3.1.1 Delivery Complications

Children born during periods of armed conflict face elevated risks of adverse birth outcomes. A meta-analysis involving over 600,000 mother–child dyads from populations exposed to war and terrorism found a higher prevalence of preterm births and low birthweight infants (under 2,500 grams; Riquelme-Gallego et al., Reference Riquelme-Gallego, Ramos-Soberbio, Leno-Duran, Martínez-Vázquez and Caparros-Gonzalez2025). These outcomes were primarily attributed to maternal malnutrition, psychosocial stress, and limited access to prenatal care in conflict-affected settings. Another systematic review of perinatal and neonatal outcomes found that refugee women from war-affected countries had worse pregnancy outcomes compared to local women in host countries. Specifically, refugee mothers had higher odds of delivering babies with low birthweight (OR = 1.8), experiencing stillbirth (OR = 1.9), and perinatal mortality (OR = 2.0). These associations correspond to substantial effect sizes, with Cohen’s d estimates ranging from 0.32 to 0.38 (Behboudi-Gandevani et al., Reference Behboudi-Gandevani, Bidhendi-Yarandi and Panahi2022).

3.1.2 Physical Growth and Development

In addition to gestation and delivery complications, malnutrition is a concern for children in both the short- and long-term aftermath of war. In armed conflicts, adequate nutrition, including sustained breastfeeding, is considered to be important for a child’s cognitive, physical, and socioemotional development (Punamäki et al., Reference Punamäki, Diab, Vänskä, AlBarqouni and Quota2025; Purkiewicz et al., Reference Purkiewicz, Regin, Mumtaz and Pietrzak-Fiećko2025). A systematic review found that proximity to armed conflict is associated with a higher risk of breastfeeding interruption, with exclusive breastfeeding rates in conflict zones falling as low as 5.5% (Rabbani et al., Reference Rabbani, Padhani, Siddiqui, Das and Bhutta2020). Displacement, maternal stress, malnourishment, partner loss, and reduced availability of mental health care were all identified as significant contributors. These disruptions were also fueled by cultural misconceptions, including the preexisting beliefs, reinforced by both communities and health workers, that maternal stress and malnutrition render breastmilk “dirty” or nutritionally inadequate (Rabbani et al., Reference Rabbani, Padhani, Siddiqui, Das and Bhutta2020).

Older children are also at risk for malnutrition. On average, 24% of war-affected children in Africa were malnourished, and 34% exhibited stunting, indicating chronic undernutrition and impaired growth trajectories (Azanaw et al., Reference Azanaw, Anley, Anteneh, Arage and Muche2023). The most significant predictors of undernutrition were low maternal education and residence in areas of active armed conflict. In contrast, average rates of wasting in sub-Saharan Africa ranged from 4% to 10%, and underweight prevalence ranged from 11% to 20% in areas that were not war-affected. The risk for food insecurity during war is further demonstrated by reports from the Gaza Strip and Ukraine. According to the Integrated Food Security Phase Classification (IPC), as of June 2025, the Gaza Strip had exceeded famine thresholds, placing more than 320,000 children at risk of starvation (IPC, 2025). Similarly, in Ukraine, over 1.5 million people living in frontline regions are struggling to access adequate food supplies and are in urgent need of humanitarian assistance (World Food Programme, 2024).

Several studies have documented the long-term health consequences of early-life exposure to starvation in child populations. Exposure to the Dutch Hunger Winter of 1944–1945 during gestation has been associated with increased risks of glucose intolerance, obesity, coronary heart disease, and hypertension in adulthood, and increased risk of breast cancer later in life (Kyle & Pichard, Reference Kyle and Pichard2006). A follow-up study of Dutch Hunger Winter survivors found that famine exposure during the first three months after conception was linked to lower DNA methylation of the IGF2 gene, which controls insulin-like growth factor – a hormone essential for human growth and development (Heijmans et al., Reference Heijmans, Tobi and Stein2008). Similar findings have emerged from an ecological study of survivors of the Holodomor, the artificial famine of 1932–1933 in what was then the Soviet Ukraine. In utero exposure to famine conditions was associated with nearly double the risk of developing type 2 diabetes mellitus in adulthood (Lumey et al., Reference Lumey, Li, Khalangot, Levchuk and Wolowyna2024). Moreover, in utero exposure to famine was associated with steeper cognitive decline in old age (Wiegersma et al., Reference Wiegersma, Boots, Roseboom and de Rooij2023).

War-related stressors and displacements disrupt pubertal development. In a systematic review, Glass and colleagues (Reference Glass, Reiches and Clarkin2024) reported that nearly two-thirds of exposed girls, with some exceptions, showed delays in the onset of menarche, ranging from several months to several years. These disruptions are likely influenced by both nutritional deficiencies and chronic psychosocial stress during sensitive developmental windows. This pattern appears to diverge from some life history theory predictions, which suggest that adverse early environments, particularly those characterized by unpredictability and resource scarcity, may result in accelerated development, leading to earlier menarche and a fast life trajectory characterized by earlier reproduction, higher fertility, and shorter lifespan (Belsky, Reference Belsky2008; Belsky et al., Reference Belsky, Steinberg and Draper1991; Chang & Lu, Reference Chang, Lu and Bornstein2025). One explanation is that extreme energy deficits, common in war-related famine and displacement, override the adaptive signaling associated with psychosocial risk (Clarkin, Reference Clarkin2019; Ellis et al., Reference Ellis, Reid and Kramer2025), which highlights the complex interaction between biological energy availability and environmental harshness in shaping reproductive timing.

3.1.3 Physical Health and Disease

Armed conflicts pose a significant threat to children’s physical safety, resulting in a wide range of injuries, including penetrating wounds, blunt force trauma, crush injuries, burns, and amputations. A UNICEF report (2022a) documented that between 2005 and 2020, at least 104,100 children were killed or seriously injured in armed conflicts, and more than 13,900 attacks were recorded against schools and hospitals. Reported trauma-related case fatality rates among children under eighteen years of age with conflict-related injuries have reached as high as 18% in some hospital- and registry-based studies, particularly following explosive and landmine injuries. In these settings, penetrating head injuries were the most common form of traumatic brain injury (see Section 5), markedly different from nonconflict settings, where blunt trauma, caused by impact with a non-sharp object, such as a fall or collision, predominates. Furthermore, the mortality rate among pediatric neurosurgical patients in these settings was reported to be as high as 24% (Kadir et al., Reference Kadir, Shenoda and Goldhagen2019).

War-related damage to hospitals, supply chains, and healthcare access also creates unsanitary living conditions and leads to major vaccine shortages. A systematic review involving data from over 590,000 children from war-affected countries between 1998 and 2018 found a decline in vaccination rates in conflict zones (Goli et al., Reference Goli, Mavisakalyan and Rammohan2022). Similarly, a review of studies of refugee children arriving in Western countries, mostly from Africa and Asia, reported elevated rates of anemia and infectious diseases, including HIV, hepatitis, and tuberculosis (Baauw et al., Reference Baauw, Kist-van Holthe and Slattery2019).

3.3.1 Direct War-Related Effects

War-related trauma impairs key aspects of cognitive development in children. A systematic review found that up to three-quarters of refugee children demonstrated lower cognitive abilities on performance tests compared with local children or estimated norms. Learning performance, executive functioning, and early math abilities were particularly lower compared to controls (Bernhardt et al., Reference Bernhardt, Le Beherec and Uppendahl2024). Hahnefeld et al. (Reference Hahnefeld, Sukale and Weigand2022) found that in Germany, a sample of three- to six-year-old refugee children from different countries scored on average more than one standard deviation below the German norm on a nonverbal IQ measure (Mean IQ = 81.5), with nearly one in four scoring below the threshold for intellectual disability. Notably, cognitive scores were not associated with PTSD symptoms, parental mental health, or duration of displacement. Instead, longer time spent in Germany predicted better learning outcomes, suggesting that the observed deficits were more reflective of disrupted early learning opportunities, cultural and language barriers, and socioeconomic deprivation. Distance learning technologies may offer a partial solution, but their efficacy in armed conflict zones is constrained by limited resources, unreliable internet access, and the absence of structured learning environments.

Evidence from studies on COVID-19-related school closures demonstrates that disruptions to formal education are associated with substantial declines in academic performance, with the magnitude of these effects varying by age group and subject domain (Storey & Zhang, Reference Storey and Zhang2024). Similarly, a study examining the academic performance of Ukrainian school students documented a significant decline in PISA (Programme for International Student Assessment) scores relative to peer countries, following the full-scale invasion. The decline seemed primarily driven by lack of sleep and stress because of air raid alerts on the night of the exam. Notably, the study did not find a significant association between war-related school closures and the observed academic decline (Mykhailyshyna, Reference Mykhailyshyna2025). However, this result should be interpreted with caution, as the sampling strategy included only schools that had resumed in-person instruction at the time of assessment, potentially limiting generalizability and underestimating the effects of prolonged or unresolved educational disruptions. These apparent contradictions highlight the complex and multifactorial nature of academic performance declines in situations of large-scale disruption, underscoring the limited understanding of the mechanisms driving educational loss in conflict-affected settings.

3.3.2 Family-Related and Intergenerational Effects

In addition to the observed associations between war-related trauma and cognitive development, armed conflict has been linked to family disruptions and declines in overall family functioning. One of the most commonly reported disruptions is the absence of fathers, due to military deployment, death, captivity, or forced displacement. These circumstances may limit children’s access to consistent emotional support, and father absence is frequently associated with reduced caregiving quality, both of which have been identified as important correlates of developmental outcomes. From a broader perspective, the effects of family disintegration might be explained by findings relating parental sensitivity to child cognitive development. A meta-analysis by Rodrigues et al. (Reference Rodrigues, Sokolovic and Madigan2021) found that parental sensitivity to a child’s needs is positively associated with broad cognitive development, with a substantial pooled effect size (r = 0.19). Father’s sensitivity was linked to better achievements across multiple cognitive domains, including language development, general cognitive ability, and executive functioning, with effect sizes ranging from r = 0.12 to 0.21.

