Abstract
Mental illness and persistent emotional dysregulation may reorganize family life even when everyday routines remain intact. Measuring total care demand alone cannot establish who absorbs that demand or at what personal cost. This critical conceptual review proposes a framework for jointly measuring care distribution, concealed costs of maintaining routines, and anticipatory emotional attention. Drawing on family stress, resource, resilience, and household labor scholarship, it develops five propositions with specified measurement approaches and conditions that would challenge them. The framework hypothesizes that, at comparable levels of overall demand, concentrated responsibility, unmet needs, and unpredictable adult availability help explain differences in child and caregiver outcomes. Its contribution lies in testing whether these measures add explanatory value beyond established indicators of symptoms, burden, and relationships, including when diagnostic criteria are not met. Existing evidence supports component associations but does not validate the integrated model. Evaluating family adaptation requires examining both the continuity of everyday life and the costs borne by those who sustain it.
Keywords: emotional dysregulation; affective instability; parental mental illness; family systems; child mental health; parentification; emotional security
Introduction
When a family member experiences persistent psychiatric symptoms or recurring difficulty regulating emotions, the consequences may extend beyond that person's distress. Others may reorganize routines, repeatedly de-escalate arguments, or remain preoccupied with anticipating another emotional upheaval. Where illness creates additional care needs, responsibilities may also include attending appointments or replacing lost income. In households with children, these responsibilities raise a developmental question: how does the organization of care affect children's access to dependable adults, ordinary activities, and support for their own difficulties? The question concerns the distribution of needs and resources within a family, rather than the diagnosis of an individual alone.
The association between parental mental illness and offspring psychopathology is well established at the population level. Uher et al. (2023) synthesized 211 independent studies, including more than three million offspring of affected parents, and identified elevated risks extending beyond the parent's specific diagnosis. Yet this evidence cannot establish that a particular family environment caused an individual child's disorder. Shared genetic liability, prenatal exposures, socioeconomic circumstances, and relationships may contribute in different combinations. Heterogeneity was substantial, and the review classified 96% of included studies as having high risk of bias. Its scale strengthens the case for preventive attention while leaving important explanatory questions unresolved.
Family systems theory provides a basis for investigating these questions because it treats relationships and individual development as interdependent (Cox & Paley, 1997). Interdependence, however, should not become an explanation that is invoked without measurement. Stating that emotional difficulties affect the whole family does not identify who undertakes additional care, what resources become scarce, which relationships change, or why some children remain well. A useful account must specify these processes and distinguish adaptation from deterioration.
This article develops a resource and relationship framework for families with dependent children. Its central proposition is that developmental risk may increase when support demands and recurring relational disruption repeatedly exceed sustainable family capacity, particularly when adults' availability diminishes or children assume responsibilities that interfere with their own development. The framework brings established literatures into a common account; it does not claim that family interdependence or resource loss is a new discovery. Its proposed contribution is to make the allocation of care, including its hidden costs, an explicit object of measurement and intervention.
The strongest evidence considered concerns parental illness and child outcomes. Research on parents caring for children with mental health difficulties and on siblings broadens the analysis, but does not establish that identical mechanisms operate across every family relationship. A child whose parent is unwell, a parent caring for an adolescent, and an adult caring for a sibling occupy different developmental and relational positions. These distinctions define the framework's scope and the limits of generalization.
Review approach and conceptual boundaries
This is a critical conceptual review based on purposively selected literature, rather than a systematic review or meta-analysis. Targeted web searches of scholarly records and publisher pages were conducted in October 2026 to identify relevant syntheses, theoretically informative studies, and methodological critiques. Searches addressed parental mental illness and offspring risk, family caregiver distress, young carers and parentification, interparental conflict, family stress, parental emotion regulation, mood instability, and preventive interventions. Article identities and selected claims were checked against accessible publisher records, abstracts, and full texts. No exhaustive database search, independent duplicate screening, formal study-level quality appraisal, or new quantitative synthesis was undertaken. The evidence supports theoretical integration and hypothesis generation, not claims of comprehensive coverage.
