational indices (rpb) reported in several studies.

Methodological overview

• Design: Qualitative systematic review following PRISMA 2020 guidelines; integrative perspective with emphasis on primary studies conducted in real-world conditions.

• Information sources: PsycArticles, PsycInfo, Medline (via EBSCO), ScienceDirect, Scopus, SpringerLink, Web of Science, plus grey literature; an external reviewer replicated the search.

• Inclusion criteria: (1) intervention involving children (4–18 years) with at least one parent diagnosed with MI; (2) child is without a diagnosed MI themselves; (3) quantitative or mixed results; (4) intervention explicit and manualized; (5) abstract available; (6) languages English/Spanish; (7) quantitative outcomes.

• Exclusion criteria: (1) exclusively narrative reviews; (2) MI limited to substance-use disorder only (dual disorder allowed); (3) children in treatment for other primary diagnoses; (4) intellectual disabilities.

• Quality appraisal: Eight criteria adapted from Cochrane and the Cuijpers framework, applied via the Joanna Briggs Institute (JBI) checklist. Criteria included parental MI diagnostic method, treatment manual, therapist training, treatment integrity, intention-to-treat reporting, adequate statistical power (≥50 per group), independent randomization, and blinded outcome assessment. Risk of bias also discussed in terms of implementation bias and reliance on self-report measures.

• Study selection: 1267 initial references, 15 studies ultimately included across several countries (Australia, USA, the Netherlands, Finland, Germany, Korea), comprising 1941 children and 1328 parents.

• Outcome and synthesis: Given heterogeneity in interventions, populations, and measures, a meta-analysis was not performed; instead, a qualitative synthesis highlighted patterns in internalizing/externalizing outcomes, mental health literacy, prosocial behavior, and parent–child relationships, with attention to follow-up duration up to 24 months.

Results: study characteristics and samples

• Included studies: Compas et al. (2010, 2015), Fernando et al. (2018), Foster et al. (2016), Fraser & Pakenham (2008), Ginsburg et al. (2015), Goodyear et al. (2009), Grové et al. (2013, 2015), Im & Heo (2021), Matthews & Nicholls (2012), Pitman & Matthey (2004), Solantaus et al. (2010), Stemmler et al. (2013), van Santvoort et al. (2014).

• Total sample: 1941 children; intervention group (IG) 1343; control group (CG) 598; child gender: 948 females, 848 males; mean age 10.28 years (SD = 1.67). A subset (Solantaus et al., 2010) did not report sex and age. Parental MI was documented across studies, with seven studies focusing on parents with major depressive disorder (MDD) totaling 991 parents.

• Parental diagnoses: beyond MDD, included anxiety disorders, bipolar disorder, schizophrenia, borderline personality disorder, PTSD, OCD, and schizoaffective disorder. In several studies, MI was experienced by one parent or both; 53 families (9%) had both parents with MI.

• Recruitment and settings: Most samples recruited from community settings; some from mental health centers or clinical samples.

Interventions: formats, targets, and components

• Intervention formats: wide heterogeneity with multicomponent designs. Formats included group-based clinical sessions, family sessions, and peer-led/peer-supported approaches; some used a DVD-based family psychoeducation component (Grove et al., 2013).

• Core components:

  • Psychoeducation about MI and coping strategies; evidence that MH literacy reaches significant improvements across measures.

  • Cognitive-behavioral (CBT) components in group formats were used in several studies (e.g., Compas et al., 2010; 2015; Ginsburg et al., 2015; Im & Heo, 2021).

  • Peer-support and resilience-building elements were central in multiple programs (e.g., Foster et al., 2016; Fraser & Pakenham, 2008; Goodyear et al., 2009; Grové et al., 2015; Matthews & Nicholls, 2012; Pitman & Matthey, 2004; van Santvoort et al., 2014).

  • Family-focused vs. child-only components: in some studies, parents participated as part of a family intervention (e.g., Fernando et al., 2018; Solantaus et al., 2010; Stemmler et al., 2013; van Santvoort et al., 2014), while others concentrated on child outcomes with parental psychoeducation delivered as a separate module or not at all.