Similar findings were obtained in a meta-analysis by Deneault et al. (Reference Deneault, Duschinsky and Van IJzendoorn2023), which reported a positive association between maternal sensitivity and child development, including cognition (r = 0.17) and language (r = 0.16). Notably, the association was stronger when sensitivity was measured during early childhood, highlighting the importance of responsive caregiving during periods of heightened neurodevelopmental plasticity. In war-affected contexts, where caregivers may be overwhelmed by trauma and loss, the capacity to maintain sensitivity is often compromised (see Section 4), potentially amplifying the developmental risks faced by children. A study of Palestinian mothers exposed to war-related trauma during the first trimester of pregnancy found that maternal mental health difficulties were a mediating factor linking prenatal trauma exposure to poorer mother–child interactions in early childhood, which in turn were associated with delayed cognitive development in children (Qouta et al., Reference Qouta, Vänskä, Diab and Punamäki2021). These findings highlight an intergenerational pathway through which maternal trauma and psychological distress can disrupt early caregiving quality, thereby impairing the developmental conditions necessary for children’s optimal cognitive outcomes.

Another population-based study suggests that parents’ exposure to war-related trauma negatively affects children’s educational outcomes. A Danish national register study involving 854,467 children found that those with at least one parent who had been exposed to war trauma or torture scored lower on school performance tests than matched controls, with ds ranging from –0.06 to –0.38 (Bager et al., Reference Bager, Munk Laursen, Skipper and Agerbo2021). In addition to lower performance, children in trauma-exposed families were also nearly five times more likely to miss the standardized tests entirely, indicating broader functional impairments and potential school disengagement. These findings underscore the cascading developmental effects of war trauma across generations, with implications for both cognitive and academic development even in high-income, host-country contexts.

3.4.1 Mental Health Problems

A large systematic review of studies by Kien et al. (Reference Kien, Sommer and Faustmann2019) on internally displaced and refugee children resettled in Europe found that, on average, 35% of children met diagnostic criteria for PTSD. The high prevalence and broad range of PTSD among children exposed to armed conflict confirmed the results of a systematic review by Attanayake et al. (Reference Attanayake, McKay and Joffres2009), with estimates ranging from 5% to 89%. The average rates for depression and anxiety disorders in a systematic review of war-exposed children resettled to Europe were 21% and 15%, respectively. Unaccompanied minors consistently show higher mental disorders prevalence rates, underscoring the protective role of parental care and family presence during trauma and displacement (Kien et al., Reference Kien, Sommer and Faustmann2019). This pattern was echoed in a recent person-centered study by Olinover & Hamama (Reference Olinover and Hamama2025) of Israeli adolescents exposed to wartime stress, which identified a multi-risk profile characterized by high trauma exposure and severe PTSD and depression symptoms. Adolescents in this profile frequently reported disrupted schooling, displacement, and reduced family support.

Since the escalation of armed conflicts, particularly in Europe, concerns have increasingly been raised about the role of direct war trauma and related experiences as significant risk factors for suicide and self-harm (Amone-P’Olak, Reference Amone-P’Olak2024). Despite the intuitive expectation of higher rates of these outcomes in war-affected child populations, the existing evidence remains limited. A study by Chemtob et al. (Reference Chemtob, Pat-Horenczyk and Madan2011) found that among Israeli adolescents aged 12–18 years, 12% reported suicidal thoughts in the two weeks preceding the assessment. Terrorism-related exposures showed varying strengths of association with mental health outcomes. Adolescents who had been injured during an attack were almost four times more likely to report suicidal thoughts (OR = 3.96), reflecting a large effect size (d ≈ 0.76). Those present but not wounded during an attack had 1.5 times higher odds (OR = 1.45), corresponding to an effect size of d ≈ 0.21, with almost similar odds for adolescents near the attack site (OR = 1.42; d ≈ 0.19), and two times higher odds if they had a closely related person injured by the attack (OR = 2.03; d ≈ 0.39). Studies in postwar Croatia support the finding of elevated suicide risks (Franić et al., Reference Franić, Kardum, Marin Prižmić, Pavletić and Marčinko2012; Fajkic et al., Reference Fajkic, Lepara and Voracek2010).

Together, these findings suggest that children (and adolescents) affected by wars and displacement are at heightened risk for mental health difficulties. At the same time, the evidence demonstrates substantial heterogeneity in reported prevalence rates. We emphasize the need for caution when interpreting prevalence estimates based solely on a single, often self-reported assessment method and the absence of causal evidence in evaluating trauma-related outcomes among war-affected children (see also Section 10).

3.5.1 Physical Growth and Health

The war in Ukraine began in 2014 with the annexation of Crimea, but evidence on its long-term consequences for children has remained limited. Available findings on growth, development, and physical and mental health largely align with research from other war-affected countries. In addition to human losses, the conflict has resulted in the destruction of daily life, supply chains, access to medical care, medications, and essential goods, challenges that became especially severe following the onset of the full-scale invasion in 2022 (Martsenkovskyi et al., Reference Martsenkovskyi, Martsenkovsky, Martsenkovska and Lorberg2022; de Alencar Rodrigues et al., Reference De Alencar Rodrigues, Lima, Neto and Uchida2022). According to the World Food Programme (2025), at least one-third of the Ukrainian population, mostly living in frontline regions, requires food and cash assistance to survive. Despite food insecurity and supply disruptions, Ukraine, compared to other conflict-affected countries, has not reported significant changes in breastfeeding. Breastfeeding practices have remained stable since 2015, with 65% of mothers initiating breastfeeding, 51% reporting exclusive breastfeeding, and 79% of children under 2 years old receiving a minimum acceptable diet (Majer et al., Reference Majer, Mbuto, Nesterova, King and Bilukha2025). These data highlight the potentially more prominent role of maternal education in sustaining positive feeding practices during crises, rather than military actions themselves. One potential explanation involves the intergenerational transmission of knowledge related to infant feeding and food preservation, rooted in the collective trauma of the Holodomor, the Stalinist-induced artificial famine, that profoundly shaped national attitudes toward food. This historical experience fostered a cultural reverence for food, often expressed through behaviors such as overeating and stockpiling (Bezo & Maggi, Reference Bezo and Maggi2015).

In contrast, vaccination coverage has been more vulnerable. Since the escalation of war, measles vaccination dropped from 88% in 2021 to 74% in 2022, raising fears of another outbreak similar to 2017–2019, when vaccine hesitancy drove an epidemic (Orsini & Martini, Reference Orsini and Martini2023). Declines in coverage were also observed for COVID-19 and other childhood immunizations, despite public health campaigns and governmental efforts to increase coverage (Habib et al., Reference Habib, Ennab and Matiashova2022). The decreasing vaccination rates were closely linked to the rise of Russian propaganda and misinformation in Ukraine following the onset of the war in 2014, which aimed to polarize society and destabilize public trust in institutions, including the healthcare system (Patel et al., Reference Patel, Moncayo, Conroy, Jordan and Erickson2020). Finally, children have also faced direct physical harm due to the active use of missiles and drones against civilian and critical infrastructure. A study in Kharkiv frontline regions further documented a significant rise in pediatric blast and shelling injuries (see Section 9), which correlated with the increase in war activities in the region (Sennersten et al., Reference Sennersten, Frogh and Falk2025).

3.5.2 Mental Health Outcomes after the Invasion in Eastern Parts of Ukraine

Alongside physical harm, the war resulted in significant mental health decline in children all over Ukraine. The earliest study by Osokina and colleagues (Reference Osokina, Silwal and Bohdanova2023) compared rates of PTSD in the war-affected Donetsk region and non-affected Kirovograd region. The study reported a PTSD prevalence of 5% among children living in the conflict-affected Donetsk region, four times higher than that observed in the neighboring, non-affected Kirovograd region (1.2%), underscoring the pronounced psychological toll of direct exposure to armed conflict. These rates may appear modest, but methodological limitations, such as limited diagnostic tools or contextual underreporting, may have contributed to underestimation (Danese & Martsenkovskyi, Reference Danese and Martsenkovskyi2023). Notably, in 2014, the Donetsk region was home to approximately 700,000 children, thus even a relatively small 5% prevalence would correspond to tens of thousands of affected children at the population level (State Statistics Service of Ukraine, 2014).

Analyses of the same sample by Sourander et al. (Reference Sourander, Silwal and Osokina2024) reported prevalences for suicidal and self-harming behavior in the war-affected region compared to the non-affected region was 43% vs 24% in girls, and 19% vs 14% in boys. These percentages suggest that living in the war-affected Donetsk region was associated with higher odds of suicidality or self-harming (OR = 2.3), reflecting a substantial effect size in girls (d = 0.46), while in boys this effect was less pronounced (d = 0.19). Notably, in girls, a higher risk of suicidality and self-harm in war-affected regions was related to PTSD (d = 0.14) and depression (d = 0.51). In boys, the associations were smaller (d = 0.29 for the association of suicidality or self-harming with PTSD, and d = 0.36 for the association with depression).

3.5.3 Mental Health Outcomes after the Full-Scale Invasion

A nationwide study using a convenience sample of 1,238 parents of children residing in Ukraine, conducted by Martsenkovskyi et al. (Reference Martsenkovskyi, Karatzias and Hyland2024a) approximately 6 months after the full-scale invasion, reported a PTSD prevalence of 18% among preschoolers. Notably, the risk of PTSD was significantly predicted by reported delays in early developmental milestones (e.g., language and motor delays; d = 0.48), parental affiliation with emergency services or the military (d = 0.42), parental PTSD (d = 0.35), and increased parental anxiety (d = 0.38). Based on the same study, McElroy et al. (Reference 78McElroy, Hyland and Shevlin2023) reported increased rates of internalizing, externalizing, and inattention symptoms in children. Using adapted versions of the Pediatric Symptom Checklist (PSC-17; Gardner et al., Reference Gardner, Murphy and Childs1999), the study found that parents most frequently observed increased worry (36%), sadness (28%), distractibility (26%), difficulties concentrating (26%), and diminished enjoyment or pleasure (25%). Higher internalizing symptoms were associated with older child age, war-related trauma exposure, and parental depression or anxiety, but were less common among children whose parents had higher educational attainment.