Four distinctions are necessary. First, a psychiatric diagnosis is not equivalent to impaired caregiving. Symptoms vary in severity, duration, functional consequences, and responsiveness to treatment. Emotional dysregulation is also not interchangeable with mental illness: difficulties regulating emotions can occur without a diagnosis, and a person with a diagnosis may provide consistent and sensitive care. Studies should measure symptoms, functioning, and caregiving behavior separately.
Second, demands and burden describe different phenomena. Caregiving demands include observable tasks, time commitments, financial expenditure, and interruptions. Perceived burden describes the person's appraisal of strain. The same number of care hours may be experienced differently depending on choice, support, health, and relationship quality. The term burden should refer to care-related strain, not to a person's value or standing within the family.
Third, family functioning should not be reduced to outward calm. A household can maintain routines while one adult loses sleep or a child suppresses their own needs. Conversely, a family can acknowledge distress and disagreements while maintaining reliable care. The relevant outcomes include safety, flexibility, equitable responsibility, and the ability to respond to each member's needs.
Fourth, ordinary contribution must be distinguished from role overload. Helping with chores or offering affection is not inherently harmful. The developmental concern is sustained responsibility that exceeds a child's capacity or compromises sleep, education, peer relationships, or access to adult care. This distinction is consistent with the heterogeneous literature on parentification, which includes both adverse outcomes and reports of competence or meaning (Dariotis et al., 2023).
Emotional volatility beyond diagnostic thresholds
A diagnosis is not the threshold at which family consequences first become important. For this framework, emotional instability refers to marked changes in emotional state over time; emotional dysregulation refers to difficulties managing emotional responses in ways suited to the situation. These constructs overlap but are not interchangeable. Broome et al. (2015) emphasize the need to distinguish features such as the intensity, frequency, and temporal pattern of mood changes. Strong emotion alone is not evidence of pathology, and emotional steadiness should not be equated with suppression or compliance.
Evidence on parental emotion regulation provides a basis for examining family processes dimensionally. Zimmer-Gembeck et al. (2022) synthesized 53 studies and found that better parental regulation was generally associated with more positive parenting, better child emotion regulation, and fewer child internalizing symptoms; evidence for externalizing behavior was less consistent. The review included community and selected-risk samples and showed substantial heterogeneity. It supports studying regulation as a feature of family life, but does not isolate the causal effects of severe volatility among people confirmed to have no disorder.
The proposed exposure is therefore a recurring pattern with observable consequences, assessed through its intensity, frequency, recovery time, and disruption of family activities or relationships. Examples might include repeated escalation during ordinary disagreements, unpredictable withdrawal from agreed responsibilities, or prolonged episodes after which others must restore household routines. These examples describe possible behaviors and demands, not diagnostic criteria. Their interpretation requires attention to stressors, relationship context, and each person's account. Anger in response to mistreatment, for example, should not automatically be classified as an individual regulation deficit.
The framework proposes that such patterns may reorganize a household even without medical appointments or formal caregiving. A partner may spend substantial time choosing words to avoid another escalation; a child may postpone asking for help while monitoring an adult's mood. The relevant research questions concern whether this vigilance persists, whether responsibilities become concentrated, and whether other members' needs go unmet. These are proposed pathways to test, not experiences that can be presumed from the description of a person as emotionally unstable.
Absence of a diagnosis must also be distinguished from demonstrated absence of a disorder: a person may never have been assessed, may have unmet clinical needs, or may experience difficulties that do not satisfy diagnostic criteria. Studies should establish diagnostic status separately from regulation difficulties and family impact. Neither a diagnostic label nor its absence determines whether a relationship is supportive, harmful, or in need of change. Emotional distress also does not excuse intimidation or coercive behavior. Supporting the person experiencing difficulty and protecting other members' needs are compatible aims.