• Parental involvement: eight studies included an intervention for parents (often alongside child-focused components); five studies emphasized joint participation or family sessions; some interventions targeted mothers specifically (e.g., Stemmler et al., 2013).

• Treatment manuals and therapist training: most programs were manualized; several included explicit therapist training and supervision; integrity checks varied across studies.

• Follow-up durations: follow-ups ranged from 2 months to 24 months; Compas et al. (2015) reported the longest follow-up at 24 months; several studies had no follow-up.

• Notable programs and study-specific notes:

  • Compas et al. (2010): family-group CBT preventive intervention; IG vs. CG with 12 sessions; significant improvements in child symptoms and parenting practices; follow-up at 12 months.

  • Compas et al. (2015): extended family-based CBT; 12 sessions; follow-up to 24 months; reductions in child depression and anxiety symptoms; lower incidence of major depressive episodes at 24 months; multiple effects persisted at 18–24 months.

  • Fernando et al. (2018): targeted MDD; multi-component intervention including individual, group, and family sessions; TAU vs TAU+Kanu-intervention; improvements in parent–child relationship and reductions in parental rejection; 6-month follow-up.

  • Solantaus et al. (2010): Finland; Family Talk Intervention (FTI) vs Let Us Talk (LT); included parents and whole family; both reduced emotional symptoms, anxiety, and hyperactivity with FTI showing earlier effects.

  • Stemmler et al. (2013): EFFEKT-E (Germany); mothers and children engaged; TAU+intervention vs TAU; improvements in mothers’ perceived competence and reductions in parental stress (IG vs CG, with larger effect in IG).

  • van Santvoort et al. (2014): Netherlands; group sessions for children and parents with family sessions; mixed results; 31.7% IG vs 18.9% CG improved in emotional/behavioral problems, but no significant overall effects.

  • Goodyear et al. (2009) and Grové et al. (2013, 2015): peer-based and psychoeducation-focused approaches; some studies reported significant improvements in coping and knowledge about MI; DVD-based family programs; mixed results on prosocial behavior.

  • Im & Heo (2021): Korea; CBT-based, group format; sessions with boosters; findings indicated some improvements but also instances of increased disruption when parental symptoms worsened.

  • Matthews & Nicholls (2012): Australian COPMI programs (Taz Club, Champs, etc.); large sample but mixed methodological quality; Taz Club showed improvements in peer connections and coping, but age-related effectiveness noted (less effective for children over 12).

  • Pitman & Matthey (2004): SMILES program; group-based resilience focus; three days of intensive sessions; mixed results.

Outcomes: internalizing, externalizing, literacy, and other domains

• Internalizing symptoms: most studies (80% measured) found statistically significant improvements post-intervention; clinically meaningful changes observed in about two-thirds of these cases; large effect sizes reported in at least one study (e.g., Ginsburg et al., 2015 with a large effect). Follow-up effects remained significant in several studies (_compas 2010, 2015; Fernando 2018; Fraser & Pakenham 2008; Ginsburg 2015; Solantaus 2010; van Santvoort 2014).

• Externalizing symptoms: measured in 47% of studies; significant improvements in four studies; follow-up significance in three (Compas 2010/2015; Fernando 2018).

• Coping styles and resilience: analyses showed coping styles changes in subset of studies; small effect sizes in some follow-ups (e.g., Fraser & Pakenham 2008; van Santvoort 2014).

• Mental health literacy (MH literacy): psychoeducation was a central component in ~87% of studies and was measured in six; significant improvements in all measured studies, with large effects in some (Pitman & Matthey, 2004; rpb values reported as high as 0.55–0.88).

• Quality of life and well-being: several studies reported improvements on life-satisfaction/quality-of-life scales; effects ranged from small to moderate, with some post-intervention improvements and mixed follow-up results (e.g., Fernando 2018; Foster 2016; Fraser & Pakenham 2008).

• Prosocial behavior: improvements observed in multiple studies, with several reporting statistically significant gains (e.g., Foster 2016; Matthews & Nicholls 2012; Goodyear 2009; Solantaus 2010).