Externalizing symptoms were more likely in children with prior emotional or behavioral difficulties, those living in single-parent households, and those with trauma-exposed parents or parental depression. Similarly, attention problems were associated with prior child difficulties, parental war trauma, and parental depression or anxiety. A nationwide study of parents of children residing in Ukraine, conducted a year after the invasion, found that 22% of preschoolers and 18% of school-aged children had probable PTSD (Naeem et al., Reference Naeem, Sikder and Wang2025). Similar to previous findings, child PTSD was predicted by parental PTSD (d = 0.65), parental depression (d = 0.99), and parental anxiety (d = 0.69). Another internet-based survey of over 8,000 adolescents aged 15 and older conducted by Goto et al. (Reference Goto, Pinchuk and Kolodezhny2024) approximately 15 months after the full-scale invasion, found that 35% of those staying in Ukraine versus 44% of those who were displaced abroad screened positive for PTSD, 25% versus 31% had moderate to severe depression, and 18% versus 25% had moderate to severe anxiety. Hence, the effect size of displacement on mental health outcomes was small for all mental issues (Cohen’s d ranging from 0.12 to 0.19).

5.1.1 Casualties among Ukrainian Civilians, Soldiers, and Children

How destructive modern warfare can be is illustrated by harrowing casualty numbers in the four years of the Russian invasion of Ukraine. On August 28, 2025, the UN Human Rights Monitoring Mission in Ukraine (HRMMU) documented the deaths of at least 13,883 civilians, including 726 children, and 35,548 injuries, including 2,234 children, since the Russian Federation launched its full-scale invasion of Ukraine on February 24, 2022. Autumn and winter 2025 saw an increase in casualties due to intensifying drone and long-range missile attacks on civilians in large Ukrainian cities across the country. Furthermore, the Russia-Ukraine War Report Card of the Harvard Kennedy School Belfer Center estimated 43,000 killed and 370,000 injured Ukrainian soldiers, and 35,000 soldiers missing, based on Ukrainian government data of December 2024. The war report card also mentions 3.7 million internally displaced Ukrainian citizens and 6.9 million international refugees. The Russian army lost more than a staggering 1 million soldiers (consulted on November 29, 2025: https://www.russiamatters.org/news-archive/russia-ukraine-war-report-card).

5.1.2 From Shell Shock to Blast-Related Traumatic Brain Injury

Modern weapons kill and inflict brain damage on their victims. They create blast waves with a force that may damage the brains of military targets and of the military that fire those weapons without sufficient head and ear protection. One tends to forget that in “modern” warfare, not only military but also civilians – children and parents – become victims of brain damage related to concussions by forceful blasts of explosives. The symptoms of such often hidden concussions might impact children’s mental health but might be attributed to the wrong etiology. Blast-related Traumatic Brain Injury (bTBI) occurs without further visible physical injuries but with potentially far-reaching neurological and psychological damage. During the World War I, “shell shock” was the term used to describe a wide variety of symptoms reported by British soldiers, including insomnia, nightmares, hallucinations, irritability, and difficulty carrying out duties, as well as psychological blindness, deafness, mutism, and paralysis. The symptoms were attributed to the soldiers’ exposure to explosions (hence the name “shell shock”). According to Zhang and colleagues (Reference Zhang, Botterbush and Bagdady2022), the term shell shock was first coined by medical officer and psychologist Charles Myers in 1915 to dispute the idea that soldiers were just cowardly simulating symptoms in order to be released from service at the frontline.

Shell shocks might indeed literally have played a larger role than even pioneering military psychologists and psychiatrists, such as Rivers (Reference Rivers1918) suspected when they emphasized that shell shock symptoms could also result from soldiers’ terrible traumatic experience of a gruesome death of a comrade in arms. From the perspective of adult attachment theory and research, unresolved loss or trauma may indeed be related to posttraumatic stress symptoms and disorder, and the empirical evidence points to such associations (see Sections 2 and 3). But this perspective may provide only a partial explanation for symptoms of trauma and may require additional explanation concerning the impact of shell shocks or, in modern terminology, blast waves or thermobaric blasts that literally “shock” the brain into disarray. Subsequent brain disturbances would create heightened vulnerability to new stressors and re-experiencing old ones, thus constituting a PTSD phenotype. One promising treatment possibility of bTBI-related PTSD is Transcranial Magnetic Stimulation (TMS; L. Sak, pers. Comm., November 4, 2025) in conjunction with behavioral or psychological therapy (see Section 9).

In a systematic review of 106 studies covering a total of 37,515 active military personnel and veterans with mild bTBI, PTSD had the highest prevalence (see also Faraji et al., Reference Faraji, Ghane and Javadzadeh2023), but sensory impairments, depression, cognitive deficits, and sleep disturbances were also often mentioned as co-morbid issues (Phipps et al., Reference Phipps, Mondello and Wilson2020). Visual impairments and headaches were very common (40% to 68%). The authors concluded that modern weaponry uses frequent blasts, and health care workers should be aware of the prevalence of bTBI and its psychological sequalae for the targeted civilians and soldiers, as well as for the operating combatants. Military personnel and civilians alike are increasingly exposed to bTBI by so-called “nonlethal,” acoustic weapons, and health care should recognize that victims may require specific treatments (see Section 9).

5.2.1 Posttraumatic Stress Disorder

War leaves deep psychological scars. It is no surprise that experiences of armed conflict are among the most gruesome or traumatic events humans can face. Traumatic events have been defined as severe challenges to the ability to adapt, requiring changes to thoughts or behavior; such challenges may pertain to threatened death, serious injury, or sexual violence (Brewin et al., Reference Brewin, Atwoli and Bisson2025). Reactions to this type of event range from chronic anxiety and depression to the more specified condition of PTSD. Posttraumatic stress disorder has been defined and measured in different ways, but four aspects that impair daily functioning have remained at its core (see Figure 3): Re-experiencing the traumatic event(s), avoidance of reminders, negative alterations in cognition or mood (e.g., numbing), and alterations in arousal or reactivity (e.g., sleep issues or startle responses). Lifetime prevalences of PTSD vary widely across studies because of differences in measures, cultural contexts, and time windows, but in the general population outside war zones, a crude estimate would be around 5% (Brewin et al., Reference Brewin, Atwoli and Bisson2025).

Figure 3

Core features of PTSD

In a meta-analysis of forty-eight studies involving 40,004 civilians living in war zones, Ahmed et al. (Reference Ahmed, Zakai and Zahid2024) reported a pooled PTSD prevalence of 24%, which is a strikingly high rate compared to 5% in populations outside of conflict regions. Findings for adolescents mirror the figure of 24% for adult civilians. Zasiekina et al. (Reference Zasiekina, Griffin, Blakemore, Hlova and Bignardi2025), drawing from twenty-one studies with 12,898 young people exposed to war-related trauma, estimated a PTSD prevalence of 29%. Lim et al. (Reference Lim, Tam and Chudzicka-Czupała2022) expanded the lens, comparing civilians and military populations exposed to war. Their meta-analysis of forty-five studies, encompassing 67,153 participants across twenty-three armed conflicts, found a similar PTSD prevalence of 24%. Notably, rates did not significantly differ between civilians and military personnel.

In some cases, PTSD rates are even higher. Tinsae et al. (Reference Tinsae, Shumet and Tadesse2024), for example, reported a prevalence of 48% among civilians in conflict-ridden regions of Ethiopia. Adverse childhood experiences (ACEs) can also amplify PTSD symptoms. As Crede et al. (2023) showed, ACEs increase vulnerability to PTSD and are linked to insecure-disorganized attachment in children and unresolved trauma in adults (see Section 2). Lim et al. (Reference Lim, Tam and Chudzicka-Czupała2022) further identified risk factors such as female gender, witnessing the murder of a loved one, losing a close family member, facing life-threatening danger, and lacking social support (see also Grover et al., Reference Grover, Schofield and Burdett2024). Many of these risk factors align closely with attachment-related issues. Carmona et al. (Reference 68Carmona, Camilo, Carvalho and Chambel2024) found similar predictors in peacekeepers, including UN Blue Helmets: Being female, longer periods post-deployment, and low social support contributed to higher PTSD rates – reaching 53% of the peacekeeper population.

5.2.2 Depression and Anxiety

Depression and anxiety surge during and after war – and they, in turn, raise the risk of PTSD. Ahmed et al. (Reference Ahmed, Zakai and Zahid2024) found that 26% of participants in war-torn areas reported elevated levels of depressive symptoms, with notably higher depression rates in civilians (35%) compared to military (21%). Anxiety was more than twice as high among civilians (39% vs. 16%). Grover et al. (Reference Grover, Schofield and Burdett2024) showed that low social support played a key role in the prediction of both depression and anxiety levels. Finally, Angelakis et al. (Reference Angelakis, Molina and Winter2025) examined sixty-two studies with over a million participants and found a two-fold increase (ORs 1.91–2.57) in suicidal ideation and suicide attempts among military personnel and veterans who experienced sexual trauma during service. Women were especially affected. Combat exposure itself was another predictor of suicidal thoughts and behaviors (ORs 1.06 and 1.27, respectively) although less strong than the effects of sexual trauma during military service.

