Determinants of Delayed Malaria Treatment-Seeking Among Under-Five Children in a Post-Conflict District of Northern Ethiopia: A Facility-Based Unmatched Case-Control Study Authors: Gebreanenia Gebretsadik Gebremamas, Fissha Brhane Mesele, Senait Alemayehu Abrha, Desalegn Massa Published: 2026 ABSTRACT BackgroundDelayed treatment seeking for malaria among under-five children remains a major public health problem in low-resource and post-conflict settings. However, there is limited evidence on its determinants in a post-conflict district. This study aimed to identify determinants of delayed malaria treatment seeking among under-five children in a post-conflict district of Northern Ethiopia, 2024.MethodsA facility-based unmatched case-control study was conducted in Samre District, Tigray, from March 18 to May 18, 2024. The study included 364 under-five children (182 cases and 182 controls) selected through systematic random sampling. Delayed malaria treatment-seeking was defined as seeking care more than 24 h after the onset of fever or malaria symptoms, whereas timely treatment-seeking was defined as seeking care within 24 h. Data were coded and entered into EpiData and analyzed using the Statistical Package for the Social Sciences version 27. Bivariable logistic regression was first used to select candidate variables at P < 0.2 for further analysis. In the multivariable analysis, variables with P < 0.05 were considered statistically significant. The Hosmer–Lemeshow test was used to assess model fit (P > 0.05).ResultsAmong the participants, delayed treatment-seeking was independently associated with caregiver’s lack of formal education (AOR = 2.7; 95% CI: 1.19–6.12), poor perception of malaria (AOR = 3.5; 95% CI: 1.71–7.37), self-medication (AOR = 8.3; 95% CI: 3.78–18.27), no previous history of malaria (AOR = 4.6; 95% CI: 2.34–8.94), and residence more than 5 km from a health facility (AOR = 7.8; 95% CI: 3.70–16.32).ConclusionDelayed malaria treatment-seeking was associated with caregiver educational level, malaria perception, self-medication, previous malaria experience, and geographic access to health facilities. In this post-conflict setting, strengthening caregiver education, reducing self-medication, and improving access to primary malaria services in remote communities should be prioritized. Introduction Malaria is a life-threatening parasitic disease caused by Plasmodium species and remains one of the world’s major public health challenges [1, 2]. Despite substantial progress in malaria prevention and control over the past two decades, the disease continues to disproportionately affect low- and middle-income countries, particularly those in sub-Saharan Africa. According to the World Malaria Report 2024, an estimated 263 million malaria cases and 597,000 deaths occurred globally in 2023, with the African Region accounting for 94% of cases and 95% of deaths. Children under five represented 76% of malaria deaths in the region [2, 3]. Prompt diagnosis and treatment within 24 h of fever onset are essential for preventing progression to severe malaria and reducing malaria-related mortality among children under five years of age [2]. However, delayed malaria treatment-seeking remains common in malaria-endemic countries, with only 26.1% of febrile children receiving prompt treatment and approximately 60% experiencing delays of more than 24 h [4, 5]. Delayed treatment substantially increases the risk of severe malaria, severe anemia, hospitalization, and death, particularly among young children [6, 7]. In Ethiopia, malaria remains a major public health concern despite earlier achievements in malaria control [8, 9]. Although reported malaria cases fell in 2021, evidence suggests this mainly reflected conflict-related disruption of surveillance and healthcare in Tigray rather than a real drop in transmission, with Samre district particularly affected by service breakdowns and increased malaria illness and deaths, especially among children under five [10]. The burden was particularly substantial in Samre District. During epidemiological weeks 19–51 of 2022, Samre reported 7,112 malaria cases, accounting for 16.6% of all regional cases, and 14 malaria-related deaths, representing 35% of regional deaths. In epidemiological weeks 1–42 of 2023, the district reported 5,220 cases and one malaria-related death, accounting for 2.34% and 6.25% of the respective regional totals. Of the 15 deaths reported across these periods, 74% occurred in the community, and 61% involved children under five years of age. These findings highlight the substantial malaria burden in Samre and the potential consequences of delayed access to diagnosis and treatment, particularly among young children [11]. Malaria treatment-seeking among caregivers in Ethiopia is influenced by socioeconomic, behavioral, and health-system factors, including limited knowledge of malaria, sociocultural beliefs, self-medication, treatment costs, and geographic distance from health facilities. In conflict-affected settings, these barriers may be further intensified by disruption of health services, damage to health facilities and infrastructure, shortages of healthcare workers, medicines and diagnostic supplies, displacement of communities, and disruption of transportation and referral systems. Such disruptions may increase travel time and costs, limit access to nearby functional health facilities, and encourage caregivers to rely on home remedies, self-medication, or other informal sources of care before seeking formal malaria treatment [12–17]. The Global Technical Strategy (GTS) for Malaria 2016–2030 aims to reduce malaria morbidity and mortality by 75% by 2025 and 90% by 2030 through universal access to prompt diagnosis, treatment, and vector control measures [18]. The Ethiopian Malaria Elimination Strategic Plan (EMESP) (2021–2025), together with the President’s Malaria Initiative (PMI) and National