Epilepsy is increasingly understood as a disorder with neurological, cognitive, psychiatric and social dimensions rather than an illness defined exclusively by recurrent seizures. Psychiatric disorders occur more frequently in people with epilepsy than in people without epilepsy. A recent systematic review and meta-analysis involving more than half a million people with epilepsy reported increased odds of several psychiatric conditions, including depression, anxiety disorders, bipolar disorder and psychotic disorders [1]. For families, the implications of epilepsy frequently extend into everyday supervision, treatment administration, protection from seizure-related injury, emergency preparedness, transport to healthcare facilities and restrictions on the caregiver's employment or social activities. The resulting strain is commonly conceptualised as caregiver burden, encompassing emotional, physical, social, occupational and financial consequences. Traditional explanations for caregiver burden have emphasised seizure frequency, duration of illness, drug resistance and dependency in activities of daily living. Such factors remain important, but they do not fully account for the variability in family experiences. In childhood intractable epilepsy, for example, behavioural problems showed a moderate-to-high correlation with maternal parenting stress, whereas seizure and demographic characteristics did not demonstrate comparable correlations [4]. The behavioural phenotype appears particularly relevant in paediatric disease. Children with epilepsy can exhibit anxiety, depressive symptoms, withdrawal, somatic complaints, aggression, attention difficulties and other internalising or externalising behaviours. In one controlled study, children with epilepsy had significantly greater emotional and behavioural symptoms than healthy controls, while parental stress correlated with total, internalising and externalising behavioural-problem scores [9]. More recent prospective evidence suggests that the pattern varies developmentally: internalising behaviour is particularly associated with stress in parents of preschool children, whereas externalising behaviour becomes more prominent among older children and adolescents [13]. A similar phenomenon is evident in adults. Gutierrez-Angel et al. observed that caregivers of people with epilepsy and psychiatric comorbidity had poorer psychological and social quality of life; after adjustment, patient aggressiveness was associated with greater caregiver burden, whereas seizure control was not significantly correlated with burden [8]. This suggests that challenging interpersonal or behavioural manifestations may impose caregiving demands that are not adequately captured by seizure counts alone. Depression and anxiety constitute another pathway through which epilepsy may influence the family. In 107 adult patient-caregiver dyads, patient anxiety and depression correlated significantly with caregiver burden. Patient cognition, stigma and quality of life were also associated with burden [12]. Dyadic research has further demonstrated that greater depressive symptoms in people with epilepsy correlate with both depressive and anxiety symptoms in their caregivers, indicating psychological interdependence between patient and caregiver [10]. These observations support a conceptual shift from a predominantly seizure-centred model toward a behavioural-psychiatric model of caregiver burden, in which neuropsychiatric symptoms, cognition, family resources and clinical disease interact to determine caregiving strain. The present systematic review was therefore undertaken to examine psychiatric, behavioural and neurocognitive comorbidities as determinants of caregiver burden and parental stress in epilepsy and to identify clinically relevant patterns across adult and paediatric populations.