5.2.3 War Experiences Related to Insecure Attachment

In a study of Israeli veterans from the 1973 Yom Kippur war, some of whom had been held captive during the war, while others fought on the same fronts but were never held captive, attachment style was measured using an attachment style questionnaire 18, 30, and 35 years after the war (Mikulincer et al., Reference Mikulincer, Ein-Dor, Solomon and Shaver2011, Reference Mikulincer, Solomon, Shaver and Ein-Dor2014). Captivity during war can involve torture, solitary confinement, and deprivation of basic physical and psychological needs, more than once deliberately aimed at breaking a prisoner’s spirit (e.g., Hunter, Reference Hunter, Wilson and Raphael1993). Eighteen years after the war, veterans who had not been held captive were more secure than those who had been, and the veterans without captivity experiences increased in security over time. Veterans who had been held captive scored higher on self-reported anxious and avoidant attachment at the initial assessment, and at the subsequent waves, their anxiety and avoidance further increased. These results show the long-term pathogenic effect of war captivity on social relationship functioning and may be related to the finding that those who were held captive had more marital problems and lower marital satisfaction than control veterans (Ein-Dor et al., Reference Ein-Dor, Doron, Solomon, Mikulincer and Shaver2010). Moreover, PTSD was associated with higher attachment insecurity scores at 18, 30, and 35 years after the war, apart from effects of war captivity. This finding suggests that attachment style covaries with broader issues than just social functioning, including the management of stressful thoughts and events. However, direction of effects cannot be derived from this study.

5.3.1 Posttraumatic Stress Disorder in Ukraine

The earliest available evidence on psychological consequences comes from a 2016 study, which found that approximately 32% of internally displaced adults met criteria for probable PTSD (Roberts et al., Reference Roberts, Makhashvili and Javakhishvili2019). Following the full-scale invasion in February 2022, violence against civilians escalated dramatically, resulting in widespread destruction of infrastructure and significant civilian casualties, including children (Haque et al., Reference Haque, Naeem and Wang2022). A rapid epidemiological study conducted within the first few weeks of the war found that 53% of respondents reported significant psychological distress (Xu et al., Reference Xu, Pavlova and Chen2023). These findings were supported by a subsequent study by Ben-Ezra et al. (Reference Ben-Ezra, Goodwin, Leshem and Hamama-Raz2023), conducted several months post-invasion, which reported that 31% of individuals met criteria for provisional PTSD. Moreover, the study found that both nationally and internationally displaced individuals exhibited higher levels of PTSD symptom severity than those who remained in place. The role of displacement was further supported by other studies, including Lushchak et al. (Reference Lushchak, Velykodna and Bolman2023), which surveyed 3,173 respondents and found that the prevalence of provisional PTSD was associated with displacement status: 51% among non-displaced individuals, 55% among internally displaced persons, and 62% among refugees who left Ukraine.

Determining whether displacement itself was a primary traumatic factor or whether displaced individuals were more severely affected due to cumulative exposure remains difficult, but evidence consistently shows that the number of traumatic events was strongly associated with increased PTSD risk. Karatzias et al. (Reference Karatzias, Shevlin, Ben-Ezra, McElroy, Redican, Vang, Cloitre, Ho, Lorberg, Martsenkovskyi and Hyland2023) conducted a study six months after the full-scale invasion and found that parents experienced a mean of 9.07 war-related stressors (range: 1–26). Among them, 41% met diagnostic criteria for PTSD. The study demonstrated that higher exposure to war-related stressors predicted more PTSD, suggesting a dose-response effect. Similar findings were reported in a study conducted fourteen to fifteen months post-invasion (Wang et al., Reference Wang, Barrett and Hicks2024a), supporting the link between cumulative trauma exposure and stress-related disorders. The duration of exposure to war-related stressors may further compound psychological risk, as indicated by data from mobile mental health clinics operating in de-occupied territories of Ukraine, where up to 51% of individuals screened positive for PTSD (Ressler et al., Reference Ressler, Hinchey and Mast2024).

Some data retrieved from the civil studies can be attributed to the military population. The exact number of Ukrainians who have served in combat roles remains unknown, but official estimates suggest that the size of the Ukrainian Armed Forces reached approximately 880,000 personnel by 2024. One of the few studies to examine mental health outcomes in this population is the Ukrainian National Initiative of Trauma Study (UNITY Study), an ongoing project assessing psychological outcomes among soldiers. An analysis involving 590 clinician-administered interviews with service members recruited from general hospitals, psychiatric institutions, and rehabilitation centers found that 68% were diagnosed with (complex) PTSD (Hyland et al., Reference Hyland, Shevlin, Karatzias, Bondjers, Scherbakova, Sulaieva, Bibikova, Dudin, Savchenko, Voznitsyna, Dosenko and Martsenkovskyi2025). Notably, officers more often had complex PTSD, suggesting that the nature and context of trauma exposure may contribute to symptom presentation.

5.3.2 Depression, Anxiety, and Grief

Depression and anxiety also constitute a substantial component of war-related psychological burden. A study conducted approximately six months after the full-scale invasion employed adapted versions of the 9-item Patient Health Questionnaire for depression (PHQ-9; Kroenke et al. Reference Kroenke, Spitzer and Williams2001) and 7-item Generalized Anxiety Disorder questionnaire (GAD-7; Spitzer et al., Reference Spitzer, Kroenke, Williams and Löwe2006) to assess changes in adult mental health symptoms. The findings indicated that 72% of respondents reported an increase in anxiety symptoms, and 62% reported elevated depressive symptoms. Notably, the severity of symptom increases was associated with the number of war-related stressors experienced, suggesting a dose-response relation between conflict exposure and internalizing psychopathology (Hyland et al., Reference Hyland, Vallières, Shevlin, Karatzias, Ben-Ezra, McElroy, Vang, Lorberg and Martsenkovskyi2023). These troubling results were largely consistent with those of similar cross-sectional epidemiological studies (e.g., Brackstone et al., Reference Brackstone, Head and Perelli-Harris2024; Lushchak et al., Reference Lushchak, Velykodna and Bolman2023; Wang et al., Reference Wang, Barrett and Hicks2024a) with higher symptom levels for displaced versus non-displaced individuals.

In a study conducted 18 months after the full-scale invasion, Martsenkovskyi et al. (Reference Martsenkovskyi, Shevlin and Ben-Ezra2024b) found that 36% of respondents met criteria for at least one mental health disorder, and 16% met diagnostic thresholds for two or more disorders. Generalized Anxiety Disorder (GAD) emerged as the most prevalent condition, reported by 21% of women and 10% of men. Depression was present in 11% of women and 6% of men. Risk for both conditions was significantly associated with female gender, financial hardship, and cumulative exposure to war-related stressors, and GAD was more prevalent among younger respondents. Notably, anxiety and depression were strongly correlated and associated with complex PTSD and prolonged grief disorder, underscoring the interconnected nature of these conditions in war contexts. Similarly, high rates of depression and its comorbidity with stress-related disorders were also reported in the UNITY Study (Hyland et al., Reference Hyland, Shevlin, Karatzias, Bondjers, Scherbakova, Sulaieva, Bibikova, Dudin, Savchenko, Voznitsyna, Dosenko and Martsenkovskyi2025), in which 34% of Ukrainian soldiers met diagnostic criteria for depression, including comorbidity rates of 45% among those with PTSD and 51% among those with complex PTSD. Similarly, Martsenkovskyi et al. (Reference Martsenkovskyi, Shevlin and Ben-Ezra2024b) found that 10% of males and 13% of females met diagnostic criteria for prolonged grief disorder, whose symptoms were strongly correlated with GAD, depression, and complex PTSD. Risk of prolonged grief disorder was associated with the loss of a loved one in the past six to twelve months, the death of a loved one due to war, and the absence of recent contact with the deceased prior to their death (Redican et al., Reference Redican, Hyland and Shevlin2024).

5.3.3 Insomnia and Short-Sleep Duration

The earliest available post-invasion data in Ukraine, collected in March 2022, from a civilian adult sample residing predominantly in areas without active ground fighting, and in which 56% of participants had children under 18 years of age, indicated that 12% of respondents reported insomnia symptoms, with female gender and residence in urban areas identified as risk factors (Xu et al., Reference Xu, Pavlova and Chen2023). A subsequent web-based survey of Ukrainian university students conducted 6–7 months after the full-scale invasion found that 49% experienced insomnia symptoms and 86% reported war-related nightmares (Pavlova & Rogowska, Reference Pavlova and Rogowska2023). An epidemiological study by Wang et al. (Reference Wang, Hicks and Barrett2024b), conducted approximately 15 months post-invasion, reported that 39% of adults experienced short sleep duration, 7% reported long sleep duration, and 38% met criteria for insomnia. Predictors of short sleep and insomnia included female gender, previous exposure to the 2014 Russian invasion, comorbid depression, and higher cumulative exposure to war-related stressors.

7.2.1 Burnout in Ukrainian Parents before the Full-Scale Russian Invasion

In this pre-full-scale invasion study in Ukraine, fathers scored lower on burnout than mothers. Perfectionism, stress, and resilience (reversed) were modestly related to burnout. Strong associations (around r = 0.50) between burnout and parental neglect and violence were found. When the usual cut-off for the 23-item PBA was used, burnout prevalence was 11% for mothers and 3% for fathers. The cut-off level implies that parents experience every symptom at least once a week (Zbrodska et al., Reference Zbrodska, Roskam, Dolynska and Mikolajczak2022). The burnout prevalence of 11% in mothers is high compared to the set of forty-two countries studied by Roskam et al. (Reference Roskam, Aguiar and Akgun2021). Of course, the invasion of Russia in the Ukrainian territory of Crimea in 2014 started a war that had already lasted for eight years at the time of the study. The authors stated that this high prevalence is “a serious challenge for researchers and clinicians in Ukraine” (p. 10).

7.2.2 Strongly Elevated Parental Burnout after the Full-Scale Russian Invasion

With a team of colleagues from Lviv National Medical University, we conducted a study in Ukraine about two years after the full-scale invasion (Chemerys et al., Reference Chemerys, Tsyhanyk, Bakermans-Kranenburg, van IJzendoorn and Nehrych2025). Of course, conditions for data collection were less than optimal to say the least, and we were assisted by the Ukrainian Gradus Research Company to create a brief online version of the PBA, the Brief Parental Burnout scale (BPBs; Aunola et al., Reference Aunola, Sorkkila and Roskam2021) with five items on parenting-related feelings of exhaustion, distancing from the children, and sense of parental ineffectiveness. This is the brief version of the 23-item PBA with demonstrated adequate psychometric properties (Aunola et al., Reference Aunola, Sorkkila and Roskam2021). Added to the online questionnaire were some items on harsh and neglecting parenting and on the use of mood-regulating medicine (e.g., antidepressants).