Malaria Strategic Plan (NMSP), prioritizes prompt diagnosis and effective treatment of malaria within 24 h of fever onset to reduce malaria morbidity and mortality and achieve malaria elimination [19, 20]. Ethiopia provides malaria diagnosis and treatment through a network of health posts, health centers, and hospitals [19–22]. However, progress toward the National Malaria Elimination Program in the Tigray Region has been severely hindered by the recent armed conflict, which damaged health facilities, disrupted healthcare services, displaced health workers, and caused shortages of antimalarial medicines and diagnostic supplies. These disruptions likely reduced timely access to malaria diagnosis and treatment, particularly among children under five years of age [23]. Most studies in Ethiopia explored the factors that affect the delayed malaria treatment seeking among under-five children in a Stable/non conflict settings [17, 24]. Although previous studies have examined delayed malaria treatment-seeking, most were conducted in relatively stable settings. The present study provides context-specific evidence from a post-conflict district of Tigray, where disruption of healthcare services, geographic access, and other conflict-related conditions may influence treatment-seeking behavior. The study therefore adds evidence that may not be captured by studies conducted in stable settings. Understanding the causes of delayed malaria treatment-seeking in post-conflict areas is critical for developing context-specific treatments to reestablish malaria services and improve health outcomes for under-five children. Therefore, the aim of the study is to assess determinants of delayed malaria treatment seeking among under-five children in a post-conflict district of Northern Ethiopia, 2024. Methods Study Setting and Period The study was conducted in Samre district, Tigray region. Samre is located in the South Eastern Zone of the Tigray Region in northern Ethiopia. It has a latitude and longitude of 13°11′N 39°12′E and an elevation of about 1,855 m above sea level. Samre town is the capital city of the Samre district, located 58 km south of Mekelle, the regional capital of Tigray region, and 787 km north of Addis Ababa, the capital city of Ethiopia. It shares borders with Selewa District to the east, Seharti District to the west, Hintalo District to the north, and the Amhara region to the south. The district receives an average annual rainfall of 400–700 millimeters, with a mean maximum annual temperature of 30 °C and a mean minimum annual temperature of 15 °C. The district has one primary hospital, two health centers, two private clinics and twelve health posts. During the study period, all selected health facilities were operational and providing malaria diagnosis and treatment services. However, healthcare service delivery had not yet fully recovered from the recent armed conflict, and some facilities continued to experience shortages of healthcare personnel and medical supplies. Despite these challenges, malaria diagnostic and treatment services remained available throughout the study period. Based on the 2007 national census conducted by the Central Statistical Agency of Ethiopia, the District is projected to have a total population of 88,604, with 43,416 men and 45,188 women and total under-five children 12,927 in 2023 [25]. In the context of Tigray, the post-conflict period was considered to have begun following the signing of the Pretoria Agreement on November 2, 2022, which established a permanent cessation of hostilities between the Ethiopian Federal Government and Tigrayan forces. The study was conducted during this post-conflict period, from March 18 to May 18, 2024 (Fig. 1).Fig. 1Map of health facilities in samre district, Tigray, Northern Ethiopia, 2024 Map of health facilities in samre district, Tigray, Northern Ethiopia, 2024 Study Design A facility-based unmatched case–control study (1:1 ratio) was conducted. An unmatched case-control design was used because the study aimed to evaluate multiple independent determinants of delayed malaria treatment-seeking. Matching was not considered necessary, as potential confounders were controlled using multivariable logistic regression analysis. Population Source Population The source population consisted of all under-five children with suspected malaria who sought care at the selected health facilities in Samre District during the study period and were confirmed to have Plasmodium infection by blood film examination. Study Population The study population consisted of sampled under-five children who met the eligibility criteria, were confirmed to have Plasmodium infection by blood film examination, and sought care at the selected health facilities during the study period. Their primary caregivers were interviewed as proxy respondents to obtain information on treatment-seeking behavior and related factors. Cases Cases were under-five children who had one or more signs and symptoms of malaria and positive for Plasmodium species through blood film examination and sought treatment after 24 h of developing sign and symptoms, particularly fever. Controls Controls were under-five children who had one or more signs and symptoms of malaria and positive for Plasmodium species through blood film examination and sought treatment within 24 h of developing sign and symptoms, particularly fever. Inclusion and Exclusion Criteria Inclusion Criteria Under-five children who were tested positive for Plasmodium species through Blood film and sought treatment at health facilities of Samre District were included in the study. Exclusion Criteria Under-five children whose caregivers or parents did not provide informed consent and children who had previously been enrolled in the study during an earlier visit to the health facility were excluded. Sample Size Determination The sample size was determined by double population proportion formula