Evidence for effects across family relationships
Parental illness and child development
Intergenerational risk should be understood as probabilistic and multiply determined. McAdams et al. (2014) reviewed children-of-twins studies and explained why parent–child associations can reflect both genetic and environmental processes. Biological parents contribute to the child's inherited liability and rearing environment, making ordinary observational associations difficult to interpret causally. Such findings justify retaining genetic and prenatal pathways alongside the relational and resource processes examined here.
Fathers' mental health belongs within this account. Zecchinato et al. (2025) found associations between paternal anxiety and offspring emotional and behavioral difficulties. This supports attention to fathers as people with their own mental health needs and as participants in children's development. It does not show that their influence operates primarily through mothers. A framework organized around a father's responsibility to produce maternal serenity would impose a directional and gendered hierarchy that these associations do not establish.
Recent scholarship also argues for examining positive outcomes and protective processes. MacBeth et al. (2026), reviewing intergenerational mental health during the first 1,000 days of life, extend a risk-focused account to include resilience and contextual influences. That early developmental emphasis is important, but cannot be assumed to explain later childhood or adolescence without additional evidence. The opportunities and demands associated with care change as children develop.
Distress among parents and siblings
Influence is not exclusively downward from parent to child. Martin et al. (2024) reviewed 32 studies of parents and carers of children and young people with mental health difficulties. Compared with controls, pooled differences indicated greater parental depression (Hedges' g = 0.24, 95% CI [0.11, 0.38]) and parenting stress (g = 0.34, 95% CI [0.20, 0.49]). These comparisons were based on only four and three studies, respectively; the depression finding did not persist in a sensitivity analysis excluding one study. Pre-existing parental difficulties and limited control for confounding constrain causal interpretation. Nevertheless, the review identifies support needs that child-focused treatment alone may not capture.
Siblings are another relevant group. Jayasinghe et al. (2023) synthesized evidence on distress, burden, and wellbeing among siblings of people with mental illness. The included samples were predominantly adult rather than child siblings. Their findings therefore support the broader importance of sibling experiences, while leaving uncertainty about effects on dependent children. Research should examine whether siblings receive less attention, take on additional responsibilities, or experience changes in their relationship with the person who is unwell, rather than inferring those experiences from family membership alone.
These findings challenge the practice of assigning one family member the permanent role of patient and another the permanent role of resource. Several people may need support simultaneously, and the person experiencing illness may also contribute substantially to household care. An adequate account should record contributions as well as needs.
Relationships and emotional security Meta-analytic research links couple relationship quality with parent–child relationship quality (Erel & Burman, 1995). Krishnakumar and Buehler (2000) also identified associations between interparental conflict and parenting behaviors. These findings support investigation of spillover between relationships. They do not imply that every episode of conflict impairs parenting or that symptoms necessarily produce destructive conflict.
Emotional security theory identifies children's responses to interparental conflict as a possible pathway to adjustment difficulties (Davies & Cummings, 1994). In a five-wave study of 232 families, Davies et al. (2016) found prospective pathways involving destructive conflict, children's insecurity, and psychological problems, with developmental differences in their pattern. Child psychological problems also predicted subsequent interparental dysphoria. This is evidence consistent with reciprocal processes; it is not proof of a universal causal sequence beginning with one parent's behavior.
Supportive caregiving may interrupt some of these associations. Vakrat et al. (2018) examined families with chronically depressed mothers and comparison families, observing interactions when children were six years old. Fathering moderated associations between maternal depression and family processes. Because fathering was not randomly assigned, the study does not establish that increasing paternal involvement reduces maternal depression. Nor does a finding concerning observed family cohesion demonstrate the absence of all adverse child outcomes. Its more defensible implication is that the quality of another caregiving relationship may matter when a parent is experiencing depression.
Young caregiving and developmental opportunity
Children may become caregivers as well as recipients of care. Bowman Grangel et al. (2025) reviewed longitudinal evidence on young carers under 25 years of age and found indications of poorer mental health, particularly among those providing more intensive care, alongside inconsistent results. The studies were conducted in high-income countries and included care for diverse conditions, not mental illness alone. This evidence informs hypotheses about caregiving intensity but cannot supply a mental-illness-specific effect for children in all settings.