Gradus recruited four groups with about 100 participants each to complete the online survey in March 2024. The four groups had the following family compositions: (1) Families with husbands serving from the start of the full-scale invasion in February 2022 and still in the army at the time of data collection (n = 100); (2) families with a veteran husband who retired from the army in 2023 (n = 107); (3) families of which the (civilian) husband never served in the army (n = 105); (4) families headed by grandmothers taking care of their grandchildren because the parents were deceased or otherwise unavailable due to the war (n = 103). At least one child in each family was younger than twelve years of age. For details of design, measures, and statistical analyses, see Chemerys et al. (Reference Chemerys, Tsyhanyk, Bakermans-Kranenburg, van IJzendoorn and Nehrych2025).

We found overall very high levels of parental burnout, amounting to 58%. This prevalence rate is much higher than any prevalence found in other countries, within or outside Europe (Roskam et a., Reference Roskam, Aguiar and Akgun2021). The percentage is almost 6 times higher than the one found just before the Russian full-scale invasion two years earlier (Zbrodska et al., Reference Zbrodska, Roskam, Dolynska and Mikolajczak2022). Of course, sampling and measures were different, but regardless of confounders the increase in burnout levels is remarkable and disturbing. During ten years of war since the Russian invasion of Crimea in 2014, Ukraine has been subjected to constant Russian threats, attacks, and killings, inducing chronic stress that elevates the levels of exhaustion and emotional distancing and diminishes the sense of efficacy in families. Not only parents with a partner in active service are at risk for burnout, but also dual-parent families with returned veterans who may suffer from traumatic brain injuries or mental health conditions. Higher levels of burnout symptoms were strongly related to self-reported harsh and neglectful parenting (r = 0.63).

7.2.3 High Levels of Grandparental Burnout

We found that grandmothers replacing parents who were unavailable due to the war reported the highest level of burnout symptoms. The large majority (70%) of grandmothers caring for their grandchildren in the absence of the parents suffered from burnout according to conventional criteria (Chemerys et al., Reference Chemerys, Tsyhanyk, Bakermans-Kranenburg, van IJzendoorn and Nehrych2025). In previous meta-analytic and empirical work, we found that grandmothers may play an important supportive and buffering role for parents in dire circumstances as long as their relationships with the parents are warm, cooperative, and harmonious (Liang et al., Reference Liang, Lin, Van IJzendoorn and Wang2021; Riem et al., Reference Riem, Bakermans-Kranenburg, Cima and Van IJzendoorn2023; Riem et al., Reference Riem, Perrykkad and Watson2024). But if grandmothers must fill the gap of absent parents, they might easily become overburdened and reach the limits of their coping capacity, a finding supported by Chan and her colleagues (Reference Chan and Piehler2024) in a narrative review covering sixty-five studies from twenty-nine countries.

Ukrainian grandmothers may struggle with anxieties, traumatic experiences, or the (unresolved) loss of the parents of their grandchildren. In wartime, two more specific but speculative explanations for the high burnout among grandmothers are the “aging health hypothesis” and the “intergenerational trauma hypothesis” (Chemerys et al., Reference Chemerys, Tsyhanyk, Bakermans-Kranenburg, van IJzendoorn and Nehrych2025). The first hypothesis refers to grandmothers’ anxiety about the considerable risk of chronic age-related diseases before their grandchildren reach the age of independent functioning. The second explanation suggests that during the ongoing war grandparents may revisit or relive past traumatic experiences with the horrors of famine (Holodomor), genocide (Holocaust), and political oppression (Stalinism) that this older generation and their parents might have undergone in Ukraine as a contested “bloodland” (Snyder, Reference Snyder2022).

7.3.1 Parental Sensitive Responsiveness in Wartime

In general, children in war-affected families who are more exposed to war seem to have to interact with parents who are less sensitive. In a systematic narrative review of eleven studies with observed parenting, Chasson et al. (Reference Chasson, Ben-Shlomo and Lyons-Ruth2025) reported the impact of parental war trauma on subsequent parenting of young children (0–3 years). War-related parental traumatic experiences predicted less sensitive responsiveness. For example, in one of the studies, higher levels of maternal PTSD symptoms in refugee mothers in the Netherlands were associated with more insensitive, unstructured, and hostile parent-child interactions (Van Ee et al., Reference Van Ee, Kleber, Jongmans, Mooren and Out2016a, Reference Van Ee, Jongmans, Van der Aa and Kleber2016b). In a similar vein, Punamaki et al. (2020) observed that maternal vocal emotion expressions of Palestinian mothers toward their children were less positive after having experienced severe traumatic war events. Dozio et al. (Reference Dozio, Feldman and Bizouerne2020) found that during a trauma narration session, mothers touched and looked at their infant less and were more absent and smiled less at their infant. In studies by Shachar-Dadon et al. (Reference Shachar-Dadon, Gueron-Sela, Weintraub, Maayan-Metzger and Leshem2017) and Cohen and Shulman (Reference Cohen and Shulman2017) mothers with traumatic war-related experiences were less emotionally available in interactions with their children.

7.3.2 War Increases Risks of Harsh Parenting

In another meta-analysis, Eltanamly et al. (Reference Eltanamly, Leijten, Jak and Overbeek2021) presented worldwide data on the association between war experiences of parents and their parenting style. The meta-analysis includes 38 studies with 54,372 families (parents and their children, average age 12 years) exposed to war, political violence, or armed conflict. War exposure was measured in terms of frequency counts of war-related incidents or severity of the experiences. Parenting style was reported by parents or their (mostly adolescent) children completing questionnaires. Studies of nonwar-related trauma and brief one-time violence instead of prolonged exposure, were excluded. Families came from the Middle East, Africa, Asia, and Eastern Europe, but Ukraine was not included. Pooling the correlations showed that war exposure weakly predicted less parental warmth (combined r = −0.02) and less parental behavioral control (combined r = −0.01), but war experiences were most importantly and substantially related to more parental harshness with a combined effect size of r = 0.12 (Eltanamly et al., Reference Eltanamly, Leijten, Jak and Overbeek2021). Study quality or design (cross-sectional or longitudinal) did not affect the results.

7.3.3 Posttraumatic Stress Associated with Lower Parenting Quality

As Chasson et al. (Reference Chasson, Ben-Shlomo and Lyons-Ruth2025) hypothesized, posttraumatic stress symptoms were a better predictor of parenting problems than actual combat experiences. In ten studies (4,304 parents), Kritikos and her colleagues (Reference Kritikos, Comer and He2019) found a combined effect size between combat-related parental PTSD and parenting problems (combined r = 0.26), and a lower combined effect size (r = 0.10) in seven studies on the association between actual combat exposure and parenting problems (5,232 parents). The authors concluded that PTSD might be a more powerful deployment-related predictor of “family stress, disequilibrium, and disorganization” (p. 143) than combat experiences per se. They suggested that alleviating these stressors and family issues might be especially effective for supporting veteran families to re-integrate into their communities.

7.4.1 More Maltreatment in Refugee Families

War-related atrocities experienced by refugees might leave their mark on the next generation because of elevated levels of family violence and other types of child maltreatment. In our Netherlands Prevalence Study of Maltreatment (NPM; Euser et al., Reference Euser, Van IJzendoorn, Prinzie and Bakermans-Kranenburg2011), we addressed the question of whether children in Dutch-immigrant families would be at increased risk of child maltreatment. We differentiated traditional immigrant families who immigrated because of economic reasons from refugees who came from unsafe countries in war, political, or ethnic conflicts. We used three sources of information about child maltreatment: 1,127 professionals working with children in various occupational branches reporting about cases of maltreatment in the last 3 months (‘sentinels’), 1,759 high school students reporting on their own experiences of maltreatment in the past year, and all data from all Dutch Child Protective Services in one year. We found that in both types of immigrant families – economic or refugee migrants – the prevalence of maltreatment was elevated compared to Dutch families without an immigration background. But controlling for sociodemographic factors, the difference in maltreatment risk remained elevated in the sentinel and CPS reports only in the refugee families. We speculated that traumatic stress and the continuing stress of their insecure asylum status in the host country would increase the risks of family violence and neglect (Euser et al., Reference Euser, Van IJzendoorn, Prinzie and Bakermans-Kranenburg2011).

7.4.2 From War-Related Violence to Intimate Partner Violence to Child Maltreatment

A meta-analysis of studies of intimate partner violence in active-duty military personnel and in veterans by Cowlishaw and colleagues (Reference Cowlishaw, Freijah and Kartal2022) shed some systematic light on the prevalence and correlates of intimate partner violence. The study included thirty-one studies involving 172,790 participants. Only two studies were conducted outside Northern America, one in the United Kingdom and one in Turkey; thus, prevalence estimates from Eastern European, Asian, South American, and African countries are missing. The authors estimated a prevalence of 13% for military personnel and veterans reporting any recent intimate partner violence perpetration (11 studies, with 131,140 participants), and a prevalence of being a victim of intimate partner violence of 21% (14 studies, with 121,649 participants). Prevalences did not differ by gender, period of service, or country of origin. Veterans showed a larger perpetration percentage (34%) than active-duty military personnel (5%), pointing to a “sleeper” effect of the consequences of war experiences emerging in postwar years. In the USA civilian populations, the prevalence of past year perpetration of intimate partner violence is estimated to be 6%, and victimization 5%, considerably lower than those found in military personnel.