• Risk of MI in offspring: two studies assessed future MI risk; Compas et al. (2015) reported an odds ratio (OR) of 2.37 with 95% CI
  95\%\ CI=[1.05, 5.35] for depression; Ginsburg et al. (2015) reported a hazard ratio (HR) of 6.60 with 95% CI
  95\%\ CI=[2.00, 21.82]; p=0.002 for anxiety disorders.

• Other variables: parent–child relationship improved in some family-intervention studies; reductions in parental distress; changes in caregiver experience and family functioning varied across studies.

Key methodological notes and limitations

• Heterogeneity: substantial variation in parental diagnoses (MDD predominance but other MI types represented), sample characteristics, intervention content, format (group vs. family vs. mixed), and outcome measures. This heterogeneity explains why meta-analysis was not feasible.

• Risk of bias: across randomized and non-randomized studies there was a notable risk of bias due to implementation factors (participants often knew they were in the intervention) and reliance on self-report measures; control groups were sometimes non-equivalent or absent.

• Follow-up: the longest reported follow-up was 24 months (Compas et al., 2015); no study followed participants beyond 24 months, limiting conclusions about long-term effects.

• Generalizability and geography: while several studies were from Anglophone contexts (Australia, USA) and Europe, there was limited representation from some regions; the review notes a lack of Spain-based projects and underscores need for country-specific research and policy translation.

• Measurement variability: a wide array of instruments was used (CBCL, YSR, K-SADS-PL, ADIS-R, CDI, SDQ, KID-KINDL-R, etc.), with some measures showing robust psychometrics and others less so; interpretation of effect sizes requires care due to instrument diversity and directionality of scales.

Family and systemic considerations

• Family involvement often amplifies impact: studies with family components or joint parent–child sessions tended to show greater and more durable effects, particularly on internalizing symptoms and parenting practices.

• Parental mood as a mediator: parental distress was a key mediator in child outcomes (e.g., Ginsburg et al., 2015), highlighting the interdependence of parent and child mental health in preventive interventions.

• Unexpected/benign paradoxes: some studies reported increases in disruptive behaviors or reduced prosocial behavior during the intervention period, potentially linked to increased awareness (self-stigma) or disclosure of family illness; improvements often emerged at follow-up.

Practical implications and recommendations for future work

• Integrated, developmentally tailored programs: evidence supports combining psychoeducation with peer-support to foster resilience; group formats with peer interactions appear particularly beneficial.

• Age-specific content: results suggest interventions may be less effective for older children/adolescents (e.g., over 12 years in Matthews & Nicholls); content should be adapted to developmental stage, with explicit adolescent-focused modules.

• Duration matters: longer, sustained engagement (as in Taz Club) may enhance peer connections and coping skills; duration appears to matter for efficacy, particularly among younger children.

• Family-wide approaches: Finland’s FTI vs LT results imply family-focused interventions can achieve earlier and more robust gains; consider family-systemic approaches as a standard option.

• Policy and dissemination: the review highlights a need for national policy action and for trials in various countries, including Spain, to inform selective prevention programs and potential scale-up.

• Methodological improvements: more RCTs with rigorous designs, standardized outcome measures, longer follow-up, and reporting of implementation fidelity are needed to establish robust, generalizable conclusions.

Conclusions

• Overall, preventive interventions for COPMI, especially those involving parents with MDD, show clinically and statistically meaningful effects, notably in reducing internalizing symptoms at follow-up (effect sizes ranging roughly from small to moderate, with some larger effects in certain studies). The evidence base supports psychoeducation and peer-supported, group-based or family-interactive formats as effective components.

• There is evidence of transmission risk reduction for certain MI outcomes in offspring, though long-term data are limited (no study beyond 24 months). The two most robust indicators of reduced risk were the depression-related and anxiety-related outcomes in the studies that conducted longer follow-ups.

• Despite heterogeneity and methodological limitations, the convergence of findings across multiple countries and designs suggests that preventive COPMI interventions can be beneficial, particularly when they promote coping, resilience, and social connectedness through group formats and family involvement.

• The field would benefit from larger, methodologically sound randomized trials with longer-term follow-up and from country-specific implementation research to guide policy, resource allocation, and program design in diverse contexts.