Parentification research further suggests that the nature and context of responsibility matter (Dariotis et al., 2023). Practical assistance, emotional caretaking, perceived fairness, and available adult support should be differentiated. A child who helps prepare a meal with adult supervision has a different role from a child who believes they must continually manage a parent's emotional state. The proposed framework treats such differences as measurable exposures, rather than assuming that all caregiving is harmful or that apparent competence demonstrates wellbeing.
A resource and relationship framework
Theoretical integration
The framework combines complementary explanations. Family systems theory directs attention to interdependence and reciprocal adaptation (Cox & Paley, 1997). Conservation of resources theory explains why threatened or actual losses of valued resources may generate stress (Hobfoll, 1989). The Family Stress Model describes pathways connecting economic pressure, caregiver distress, relationships, parenting, and child development (Masarik & Conger, 2017). Emotional security theory specifies one process through which children may respond to relational threat (Davies & Cummings, 1994). These perspectives contribute different elements; none independently verifies the integrated framework.
The organizing sequence is as follows: psychiatric impairment or disruptive emotional patterns can create additional support demands; families and services respond by reallocating resources; those responses may preserve or disrupt caregiving relationships and developmental opportunities; changes in family members' functioning can then alter subsequent needs. Figure 1 summarizes the proposed pathways. The sequence is a set of hypotheses about processes over time, not an estimated causal model.
Figure 1. Proposed resource and relationship pathways. Dashed arrows indicate hypotheses, not established causal effects. Support may modify several pathways. The diagram is an organizing schematic rather than a causal identification graph; not all reciprocal or alternative paths are shown.
Care demands and available resources
We propose assessing the fit between care demands and resources repeatedly, without treating the presence or absence of a diagnosis as a sufficient measure of family strain. Relevant demands may include supervision, treatment coordination, additional domestic labor, recurrent efforts to manage conflict, and uncertainty about when support will be needed. Practical caregiving, voluntary emotional support, and accommodation motivated by fear should be measured separately; describing their costs does not imply that family members are obliged to absorb them. Resources include time, income, rest, practical assistance, accessible clinical care, and dependable relationships. These domains should initially be measured separately. Combining them into a single score would require evidence that the score has a coherent meaning and predicts outcomes beyond its components.
A specific candidate mechanism is the diversion of emotional attention. Even when no active task is being performed, a caregiver or child may remain preoccupied with the possibility of another difficult episode. The hypothesis is that this anticipatory attention can reduce engagement with other relationships and activities. It should not be inferred from symptoms or care hours alone. Repeated reports of intrusive concern, interrupted activity, and perceived availability could test whether it contributes information beyond general distress.
Resources are not necessarily depleted whenever symptoms or emotional difficulties occur. Effective treatment, practical assistance, and negotiated responsibility may allow a family to meet additional needs without sustained disruption. Caregiving can also be meaningful. The empirical question is when care remains sustainable and when its costs accumulate or fall disproportionately on members with limited choice.
Distribution of care and concealed costs
The framework distinguishes total care demand from its distribution. Two households may provide similar amounts of support while allocating it very differently. In one, several adults and services share responsibilities. In another, one partner provides nearly all additional care and a child undertakes responsibilities that adults can no longer meet. We propose that the concentration of responsibility may help explain differences in family adaptation even after total demand is considered.
This distributional emphasis adds a testable qualification to standard measures of functioning. Preserved meal schedules, school attendance, or low overt conflict may coexist with exhaustion, lost opportunities, or concealed distress. Family assessment should therefore examine the costs of maintaining routines and ask each member about unmet needs. A reduction in one person's symptoms is valuable, but cannot by itself establish that responsibility has become sustainable for everyone else.
The same logic applies to recovery. Family members may continue monitoring for danger or maintaining restrictive roles after symptoms improve. We propose that restoration of ordinary relationships and opportunities may follow a different timescale from clinical symptom change. This possibility requires longitudinal testing; it should not be assumed in every recovering family.