Study Design and Reporting Framework The review was designed as a systematic qualitative synthesis and reported according to the principles of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement [2]. Review Question The principal review question was: Which psychiatric, behavioural and neurocognitive characteristics of people with epilepsy are associated with greater burden, stress or psychological morbidity among their informal caregivers? Population: Children, adolescents or adults with epilepsy and their informal/family caregivers. Exposure: Psychiatric or behavioural characteristics in the person with epilepsy, including depression, anxiety, aggression, internalising behaviour, externalising behaviour, behavioural dysregulation, neuropsychiatric comorbidity and cognitive impairment. Outcome: Caregiver burden, parental stress, family burden, caregiver quality of life or caregiver psychological distress. Information Sources and Search Strategy A new structured search was performed for this review using PubMed/MEDLINE and a multidisciplinary peer-reviewed academic literature index, with supplementary reference searching. Publications available up to January 2026 were considered. The core search strategy combined: (epilepsy OR seizure disorder) AND (psychiatric OR neuropsychiatric OR depression OR anxiety OR aggression OR behavioural OR behavioral OR internalising OR externalising OR cognition) AND (caregiver OR parent OR family) AND (caregiver burden OR family burden OR parental stress OR caregiver distress OR caregiver quality of life). No restriction was applied solely on the basis of adult versus paediatric population. Eligibility Criteria Original studies were eligible when they involved clinically diagnosed epilepsy; included an informal caregiver or parent outcome; assessed psychiatric, emotional, behavioural or cognitive characteristics of the patient; and examined their relationship with caregiver burden, parental stress, family impact, caregiver quality of life or psychological distress. Reviews, meta-analyses, commentaries, editorials and case reports were excluded. Studies of functional or psychogenic non-epileptic seizures without epilepsy were excluded. Studies evaluating caregiver distress without an analysable patient psychiatric, behavioural or cognitive exposure were not included in the principal synthesis. Study Selection The search strategy generated 38 records. Two duplicate publications were removed, leaving 36 unique records for title and abstract screening. Twenty records did not satisfy the review question and were excluded. The full text of 16 reports was assessed. Four reports were excluded at this stage: one was secondary/non-primary research, one did not report an eligible caregiver burden or stress outcome, one did not evaluate an appropriate psychiatric or behavioural patient exposure, and one concerned an ineligible seizure population. Accordingly, 12 studies were included in the qualitative synthesis. Data Extraction The following information was extracted from eligible studies: year of publication, study design, adult or paediatric population, patient and caregiver sample size, psychiatric/behavioural/cognitive exposure, caregiver-outcome measure and principal relationship between the exposure and caregiver burden or stress. Methodological Appraisal Methodological appraisal focused on sampling, measurement of exposure and caregiver outcomes, use of validated questionnaires, handling of potential confounding factors and appropriateness of statistical analyses. Particular attention was given to whether studies differentiated psychiatric and behavioural influences from traditional epilepsy variables such as seizure frequency and treatment intensity. Data Synthesis Quantitative meta-analysis was considered inappropriate because studies differed substantially in participant age, epilepsy severity, psychiatric constructs, behavioural scales, caregiver measures and statistical effect estimates. A structured narrative synthesis was therefore performed.

Characteristics of Included Studies The 12 included studies represented research conducted between 2007 and 2025. Seven primarily addressed paediatric epilepsy, while five examined adult epilepsy or adult patient-caregiver dyads. Most studies were cross-sectional; one recent paediatric study employed a prospective observational design. Caregiver outcomes included Parenting Stress Index scores, parental stress scales, Zarit Caregiver Burden scores, caregiver quality of life and caregiver anxiety/depression. Table 1. Characteristics of studies included in the systematic review Study Population/sample Principal exposure Caregiver outcome Main finding Rodenburg et al., 2007 91 parents of children with epilepsy Child-related stressors, temperament and coping context Parenting stress Stressors had relatively large effects on parenting stress; stress partly mediated parenting behaviour Wirrell et al., 2008 52 mothers of children with intractable epilepsy Child behavioural problems Parenting Stress Index 63% had