Trauma and mental health burden of Gaza’s displaced children during war: a cross-sectional study

Abstract

Background

The ongoing war in Gaza since 2023 has caused unprecedented trauma and widespread displacement, with nearly 200,000 casualties and over 90% of the population displaced into crowded shelters and makeshift tents. Evidence on the mental health effects on children during this crisis remains scarce.

Methods

A cross-sectional study was conducted in May 2025 among displaced children aged 3–12 years living in shelters and tented communities across the inhabited areas of the Gaza Strip. The study used a multistage purposive and convenience sampling strategy. Caregivers filled out questionnaires assessing sociodemographic characteristics, war exposures, forced migrations, and caregiver PTSD, as well as the caregiver-reported versions of the Child and Adolescent Trauma Screen (CATS) for PTSD symptoms and the Strengths and Difficulties Questionnaire (SDQ) for psychosocial functioning. Multivariable logistic regression analyses were conducted to examine associations between sociodemographic factors, war exposures, and child PTSD.

Results

The final sample included 933 children (50.4% boys, 49.6% girls), with an average age of 7.6 years (SD = 2.8). Caregivers reported that children experienced an average of 6.7 forced displacements and 6.6 out of 10 surveyed war-related exposures. Nearly all experienced hunger (98%) and house destruction (95%). Based on age-appropriate CATS cutoffs, 57.8% of children met criteria for probable PTSD. SDQ results showed high rates of emotional and behavioral problems, with over 46.3% classified as abnormal on the Total Difficulties score. Boys were significantly more likely than girls to have experienced the loss of a close family member (57.0% vs. 49.9%, p = 0.035), although no other significant sex differences were found in the average number of traumas experienced or in the rates of other surveyed war exposures. Additionally, boys were significantly more likely to have more conduct problems and to score lower on prosocial behaviors (p < 0.001). In regression analyses, greater psychosocial difficulties (SDQ; OR = 1.18, 95% CI: 1.14–1.22), higher trauma exposure (OR = 1.17, 95% CI: 1.09–1.26), and more severe parental PTSD symptoms (OR = 1.26, 95% CI: 1.12–1.41) were consistently associated with increased odds of child PTSD. Non-marital parental status (OR = 1.62, 95% CI: 1.04–2.51) and non-maternal caregiving (OR = 2.61, 95% CI: 1.40–4.88) were also associated with higher odds. In the backward elimination model, older age showed a modest protective effect (OR = 0.95, 95% CI: 0.90–1.00), while female sex was associated with higher odds of PTSD (OR = 1.36, 95% CI: 1.02–1.82).

Conclusions

Displaced children in Gaza are experiencing extraordinarily high levels of trauma exposure, PTSD symptoms, and psychosocial difficulties. The findings highlight both the acute and generational mental health burden of war on children and underscore the urgent need for scalable, context-sensitive psychosocial interventions.

Background

Armed conflict and forced displacement are among the most impactful experiences that can threaten children’s psychological well-being and development. The Gaza Strip is home to approximately 2.2 million people, nearly half of whom are children under 18 years of age, making it one of the youngest populations globally [1]. In contexts of war, children are often exposed to a constellation of traumatic experiences, including bombardment, home destruction, forced migration, loss of relatives, chronic insecurity, and the breakdown of family cohesion and stability [2–4]. Furthermore, children not only directly experience violence but also inherit the psychological burdens carried by their families and communities, compounded by limited access to mental health support—or the complete absence of it [2, 5].

Evidence from earlier studies underscores the high prevalence of trauma-related psychopathology in Gaza’s children. A systematic review and meta-analysis in 2021 estimated a pooled post-traumatic stress disorder (PTSD) prevalence of 36% among Palestinian children exposed to political violence [2]. Other longitudinal research highlights that prolonged and repeated exposure to war-related trauma predicts not only PTSD but also comorbid anxiety, depression, and impaired social functioning [5]. While these findings provide important insights, they were conducted in periods before the unprecedented and protracted escalation of hostilities since October 2023.

Studies from other contemporary conflict settings similarly highlight the profound psychological impact of ongoing war on children. Recent studies among war-affected children in Ukraine and Sudan have reported high rates of PTSD symptoms, anxiety, and functional impairment, particularly among those exposed to displacement, family separation, and repeated traumatic events [6–8]. Across conflict contexts, cumulative trauma exposure, caregiver psychological distress, and family disruption consistently emerge as key predictors of child PTSD, underscoring the role of both direct exposure and the caregiving environment in shaping children’s mental health under conditions of chronic insecurity [9].