Connections to broader concepts

• This review aligns with a public health framework that emphasizes prevention, early intervention, and the strengthening of protective factors (e.g., coping, social support) to mitigate the development of MI in at-risk youth.

• The role of parental distress as a mediator underscores family systems theory: child outcomes are shaped by the parent–child dyad and broader family functioning, not by the child in isolation.

• Psychoeducation as a core, cross-cutting component resonates with health literacy and empowerment approaches, suggesting that increasing children’s knowledge about MI improves communication, reduces fear/stigma, and promotes resilience.

• The emphasis on peer-based group formats reflects social learning and social-cognitive theory, highlighting the importance of positive peer influence and social modeling in protective development.

Key formulas and statistics (selected)

• Odds ratio for depression risk in offspring (Compas et al., 2015):
  OR = 2.37, \ 95\% \ CI \ [1.05, 5.35]

• Hazard ratio for anxiety risk (Ginsburg et al., 2015):
  HR = 6.60, \ 95\% \ CI \ [2.00, 21.82], \ p = 0.002

• Cohen’s d for effect sizes (various studies): observed values ranged from around |d| \approx 0.28-0.57 for post-test/internalizing outcomes, with some larger effects (e.g., d\approx -0.64 or similar in specific measures indicating improvement with lower scores). Note that directionality depends on the specific instrument; negative d can indicate improvement for some scales (e.g., quality of life measures where higher scores reflect worse status).

• Psychoeducation effects: large effects reported in MH literacy measures in several studies, e.g., Pitman & Matthey (2004) reported a large effect in MH literacy, with rpb values between 0.55 and 0.88.

• Follow-up durations: longest follow-up reported was 24 ext{ months} (Compas et al., 2015); several studies had follow-ups of 6–12 months or none at all.

References to specific studies (highlights)

• Compas et al. (2010, 2015): family-based CBT preventive interventions; multiple outcomes including internalizing/externalizing symptoms and parenting behaviors; long follow-up up to 24 months.

• Fernando et al. (2018): MDD-focused intervention; mixed components; improvements in parent–child relationships; 6-month follow-up.

• Solantaus et al. (2010): Finland; Family Talk Intervention vs LT; early and sustained benefits in emotional symptoms and prosocial behavior.

• Ginsburg et al. (2015): randomized trial reducing onset of anxiety disorders in offspring of anxious parents; medium-to-large effects; parental distress as mediator.

• Matthews & Nicholls (2012); van Santvoort et al. (2014); Goodyear et al. (2009); Grové et al. (2013, 2015): varied formats including Taz Club, Champs, Champs-related platforms, and DVD-based family psychoeducation; mixed efficacy patterns.

Ethical and practical implications

• Stigma and disclosure: some interventions may temporarily raise self-stigma or awareness-related distress in children; mindful support and supervision are essential during psychoeducation components.

• Social inclusion and belonging: peer-supported formats and group-based interventions promote social connectedness and resilience, potentially mitigating isolation and stigma.

• Equity and access: disparities in who receives preventive services (e.g., access to mental health care, language, cultural adaptation) require policy attention to ensure COPMI across diverse populations have opportunities to participate in prevention programs.

• Implications for practice: clinicians and service planners should consider family-centered approaches, incorporate psychoeducation, and design age-appropriate content; longer-duration programs may yield more durable benefits, especially for younger children.

Limitations and future directions (summarized)

• Limitations: heterogeneity in MI types among parents, participant ages, intervention components, and outcome measures; lack of long-term follow-up beyond 2 years; variable methodological quality across studies.

• Future research directions: conducting more RCTs with standardized measures, longer-term follow-up, and cross-cultural replication; examining differential efficacy by parental MI type, single vs. dual parental MI, and child age group; investigating mechanisms (e.g., parental distress as mediator) and potential moderators (e.g., family functioning, social support networks).

Takeaway for exam preparedness

• The preventive COPMI literature supports the efficacy of psychoeducation plus peer-support and family-inclusive interventions to reduce internalizing symptoms in children and to promote resilience and coping. Effects are most robust for internalizing symptoms at follow-up and when interventions involve both children and parents in a structured, manualized format. Long-term outcomes require further study, and program design should be developmentally sensitive and culturally adaptable.