clinical-range total stress; child behavioural problems showed moderate-to-high correlation with stress Shatla et al., 2011 23 children and families Depression, anxiety, internalising/externalising problems Parenting stress Clinical behavioural problems were common and intractable epilepsy was associated with high parental stress Farrace et al., 2013 26 parents of children with epilepsy; 31 controls CBCL behavioural syndromes Parenting Stress Index Epilepsy group showed greater behavioural symptoms and parenting stress; behavioural profile predicted stress Karakis et al., 2014 48 PWE-caregiver pairs Cognition, psychiatric/psychosocial characteristics, QOL Caregiver burden/QOL Poorer patient neuropsychological performance and lower patient QOL were associated with greater burden Gutierrez-Angel et al., 2018 151 caregivers Psychiatric comorbidity; aggression Zarit burden; caregiver QOL Aggression independently associated with higher burden; psychiatric comorbidity related to poorer caregiver psychological/social QOL Operto et al., 2021 103 children with epilepsy; 93 controls Internalising, externalising and total behavioural symptoms Parenting Stress Index All major behavioural domains correlated with total parental stress João et al., 2022 548 adults with epilepsy; 191 caregivers; 120 dyads Patient depression and anxiety Caregiver depression/anxiety Patient depression correlated with caregiver depression and anxiety Wu et al., 2023 288 parents Neuropsychiatric comorbidity and child QOL Family burden; parental depression/anxiety High family burden and parental psychiatric symptoms were common; child comorbidities strongly influenced child QOL Yeni et al., 2024 107 patient-caregiver dyads Anxiety, depression, cognition, stigma and QOL Zarit caregiver burden Patient anxiety, depression, cognition, stigma and QOL correlated significantly with burden Sirisha et al., 2025 314 parents Internalising and externalising behavioural problems Parental Stress Scale Internalising symptoms predicted stress in preschoolers; externalising symptoms predominated in older groups Correale et al., 2025 117 children; 149 caregivers Behavioural symptoms and cognition Parenting Stress Index-Short Form Clinical behavioural symptoms, especially internalising problems, and cognitive impairment were associated with greater stress Behavioural Problems as Determinants of Parenting Stress The strongest and most consistent evidence was found in paediatric epilepsy. Earlier work by Rodenburg et al. demonstrated that stressors related to the child and family context had comparatively large effects on parenting stress, with parenting stress mediating some of the relationship between stressors and parenting behaviour [3]. Wirrell et al. subsequently examined mothers of children with intractable epilepsy. Sixty-three percent scored in the clinical range for total parenting stress and 75% for the Child Domain. Behavioural problems in the child showed a moderate-to-high correlation with total stress, whereas other seizure and demographic variables were not significantly correlated with stress [4]. Shatla et al. reported clinically elevated internalising and total behavioural-problem scores among children with epilepsy. The investigation incorporated measures of child depression, anxiety, behaviour and cognition and demonstrated high parental stress, particularly in the context of difficult-to-control epilepsy [5]. Similarly, Farrace et al. found higher Parenting Stress Index and Child Behavior Checklist scores among families affected by epilepsy than among healthy controls. Behavioural symptoms differed in their ability to predict parenting stress, reinforcing the importance of behavioural phenotype rather than diagnosis alone [6]. Internalising and Externalising Symptoms More recent studies allow greater differentiation between behavioural domains. Operto et al. found significant correlations between total parental stress and total problems, internalising problems and externalising problems on the Child Behavior Checklist. Internalising difficulties were largely represented by anxiety and depressive symptoms [9]. Sirisha et al. provided evidence of developmental variation. Among 314 children and adolescents, withdrawal and other internalising characteristics were strongly associated with parental stress in children below five years. By contrast, externalising characteristics showed stronger correlations with parental stress after five years of age and during adolescence. Poor school performance and requirements for special schooling were also associated with higher parental stress [13]. Correale et al. extended these observations by considering behavioural symptoms together with cognition and treatment characteristics. Clinical-range behavioural symptoms, especially internalising problems, were associated with elevated stress across Parenting Stress Index domains. Moderate intellectual disability, polytherapy, drug resistance and lower caregiver educational attainment contributed additional stress. The authors concluded that cognitive