The current war in Gaza has generated a humanitarian crisis of unparalleled magnitude. By mid-2025, over 62,000 Palestinians have been killed, more than twice that number wounded, and the entire population lives under conditions characterized by poor sanitation, inadequate nutrition, lack of education, and severely limited access to healthcare, creating an environment of extreme vulnerability [10–13]. Humanitarian and UN organizations estimate that more than 90% of the population across the Gaza Strip has been displaced during the war, often repeatedly. Moreover, an estimated 600,000 were displaced between March and May 2025 alone as Israeli forces resumed ground operations after a temporary six-week pause in fighting, including over 200,000 displaced in the second half of May alone [14, 15]. The majority of the displaced population resided in shelters and temporary tented communities set up in schools, public buildings, and open fields that had been converted into emergency accommodations [16]. The living conditions were marked by limited privacy and inadequate access to clean water, sanitation, and electricity. Furthermore, tented communities were especially vulnerable to harsh weather conditions and lacked reliable infrastructure [17].

Recent wartime surveys demonstrated the magnitude of the mental health crisis in Gaza. In a 2024 cross-sectional study of Gaza medical students, 97% reported at least mild depressive symptoms, 84% anxiety, 91% stress, and 63% met PTSD criteria [18]. Similarly, a November 2024 study of adult internally displaced persons (IDPs) documented very high rates of anxiety, depression, and PTSD: moderate or higher levels of anxiety were reported by 79.3%, and depression by 84.5%, while the rate of symptomatic PTSD was 67.8%. Notably, 63.1% of participants experienced significant symptoms of all three conditions [17]. Displacement-related stressors such as injury, home destruction, bereavement, and living in tents or shelters were strongly linked to poorer mental health outcomes. Together, these findings highlight the unprecedented psychological toll of the current escalation and emphasize the urgent need for systematic evidence on the mental health of Gaza’s children.

Yet, despite the scale of the crisis, systematic data on the mental health of Gaza’s children during the current war remains limited. Existing studies have predominantly focused on adults, leaving younger children—arguably the most vulnerable—underrepresented in research and policy response. Moreover, while prior studies often examined PTSD in isolation, less is known about the co-occurrence of PTSD with broader psychosocial difficulties, such as conduct, emotional, and peer-related problems, that can compromise daily functioning and long-term adjustment.

The primary aim of this study was to estimate the prevalence and severity of PTSD symptoms among displaced children living in shelters and tents in Gaza in May 2025. The secondary aims were to explore sociodemographic and trauma-related factors linked to PTSD among war-exposed children, and to evaluate their emotional and behavioral functioning along with overall psychosocial difficulties.

Methods

Design

This cross-sectional study was conducted in May 2025 using a multistage purposive and convenience sampling approach. Eligible participants included children aged 3–12 years living in displacement shelters or tents. This age range was selected to focus on early and middle childhood, developmental stages characterized by distinct emotional, behavioral, and trauma-related profiles that can be reliably assessed using caregiver-reported instruments. Adolescence was not included because psychological responses, symptom expression, and appropriate measurement tools differ substantially at that developmental stage. The child’s primary caregiver must be a parent or an individual who has served as the child’s primary caregiver for at least three months, be able to provide informed consent, and complete the questionnaires. Children outside the eligible age range, those without an eligible caregiver, cases in which the caregiver was unable to complete the survey reliably, and questionnaires with missing outcome data were excluded.

Sampling procedure and sample size

In May 2025, nearly 82% of the Gaza Strip’s area was included in Israeli military evacuation orders and was undergoing ground offensives [11, 15]. To ensure representativeness, data were collected in displacement shelters and tented communities across the inhabited areas of the Gaza Strip during that month. A multistage purposive and convenience sampling strategy was used.

In the first stage, purposive sampling ensured geographic representation of the inhabited areas, covering three out of the Strip’s five governorates: the Gaza, Middle, and Khan Younis governorates. The North Gaza and Rafah governorates, along with the eastern neighborhoods of Gaza City and Khan Younis, were not included due to large-scale evacuations and restricted access enforced by the ongoing Israeli ground offensive. Within each included governorate, shelters and displacement camps were purposefully selected to cover all populated neighborhoods, and a similar strategy was applied in each neighborhood based on the team’s knowledge and assessment of that area. For example, in the Gaza Governorate, recruitment took place in the Sheik Radwan and Al-Nasser neighborhoods (northwest and west Gaza City), Aldaraj (central Gaza City), and Al-Remal and Tel Al-Hawa, and Al-Sabra (west and southwest Gaza City). In the Middle Governorate, participants were recruited from the Al-Nuseirat and Al-Maghazi refugee camps, the Al-Zawaida town, and the Deir Al-Balah city. In Khan Younis, recruitment occurred in shelters and camps located in the western parts of the city up to the Israeli-designated “safe zone” in Al-Mawasi. In each neighborhood, between three and four shelters or camps were selected. Shelters based in UNRWA schools were excluded because obtaining administrative approval from the agency proved time-consuming. However, living conditions were similarly harsh across all shelters, making this detail unlikely to predispose significant bias. Administration at all approached shelters/communities granted permission to approach its residents.

In the second stage, convenience sampling was used within each shelter or camp to recruit participants. Approximately 25 children were recruited from each site. To improve representativeness, a sampling sheet guided recruitment to ensure that the sample reflected the age- and sex-distribution of children in the Gaza Strip, as defined by the official population pyramid. Overall, participants were recruited from 31 different shelters and camps.