for case control study by using EpiInfo 7.0 stat calc. The calculation was computed by taking three variables; fear of side effect, cost of transportation and history of child death which were significantly associated with delayed malaria treatment seeking in the studies [17, 24]. The final sample size was calculated by using the assumption of “fear of side effect” with the Odds ratio of 2.07, percent of controls exposed 21.1%, percent of cases exposed 35.76% with 95% confidence interval, power being 80%, case to control ratio of 1:1. The study’s initial sample size was 328. After considering 10% non-response rate, the final sample size became 364, consisting of 182 cases and 182 controls (Table 1).Table 1Sample size determination for samre district, Tigray, Northern Ethiopia, 2024VariablesCIPower% of controls exposed% of cases exposedCases to control ratioORTotal sample sizeFear of side effect(yes)95%80%21.135.7612.07328 [19].Cost of transportation(Expensive)95%80%10.319.712.53320 [12].History of child death(No)95%80%77.790.1912.63304 [20].N.B. Total sample size =328*10%=364(with 182 cases and 182 controls) by adding 10% of non-response rate. Table 1Sample size determination for samre district, Tigray, Northern Ethiopia, 2024VariablesCIPower% of controls exposed% of cases exposedCases to control ratioORTotal sample sizeFear of side effect(yes)95%80%21.135.7612.07328 [19].Cost of transportation(Expensive)95%80%10.319.712.53320 [12].History of child death(No)95%80%77.790.1912.63304 [20].N.B. Total sample size =328*10%=364(with 182 cases and 182 controls) by adding 10% of non-response rate. Sample size determination for samre district, Tigray, Northern Ethiopia, 2024 N.B. Total sample size =328*10%=364(with 182 cases and 182 controls) by adding 10% of non-response rate. Sampling Technique and Procedure The district was selected purposively based on malaria endemicity and its elevated occurrences of malaria morbidity and mortality rates in 2022 [26]. The study area encompasses a total of five health facilities, comprising of three public health facilities and two private clinics. All healthcare facilities were included in the study. The overall sample size was proportionally allocated to each health facility based on the monthly caseloads of children under five from the previous year (March-April 2023). Study subjects were selected using systematic random sampling technique, with the sampling interval [k] calculated for each health facilities based on the average monthly caseloads of the health facilities (Fig. 2).Fig. 2Schematic presentation of the sampling procedures in samre district, Tigray, Northern Ethiopia, 2024 Schematic presentation of the sampling procedures in samre district, Tigray, Northern Ethiopia, 2024 Study Variables Dependent Variable Delayed Malaria Treatment Seeking among under-five children. Independent Variables Socio-demographic variables Age of child, sex of child, age of caregiver, head of household, family size, number of u-5 children within the household, residence, marital status, religion, ethnicity, educational status and occupational status. Behavioral variables Knowledge, perception, history child death, previous history of malaria in the household, complain of treatment side effects, history of first place to seek treatment, history of taking traditional medicine and history of practicing self-medication. Accessibility variables Presence of a road from the village/kebele to the health facility, access to transportation, mode of transportation, travel time to the health facility, distance to the health facility, and transportation cost. Health-system variables Women Development Army membership, membership in free-waiver/exempted services, and previous waiting time at health facilities. Operational Definitions and Measurements Malaria treatment-seeking was defined as caregiver obtained expert opinion or treatment for malaria from health care provider when the child is sick [27, 28]. Timely treatment seeking defined as seeking care within 24 h of the onset of malaria symptoms, particularly fever [27, 28]. Delayed treatment seeking defined as seeking care more than 24 h after the onset of symptoms [27, 28]. Caregiver was defined as any person primarily responsible for the care of the child, including parents or other family members [17]. Self-medication refers to the practice of treating a child at home without consulting a healthcare provider [24]. Visiting traditional healers refers to caregivers who visited a traditional healer at least once before seeking care at a formal health facility [24]. Distance to health facility defined as the estimated distance between the caregiver’s residence and the nearest health facility, categorized as ≤ 3 km, 3–5 km, and > 5 km [17]. Self-efficacy refers to the caregiver’s confidence in their ability to take a child with suspected malaria to a health facility within the recommended time [29]. Cue to action refers to factors or triggers that motivate caregivers to seek timely treatment for their sick child [29]. Knowledge of malaria defined as eleven questions totaling sixty-six yes/no items were used to gauge the caregivers’ knowledge of malaria. In particular, the caregivers were questioned about how they learned about malaria, its signs and symptoms, its modes of transmission, its seasonal patterns, the risks and complications these vulnerable groups may face, how to prevent malaria, how to treat it, where to get treatment, mosquito breeding sites, and when mosquitoes bite. A scoring system was used, with a score of 1 for each right response and a score of 0 for each wrong response. After adding together all of the scores, they were categorized as “good knowledge” or “inadequate knowledge” according to the mean score. Caregivers were classed as having strong understanding of malaria if they scored at or above the mean, and as having inadequate knowledge if they scored below the mean [30]. Perception of malaria defined as the HBM states that perceived vulnerability, perceived severity, perceived treatment advantages, perceived treatment hurdles, cue to action, and self-efficacy all influence an individual’s chance of engaging in a particular health (improving) behavior. Five questions about perceived severity, six questions about perceived susceptibility, five questions about perceived advantages, eight questions about perceived barriers, seven questions about cues (prompts to action), and three questions about self-efficacy were used to gauge the caregivers’ perceptions. Depending on the type of questions and the caregivers’ responses, the responses were ranked using a 5-point Likert scale: 1-strongly disagree, 2-disagree, 3-Neutral, 4-agree, and 5-strongly agree, or 1-strongly agree, 2-agree, 3-Neutral, 4-disagree, and 5-strongly disagree. The mean scores for each of the 34 constructs were calculated and divided into two categories: good and bad perception. Caregivers were categorized as having a positive perception of malaria if their answers to the perception-related questions were equal to or higher than the mean. Conversely, caregivers who scored lower than the average on these questions were thought to have a poor understanding of the illness. [14]. Data Collection Tool Data were collected using a structured interviewer-administered questionnaire adapted from previous studies and modified to fit the objectives of the study [17, 24]. The questionnaire was originally prepared in English, translated into the local language (Tigrigna), and back-translated into English to ensure consistency. It consisted of four sections: participant information and informed consent, socio-demographic characteristics, behavioral factors, and physical accessibility and health system factors. Data Collection Procedure Data were collected through face-to-face interviews with caregivers and review of medical records using a structured questionnaire. After malaria was confirmed by blood film examination, eligible caregivers were interviewed in a private room to ensure confidentiality. Data collection was conducted by four clinical nurses working in the under-five OPDs of the participating health centers and private clinics and one BSc nurse at Samre Primary Hospital. Data Quality Control. To ensure data quality, one supervisor and the data collectors were recruited from the District Health Office and local health facilities. They received two days of training on the study objectives, interview techniques, ethical considerations, standardized data collection procedures, and the application of the inclusion and exclusion criteria. To minimize recall bias, a six-month recall period was used. During data collection, trained data collectors carefully probed caregivers about the onset and sequence of malaria-related symptoms and treatment-seeking events to improve the accuracy of reported timing. A pretest was conducted on 5% of the sample outside the study area to assess the clarity, consistency, and applicability of the questionnaire. The internal consistency of the questionnaire was evaluated using Cronbach’s alpha, yielding a coefficient of 0.938, indicating excellent reliability. The supervisor closely monitored the data collection process and checked the completeness and consistency of the completed questionnaires daily. Data Processing and Analysis The coded data were checked for completeness and consistency before being entered into EpiData version 4.4.3.1 and exported to Statistical Package for the Social Sciences (SPSS) version 27 for analysis. Descriptive statistics, including frequencies, percentages, means, standard deviations, and cross-tabulations, were used to summarize the study variables. Bivariable logistic regression analysis was performed to identify candidate variables associated with delayed malaria treatment-seeking. Variables with a p-value < 0.20 were entered into the multivariable logistic regression model. Before model fitting, multicollinearity was assessed using the Variance Inflation Factor (VIF) and tolerance values. A VIF < 5 and tolerance > 0.20 indicated the absence of problematic multicollinearity. The goodness-of-fit of the final model was assessed using the Hosmer–Lemeshow goodness-of-fit test. A p-value greater than 0.05 indicated adequate model fit. Model diagnostics included assessment of influential observations using standardized residuals and Cook’s distance. No influential outliers were identified, and no missing data requiring imputation were observed. Adjusted odds ratios (AORs) with 95% confidence intervals (CIs) were calculated, and variables with a p-value < 0.05 in the multivariable analysis were considered statistically significant. Results Socio-Demographic Characteristics A total of 364 caregiver–child pairs (182 cases and 182 controls) participated in the study, resulting in a 100% response rate. The mean age of the children was 29.95 (SD ± 13.34) months among cases and 26.57 (SD ± 14.12) months among controls. Most caregivers in both groups were mothers. Rural residence was reported by 75.3% of caregivers of cases and 40.1% of caregivers of controls, while no formal education was reported by 35.2% and 23.1% of caregivers, respectively (Table 2).Table 2Socio-demographic characteristics of caregivers and children under five in Samre District, Tigray, Northern Ethiopia, 2024S. NoVariablesResponseCases(n = 182)Control(n = 182)Total(n = 364)Count (%)Count (%)Count (%)1Sex of childMale72(39.6%)84(46.2%)156(42.86%)Female110(60.4%)98(53.8%)208(57.14%)2Age of the child(months)≤ 1217(9.3%)30(16.5%)47(12.91%)13–2457(31.3%)73(40.1)130(35.71)> 24108(59.3%)79(43.4%)187(51.37%)3Caregivers age(years)≤ 20 years19(10.4%)20(11%)39(10.71%)20–29 years34(18.7%)72(39.6%)106(29.12%)30–39 years87(47.8%)55(30.2%)142(39.01%)≥ 40 years42(23.1%)35(19.2%)77(21.15%)4Relationship of childMother167(91.8%)170(93.4%)337(92.58%)1other15(8.2%)12(6.6%)27(7.42%)5Head