For civilians, associations between experiences of political or war-related violence and heightened levels of intimate partner violence have been documented (e.g., Saile et al., Reference Saile, Ertl, Neuner and Catani2014). In their study of 1,185 Ukrainian adults from the Donbas region, Botchkovar et al. (Reference Botchkovar, Antonaccio and Timmer2025) reported that daily hassles, negative affect, and excessive alcohol abuse predicted involvement in intimate partner violence for both males and females. War exposure itself did not seem to have a direct impact on intimate partner violence. This unexpected result might be partly due to the absence of a matched comparison group of participants neither directly nor indirectly exposed to war, thus reducing variation in war experiences. As a result, war experiences may be substantively relevant but lacking sufficient variation to achieve significance.

Importantly, in a quantitative umbrella synthesis of meta-analyses of child maltreatment antecedents, we found intimate partner violence among the risk factors for elevated levels of child abuse and neglect (Van IJzendoorn et al., Reference Van IJzendoorn, Bakermans-Kranenburg and Duschinsky2020). The umbrella synthesis covered about 1.5 million participants. Intimate partner violence as an antecedent of emotional abuse had an effect size of d = 0.47 (six studies including 5,500 families), and for physical abuse, the combined effect size was d = 0.35 (fifteen studies including 8,500 families). In combination, these findings indicate how war violence may trigger intimate partner violence that, in turn, is related to child maltreatment.

7.4.3 Unresolved Loss or Trauma and Insecure Attachment in Veterans and Refugees

Most studies of attachment insecurity and elevated risk of child maltreatment have used self-reports of attachment style that have limited validity compared to the Adult Attachment Interview and other interview or observational assessments (Van IJzendoorn & Bakermans-Kranenburg, Reference Van IJzendoorn and Bakermans-Kranenburg2024). However, in a study on Dutch veterans with deployment-related trauma, we documented a strong association between unresolved attachment as assessed with the AAI and level of PTSD symptoms measured with the Clinician Administered PTSD Scale (CAPS; Harari et al., Reference Harari, Bakermans-Kranenburg and De Kloet2009). Meta-analytically, we found unresolved loss or trauma in combination with insecure-preoccupied representations of attachment overrepresented in samples with PTSD (see Section 2), as well as in samples struggling with addiction (Bakermans-Kranenburg et al., Reference Bakermans-Kranenburg, Dagan, Cárcamo and Van IJzendoorn2025). These findings point to the potential role of unresolved attachment representations in veterans who bring their war-related traumatic experiences home and into their families. Ukrainian veterans might be expected to return home not only depleted, exhausted, and burned out as soldiers and parents, but they also might have to cope with unresolved losses and other war-related traumatic experiences. In a narrative systematic review of refugee samples, Egger et al. (Reference Egger, Buchheim and Gander2025) reported that insecure and unresolved attachment were consistently linked to higher psychological distress, particularly PTSD, in adults. In refugee children, insecure attachment was associated with parental mental health problems and dysfunctional parenting, whereas secure attachment might buffer the effects of parental PTSD.

9.1.1 Developmental Damage in Residential Care

Residential care harms child development in peace and war (Dozier et al., Reference Dozier, Smetana and Allen2025). Our meta-analysis of more than 300 studies and >100,000 children across more than sixty countries found that group care delays or damages physical/brain growth, health, cognition, attention, and socio-emotional development (Van IJzendoorn et al., Reference Van IJzendoorn, Bakermans-Kranenburg and Duschinsky2020). Effects were large for young children (under 3.5 years) in terms of physical growth, health, head circumference, and cognitive development. Attention and socio-emotional development also lagged behind but to a lesser extent. Adolescents (10–16 years) still showed sizeable delays, about half a SD on average across the various physical and mental domains. Although the effects were slightly smaller than in early childhood, perhaps because adolescents exert more active influence on their environment due to gene–environment correlation, confidence intervals mostly overlapped with those of younger children, suggesting no conclusive mitigation with age.

9.1.2 Small-Group Care: SOS Children Villages

Small homes are not a remedy. We reviewed reputedly high-quality small-group models such as SOS Children’s Villages (global NGO; 2018 budget a staggering ≈ USD 1.37 billion; also active in Ukraine; Van IJzendoorn & Bakermans-Kranenburg, Reference Van IJzendoorn and Bakermans-Kranenburg2022). Despite the motto of the founder of SOS Children Villages some seventy years ago, Hermann Gmeiner: “Tear down your institutions,” SOS institutions have been sprawling across the globe. Only eight evaluative studies on the effects of SOS institutions met meta-analytic criteria. Meta-analysis showed that SOS children did better than those in large institutions on mental health but worse on physical growth, and inferior compared to family-reared peers on both. Small-group homes are therefore not a viable substitute for family-based care. De-institutionalization, also of SOS Children Villages, is urgently needed. We acknowledge that economic hardship and disruption of local social infrastructure during and after armed conflicts can limit the capacity of (extended) families or foster families to take on the care of additional children without available parents. Increasing the number of alternative family-based care settings might be hampered by problems of recruiting and supporting enough family caregivers, particularly with large-scale displacements such as in Ukraine. The serious war-related barriers to rapid implementation of family-based alternatives, however, should not siphon resources away from responsible de-institutionalization to (re-)establishing institutions. Elsewhere, we discussed potential transitional solutions with intermediate steps toward the goal of family-based care for all children (Goldman et al., Reference Goldman, Bakermans- Kranenburg and Bradford2020).

A halt to SOS Children Villages is the more urgent as independent inquiries report serious sexual, physical, and emotional abuse in multiple SOS villages across Africa, Asia, and Central America (Independent Special Commission; Van Wijk et al., 2023). Substantiated allegations also implicate founder Hermann Gmeiner in abuse within Austrian homes (SOSCV statement of apologies on October 23, 2025). These findings echo systemic maltreatment uncovered in institutional care for indigenous children in Canada, the USA, Greenland, and elsewhere. Harm extends to staff: residential work is “a challenging job” with frequent experiences of threats and violence (Alink et al., Reference Alink, Euser, Bakermans-Kranenburg and Van IJzendoorn2014). Elevated maltreatment may stem from mixed-age/mixed-gender groupings (Euser et al., Reference Euser, Alink, Tharner, Van IJzendoorn and Bakermans-Kranenburg2014), deviant-peer contagion (Dishion & Tipsord, Reference Dishion and Tipsord2011), and, especially, unstable caregiving without long-term attachment relationships, consistent but sensitive limit-setting, and empathic monitoring. Low training, low status and low salaries lead to high staff turnover. While maltreatment also occurs in the general youth population and in alternative family-based care, risks are consistently higher in residential settings.

9.2.1 PTSD Medication

As was highlighted above (see Section 3), PTSD is among the most frequently reported mental health outcomes in trauma-affected children. Yet, the existing evidence for medication use in pediatric PTSD is limited. Most available data are derived from case reports, open-label studies, or a small number of controlled trials with modest sample sizes, which restricts the ability to draw firm conclusions about efficacy and safety (Bandelow et al., Reference Bandelow, Allgulander and Baldwin2022).

Several studies have highlighted the role of increased noradrenergic activity in the pathophysiology of PTSD, which has drawn attention to agents such as clonidine and guanfacine. These alpha-2 agonists act through presynaptic receptor stimulation, reducing central noradrenergic tone and thereby attenuating hypervigilance, irritability, and autonomic hyperarousal. Alpha-1 antagonists reduce central noradrenergic activity by blocking postsynaptic alpha-1 receptors, particularly in locus coeruleus, improving hyperarousal, nightmares, and sleep disturbances (Belkin & Schwartz, Reference Belkin and Schwartz2015; Slavova et al., Reference Slavova, Ortiz and Blaise2024). Some evidence, though limited, suggests that these agents may be useful for managing hyperarousal and sleep disturbances in children with trauma-related disorders.

A small 8-week open-label study of extended-release guanfacine downregulating noradrenergic tone in traumatized children (n = 17) demonstrated significant posttreatment improvement on the parent-reported UCLA PTSD Reaction Index (with a suspiciously large effect size d = 1.28; Connor et al., Reference Connor, Grasso, Slivinsky, Pearson and Banga2013). Notably, nearly 90% of participants also met diagnostic criteria for ADHD, which warrants caution in interpretation due to potential bias arising from overlapping symptom domains of ADHD and PTSD in children (Daud et al., Reference Daud and Rydelius2009). This overlap could be amplified when assessments rely primarily on parental rather than clinician-rated measures. The evidence base remains insufficient to support routine use, as most reports lack control conditions, standardized outcome measures, and long-term follow-up.

9.2.2 Antidepressant Medication

Several studies have explored the use of antidepressants in pediatric burn victims, either with the aim of preventing the onset of PTSD or reducing symptoms of acute stress disorder. While these early trials suggested potential benefit, their interpretation requires caution in light of the current conceptualization of acute stress disorder as a short-lasting, largely self-limiting reaction to trauma with low predictive value for later PTSD (Maercker & Eberle, Reference Maercker and Eberle2022). For example, a 24-week randomized placebo-controlled trial in children with PTSD symptoms (n = 26) reported a (too) large parent-reported improvement compared with placebo (d = 1.27), while no differences were observed on child-reported measures (Stoddard et al., Reference Stoddard, Luthra and Sorrentino2011). Likewise, a 3-week RCT comparing fluoxetine, imipramine, and placebo in a small sample of pediatric burn victims (n = 60) failed to demonstrate any clinical advantage of antidepressants over placebo (Robert et al., Reference Robert, Tcheung and Rosenberg2008). Notably, in addition to the studied medications, children were not limited in receiving morphine, hydroxyzine, diazepam, etc. Thus, despite the existing evidence on the efficacy of antidepressants, particularly SSRIs, in adults with PTSD, the existing evidence in children is limited to some RCTs, which do not support the prescription of these agents for routine use.