The minimum required sample size was estimated at 384 participants, based on a 95% confidence level, a 5% margin of error, and the assumption of an infinite population. To increase the precision of prevalence estimates and enable meaningful subgroup analyses, the target sample size was pragmatically increased to 900 children.

Measures

A structured questionnaire was used in the study, consisting of three main sections. The first section gathered sociodemographic information, including the child’s age, sex, parental marital status, caregiver’s relationship to the child, caregiver’s education level, parental occupation, household income, and both place of origin and current residence. Additional items assessed household composition and the number of forced displacements.

Traumatic exposure was assessed using a context-adapted checklist derived from Gaza- and Palestine-specific trauma instruments developed by the Gaza Community Mental Health Programme during the First and Second Intifadas, as well as from more recent work by Veronese et al. [21], which have been widely used and validated in Palestinian child populations [19–21]. Surveyed exposures included house destruction, passing through checkpoints, being directly targeted by gunfire, rescue from rubble or a bombed house, injuries, hunger, the loss of close relatives, and witnessing dead or wounded individuals.

Finally, caregivers completed the Primary Care PTSD Screen for DSM-5 (PC-PTSD-5) [22]. This 5-item questionnaire requires yes/no responses and yields a total score ranging from 0 to 5. A score of 3 or higher indicates a positive screening for probable PTSD.

The second section included the caregiver-reported version of the Child and Adolescent Trauma Screen (CATS), which evaluates child PTSD based on DSM-5 criteria [23]. It comprises 20 symptom items, each rated on a 4-point Likert scale (0 = “never” to 3 = “almost always”), yielding a total score ranging from 0 to 60. Probable PTSD was identified using two methods [24]:

1. Cutoff method: Children aged ≤ 6 years were considered positive if their scores were ≥ 16; those aged ≥ 7 years were considered positive if their scores were ≥ 21.

2. DSM-5 algorithm: A symptom was counted if rated ≥ 2. Diagnostic criteria required at least one intrusion symptom (items 1–5), at least one avoidance symptom (items 6–7), at least two negative mood or cognition symptoms (8–14), and at least two arousal symptoms (15–20), along with endorsement of at least one functional impairment domain.

The third section evaluated the children’s broader psychosocial functioning using the caregiver-reported version of the 25-item parent-report version of the Strengths and Difficulties Questionnaire (SDQ) for ages 4–17 [25, 26]. Items are rated on a 3-point scale (0 = “not true,” 1 = “somewhat true,” 2 = “certainly true”). The SDQ produces five subscales: emotional symptoms, conduct problems, hyperactivity/inattention, peer problems, and prosocial behavior (5 items each). A Total Difficulties Score is calculated by summing all subscales except prosocial behavior (range 0–40). For categorical interpretation, Goodman’s UK parent-report cutoffs were used [27]:

• Emotional symptoms: normal 0–3, borderline 4, abnormal 5–10.

• Conduct problems: normal 0–2, borderline 3, abnormal 4–10.

• Hyperactivity: normal 0–5, borderline 6, abnormal 7–10.

• Peer problems: normal 0–2, borderline 3, abnormal 4–10.

• Prosocial behavior: normal 6–10, borderline 5, abnormal 0–4.

• Total difficulties: normal 0–13, borderline 14–16, abnormal 17–40.

Questionnaires were administered in Arabic by interviewers to ensure clarity and to overcome literacy barriers, with caregivers responding on behalf of their children. Data were collected electronically using the Zoho application, enabling secure, real-time entry and response monitoring. The average interview duration was about 20 min.

Ethical considerations

The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Approval was obtained from the IRB at the Islamic University of Gaza prior to data collection. Written informed consent was obtained from all participating caregivers, who were assured that participation was voluntary and that they could withdraw at any time without penalty. To protect confidentiality, all identifying information was removed, and data were stored securely with access restricted to the research team. To ensure methodological rigor and cultural sensitivity, data collectors underwent intensive training led by a specialized professor in child and adolescent mental health and a psychiatrist with two board certifications. A comprehensive training package was developed in three versions: the original English, an officially translated Arabic version, and an adapted version using the local Gazan dialect. Trainees were specifically instructed to employ the Gazan dialect when interacting with participants to maximize clarity, rapport, and cultural appropriateness. Given the study’s focus on displaced children exposed to war-related trauma, participants identified as experiencing acute distress were referred to psychosocial support services coordinated by humanitarian agencies active in Gaza.