and behavioural characteristics, together with treatment complexity, may play a greater role in parental stress than core epilepsy variables alone [14]. Psychiatric Comorbidity in Adult Epilepsy Among adults, psychiatric comorbidity had important implications for family caregivers. Gutierrez-Angel et al. reported psychiatric comorbidity in 112 of 151 patients represented by their caregivers. Caregivers of patients with psychiatric comorbidity had significantly poorer psychological and social quality-of-life scores. In multivariable analysis, aggressiveness was associated with greater caregiver burden (p=0.008). Notably, seizure control did not significantly correlate with caregiver burden in this study [8]. Yeni et al. provided broader evidence across several psychological domains. In 107 adult patient-caregiver dyads, patient cognitive function, stigma, anxiety, depression and quality of life all correlated with caregiver burden. Anxiety was associated at p=0.001, depression at p=0.005, and cognition and stigma at p<0.001. Their multivariable model explained a substantial proportion of burden variation, although caregiver and treatment variables also remained relevant [12]. Karakis et al. likewise demonstrated that poorer patient neuropsychological functioning and lower patient quality of life were associated with greater caregiver burden. Greater burden was in turn inversely related to caregiver quality of life, with a particularly strong relationship involving the mental-health component [7]. Patient Depression and Psychological Spillover to Caregivers Evidence from dyadic studies suggests that psychiatric morbidity may affect caregivers through emotional spillover in addition to increasing practical care requirements. João et al. studied 548 adults with epilepsy and 191 caregivers, including 120 matched patient-caregiver dyads. The severity of depressive symptoms in patients correlated with caregiver depression (r=0.35, p<0.001) and caregiver anxiety (r=0.25, p=0.01), while patient anxiety did not show the same dyadic relationship [10]. This distinction suggests that depression in the person with epilepsy may influence caregiver emotional health through changes in motivation, independence, interpersonal interaction and perceived prognosis, although causal mechanisms cannot be established from cross-sectional data. Family Burden and Neuropsychiatric Comorbidity Wu et al. examined 288 parents of children with epilepsy and reported that 94.8% of families had high disease-burden scores, while parental anxiety and depression states were also common. More than half of the children had comorbid neuropsychiatric symptoms. Family burden was the strongest predictor of parental depression, while epilepsy comorbidities strongly influenced child quality of life, demonstrating a multidirectional relationship between the child's health, family burden and parental psychological state [11]. These findings support the view that caregiver burden is best conceptualised as part of a family system rather than as an isolated caregiver outcome.

The principal finding of this systematic review is that psychiatric and behavioural comorbidities represent meaningful determinants of caregiver burden in epilepsy, and their importance is particularly clear when behavioural measures are considered independently of seizure severity. The paediatric evidence consistently indicates that how a child behaves may be at least as relevant to the caregiver as how frequently the child has seizures. The 2008 study of mothers caring for children with intractable epilepsy demonstrated a moderate-to-high relationship between child behavioural problems and parental stress while failing to identify significant associations for several seizure and demographic variables [4]. Nearly two decades later, Correale et al. similarly concluded that behavioural and cognitive characteristics together with treatment complexity may contribute more strongly to parental stress than traditional epilepsy variables [14]. The consistency of these observations across different periods and settings strengthens the argument for routine behavioural assessment. The distinction between internalising and externalising symptoms is clinically important. Internalising disorders may manifest as anxiety, sadness, withdrawal or somatic complaints and may require increased reassurance, supervision and healthcare use. Externalising behaviour may include impulsivity, oppositional behaviour, aggression or behavioural dysregulation and can create direct interpersonal strain. Operto et al. showed associations across both behavioural domains, while Sirisha et al. demonstrated a developmental shift from internalising correlates of parental stress in preschool children toward externalising correlates in later childhood and adolescence [9,13]. The adult literature identifies a somewhat different pattern. Aggression appears especially burdensome because it directly affects caregiver safety, interpersonal functioning and manageability. Gutierrez-Angel et al. found aggression