9.2.3 Antipsychotic Medication

Although primarily developed for the treatment of psychotic disorders, antipsychotics are increasingly prescribed in pediatric populations, with a steady rise observed across multiple countries in recent years (Klau et al., Reference Klau, Bernardo, Gonzalez-Chica, Raven and Jureidini2022; Radojčić et al., Reference Radojčić, Pierce and Hope2023; Waszak et al., Reference Waszak, Zagożdżon, Pierucka and Kubanek2018; Zhaojian et al., Reference Zhaojian, Meizhu and Jun2024). These medications are also used off-label to manage anxiety, mood, and behavioral disorders in children, as well as in high-income countries. The trend is particularly pronounced among vulnerable populations, such as foster children, and in remote or underserved areas where access to specialist mental health care is limited (Klau et al., Reference Klau, Bernardo, Gonzalez-Chica, Raven and Jureidini2022; Keefe et al., Reference Keefe, Cummings, Smith, Greeley and Van Horne2023). But evidence for the use of antipsychotics in pediatric PTSD is limited to case reports and small case series, which mainly describe reductions in trauma-related aggression, disruptive behavior, self-harm, and insomnia (Meighen et al., Reference Meighen, Hines and Lagges2007; Stathis et al., Reference Stathis, Martin and McKenna2005).

Overall, while some preliminary observations indicate potential benefits of antipsychotics for severe externalizing symptoms in pediatric PTSD, these agents should be reserved for treatment-resistant cases as a last resort. The absence of randomized controlled trials and the risk of adverse effects should drastically limit routine use in clinical practice.

9.3.1 Neurofeedback

Adult studies have demonstrated the efficacy of neurofeedback, particularly as an adjunct to trauma-focused psychotherapy (Voigt et al., Reference Voigt, Mosier and Tendler2024) in reducing core PTSD symptoms. However, evidence in pediatric populations remains limited. A 12-week randomized controlled trial (24 sessions; n = 32) of neurofeedback training (NFT) versus a wait-list control (treatment as usual) in children, conducted by Rogel and colleagues (2020), found that at the treatment endpoint only four of sixteen participants in the NFT group, compared to ten of sixteen in the wait-list group, continued to meet diagnostic criteria for PTSD on the K-SADS-PL. However, group differences were no longer statistically significant at the 1-month follow-up. At the same time, children who received NFT demonstrated larger between-group improvements on the Child Behavior Checklist (CBCL), with effect sizes ranging from d = 0.48 to d = 1.15 for internalizing and externalizing behavior problems at follow-up.

Another 6-week randomized controlled trial (two sessions per week; n = 77) compared a game-based meditation application using a brain-sensing neurofeedback headband (Muse) with treatment as usual (TAU) in traumatized children residing in institutional care (Schuurmans et al., Reference Schuurmans, Nijhof, Scholte, Popma and Otten2021). At posttreatment, participants in the TAU group exhibited a longer basal pre-ejection period and greater skin-conductance response to acute stress, indicating lower resting sympathetic activation and more adaptive physiological reactivity to stress stimuli. Although these differences reached statistical significance, the corresponding effect sizes were small, suggesting uncertain clinical relevance. The headband might, however be an effective placebo (see Section 11).

Existing trials are few, underpowered, and methodologically heterogeneous, underscoring the need for further investigation before neurofeedback can be recommended as a standard or adjunctive treatment for childhood PTSD.

9.3.2 Repetitive Transcranial Magnetic Stimulation (rTMS)

Repetitive Transcranial Magnetic Stimulation (rTMS) is a noninvasive technique that uses rapidly changing magnetic fields to stimulate specific regions of the cerebral cortex through the skull (Sak et al., in prep.), see Figure 5. By modulating neural activity in targeted networks, rTMS is hypothesized to alter cognitive, emotional, or motor processes and is increasingly used in neuroscience research and experimental clinical treatments of adults suffering from concussion syndrome, depression, or PTSD. A meta-analysis by Belsher et al. (Reference Belsher, Beech and Reddy2021), including 549 participants across available randomized trials, found that rTMS was superior to sham stimulation in reducing PTSD symptoms (an incredibly large effect size of d = −1.13). However, most of the included studies were underpowered, with small sample sizes and considerable heterogeneity in stimulation protocols and treatment parameters, limiting the robustness of this conclusion. A more recent Cochrane Review by Brown et al. (2024) reported little to no difference in core PTSD symptoms immediately following rTMS treatment or at short-term follow-up.

Figure 5

Mechanism of transcranial magnetic stimulation

In pediatric populations, existing data indicate that rTMS is largely safe and well-tolerated, with some evidence of benefit for certain psychiatric conditions. Nevertheless, the research base remains preliminary, with most pediatric studies focusing on neurodevelopmental disorders and depression, in the absence of randomized trials of rTMS in pediatric PTSD (Gallop et al., Reference Gallop, Westwood and Hemmings2025). Taken together, the current evidence base remains insufficient to recommend rTMS for routine clinical practice, and emphasizes the need for larger, methodologically rigorous studies.

10.2.1 CBT-TR

Parenting programs with a CBT component in humanitarian contexts often combine various traditions and methods in an eclectic way (Betancourt, Reference Betancourt2025). In the culturally adapted CBT-TR approach of O’Callaghan (Reference O’Callaghan, McMullen, Shannon, Rafferty and Black2013), for example, cognitive coping with trauma is central, with identifying and changing unhelpful cognitions and awareness of relations among thoughts, feelings, and behaviors. But psychoeducation components and relaxation techniques also are part of these programs. An adaptation to female adolescents in the Democratic Republic of Congo who had suffered from sexual abuse by militia contained modules with psychoeducation on rape and trauma; stress management by controlled breathing and progressive muscle relaxation; and cognitive behavior therapy components consisting of cognitive coping and addressing trauma narratives (O’Callaghan et al., Reference O’Callaghan, McMullen, Shannon, Rafferty and Black2013). Most modules were delivered in group sessions, combined with three individual sessions on re-orienting cognitions of trauma narratives. Much work was invested in cultural adaptations of the program with practical guidance on protection against sexual violence (e.g., fetching firewood in the evening with a friend instead of alone) and the use of local games and songs. To counter the prejudice of the social environment, social workers tried to facilitate the repair of relationships with family and friends. In three caregiver sessions, the girls’ parents were introduced to the program, made aware of the impact of trauma on their child, and other psychoeducation components. In the small trial (N = 52), the effects of this eclectic approach on the reduction of trauma symptoms, conduct problems, and depression were very large (ds ≥ 1.50, dependent on the outcome). The authors mention high initial distress levels, which might have allowed for substantial improvements. In addition, the participants’ belief in the intervention’s potential to help them (placebo, see General Discussion) may have contributed to the large (self-reported) effects.

10.2.2 Social Learning: STEP

Systematic Training for Effective Parenting (STEP; Dinkmeyer & McKay, Reference Dinkmeyer and McKay1997) is based on Adlerian ideas about family dynamics and democratic parenting, but the methods are clearly based on social learning ideas about modeling and reinforcement of positive behavior with contingent praise and avoiding coercive cycles in which parents and children might become entangled (see also VIPP-SD; Juffer et al., Reference Juffer, Bakermans-Kranenburg and Van IJzendoorn2017). STEP is usually delivered in 7–9 weeks in group workshops of about 2 hours each, with an optimal group size of 6–14 parents. The approach is didactic and includes video-recorded information about parental behavior and parenting style, exploring alternatives and expressing different ideas and feelings in role plays, understanding the reasons for children’s misbehavior, and promoting children’s responsibilities and confidence. The program also includes videos that demonstrate examples of effective and ineffective family interactions. The STEP parent handbook (Dinkmeyer & McKay, Reference Dinkmeyer and McKay1997) discusses the following ambitious goals for the parents: understanding themselves and their child; understanding children’s misbehavior and choosing an effective response; promoting positive child behavior; knowing how to listen and talk to the child; stimulating child cooperation; and using sensible limit setting. The small pilot RCT of Sangawi et al. (Reference Sangawi, Adams and Reissland2018) with 17 Kurdish mothers and their children, living in the Kurdistan region of Iraq, was a proof of principle that STEP trials could be implemented in a contested and violent environment.

10.2.3 Attachment-Based: CONNECT

Because attachment research originally was focused on the development of attachment relationships in infancy and early childhood, parenting support programs for preteens still are scarce. The CONNECT Program, starting from age eight years, fills that gap (Moretti et al., Reference Moretti, Pasalich, O’Donnell, Bartolo, Brown, Goulter and Obsuth2024). It is a ten-week manualized group program for parents and their children with behavior problems designed to enhance parental reflective function, empathy, mindfulness, and sensitivity to the attachment needs (Moretti et al., Reference Moretti, Obsuth and Craig2015; Moretti et al., Reference Moretti, Pasalich, O’Donnell, Bartolo, Brown, Goulter and Obsuth2024). In a randomized controlled trial with 120 Somalian refugee families in Sweden, Osman and colleagues (Reference Osman, Salari, Klingberg-Allvin, Schön and Flacking2017b) developed and tested a culturally tailored variant of the CONNECT program. The adapted parenting support program consisted of twelve 1–2-hour sessions with groups of 12–17 parents and two coaches. The first two sessions provided parenting education related to family life in a foreign country, whereas the remaining ten sessions used CONNECT methods such as role playing to stimulate parents’ reflection on their behavior in interactions with their children, to promote sensitive parenting behaviors and to refrain from coercive discipline practices. CONNECT is a promising attachment-based parenting support program for families with preteens and adolescents and deserves more observational outcome assessments in future RCTs with refugee families coping with war-related mental health problems.

10.3.1 Safe Care

First, a safe caregiving environment is required, free from caregiver violence and child maltreatment. Children have the right to grow up in an environment that serves their basic needs for sufficient food, shelter, and freedom from violence, oppression, or neglect by their parents or the wider society. The United Nations Security Council (2025) listed six grave violations against children in times of war: Killing and maiming of children; recruitment or use of children in armed forces and armed groups; attacks on schools or hospitals; rape or other grave sexual violence; abduction of children; and denial of humanitarian access for children (UNICEF, 2022). Between 2005 and 2023, more than 347,000 grave violations were verified against children, committed by conflicting parties in more than thirty regions across Africa, Asia, the Eastern Mediterranean, and Latin America. It is evident that in the Russian war against Ukraine, the Russians committed several such violations on a large scale (about 2,500 killed or maimed until mid-2025, about 20,000 subjected to forced abduction).