Predictability and relational availability
Children's experiences may depend on both the amount and predictability of adult availability. Davis and Glynn (2024) review developmental evidence concerning unpredictability across multiple timescales, including patterns of parental signals. That literature motivates a distinction between consistently limited availability and highly variable availability. It does not establish that a particular psychiatric diagnosis produces either pattern.
The framework predicts that uncertainty about care may matter even when average caregiving time is preserved. For example, adults could provide substantial time overall while repeatedly cancelling promised activities during acute episodes. Whether such variability affects security independently of symptom severity and conflict is an empirical question. Measurement should distinguish ordinary flexibility, developmentally necessary changes, and recurring disruption that children experience as threatening or difficult to understand.
Feedback and alternative explanations Feedback can occur at several points. Distress may increase care demands; disrupted sleep or employment may reduce a caregiver's resources; child difficulties may require further appointments and support. Improvement may also be reciprocal, as restored functioning allows relationships and resources to recover. These are plausible extensions of transactional accounts, not evidence that illness spreads automatically through a family.
Competing explanations must remain visible. Economic hardship may precede both illness and family strain. Genetic liability may contribute to symptoms across generations. Relationship difficulties may precede an illness episode, and treatment access may be associated with resources that independently predict child outcomes. A model that attributes every association to resource depletion or one person's emotional volatility would reproduce the causal overreach it is intended to correct. Strong tests must compare the proposed pathways with these alternatives.
Testable propositions and research design
Proposition one concerns changes in resources
Within a family, increases in support demands are hypothesized to precede reductions in caregiver rest, available time, or financial security, followed by changes in relational availability and child functioning. Tests should measure baseline levels and repeated changes in all domains, assess plausible common causes, and compare alternative temporal orderings. If resource changes provide no explanatory information beyond symptoms and prior family conditions, the proposed resource pathway would receive less support.
Proposition two concerns unpredictability
Variability in adult availability is hypothesized to predict children's insecurity beyond average availability, symptom severity, and destructive conflict. Testing this proposition requires measuring emotional volatility and diagnostic status separately, including among participants who do not meet diagnostic criteria. Daily diaries and repeated observations could distinguish within-family variability from differences between households. The proposition would be weakened if predictability measures add no reproducible information once these established exposures and measurement error are considered.
Proposition three concerns the distribution of care
At comparable levels of care demand, concentration of responsibility in one person is hypothesized to be associated with greater unmet needs elsewhere in the family. Records should identify who completes each task, whether backup exists, and whether responsibilities are chosen or imposed. The hypothesis concerns supported and sustainable allocation, not the number or gender of adults in the household. A single-parent family with reliable external assistance may have greater practical capacity than a larger household in which care is concentrated in one person.
Proposition four concerns child role overload
Responsibility that is developmentally inappropriate or interferes with sleep, learning, and peer relationships is hypothesized to predict later difficulties more consistently than occasional, supported contribution. Children's reports of choice and responsibility should be considered alongside task frequency. Distinct measures of emotional caretaking and practical assistance are necessary. Evidence that age-appropriate involvement predicts equivalent difficulties, independent of adversity and prior functioning, would challenge this distinction.
Proposition five concerns recovery across outcomes
Interventions that reduce symptoms may improve family outcomes, but improvements in caregiving distribution and child opportunity are hypothesized to provide additional benefits in families with persistent unmet needs. A feasible test would compare appropriate clinical care with the same care plus a clearly specified family-support component. Prespecified outcomes should include patient functioning, caregiver strain, children's symptoms, and participation in ordinary activities. A benefit for one member accompanied by increased responsibility for another should be recorded as a mixed outcome.
Design requirements Recruitment should include community families as well as clinical samples so that family impact is not defined by service contact. Structured diagnostic assessment can distinguish unassessed difficulties from confirmed absence of a current disorder; dimensional measures can then test whether regulation difficulties contribute information beyond diagnostic status. These propositions require designs that distinguish between-family differences from change within the same family. Conventional cross-lagged panel models can conflate stable individual differences with within-person processes (Hamaker et al., 2015). Random-intercept models or intensive longitudinal approaches may address aspects of that problem, but do not independently remove time-varying confounding or establish causality. Measurement intervals should correspond to the proposed mechanism: daily changes in attention and sleep may not be detectable in annual assessments.