Statistical analysis

Data were analyzed using IBM SPSS Statistics version 28 (IBM Corp., Armonk, NY, USA). Descriptive statistics were used to summarize sociodemographic variables, trauma exposures, and outcome measures (PC-PTSD-5, CATS, and SDQ). Group differences between boys and girls were examined using chi-square tests for categorical variables and independent-samples t tests for continuous variables. Prevalence estimates of probable PTSD were calculated using both CATS cutoff scores and DSM-5 diagnostic algorithms with impairment criteria. Agreement between these methods was assessed using Cohen’s kappa statistic. To identify factors associated with probable PTSD, stepwise logistic regression analyses were conducted using both forward and backward likelihood ratio approaches. Predictor variables included child age, sex, total number of traumatic events, caregiver type, parental marital status, caregiver PTSD symptoms, and SDQ total difficulties scores. Model performance was evaluated using the omnibus chi-square test, Nagelkerke R², − 2 log likelihood, classification accuracy, sensitivity, and specificity. Statistical significance was set at p < 0.05, and all tests were two-tailed.

Results

Sociodemographic information and war trauma exposures

A total of 933 children were included, with an overall mean age of 7.66 years (SD = 2.79; range = 3–12). The sample was almost evenly split by sex, with 50.6% boys (n = 472) and 49.4% girls (n = 461) (see Table 1). More than half (58.2%) lived in tents, while 41.8% were displaced in shelters. Regarding family context, most children (82.9%) had married parents, whereas 2.4% had a deceased mother and 10.3% had a deceased father. A small proportion (0.8%) had lost both parents. Most children were primarily cared for by their mothers (90.3%), followed by fathers (3.8%), grandparents (3.0%), and other relatives (2.9%). Caregivers’ education, parental occupation, and income status are further illustrated in Table 1.

Table 1.

Sociodemographic information of the cohort

Variable	Category	N	%
Age (years)	Total cohort: Mean = 7.66 years, SD = 2.79, Range = 3–12	933	100
	Boys: Mean = 7.57 years, SD = 2.80	472	50.6
	Girls: Mean = 7.74 years, SD = 2.78	461	49.4
Parents’ marital status	Married	773	82.9
	Divorced	28	3.0
	Mother deceased	22	2.4
	Father deceased	96	10.3
	Both parents deceased	7	0.8
	Missing information	7	0.8
Primary caregiver	Mother	843	90.3
	Father	35	3.8
	Grandparent	28	3.0
	Other relative	27	2.9
Primary caregiver’s education	No formal education	6	0.6
	Primary schooling	40	4.3
	Secondary schooling	101	10.8
	High school	352	37.7
	Diploma/Bachelor	405	43.4
	Postgraduate studies	29	3.1
Father’s Occupation	Not working	388	41.6
	Self-employment	173	18.5
	Government employee	206	22.1
	Non-governmental employee	74	7.9
	Father deceased/missing info	92	9.9
Mother’s Occupation	Not working	757	81.1
	Self-employment	21	2.3
	Government employee	61	6.5
	Non-governmental employee	59	6.3
	Mother deceased/missing info	35	3.8
Residence	Tent	543	58.2
	Shelter	390	41.8
Income	0–800 NIS*	653	70.0
	801–1600 NIS	192	20.6
	1601–3200 NIS	61	6.5
	> 3200 NIS	27	2.9
Original Governorate	North	252	27.0
	Gaza	287	30.8
	Middle	142	15.2
	Khan Younis	150	16.1
	Rafah	102	10.9
Current Governorate	North	6	0.6
	Gaza	419	44.9
	Middle	287	30.8
	Khan Younis	221	23.7

*New Israeli Shekels (NIS; 1 NIS ≈ 0.27 USD at the time of the study)

Finally, displacement patterns were evident in the sample. Originally, families were distributed across the northern (27.0%), Gaza (30.8%), middle (15.2%), Khan Younis (16.1%), and Rafah (10.9%) governorates. At the time of assessment, however, nearly half (44.9%) were residing in Gaza governorate, 30.8% in the middle region, 23.7% in Khan Younis, and less than 1% in the North.

Children in the cohort experienced extremely high levels of displacement and war-related trauma (see Table 2). On average, children experienced 6.73 forced migrations until the study period (SD = 5.40; median = 5, IQR = 4–8). Nearly all children (95.3%) had their homes destroyed totally or partially, and 98.4% were experiencing hunger. Direct personal threats were also common: 42.2% were directly shot at, 15.1% had been evacuated from under rubble or from bombed buildings, and 17.6% had sustained wounds. Visual exposure to violence was also common, with 68.3% having seen dead bodies and 80.6% having witnessed wounded persons. Additionally, 32.0% had been forced to pass through military checkpoints. The total number of war experiences averaged 6.63 events (median = 7). When comparing by sex, boys were significantly more likely than girls to report the loss of a close family member (57.0% vs. 49.9%, p = 0.035). All other war exposure variables, including the number of forced migrations as well as the individual and total trauma experiences, did not differ significantly between boys and girls.

Table 2.