Safety should not be confused with security in the sense used in attachment theory and research. Insecure attachment might be good enough for adaptive development of children into adulthood as long as the basic needs of safety and protection, stability of care arrangement, and shared care in an attachment network are met (see Section 2).

10.3.2 Stable Care

Second, stability of care arrangements is crucial for children lest they are exposed to fragmented care and breaking of bonds. Instability of care elevates the risk of overwhelming stress in response to threatening events, enhances extreme vigilance for dangers, and distracts from exploring the wider social and physical environment that promotes developmentally important experiences (Palacios et al., Reference Palacios, Adroher and Brodzinsky2019). Stability of care arrangement does not imply the constant presence of a primary caregiver, as is suggested by the so-called “attachment parenting” movement (Sears & Sears, Reference Sears and Sears2001), which is a pseudo-scientific ideology. Stability of care is needed for both children and parents or caregivers to develop expectations about each other’s whereabouts and availability in times of need. Institutionalized care often falls at the extreme end of the dimension of fragmented care, with high staff turnover because of 24/7 care in shifts and the high burden of the profession of caregiver in congregate care (Van IJzendoorn et al., Reference Van IJzendoorn, Bakermans-Kranenburg and Duschinsky2020).

But alternative family-based care is not always substantially more stable because of the high risk of breakdown of the child–foster carer relationship (Eltink et al., Reference Eltink, Waaijenberg, Broers, van Anrooij, van Rooij, Stams and Assink2025) and maybe to a lesser extent breakdown of the relationship of the adopted child to the adopting parents (Bornstein & Suwalsky, Reference Bornstein and Suwalsky2021; Palacios et al., Reference Palacios, Adroher and Brodzinsky2019). During and after times of war or armed conflicts, children might become separated from their parents and other caregivers due to loss through death or to seeking refuge from war or armed conflict with part of the family. Displaced families may be at greater risk of falling apart due to stress and burnout and long-term living apart triggering alienation and divorce between parents with very different war-related experiences. Families living through war or armed conflict, as well as alternative care families in case no other viable options are available, need parenting support that is focused on increasing the stability of the care arrangement. Some of the parenting intervention programs discussed in this section might through their focus on effective interaction or reflective functioning, also elevate the chance that parents stick together even under dire wartime conditions or in the aftermath of war.

Importantly, policy and jurisdiction should only, in exceptional cases, decide to separate children from their parents or other caregivers such as grandparents, foster or adoptive parents. We argued elsewhere that family courts should refrain as much as possible from disrupting existing child–caregiver relationships, even when they may not be in the “best interest” of the child but provide good-enough care. In case of unavoidable divorce, family courts might prioritize verdicts that stimulate access to both parents for the child even though the time investment and shared activities with one of the parents seem to point at much less value of that relationship to the child. As long as safety can be guaranteed and the parents or caregivers are open to parenting support, existing attachment relationships should get a chance to be continued and to show improvement (Forslund et al., Reference Forslund, Granqvist and Van IJzendoorn2022; Van IJzendoorn et al., Reference Van IJzendoorn, Bakermans-Kranenburg and Duschinsky2020). In our study of Ukrainian families compared to institutional settings, we found evidence for better developmental prospects for children in families even when parents were addicted and their children were born with HIV infection, despite better medical care in institutional settings (Dobrova-Krol et al., Reference Dobrova-Krol, Bakermans-Kranenburg, Van IJzendoorn and Juffer2010a, Reference Dobrova-Krol, Van IJzendoorn, Bakermans-Kranenburg and Juffer2010b).

10.3.3 Shared Care

Third, shared care for developing offspring seems evolutionarily ingrained and historically and culturally predominant in humans (Bakermans-Kranenburg, Reference Bakermans-Kranenburg2021). The idea of “monotropy” is wrong if it is understood to imply that the care of children would be preferably provided by one exclusive caregiver, mostly assumed to be the biological mother (see Section 2). This misconception has long been standing in the way of the study of attachment networks that are found to exist in the large majority of 186 nonindustrial societies (Weisner & Gallimore, Reference Weisner and Gallimore1977). “Allomothers” are badly needed to support the mother in providing the estimated 13 million calories to bring a newborn to nutritional independence (Kaplan, Reference Kaplan1994), therefore “cooperative breeding” has been the rule instead of the exception since ancient times according to Hrdy (Reference Hrdy2009). Allomothers were indispensable when the risks of death during childbirth were high, which, for a long time in human evolution, was a dreadful collateral damage of reproduction. Alloparents also create opportunities for inexperienced parents-to-be such as older siblings, to practice parenting skills (Mooya et al., Reference Mooya, Sichimba and Bakermans-Kranenburg2016).

Grandparents can be an important part of the shared care arrangement. The Survey of Health, Ageing and Retirement in Europe (SHARE; Zanasi et al., Reference Zanasi, Arpino, Bordone and Hank2023) documented substantial grandparental care in 26 European countries, amounting to weekly care provision in 25% of the families on average. In China, a 2014 survey (Song et al., Reference Song, Wang and Qin2018) found that 60% of the elderly were taking care of their grandchildren, and half of them were providing care more than 9 hours a day. Grandmothers who were often quietly watching mothers’ parenting of their infants stimulated secure infant-mother attachment relationships, and fewer externalizing behavior problems later in development, presumably because they promoted feelings of parenting efficacy in the mothers (Liang et al., Reference Liang, Lin, Van IJzendoorn and Wang2021). When tensions emerge in the relationship between grandparent and parent, however, the grandparental influence might become negative, for the parent as well as the children (Liang et al., Reference Liang, Lin, Van IJzendoorn and Wang2021; Riem et al., Reference Riem, Bakermans-Kranenburg, Cima and Van IJzendoorn2023).

Grandparents might function as attachment figures and a major safe haven for their children – now parents – to cope with the heavy burden of child-rearing in an unsafe and unpredictable context. We found that support from grandparents was related to better maternal mental health and more adequate caregiving practices during the COVID-19 lockdown in China (Guo et al., Reference Guo, De Carli, Lodder, Bakermans-Kranenburg and Riem2021; Riem et al., Reference Riem, Lodder and Guo2021). In our meta-analysis of eleven studies (N = 3,381 participants), we found that involved grandparents fulfilled a protective role against postpartum mental health problems of the mothers (Riem et al., Reference Riem, Bakermans-Kranenburg, Cima and Van IJzendoorn2023). In times of war or armed conflict, grandparents might play an even more important role than in “normal” times, in the network of attachment relationships around their grandchildren, as substitute parents and attachment figures when the parents themselves are unavailable because of death, active service, or physical or mental health problems including substance misuse and depression. Moreover, an active role of grandmothers in co-rearing their grandchildren seems to decrease the grandparents’ depressive symptoms (Yang et al., Reference Yang, Evandrou and Vlachantoni2022). However, complete responsibility for grandchildren might lead to exhaustion and burnout (Chemerys et al., Reference Chemerys, Tsyhanyk, Bakermans-Kranenburg, van IJzendoorn and Nehrych2025, see Section 7). Because of their pivotal role as safe haven or as substitutes for parents, public policy measures and (preventive) interventions might prioritize support of grandparent-headed households during and after war or armed conflicts.

10.4.1 Incomplete Evidence-Base for Professionals working with Children during or after War

There is a growing need for child and family support and treatment, but pharmacological interventions, neurological treatments, and several popular psychosocial and behavioral interventions are still lacking a robust body of randomized controlled trial evidence supporting their efficacy for war-related symptoms in children and their parents. Despite this lack of evidence, medications are frequently prescribed off-label, often in complex polypharmaceutical regimens, and neurofeedback is increasingly promoted as a “cutting-edge” intervention. The same goes for psychosocial and behavioral interventions for which a suite of trainings and certifications has been implemented (e.g., ADAPT; Gewirtz, Reference Gewirtz2018). This pattern of practice risks normalizing the use of interventions with questionable cost–benefit profiles, especially in vulnerable or poorly studied groups such as those with war-related exposures. In the absence of strong RCT evidence, routine prescription and implementation should be approached with caution, prioritizing transparent communication with families, rigorous monitoring of adverse effects, and, where possible, participation in carefully designed clinical trials rather than widespread, uncritical use in routine care.

10.4.2 Pressures on Professionals Working with Children and Their Families

Based on attachment research, we suggest that diagnostics, interventions, and therapies should create a safe haven and secure base for the participants’ exploration of mental or physical suffering (Van IJzendoorn et al., Reference Van IJzendoorn, Bakermans-Kranenburg, Sak and Chaban2025). This is not a call for regression to an earlier stage of development but is based on the well-researched assumption that attachment is lifelong. Attachment continues to play an important role “from cradle to grave,” especially in frightening or stressful situations involving (mental) health issues, and not only with one’s own parents or partners but also with social workers, psychologists, or medical professionals in treatment trajectories (Bowlby, Reference Bowlby1988; Byng-Hall, Reference Byng-Hall1995).

It is impossible for health professionals to avoid forming a supportive relationship with clients, precisely because, according to Bowlby (Reference Bowlby1988) in distress humans tend to cling to any straw of possible reassurance and trust. Attachment is probably a generic dimension in any intervention or treatment process. A bond develops between the therapist or intervener and children, parents or clients, whether they want it or not. The bond may happen even in rather short-term interventions and therapies (Dozier & Bernard, Reference Dozier and Bernard2019; Juffer et al., Reference Juffer, Bakermans-Kranenburg and Van IJzendoorn2017). Therefore, it is better to acknowledge this relational dimension and consciously apply it in communication with the children, parents or patients, in the service of the child (see Figure 7). Insight into the emergence and development of attachment can be helpful here, including reflection on the professionals’ own biography. At least, that seems to be the implicit message of British brain surgeon Henry Marsh (Reference Marsh2022) in And Finally, his fascinating autobiography reflecting his more than thirty years volunteering in Ukraine.