Measurement should also separate the proposed causes, mediators, and outcomes. A burden scale that includes depressive symptoms cannot cleanly establish burden as a distinct predictor of depression. Reports from children, multiple caregivers, and observers can reduce dependence on a single informant, although disagreement may itself be informative. A caregiver's view that routines are stable and a child's report of persistent vigilance need not be treated as interchangeable measurements of the same experience.
Genetically informative designs can help distinguish inherited liability from environmental associations (McAdams et al., 2014). Intervention studies and carefully justified natural experiments can test whether changing specific supports alters subsequent outcomes. Neither a statistically significant indirect association nor improvement after a multicomponent intervention proves that the proposed mediator caused the outcome. Analyses should specify their causal assumptions, handle missing data transparently, and examine whether attrition disproportionately removes families under greatest strain.
Implications for care and policy
Evidence supports cautious optimism about prevention. Lannes et al. (2021) identified 20 randomized trials involving 2,689 children of parents with mental illness. The pooled relative risk of incident mental disorder was 0.53 (95% CI [0.34, 0.84]); effects on internalizing symptoms were small, and the externalizing estimate did not reach statistical significance. These are relative and average intervention effects, not a guarantee for an individual child or proof of the mechanisms proposed here. The findings support making preventive help available without assuming that every exposed child will develop a disorder.
Caregivers may also benefit from support directed at their own wellbeing. Hansen et al. (2022) reported a small pooled reduction in psychological distress across caregiver interventions, with substantial heterogeneity. Their review encompassed varied care-recipient conditions, including dementia, so the aggregate estimate should not be transferred directly to parents or partners in the present framework. It nevertheless illustrates why caregiver outcomes should be evaluated in their own right.
The framework suggests complementing individual assessment with a practical account of family responsibilities. Persistent disruption or unmet needs warrant attention even when diagnostic criteria are not met; support goals can concern regulation skills, reliable routines, boundaries, and communication rather than presumed treatment of an unestablished disorder. Clinicians could establish who depends on the person receiving care, who provides additional support, which ordinary activities have been interrupted, and whether children believe they are responsible for keeping adults well. This is a proposed assessment orientation, not a validated screening instrument. Family involvement should be negotiated with attention to consent, confidentiality, developmental needs, and safety. Where coercion or abuse is present, joint work must not be presumed appropriate, and psychiatric diagnosis must not be used as a proxy for dangerousness.
Fathers should be included as caregivers and potential recipients of support, alongside mothers and other significant adults. The relevant question is who can provide safe, reliable care without being overextended. The aim is not to assign one adult responsibility for regulating another's emotions. For children, age-appropriate explanations and access to a trusted adult can be considered alongside treatment and practical assistance; any support plan should preserve their right to have needs of their own.
Policy implications follow as priorities for evaluation rather than proven consequences of the proposed model. Accessible treatment, flexible work, income support, respite, and coordination between adult and child services may change the resources available for care. Evaluations should record family time, out-of-pocket costs, unmet needs, and distribution of responsibility alongside clinical outcomes. Services that depend on unpaid household labor should make that labor visible when assessing feasibility and benefit.
Limits and directions for refinement
This synthesis is selective, and the proposed framework has not been validated as an integrated model. Evidence from parental depression, paternal anxiety, emotion regulation difficulties, interparental conflict, young caregiving, and adult sibling experiences cannot be treated as interchangeable. Effect estimates from diagnosed populations cannot be applied directly to people with emotional volatility but no established disorder. The broader scope requires testing, not an assumption that these groups face equal risks. Studies differ in diagnosis, age, follow-up, measurement, and social context. Overlapping primary studies across reviews also mean that multiple syntheses should not be counted as independent replications.