Trauma exposure and war impacts on participants

Variable	Category/value	N	%
Forced migrations	Total cohort: Mean = 6.73 forced migrations, SD = 5.40, Median = 5, IQR = 4–8, Range = 1–65
	Boys: Mean = 6.72 forced migrations, SD = 4.98, Median = 5, IQR = 4–8
	Girls: Mean = 6.74 forced migrations, SD = 5.81, Median = 5, IQR = 4–8
Traumatic war experiences	House destruction	946	95.3
	Passing through military checkpoint	318	32.0
	Shot at directly	419	42.2
	Evacuated from under rubble/bombed building	150	15.1
	Wounded	175	17.6
	Hunger	918	98.4
	Lost a close family member*	293	29.5
	Lost other relatives	723	72.9
	Saw dead bodies	678	68.3
	Saw wounded persons	800	80.6
Total traumatic war experiences	Total cohort: Mean = 6.63/10 surveyed experiences, Median = 7
	Boys: Mean = 6.76/10 surveyed experiences, Median = 7
	Girls: Mean = 6.50/10 surveyed experiences, Median = 7
Caregiver PTSD (PC-PTSD-5)	Total score: Mean = 3.55, Median = 4, Range = 0–5
	Positive screen (score ≥ 3)	774/992	78.1
	Mothers	700/897	78.0
	Fathers	28/35	80.0
	Grandparents	28/31	90.3
	Other relatives	18/29	62.1

*Indicates a significant sex difference

On the PC-PTSD-5, caregivers scored an average of 3.55 (median = 4; range = 0–5). More than three-quarters (78.1%) screened positive for probable PTSD (≥ 3). Prevalence rates varied by caregiver type: 78.0% among mothers, 80.0% among fathers, 90.3% among grandparents, and 62.1% among other relatives. Table 2 summarizes the exposure rates to the ten surveyed war traumas among participating children as reported by their primary caregivers.

3.2. Child PTSD prevalence and sex differences

Based on the CATS results, 57.8% of children met the cutoff for probable PTSD. When disaggregated by sex, 55.7% of boys and 59.9% of girls screened positive, χ²(1, N = 933) = 1.48, p = 0.224. Using the more stringent DSM-5 diagnostic algorithm combined with functional impairment, 15.6% of children met criteria for PTSD. This prevalence among boys was 16.7%, and among girls 14.5%, χ²(1, N = 933) = 0.70, p = 0.403.

Comparison of the two scoring methods revealed only fair agreement (Cohen’s κ = 0.24). Specifically, while no children with scores below the cutoff met DSM-5 PTSD criteria, 146 children (15.6%) who exceeded the cutoff also met full DSM-5 algorithm criteria, whereas 393 (42.1%) exceeded the cutoff but did not fulfill full DSM-5 diagnostic requirements. Table 3 presents the prevalence estimates and sex differences in PTSD screening outcomes.

Table 3.

Prevalence of positive PTSD screening across CATS scoring criteria, by sex

Criterion	Total positive (%)	Boys positive (%)		Girls positive (%)		p-value
CATS score ≥ cutoff	57.8	55.7		59.9		0.2237
DSM-5 Algorithm + Impairment	15.6	16.7		14.5		0.4031

Agreement between scoring methods (age-appropriate CATS cutoff score vs. DSM-5 algorithm)*
	DSM-5 negative		DSM-5 positive		Total
Cutoff negative	394		0		394
Cutoff positive	393		146		539
Total	787		146		933

*Cohen’s κ = 0.24 → Fair agreement

Factors associated with child PTSD

Results from both forward and backward stepwise regression analyses converged on similarly strong final models with significant improvements in fit and comparable classification performance (Table 4). In both solutions (see Tables 5 and 6), greater overall difficulties, higher trauma exposure, and more severe parental PTSD symptoms were linked to increased odds of developing PTSD. Additionally, non-married parental status and non-maternal caregiving were associated with higher odds of child PTSD. In the Backward solution, older age showed a small inverse association (modestly protective), and female sex was linked to higher odds of PTSD. Overall, the models explained approximately 19–26% of the variance (Nagelkerke R²) and improved classification accuracy to approximately 70%.

Table 4.

Model fit and classification performance

Model (final step)	χ² (df)	−2LL	Cox–Snell R²	Nagelkerke R²	Accuracy (%)	Sensitivity (%)	Specificity (%)
Forward LR (Step 5)	187.73 (5)*	1081.41	0.184	0.246	69.9	77.4	60.2
Backward LR (Step 4)	195.50 (7)*	1073.64	0.190	0.255	70.7	78.9	60.2

Statistics are model-level; effect estimates appear in Tables 5 and 6

χ² = omnibus model test. R² = Nagelkerke

*p < 0.001

Table 5.

Forward stepwise logistic regression (final step 5): predictors of probable PTSD

Predictor	B	SE	Wald	OR	95% CI	p
Total difficulties	0.163	0.017	90.233	1.177	1.138–1.217	< 0.001
Total traumatic exposures (n)	0.153	0.036	18.197	1.165	1.086–1.251	< 0.001
Parental PTSD symptoms (PC-PTSD-5)	0.232	0.058	16.119	1.261	1.126–1.413	< 0.001
Parental marital status (others vs. married)	0.479	0.224	4.571	1.614	1.041–2.504	0.033
Caregiver type (others vs. mother)	0.954	0.318	9.002	2.596	1.392–4.842	0.003
Constant	- 4.771	0.456	109.484	–	–	< 0.001

Reference categories and coding are described in methods

OR: odds ratio; CI: 95% confidence interval

p values are two-tailed

Table 6.