of householdMale158(86.8%)147(80.8%)305(83.79%)Female24(13.2%)35(19.2%)36(16.21%)6Number of under-five children1 child82(45.1%)84(46.2%)166(45.6%)2 children54(29.7)68(37.4%)122(33.52)> 2 children46(25.3%)30(16.5%)76(20.88%)7Family size≤ 319(10.4%)47(25.8%)66(18.13%)4–7103(56.6%)105(57.7%)208(57.14%)≥ 760(33%)30(16.5%)90(24.73%)8ResidencyRural137(75.3%)73(40.1%)210(57.69%)Urban45(24.7%)109(59.9%)154(42.31%)9Marital statusMarried162(89%)158(86.8%)320(87.91%)2Other20(11%)24(13.2%)44(12.1%)10ReligionOrthodox166(91.2%)165(90.7%)331(90.93%)Islam16(8.8%)17(9.3%)33(9.07%)11EthnicityTigray170(93.4%)165(90.7%)335(92.03%)Agew12(6.6%)17(9.3%)29(7.97%)12Caregivers educational statusNo formal education64(35.2%)42(23.1%)106(29.04%)Primary education74 (40.7%)79(43.4%)153(42.03%)secondary education17(9.3%)18(9.9%)35(9.59%)college and above27(14.8%)43(23.6%)70(19.18%)13Partner educational statusNo formal education58(36.7%)79(45.7%)137(37.64%)primary education46(29.1%)50(28.9%)96(26.37%)secondary education29(18.4%)25(14.5%)54(14.84%)college and above25(15.8%)19(11%)44(12.09%)14Caregivers occupational statusHouse wife114(62.6%)124(68.1%)238(65.4%)Employed25(13.7%)30(16.5%)55(15.11%)Merchant22(12.1%)18(9.9%)40(10.99%)3Other21(11.5%)10(5.5)31(8.52%)15Partners occupational statusFarmer94(59.1%)109(63%)203(53.85%)Employed29(18.2%)37(21.4%)66(18.13%)Merchant31(19.5%)23(13.3%)54(14.84%)4Other5(3.1%)4(2.3%)9(2.47%)N.B: 1other (Father, brother and sister), 2Other (Single, divorced, widowed and separated), 3Others (Student, Daily Laborers), 4Others (Student, Daily Laborers). Table 2Socio-demographic characteristics of caregivers and children under five in Samre District, Tigray, Northern Ethiopia, 2024S. NoVariablesResponseCases(n = 182)Control(n = 182)Total(n = 364)Count (%)Count (%)Count (%)1Sex of childMale72(39.6%)84(46.2%)156(42.86%)Female110(60.4%)98(53.8%)208(57.14%)2Age of the child(months)≤ 1217(9.3%)30(16.5%)47(12.91%)13–2457(31.3%)73(40.1)130(35.71)> 24108(59.3%)79(43.4%)187(51.37%)3Caregivers age(years)≤ 20 years19(10.4%)20(11%)39(10.71%)20–29 years34(18.7%)72(39.6%)106(29.12%)30–39 years87(47.8%)55(30.2%)142(39.01%)≥ 40 years42(23.1%)35(19.2%)77(21.15%)4Relationship of childMother167(91.8%)170(93.4%)337(92.58%)1other15(8.2%)12(6.6%)27(7.42%)5Head of householdMale158(86.8%)147(80.8%)305(83.79%)Female24(13.2%)35(19.2%)36(16.21%)6Number of under-five children1 child82(45.1%)84(46.2%)166(45.6%)2 children54(29.7)68(37.4%)122(33.52)> 2 children46(25.3%)30(16.5%)76(20.88%)7Family size≤ 319(10.4%)47(25.8%)66(18.13%)4–7103(56.6%)105(57.7%)208(57.14%)≥ 760(33%)30(16.5%)90(24.73%)8ResidencyRural137(75.3%)73(40.1%)210(57.69%)Urban45(24.7%)109(59.9%)154(42.31%)9Marital statusMarried162(89%)158(86.8%)320(87.91%)2Other20(11%)24(13.2%)44(12.1%)10ReligionOrthodox166(91.2%)165(90.7%)331(90.93%)Islam16(8.8%)17(9.3%)33(9.07%)11EthnicityTigray170(93.4%)165(90.7%)335(92.03%)Agew12(6.6%)17(9.3%)29(7.97%)12Caregivers educational statusNo formal education64(35.2%)42(23.1%)106(29.04%)Primary education74 (40.7%)79(43.4%)153(42.03%)secondary education17(9.3%)18(9.9%)35(9.59%)college and above27(14.8%)43(23.6%)70(19.18%)13Partner educational statusNo formal education58(36.7%)79(45.7%)137(37.64%)primary education46(29.1%)50(28.9%)96(26.37%)secondary education29(18.4%)25(14.5%)54(14.84%)college and above25(15.8%)19(11%)44(12.09%)14Caregivers occupational statusHouse wife114(62.6%)124(68.1%)238(65.4%)Employed25(13.7%)30(16.5%)55(15.11%)Merchant22(12.1%)18(9.9%)40(10.99%)3Other21(11.5%)10(5.5)31(8.52%)15Partners occupational statusFarmer94(59.1%)109(63%)203(53.85%)Employed29(18.2%)37(21.4%)66(18.13%)Merchant31(19.5%)23(13.3%)54(14.84%)4Other5(3.1%)4(2.3%)9(2.47%)N.B: 1other (Father, brother and sister), 2Other (Single, divorced, widowed and separated), 3Others (Student, Daily Laborers), 4Others (Student, Daily Laborers). Socio-demographic characteristics of caregivers and children under five in Samre District, Tigray, Northern Ethiopia, 2024 Age of the child (months) N.B: 1other (Father, brother and sister), 2Other (Single, divorced, widowed and separated), 3Others (Student, Daily Laborers), 4Others (Student, Daily Laborers). Behavioral Factors The mean knowledge and perception scores of caregivers were 30.51 and 92.85, respectively. Inadequate knowledge of malaria was observed among 73.6% of caregivers of cases and 66.5% of caregivers of controls. Poor perception of malaria was reported by 70.9% of caregivers of cases and 63.2% of caregivers of controls. A history of under-five child death was reported by 6.6% of caregivers of cases and 9.9% of caregivers of controls (Table 3).Table 3Behavioral factors of caregivers of under five children in Samre District, Tigray, Northern Ethiopia, 2024S.NOvariablesResponseCase(n = 182)Control(n = 182)Total(n = 364)Count (%)Count (%)Count (%)1knowledgePoor134(73.6%)121(66.5%)255(70.05%)Good48(26.4%)61(33.5%)109(29.95%)2PerceptionPoor129(70.9%)115(63.2%)244(67.03%)Good53(29.1%)67(36.8%)120(32.97%)3Death of under-five childYes12(6.6%)18(9.9%)30(8.24%)No170(93.4%)164(90.1%)334(91.76%)4Previous history of malariaYes122(67%)148(81.3%)270(74.18%)No60(33.0%)34(18.7%)94(25.82%)5Experienced treatments S/EYes77(42.3%)44(24.2%)121(33.24%)No105(57.7%)138(75.8%)243(66.76%)6Choice of Health facility for malaria treatment1Modern Hfs139(76.4%)151(83%)290(79.67%)private pharmacy22(12.1%)17(9.3%)39(10.71%)Traditional21(11.5%)14(7.7%)35(9.62%)7Take traditional medicineYes89(48.9%)43(23.6%)132(36.26%)No93(51.1%)139(76.4%)232(63.74%)8Practiced of self-medicationYes99(54.4%)42(23.1%)141(38.74%)No83(45.6%)140(76.9%)223(61.26%)N.B: 1Modern (health post, Health center, Hospital and private clinics). Table 3Behavioral factors of caregivers of under five children in Samre District, Tigray, Northern Ethiopia, 2024S.NOvariablesResponseCase(n = 182)Control(n = 182)Total(n = 364)Count (%)Count (%)Count (%)1knowledgePoor134(73.6%)121(66.5%)255(70.05%)Good48(26.4%)61(33.5%)109(29.95%)2PerceptionPoor129(70.9%)115(63.2%)244(67.03%)Good53(29.1%)67(36.8%)120(32.97%)3Death of under-five childYes12(6.6%)18(9.9%)30(8.24%)No170(93.4%)164(90.1%)334(91.76%)4Previous history of malariaYes122(67%)148(81.3%)270(74.18%)No60(33.0%)34(18.7%)94(25.82%)5Experienced treatments S/EYes77(42.3%)44(24.2%)121(33.24%)No105(57.7%)138(75.8%)243(66.76%)6Choice of Health facility for malaria treatment1Modern Hfs139(76.4%)151(83%)290(79.67%)private pharmacy22(12.1%)17(9.3%)39(10.71%)Traditional21(11.5%)14(7.7%)35(9.62%)7Take traditional medicineYes89(48.9%)43(23.6%)132(36.26%)No93(51.1%)139(76.4%)232(63.74%)8Practiced of self-medicationYes99(54.4%)42(23.1%)141(38.74%)No83(45.6%)140(76.9%)223(61.26%)N.B: 1Modern (health post, Health center, Hospital and private clinics). Behavioral factors of caregivers of under five children in Samre District, Tigray, Northern Ethiopia, 2024 N.B: 1Modern (health post, Health center, Hospital and private clinics). Physical Accessibility and Health Care System Factors Access to transportation from the caregivers’ kebelle or village to a health facility was reported by 45.1% of caregivers of cases and 52.2% of caregivers of controls. Living more than 5 km from the nearest health facility was reported by 79.7% of caregivers of cases and 26.4% of caregivers of controls. A previous waiting time of 30–60 min at a health facility was reported by 50.5% of caregivers of cases and 53.3% of caregivers of controls (Table 4).Table 4Physical accessibility and healthcare system factors of caregivers of under-five children in Samre District, Tigray, Northern Ethiopia, 2024S.NOVariablesResponseCase(n = 182)Control(n = 182)Total(n = 364)Count (%)Count (%)Count (%)1Presence of road from kebelle to HfsYes106(58.2%)120(65.9%)226(62.09%)No76(41.8%)62(34.1%)138(37.91%)2Transportation access from kebelle to HfsYes82(45.1%)95(52.2%)177(48.63%)No100(54.9%)87(47.8%)187(51.37%)3Transportation mechanismon foot105(57.7%)45(24.7%)150(41.21%)Public77(42.3%)137(75.3%)214(58.79%)4Time taken from kebelle to reach Hfs≤ 160(33.0%)160(87.9%)220(60.44%)> 1122(67.0%)22(12.1%)144(39.56%)5Distance to HF≤ 3 km18(9.9%)65(35.7%)83(22.8%)3–5 km19(10.4%)69(37.9%)88(24.18%)> 5 km145(79.7%)48(26.4%)193(53.02%)6Cost of transportationcheap26(28.3%)36(28.1%)62(26.10%)optimum24(26.1%)49(38.3%)73(16.76%)Expensive42(45.7%)43(33.6%)85(17.58%)7Member of WDAYes28(15.4%)45(24.7%)73(20.05%)No154(84.6%)137(75.3%)291(79.95%)8Free waiver exempted servicesYes8(4.4%)4(2.2%)12(3.30%)No174(95.6%)178(97.8%)352(96.70%)9Prior waiting time at Hfs(minutes)< 306(3.3%)21(11.5%)27(7.42%)30–6092(50.5%)97(53.3%)189(51.92%)> 6084(46.2%)64(35.2%)148(40.66%) Table 4Physical accessibility and healthcare system factors of caregivers of under-five children in Samre District, Tigray, Northern Ethiopia, 2024S.NOVariablesResponseCase(n = 182)Control(n = 182)Total(n = 364)Count (%)Count (%)Count (%)1Presence of road from kebelle to HfsYes106(58.2%)120(65.9%)226(62.09%)No76(41.8%)62(34.1%)138(37.91%)2Transportation access from kebelle to HfsYes82(45.1%)95(52.2%)177(48.63%)No100(54.9%)87(47.8%)187(51.37%)3Transportation mechanismon foot105(57.7%)45(24.7%)150(41.21%)Public77(42.3%)137(75.3%)214(58.79%)4Time taken from kebelle to reach Hfs≤ 160(33.0%)160(87.9%)220(60.44%)> 1122(67.0%)22(12.1%)144(39.56%)5Distance to HF≤ 3 km18(9.9%)65(35.7%)83(22.8%)3–5 km19(10.4%)69(37.9%)88(24.18%)> 5 km145(79.7%)48(26.4%)193(53.02%)6Cost of transportationcheap26(28.3%)36(28.1%)62(26.10%)optimum24(26.1%)49(38.3%)73(16.76%)Expensive42(45.7%)43(33.6%)85(17.58%)7Member of WDAYes28(15.4%)45(24.7%)73(20.05%)No154(84.6%)137(75.3%)291(79.95%)8Free waiver exempted servicesYes8(4.4%)4(2.2%)12(3.30%)No174(95.6%)178(97.8%)352(96.70%)9Prior waiting time at Hfs(minutes)< 306(3.3%)21(11.5%)27(7.42%)30–6092(50.5%)97(53.3%)189(51.92%)> 6084(46.2%)64(35.2%)148(40.66%) Physical accessibility and healthcare system factors of caregivers of under-five children in Samre District, Tigray, Northern Ethiopia, 2024 Determinants of Delayed Malaria Treatment Seeking Among Under-5 Children A bivariable logistic regression analysis was conducted to identify candidate variables for the inclusion in the multivariable logistic regression model. Eighteen variables met the inclusion criteria, defined by a p-value of equal or less than 0.2, making them candidates for further analysis. These variables were included in the multivariable analysis to control for potential confounding factors. The analysis identified five independently associated factors with delayed malaria treatment seeking among under-five children: Children whose caregivers had no formal education were 2.7 times more likely to delay in seeking treatment for under-five children compared to those who have completed college or higher education (AOR = 2.7; 95% CI: 1.19, 6.12). Caregivers with a poor perception of malaria were 3.5 times more likely to delay in seeking treatment for under-five children compared to those with a good perception of malaria (AOR = 3.5; 95% CI: 1.71, 7.37). Caregivers who practiced self-medication with anti-malarial drugs were 8.3 times more likely to delay in seeking treatment for under-five children compared to those who did not self-medicate (AOR = 8.3; 95% CI: 3.78,18.27). Caregivers who had no previous history of malaria were 4.6 times more likely to delay in seeking treatment for under-five children compared to those with a history of malaria infection (AOR = 4.6; 95% CI: 2.34, 8.94). Finally, caregivers living more than five kilometers from a health facility were 7.8 times more likely to delay in seeking treatment for under-five children compared to those living within three kilometers of a health facility (AOR = 7.8; 95% CI: 3.7, 16.32) (Table 5).Table 5Bivariable and multivariable analysis of determinants of delayed malaria treatment seeking among under-five children, Samre District, Tigray, Northern Ethiopia, 2024S.NOVariableResponseCasesControlCOR (95% C.I)Sig.AOR (95% C.I)Count (%)Count (%)1Member of WDAYes28(15.4%)45(24.7%)0.554(0.327,0.936)0.3570.71(0.35,1.47)No154(84.6%)137(75.3%)112Family size≤ 319(10.4%)47(25.8%)0.202(0.101,0.403)0.9321.05(0.32,3.49)4–7103(56.6%)105(57.7%)0.490(0.293,0.821)0.5181.32(0.57,3.05)> 760(33%)30(16.5%)113Treatment side effectsYes77(42.3%)44(24.2%)2.300(1.468,3.604)0.7960.87(0.29,2.54)No105(57.7%)138(75.8%)114ResidencyRural137(75.3%)73(40.1%)4.546(2.903,7.120)0.2250.62(0.28,1.35)Urban45 (24.7%)109(59.9%)115Traditional medicineYes89(48.9%)43(23.6%)3.094(1.975,4.846)0.2850.48 (0.13,1.83)No93(51.1%)139(76.4%)116Prior waiting time at HF(minutes)< 306(3.3%)21(11.5%)0.218(0.083,0.571)0.0990.35(0.1,1.22)30–6092(50.5%)97(53.3%)0.723(0.469,1.114)0.5570.84(0.47,1.5)> 6084(46.2%)64(35.2%)117Caretakers Educational statusNo formal education64(35.2%)42(23.1%)2.43(1.307,4.507)0.0182.7(1.19,6.12)*Primary education74(40.7%)79(43.4%)1.49(0.84,2.66)0.4131.368(0.65,2.89)Secondary education17(9.3%)18(9.9%)1.5(0.66,3.41)0.4621.495(0.51,4.37)College and above27(14.8%)43(23.6%)118Caretaker occupational statusHouse wife114(62.6%)124(68.1%)1.103(0.612,1.987)0.0720.28 (0.072,1.12)Merchant22(12.1%)18(9.9%)1.467(0.647,3.325)0.1480.37(0.097,1.42)3Other21(11.5%)10(5.5%)2.52(1.003,6.332)0.1840.38(0.09,1.59)Employed25(13.7%)30(16.5%)119Age of caretakers(years)≤ 2019(10.4%)20(11%)1120–2934(18.7%)72(39.6%)0.497(0.235,1.05)0.8800.93(0.33,2.56)30–3987(47.8%)55(30.2%)1.67(0.82,3.397)0.1562(0.77,5.24)≥ 4042(23.1%)35(19.2%)1.26(0.58,2.73)0.8290.897(0.33,2.4210Previous history of malariaYes122(67%)148(81.3%)11NO60(33%)34(18.7%)2.141(1.319,3.474)0.004.57(2.34,8.94)*11Practice of self-medicationYes99(54.4%)42(23.1%)3.976(2.531,6.246)0.008.3(3.78,18.27)*No83(45.6%)140(76.9%)1112Transport mechanismOn foot105(57.7%)45(24.7%)4.152(2.655,6.492)0.2241.58(0.76,3.29)Public transport77(42.3%)137(75.3%)1113Distance to HFs≤ 3 km18(9.9%)65(35.7%)113–5 km19(10.4%)69(37.9%)0.994(0.480,2.060)0.3860.69(0.3,1.59)> 5 km145(79.7%)48(26.4%)10.91(5.89,20.190)0.0077.77(3.7,16.32)*14knowledgePoor134(73.6%)121(66.5%)1.41(0.89,2.21)0.2170.2.02(0.66,6.24)Good48(26.4%)61(33.5%)1115Number of under five children1 child82(45.1%)84(46.2%)0.637(0.367,1.105)0.8781.09(0.57,2.1)2 child54(29.7)68(37.4%)0.518(0.289,0.927)0.0810.49(0.22,1.09)> 2child46(25.3%)30(16.5%)0.64(0.37,1.11)1116PerceptionPoor129(70.9%)115(63.2%)1.42(0.91,2.20)0.0013.55(1.71,7.37)*Good53(29.1%)67(36.8%)1117Age of children(months)≤ 1217(9.3%)30(16.5%)1113–2457(31.3%)73(40.1%)1.38(0.69,2.74)0.8891.07(0.43,2,65)> 24108(59.3%)79(43.4%)2.41(1.24,4.67)0.1861.81(0.75,4.36)18Head HHMale158(86.8%)147(80.8%)1.567(0.89,2.76)0.3521.45(0.64,3.31)Female24(13.2%)35(19.2%)11N.B: variables with *P value<0.05 refers independent determinants of delayed malaria treatment seeking among under five children. 1other (Student, Daily Laborers) Table 5Bivariable and multivariable analysis of determinants of delayed malaria treatment seeking among under-five children, Samre District, Tigray, Northern Ethiopia, 2024S.NOVariableResponseCasesControlCOR (95% C.I)Sig.AOR (95% C.I)Count (%)Count (%)1Member of WDAYes28(15.4%)45(24.7%)0.554(0.327,0.936)0.3570.71(0.35,1.47)No154(84.6%)137(75.3%)112Family size≤ 319(10.4%)47(25.8%)0.202(0.101,0.403)0.9321.05(0.32,3.49)4–7103(56.6%)105(57.7%)0.490(0.293,0.821)0.5181.32(0.57,3.05)> 760(33%)30(16.5%)113Treatment side effectsYes77(42.3%)44(24.2%)2.300(1.468,3.604)0.7960.87(0.29,2.54)No105(57.7%)138(75.8%)114ResidencyRural137(75.3%)73(40.1%)4.546(2.903,7.120)0.2250.62(0.28,1.35)Urban45 (24.7%)109(59.9%)115Traditional medicineYes89(48.9%)43(23.6%)3.094(1.975,4.846)0.2850.48 (0.13,1.83)No93(51.1%)139(76.4%)116Prior waiting time at HF(minutes)< 306(3.3%)21(11.5%)0.218(0.083,0.571)0.0990.35(0.1,1.22)30–6092(50.5%)97(53.3%)0.723(0.469,1.114)0.5570.84(0.47,1.5)> 6084(46.2%)64(35.2%)117Caretakers Educational statusNo formal education64(35.2%)42(23.1%)2.43(1.307,4.507)0.0182.7(1.19,6.12)*Primary education74(40.7%)79(43.4%)1.49(0.84,2.66)0.4131.368(0.65,2.89)Secondary education17(9.3%)18(9.9%)1.5(0.66,3.41)0.4621.495(0.51,4.37)College and above27(14.8%)43(23.6%)118Caretaker occupational statusHouse wife114(62.6%)124(68.1%)1.103(0.612,1.987)0.0720.28 (0.072,1.12)Merchant22(12.1%)18(9.9%)1.467(0.647,3.325)0.1480.37(0.097,1.42)3Other21(11.5%)10(5.5%)2.52(1.003,6.332)0.1840.38(0.09,1.59)Employed25(13.7%)30(16.5%)119Age of caretakers(years)≤ 2019(10.4%)20(11%)1120–2934(18.7%)72(39.6%)0.497(0.235,1.05)0.8800.93(0.33,2.56)30–3987(47.8%)55(30.2%)1.67(0.82,3.397)0.1562(0.77,5.24)≥ 4042(23.1%)35(19.2%)1.26(0.58,2.73)0.8290.897(0.33,2.4210Previous history of malariaYes122(67%)148(81.3%)11NO60(33%)34(18.7%)2.141(1.319,3.474)0.004.57(2.34,8.94)*11Practice of self-medicationYes99(54.4%)42(23.1%)3.976(2.531,6.246)0.008.3(3.78,18.27)*No83(45.6%)140(76.9%)1112Transport mechanismOn foot105(57.7%)45(24.7%)4.152(2.655,6.492)0.2241.58(0.76,3.29)Public transport77(42.3%)137(75.3%)1113Distance to HFs≤ 3 km18(9.9%)65(35.7%)113–5 km19(10.4%)69(37.9%)0.994(0.480,2.060)0.3860.69(0.3,1.59)> 5 km145(79.7%)48(26.4%)10.91(5.89,20.190)0.0077.77(3.7,16.32)*14knowledgePoor134(73.6%)121(66.5%)1.41(0.89,2.21)0.2170.2.02(0.66,6.24)Good48(26.4%)61(33.5%)1115Number of under five children1 child82(45.1%)84(46.2%)0.637(0.367,1.105)0.8781.09(0.57,2.1)2 child54(29.7)68(37.4%)0.518(0.289,0.927)0.0810.49(0.22,1.09)> 2child46(25.3%)30(16.5%)0.64(0.37,1.11)1116PerceptionPoor129(70.9%)115(63.2%)1.42(0.91,2.20)0.0013.55(1.71,7.37)*Good53(29.1%)67(36.8%)1117Age of children(months)≤ 1217(9.3%)30(16.5%)1113–2457(31.3%)73(40.1%)1.38(0.69,2.74)0.8891.07(0.43,2,65)> 24108(59.3%)79(43.4%)2.41(1.24,4.67)0.1861.81(0.75,4.36)18Head HHMale158(86.8%)147(80.8%)1.567(0.89,2.76)0.3521.45(0.64,3.31)Female24(13.2%)35(19.2%)11N.B: variables with *P value<0.05 refers independent determinants of delayed malaria treatment seeking among under five children. 