to be independently associated with burden even when seizure control was not [8]. Depression and anxiety, however, may produce more subtle but equally important caregiving consequences. Yeni et al. demonstrated significant correlations of both patient depression and anxiety with caregiver burden, while patient cognition, stigma and quality of life also contributed [12]. Cognition deserves particular attention because cognitive impairment may bridge neurological and psychiatric determinants of burden. Poor executive function, memory impairment or intellectual disability can reduce patient independence even in the absence of frequent seizures. Karakis et al. associated poorer neuropsychological performance with higher caregiver burden, and Correale et al. found greater parental stress in caregivers of children with moderate intellectual disability [7,14]. These findings suggest that neurocognitive assessment can provide information relevant not only to patient functioning but also to expected caregiver needs. The review also highlights the dyadic nature of psychological distress. João et al. demonstrated that greater depressive symptoms in patients were accompanied by greater depressive and anxiety symptoms in their caregivers [10]. This does not prove that patient depression causes caregiver psychopathology; cross-sectional relationships could be reciprocal or influenced by shared contextual factors. Nonetheless, treating depression in one member of the dyad without evaluating the psychological health of the other may leave an important component of disease burden unaddressed. Caregiver characteristics and social context further modify these relationships. Educational level, income, caregiving time, stigma, treatment intensity and family resources have repeatedly emerged alongside psychiatric variables. Yeni et al. found associations between burden and several caregiver sociodemographic characteristics as well as daily caregiving time [12]. Caregiver burden may therefore function simultaneously as a consequence and amplifier of family psychological strain. This has implications for how caregiver burden should be evaluated clinically. Seizure counts and medication lists alone are unlikely to identify all high-risk families. An apparently well-controlled patient may still generate substantial caregiver strain if severe depression, aggression, cognitive impairment, stigma or behavioural dysregulation remains untreated. CLINICAL IMPLICATIONS Epilepsy services should incorporate brief psychiatric and behavioural screening into routine follow-up, especially when caregivers report exhaustion or disproportionate difficulty despite acceptable seizure control. Paediatric services should screen for internalising and externalising behaviour, school functioning and cognition, whereas adult services should pay particular attention to depression, anxiety, aggression, cognitive impairment and stigma. Assessment should also extend to the caregiver. Significant burden should trigger evaluation for caregiver anxiety, depression, sleep disruption, social isolation and financial or occupational consequences. Management may require collaboration between neurology, psychiatry, psychology, paediatrics, social work and rehabilitation services rather than escalation of antiseizure treatment alone. STRENGTHS This review differs from broader reviews of epilepsy caregiving by focusing specifically on psychiatric, behavioural and neurocognitive determinants of caregiver burden. It integrates early and contemporary evidence across almost two decades, allowing identification of consistently replicated behavioural findings. The inclusion of adult and paediatric populations also illustrates how the determinants of burden vary across developmental stages. LIMITATIONS Most included studies were cross-sectional, and consequently temporal direction and causality cannot be established. Many originated from tertiary-care centres, which may over-represent treatment-resistant epilepsy, psychiatric comorbidity or highly burdened families. Measurement heterogeneity was substantial. Behavioural exposure ranged from total Child Behavior Checklist scores to specific internalising or externalising domains, while adult studies used measures of depression, anxiety, aggression, cognition, stigma or psychiatric history. Outcomes also differed between Zarit caregiver burden, Parenting Stress Index, family burden, caregiver quality of life and caregiver psychiatric symptoms. Several studies involved small samples; the 2011 and 2013 paediatric investigations included only 23 families and 26 epilepsy caregivers, respectively [5,6]. Conversely, larger contemporary studies employed different instruments, limiting statistical pooling. Finally, caregiver burden is influenced by socioeconomic resources, seizure severity, dependency, treatment burden and family structure. Even when psychiatric or behavioural associations remain statistically significant, residual confounding is likely. For these reasons, a formal meta-analysis was not undertaken.