The account emphasizes resources and relationships without claiming to explain all intergenerational risk. Prenatal and other biological mechanisms require additional attention, as do disorder-specific processes. Cross-cultural research should measure the meaning, choice, and consequences of responsibility rather than assume that a particular household arrangement is universally adaptive. Evidence from young carers in high-income countries is especially limited as a basis for global conclusions.
Resilience also requires fuller measurement. Walsh's (2016) framework treats family resilience as a dynamic process shaped by context. Applied here, that perspective directs attention to recovery, flexible roles, and accessible support rather than treating families as permanently damaged. Research developed with people experiencing mental illness, partners, and young people could clarify which responsibilities feel meaningful, which are excessive, and which outcomes services currently overlook.
Conclusion
Mental illness and persistent emotional dysregulation may create demands that extend across a family. A diagnostic threshold is not a prerequisite for meaningful relational disruption, but neither diagnosis nor emotional variability determines the family's developmental future. The proposed framework locates a tractable research question in the fit between care demands, available resources, and the allocation of responsibility. It asks whether apparent stability is sustained at another member's expense and whether recovery restores opportunities as well as reduces symptoms. A credible family perspective must recognize the person experiencing difficulty as a participant in relationships, caregivers as people with needs, and children as entitled to dependable care.
References
Bowman Grangel, A., McMahon, J., Dunne, N., O'Donoghue, F., & Gallagher, S. (2025). Health outcomes and psychosocial determinants in young carers: A systematic review. The Lancet Public Health , 10(7), e625–e639. https://doi.org/10.1016/S2468-2667(25)00099-4
Broome, M. R., Saunders, K. E. A., Harrison, P. J., & Marwaha, S. (2015). Mood instability: Significance, definition and measurement. The British Journal of Psychiatry , 207(4), 283–285. https://doi.org/10.1192/bjp.bp.114.158543
Cox, M. J., & Paley, B. (1997). Families as systems. Annual Review of Psychology , 48, 243–267. https://doi.org/10.1146/annurev.psych.48.1.243
Dariotis, J. K., Chen, F. R., Park, Y. R., Nowak, M. K., French, K. M., & Codamon, A. M. (2023). Parentification vulnerability, reactivity, resilience, and thriving: A mixed methods systematic literature review. International Journal of Environmental Research and Public Health , 20(13), 6197. https://doi.org/10.3390/ijerph20136197
Davies, P. T., & Cummings, E. M. (1994). Marital conflict and child adjustment: An emotional security hypothesis. Psychological Bulletin , 116(3), 387–411. https://doi.org/10.1037/0033-2909.116.3.387
Davies, P. T., Martin, M. J., Coe, J. L., & Cummings, E. M. (2016). Transactional cascades of destructive interparental conflict, children's emotional insecurity, and psychological problems across childhood and adolescence. Development and Psychopathology , 28(3), 653–671. https://doi.org/10.1017/S0954579416000237
Davis, E. P., & Glynn, L. M. (2024). Annual Research Review: The power of predictability—patterns of signals in early life shape neurodevelopment and mental health trajectories. Journal of Child Psychology and Psychiatry , 65(4), 508–534. https://doi.org/10.1111/jcpp.13958
Erel, O., & Burman, B. (1995). Interrelatedness of marital relations and parent-child relations: A meta-analytic review. Psychological Bulletin , 118(1), 108–132. https://doi.org/10.1037/0033-2909.118.1.108
Hamaker, E. L., Kuiper, R. M., & Grasman, R. P. P. P. (2015). A critique of the cross-lagged panel model. Psychological Methods , 20(1), 102–116. https://doi.org/10.1037/a0038889