Backward stepwise logistic regression (final step 4): predictors of probable PTSD

Predictor	B	SE	Wald	OR	95% CI	p
Age (years)	-0.053	0.027	3.919	0.948	0.899–1.000	0.048
Sex (female vs. male)	0.307	0.149	4.244	1.359	1.015–1.820	0.039
Total difficulties	0.168	0.017	92.377	1.183	1.144–1.223	< 0.001
Total traumatic exposures (n)	0.157	0.036	18.773	1.170	1.090–1.256	< 0.001
Parental PTSD symptoms (PC-PTSD-5)	0.229	0.058	15.731	1.257	1.122–1.409	< 0.001
Parental marital status (others vs. married)	0.481	0.224	4.608	1.618	1.043–2.509	0.032
Caregiver type (others vs. mother)	0.961	0.318	9.115	2.614	1.402–4.876	0.003
Constant	-4.627	0.501	85.151	–	–	< 0.001

Reference categories and coding are described in methods

OR: odds ratio; CI: 95% confidence interval

p values are two-tailed

Strengths and difficulties questionnaire results and sex-based differences

Analysis of the SDQ results revealed substantial psychosocial difficulties among participating children (see Tables 7 and 8). Using the Goodman (2009) updated banding cutoffs, 46.3% of children fell within the abnormal range for total difficulties, while only 29.6% were categorized as normal. Domain-level analysis revealed particularly high rates of emotional symptoms and peer-related problems: 43.9% of children were rated abnormal for emotional symptoms and 55.6% for peer problems. Similarly, 41.7% were considered abnormal for conduct problems. In contrast, hyperactivity had lower abnormal rates (17.5%), although nearly one in five children was still affected. Prosocial behavior, which is scored in reverse (lower scores indicate poorer functioning), showed that 45.9% of children were classified as abnormal, indicating significant impairments in social functioning. Mean scores across domains were highest for prosocial behavior (M = 5.90, SD = 2.39) and emotional symptoms (M = 5.15, SD = 2.21), with an overall mean of 19.16 (SD = 4.79) on the total difficulties scale.

Table 7.

Strengths and difficulties questionnaire (SDQ) domain classifications and mean scores among children

Domain	Normal (%)	Borderline (%)	Abnormal (%)	Overall mean (SD)
Emotional symptoms	39.0	17.0	43.9	5.15 (2.21)
Conduct Problems	33.4	24.9	41.7	4.29 (1.66)
Hyperactivity	64.7	17.8	17.5	4.98 (1.68)
Peer problems	25.6	18.8	55.6	4.73 (1.88)
Prosocial behavior	38.3	15.9	45.9	5.90 (2.39)
Total difficulties	29.6	24.1	46.3	19.16 (4.79)

Table 8.

Sex differences in strengths and difficulties questionnaire (SDQ) scores and classifications across domains

Domain	Boys normal/borderline/abnormal (%)	Boys mean (SD)	Girls normal/borderline/abnormal (%)	Girls mean (SD)	p (abnormal %)	p (mean)
Emotional symptoms	23.7%/16.1%/60.2%	5.03 (2.15)	20.6%/17.6%/61.8%	5.28 (2.27)	0.653	0.085
Conduct problems	9.3%/17.4%/73.3%	4.53 (1.61)	17.6%/22.8%/59.7%	4.06 (1.67)	< 0.001	< 0.001
Hyperactivity	64.8%/18.9%/16.3%	4.97 (1.63)	64.6%/16.7%/18.7%	4.98 (1.73)	0.392	0.880
Peer problems	10.0%/15.3%/74.8%	4.77 (1.81)	16.1%/10.0%/74.0%	4.69 (1.96)	0.833	0.555
Prosocial behavior	49.8%/19.7%/30.5%	5.64 (2.39)	58.6%/17.8%/23.6%	6.16 (2.35)	0.022	0.001
Total difficulties	11.7%/14.8%/73.5%	19.29 (4.60)	14.3%/18.4%/67.2%	19.02 (4.99)	0.043	0.378

Sex-based analyses (Table 8) indicated nuanced differences across domains. Boys demonstrated significantly higher rates of conduct problems compared to girls (73.3% vs. 59.7% abnormal; χ² = 12.45, p < 0.001), and their mean conduct scores were also elevated (M = 4.53, SD = 1.61 vs. M = 4.06, SD = 1.67, p < 0.001). Conversely, girls exhibited stronger prosocial behavior, with significantly fewer classified as abnormal (23.6% vs. 30.5% for boys; χ² = 5.25, p = 0.022) and higher mean scores (M = 6.16, SD = 2.35 vs. M = 5.64, SD = 2.39, p = 0.001). No statistically significant sex differences were observed for emotional symptoms, hyperactivity, or peer problems; however, girls had slightly higher mean scores for emotional symptoms. Figure 1 further demonstrates significantly higher rates of conduct problems and total difficulties among boys, as well as stronger prosocial behavior among girls.