Psychiatric, behavioural and cognitive comorbidities are important determinants of caregiver burden in epilepsy. Across paediatric studies, behavioural problems repeatedly showed strong relationships with parental stress, with internalising symptoms particularly relevant in younger children and externalising behaviour more prominent later in childhood and adolescence. In adult epilepsy, aggression, depression, anxiety, cognitive impairment and stigma were associated with greater caregiver burden or adverse psychological outcomes. The findings challenge a purely seizure-centred interpretation of family burden. Effective epilepsy care should simultaneously assess seizure control, psychiatric symptoms, behaviour, cognition and caregiver well-being. Future longitudinal studies should clarify temporal relationships and determine whether targeted treatment of patient psychiatric and behavioural comorbidity produces measurable reductions in caregiver burden.

1. Kwon CS, Rafati A, Ottman R, et al. Psychiatric comorbidities in persons with epilepsy compared with persons without epilepsy: a systematic review and meta-analysis. JAMA Neurol. 2025;82(1):72-84. doi:10.1001/jamaneurol.2024.3976. 2. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi:10.1136/bmj.n71. 3. Rodenburg R, Meijer AM, Deković M, Aldenkamp AP. Parents of children with enduring epilepsy: predictors of parenting stress and parenting. Epilepsy Behav. 2007;11(2):197-207. doi:10.1016/j.yebeh.2007.05.001. 4. Wirrell EC, Wood L, Hamiwka LD, Sherman EMS. Parenting stress in mothers of children with intractable epilepsy. Epilepsy Behav. 2008;13(1):169-173. doi:10.1016/j.yebeh.2008.02.011. 5. Shatla R, Sayyah HES, Azzam H, Elsayed RM. Correlates of parental stress and psychopathology in pediatric epilepsy. Ann Indian Acad Neurol. 2011;14(4):252-256. doi:10.4103/0972-2327.91938. 6. Farrace D, Tommasi M, Casadio C, Verrotti A. Parenting stress evaluation and behavioral syndromes in a group of pediatric patients with epilepsy. Epilepsy Behav. 2013;29(1):222-227. doi:10.1016/j.yebeh.2013.07.020. 7. Karakis I, Cole AJ, Montouris GD, San Luciano M, Meador KJ, Piperidou C. Caregiver burden in epilepsy: determinants and impact. Epilepsy Res Treat. 2014;2014:808421. doi:10.1155/2014/808421. 8. Gutierrez-Angel AM, Martinez-Juarez IE, Hernandez-Vanegas LE, Crail-Melendez D. Quality of life and level of burden in primary caregivers of patients with epilepsy: effect of neuropsychiatric comorbidity. Epilepsy Behav. 2018;81:12-17. doi:10.1016/j.yebeh.2018.01.034. 9. Operto FF, Pastorino GMG, Pippa F, et al. Psychiatric symptoms and parental stress in children and adolescents with epilepsy. Front Neurol. 2021;12:778410. doi:10.3389/fneur.2021.778410. 10. João RB, Nogueira MH, Morita-Sherman ME, et al. The relationship between depression and anxiety symptoms of adult PWE and caregivers in a tertiary center. Front Neurol. 2022;13:766009. doi:10.3389/fneur.2022.766009. 11. Wu Y, Zhang R, Tang J, et al. Multi-dimensional influence of pediatric epilepsy on children and their families: a cross-sectional study. Epilepsy Behav. 2023;146:109360. doi:10.1016/j.yebeh.2023.109360. 12. Yeni K, Tulek Z, Cavusoglu A, et al. Caregiver burden and its predictors in adult epilepsy patients. Epilepsy Behav. 2024;153:109685. doi:10.1016/j.yebeh.2024.109685. 13. Sirisha S, Jala S, Vooturi S, Patil A, Somayajula S, Jayalakshmi S. Association between behavioral problems and parental stress in children and adolescents with epilepsy. Epilepsy Behav. 2025;163:110229. 14. Correale C, Mercier M, Cappelletti S, et al. Neurocognitive, behavioral, and treatment burden as key predictors of parental stress in pediatric epilepsy. Epilepsia. 2025;66(12):4960-4971. doi:10.1111/epi.18580.