Hansen, N. H., Bjerrekær, L., Pallesen, K. J., Juul, L., & Fjorback, L. O. (2022). The effect of mental health interventions on psychological distress for informal caregivers of people with mental illness: A systematic review and meta-analysis. Frontiers in Psychiatry , 13, 949066. https://doi.org/10.3389/fpsyt.2022.949066
Hobfoll, S. E. (1989). Conservation of resources: A new attempt at conceptualizing stress. American Psychologist , 44(3), 513–524. https://doi.org/10.1037/0003-066X.44.3.513
Jayasinghe, A., Wrobel, A., Filia, K., Byrne, L. K., Melvin, G., Murrihy, S., Moller, C., Berk, L., Berk, M., & Cotton, S. (2023). Distress, burden, and wellbeing in siblings of people with mental illness: A mixed studies systematic review and meta-analysis. Psychological Medicine , 53(15), 6945–6964. https://doi.org/10.1017/S0033291723001733
Krishnakumar, A., & Buehler, C. (2000). Interparental conflict and parenting behaviors: A meta-analytic review. Family Relations , 49(1), 25–44. https://doi.org/10.1111/j.1741-3729.2000.00025.x
Lannes, A., Bui, E., Arnaud, C., Raynaud, J.-P., & Revet, A. (2021). Preventive interventions in offspring of parents with mental illness: A systematic review and meta-analysis of randomized controlled trials. Psychological Medicine , 51(14), 2321–2336. https://doi.org/10.1017/S0033291721003366
MacBeth, A., Morales, M. F., & Golds, L. (2026). Intergenerational transmission of mental health during the first 1,000 days of life. Nature Reviews Psychology , 5, 105–119. https://doi.org/10.1038/s44159-025-00521-2
Martin, F., Dahmash, D., Wicker, S., Glover, S.-L., Duncan, C., Anastassiou, A., Docherty, L., & Halligan, S. (2024). Psychological well-being and needs of parents and carers of children and young people with mental health difficulties: A quantitative systematic review with meta-analyses. BMJ Mental Health , 27(1), e300971. https://doi.org/10.1136/bmjment-2023-300971
Masarik, A. S., & Conger, R. D. (2017). Stress and child development: A review of the Family Stress Model. Current Opinion in Psychology , 13, 85–90. https://doi.org/10.1016/j.copsyc.2016.05.008
McAdams, T. A., Neiderhiser, J. M., Rijsdijk, F. V., Narusyte, J., Lichtenstein, P., & Eley, T. C. (2014). Accounting for genetic and environmental confounds in associations between parent and child characteristics: A systematic review of children-of-twins studies. Psychological Bulletin , 140(4), 1138–1173. https://doi.org/10.1037/a0036416
Uher, R., Pavlova, B., Radua, J., Provenzani, U., Najafi, S., Fortea, L., Ortuño, M., Nazarova, A., Perroud, N., Palaniyappan, L., Domschke, K., Cortese, S., Arnold, P. D., Austin, J. C., Vanyukov, M. M., Weissman, M. M., Young, A. H., Hillegers, M. H. J., Danese, A., … Fusar-Poli, P. (2023). Transdiagnostic risk of mental disorders in offspring of affected parents: A meta-analysis of family high-risk and registry studies. World Psychiatry , 22(3), 433–448. https://doi.org/10.1002/wps.21147
Vakrat, A., Apter-Levy, Y., & Feldman, R. (2018). Fathering moderates the effects of maternal depression on the family process. Development and Psychopathology , 30(1), 27–38. https://doi.org/10.1017/S095457941700044X
Walsh, F. (2016). Applying a family resilience framework in training, practice, and research: Mastering the art of the possible. Family Process , 55(4), 616–632. https://doi.org/10.1111/famp.12260
Zecchinato, F., Ahmadzadeh, Y. I., Kreppner, J. M., & Lawrence, P. J. (2025). A systematic review and meta-analysis: Paternal anxiety and the emotional and behavioral outcomes in their offspring. Journal of the American Academy of Child & Adolescent Psychiatry , 64(2), 172–197. https://doi.org/10.1016/j.jaac.2024.04.005
Zimmer-Gembeck, M. J., Rudolph, J., Kerin, J., & Bohadana-Brown, G. (2022). Parent emotional regulation: A meta-analytic review of its association with parenting and child adjustment. International Journal of Behavioral Development , 46(1), 63–82. https://doi.org/10.1177/01650254211051086