Fig. 1.

Proportion of abnormal Strengths and Difficulties Questionnaire scores by sex

Discussion

This study presents timely evidence on the mental health of displaced children in Gaza during the ongoing war since 2023. The findings reveal extremely high levels of war-related trauma exposure, with nearly all children reporting home destruction (95.3%), hunger (98.4%), and multiple forced displacements (mean = 6.7). While the cumulative adversities align with previous global evidence showing that children in conflict zones are repeatedly exposed to overlapping traumatic events [2–4], the intensity and pervasiveness of exposures in this group surpass those reported in most conflict settings. For example, while studies among Syrian refugee children documented home destruction in about 40–60% of cases and forced displacement typically once or twice [28, 29], nearly all children in this study (95.3%) lost their homes and experienced an average of nearly seven forced migrations. Similarly, whereas studies of war-affected populations in Iraq and Lebanon often report hunger and food insecurity affecting one-third to one-half of children [30], almost every child in our sample (98.4%) reported hunger. Direct exposure to gunfire (42.2%), rescue from rubble (15.1%), and visual encounters with dead bodies (68.3%) or wounded individuals (80.6%) were also exceptionally high, indicating that extreme, repeated, and multi-layered trauma exposures are the norm rather than the exception for Gaza’s displaced children.

The prevalence of symptomatic PTSD in this study was remarkable: 57.8% of children screened positive using CATS cutoff scores, while 15.6% met the more stringent DSM-5 algorithm with impairment criteria. These findings are significantly higher than pooled estimates from previous Palestinian child studies, where the prevalence of PTSD was estimated at 36% [2], and they also surpass rates reported in Syrian refugee children (ranging 30–50%) [28, 29]. The discrepancy likely reflects the acute severity of the current war, marked by large-scale displacement, infrastructure destruction, and limited access to basic necessities.

Sex-based differences were nuanced. Girls exhibited slightly higher rates of PTSD by cutoff criteria, whereas boys showed higher conduct problems and overall difficulties on the SDQ. This aligns partially with prior findings in conflict-affected populations, where girls are often more prone to internalizing symptoms such as anxiety and PTSD, while boys exhibit more externalizing difficulties [31, 32].

The identified predictors of child PTSD in this study align with well-established risk mechanisms in conflict-affected areas, while also highlighting the exceptional severity of the current context in Gaza. Cumulative exposure to war-related adversities proved to be a strong predictor, consistent with evidence from other conflicts in the Middle East, where repeated and overlapping traumatic experiences have additive effects on children’s psychological functioning [32].

Caregiver PTSD symptoms were among the most robust predictors of child PTSD, echoing longitudinal evidence that parental mental health strongly influences children’s outcomes in war contexts [33, 34]. This finding aligns with models of intergenerational transmission of trauma, whereby caregiver distress impairs emotional availability, increases household stress, and limits effective buffering of children’s fear and insecurity [17]. In the context of widespread adult trauma in Gaza—where nearly four in five caregivers screened positive for probable PTSD—the capacity of families to provide psychological containment is profoundly compromised. Similarly, non-maternal caregiving and non-married parental status were associated with higher odds of child PTSD, suggesting that family disruption and loss of primary attachment figures further exacerbate risk under displacement.

Elevated psychosocial difficulties, as measured by the SDQ, independently predicted PTSD, indicating that emotional, behavioral, and peer-related problems are not merely comorbid outcomes but may also signal heightened susceptibility to trauma-related psychopathology. Although the per-unit effect sizes were modest, these associations reflect cumulative vulnerability across multiple domains of functioning. In a population with a high prevalence of psychosocial difficulties, even incremental increases in difficulty may translate into meaningful differences in risk at the population level.

Taken together, these predictors depict a pattern of compounded risk driven by ongoing trauma exposure, caregiver distress, and family instability, highlighting the importance of interventions that address both child symptoms and the broader family system within humanitarian response frameworks.

The SDQ findings indicate that nearly half of the children (46.3%) scored in the abnormal range for total difficulties, with particularly high rates of emotional (43.9%), conduct (41.7%), and peer problems (55.6%). These rates are significantly higher than international norms [25, 34] and exceed those reported in war-affected Syrian and Iraqi children [30]. Peer problems were especially pronounced in our cohort, reflecting the breakdown of social networks and the loss of stable schooling environments. Interestingly, while hyperactivity was less affected (17.5% abnormal), prosocial functioning was severely impaired (45.9% abnormal), suggesting that the social dimensions of war trauma may be especially salient for displaced children.

The divergence between the two CATS scoring methods (κ = 0.24) warrants further consideration. While the cutoff method identified a majority as symptomatic PTSD, only a fraction met DSM-5 algorithmic criteria with impairment. This discrepancy may reflect both the heightened sensitivity of cutoff scores and the complex manifestation of PTSD in younger children, whose distress may not fully map onto DSM-5 categories. Similar discrepancies have been reported in other pediatric trauma studies [23], underscoring the importance of using multiple approaches to avoid underestimating or overestimating burden.

Together, these findings reveal an alarming convergence of factors—extreme trauma exposure, high caregiver distress, and pervasive psychosocial dysfunction—that place Gaza’s children at profound risk of long-term impairment. They also highlight the urgent need for context-specific interventions that address both individual symptoms and family systems, while acknowledging the unique sociopolitical constraints of Gaza.

Evidence from conflict-affected settings suggests that trauma-focused interventions can be implemented even under conditions of chronic insecurity, as long as they are adapted for feasibility and scalability [35]. Programs such as Teaching Recovery Techniques (TRT), implemented in Gaza and other humanitarian settings [36], and trauma-focused cognitive behavioral therapy (TF-CBT), including recent adaptations in Ukraine [37], demonstrate that structured, evidence-based methods can reduce trauma-related symptoms when delivered in group settings or through task-shifting models. However, the exceptionally high burden of distress found in this study highlights the need for stepped-care strategies, where low-intensity screening and psychosocial support are widely deployed, with referral routes to more specialized trauma-focused treatment for children with severe or persistent symptoms [38].

The strong links between child PTSD, caregiver PTSD, and family disruption further suggest that interventions focused only on the child may be inadequate. Family-centered and caregiver-inclusive approaches are likely essential, especially where parental distress is common and caregiving capacity is compromised by displacement and material deprivation. Addressing the significant treatment gap in Gaza will therefore require investment not just in specialized mental health services but also in training frontline health workers, educators, and community providers in trauma-informed screening, basic psychosocial support, and referral processes. Incorporating child mental health into humanitarian response systems—alongside protection, shelter, and food security—may be essential to mitigate the long-term intergenerational consequences of extended war exposure.

Limitations

Several limitations should be acknowledged. First, the cross-sectional design of the study precludes conclusions about causality or the persistence and trajectories of psychological distress over time, highlighting the need for longitudinal follow-up. Second, the assessment relied on caregiver reports rather than child self-reports, which may have introduced reporting bias, particularly for internalizing symptoms such as PTSD. This approach was adopted due to the inclusion of young children and the logistical constraints of data collection in displacement settings; nevertheless, it may have limited the capture of children’s subjective experiences.

Third, the study focused on PTSD symptoms and psychosocial difficulties and did not assess other trauma-related disorders, such as depression or anxiety, which are also highly prevalent and clinically relevant in conflict-affected populations. Fourth, the use of screening instruments rather than clinical diagnostic interviews may have inflated prevalence estimates; however, the application of established cutoffs and algorithmic scoring enhances the robustness and comparability of findings.

Fifth, while measures were carefully adapted linguistically and culturally, subtle nuances in children’s symptom expression may not have been fully captured. Sixth, the war-exposure score represented the number of endorsed exposure types and did not weight exposures by severity or capture incident frequency, thereby reflecting exposure breadth rather than intensity. Finally, the use of convenience sampling within shelters and camps may have introduced selection bias, potentially limiting the generalizability of the findings.

Conclusion

This study provides robust, large-scale, and urgently needed evidence of the extraordinary mental health burden borne by Gaza’s displaced children during the war since 2023. With more than half screening positive for PTSD, and nearly half displaying severe psychosocial difficulties, the findings underscore the acute vulnerability of this population. These results have important implications for both immediate humanitarian response and the development of longer-term psychosocial support strategies. The strong associations between child PTSD and cumulative trauma exposure highlight the central importance of safety and protection as prerequisites for psychological recovery in conflict settings. Moreover, the links with caregiver PTSD symptoms and family disruption emphasize the interdependent nature of child and family well-being under conditions of forced displacement and chronic insecurity. Together, these findings call for urgent humanitarian action that integrates child mental health into relief efforts, alongside the implementation of sustainable, scalable, culturally sensitive, and family-centered psychosocial interventions. Without immediate and long-term responses, the cumulative psychological toll risks perpetuating cycles of trauma across generations.

Author contributions

BA, AA, and LAD: conceptualization and study design. All authors contributed to the literature review and tool development. AA and LAD supervised pre-data collection training. BA led sampling and data collection. All authors except the first and last authors contributed to the data collection. BA and LAD contributed to data analysis and interpretation, as well as to the writing of the manuscript. All authors approved the final version.

Funding

The study was not funded.

Data availability

Available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Ethical approval for this study was obtained from the Institutional Review Board (IRB) at the Islamic University of Gaza. Written informed consent was obtained from all participating caregivers, who were assured of the voluntary nature of participation and their right to withdraw at any time without penalty. The confidentiality of the data was maintained throughout the data collection and analysis process.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.