1other (Student, Daily Laborers) Bivariable and multivariable analysis of determinants of delayed malaria treatment seeking among under-five children, Samre District, Tigray, Northern Ethiopia, 2024 N.B: variables with *P value<0.05 refers independent determinants of delayed malaria treatment seeking among under five children. 1other (Student, Daily Laborers) Discussion This study identified key independent determinants of delayed malaria treatment seeking among under-five children. Specifically, children whose caregivers had no formal education, poor perception, practiced of self-medication, had no previous history of malaria, and living in villages located more than five kilometers from health facilities were associated with an increased likelihood of delayed malaria treatment seeking among under-five children. Caregivers with no formal education were more likely to delay seeking malaria treatment for their children than those with college-level education or above. This finding is consistent with studies conducted in Mozambique [31]. Caregivers with higher educational attainment are more likely to access malaria-related information through health education, community programs, and other communication channels, enabling them to recognize malaria symptoms early and seek prompt treatment [32, 33]. In contrast, this finding differs from a study conducted in Gambella Town, Ethiopia [24]. The discrepancy may be explained by differences in healthcare accessibility. Unlike the post-conflict setting of the present study, Gambella has relatively functional primary healthcare services and outreach programs that may facilitate timely treatment regardless of caregivers’ educational level [34, 35]. Poor caregiver perception of malaria was also associated with delayed treatment-seeking. Caregivers with poor perceptions may underestimate the severity of malaria, fail to recognize early symptoms, or perceive less benefit from seeking prompt treatment. Consequently, they may delay deciding to seek care until the child’s condition worsens [36]. Self-medication was another important determinant of delayed treatment-seeking. Similar findings have been reported from Dambi Dollo and Sayyo districts, Zambia, and Equatorial Guinea [37–39]. Self-medication was strongly associated with delayed malaria treatment-seeking. This relatively large association should be interpreted with caution because self-medication and delayed formal treatment-seeking are closely related aspects of caregivers’ treatment-seeking behavior. Caregivers who initially attempt to manage a child’s illness at home may postpone seeking formal healthcare, which may partly explain the magnitude of the association. In addition, because both behaviors were assessed retrospectively, the temporal relationship between self-medication and delayed formal care-seeking cannot be established with certainty. Other barriers, including distance to health facilities, transportation difficulties, treatment costs, and disruption of health services in the post-conflict setting, may also influence both behaviors. Therefore, the observed association should be understood as a statistical association rather than evidence of a causal relationship [40]. However, this finding contrasts with studies from Bench-Maji Zone and Gambella Town [17, 24],.These differences may reflect better access to functional health facilities and stronger community health education programs in those settings, reducing reliance on self-medication [10]. Caregivers without previous experience of malaria were more likely to delay seeking treatment than those with prior experience. This finding is consistent with studies conducted in Oromia Region and Dera, Ethiopia [12, 41]. Previous exposure to malaria may improve caregivers’ ability to recognize symptoms, appreciate the severity of the disease, and understand the importance of seeking prompt treatment based on earlier interactions with healthcare providers [42]. Caregivers residing in a village located more than five kilometers away from a health facilities are more likely to delay in seeking malaria treatment for under-five children compared to caregivers living within a three-kilometer radius of healthcare facilities. This finding is consistent with studies from Bench-Maji, Zambia, and Burkina Faso [17, 39, 43]. Long distances to health facilities increase travel time and transportation costs, creating barriers to timely access to malaria diagnosis and treatment. Similar findings have been reported in Ethiopia, where distance from health facilities and limited transportation were identified as important determinants of delayed malaria treatment-seeking [44, 45]. In contrast, a study from Gambella Town reported no such association [24]. This difference may be explained by the urban setting of Gambella, where health facilities and transportation are more accessible. In the present post-conflict setting, damage to health infrastructure and transportation systems may have amplified the effect of distance on treatment-seeking [46]. These findings have important implications for malaria control in post-conflict settings. The associations of delayed treatment-seeking with long distance to health facilities, self-medication, and poor caregiver perception suggest that disruption of health services following the conflict continues to affect timely access to malaria care. Strengthening the rehabilitation of damaged health facilities, restoring community-based malaria services, ensuring uninterrupted availability of malaria diagnostics and antimalarial medicines, and implementing targeted health education programs for caregivers should be prioritized. Improving access to malaria services for remote communities through outreach services and strengthened health extension programs may also help reduce treatment delays. Limitation of the Study This study has several limitations. First, because it employed an unmatched case-control design, a causal relationship between the identified determinants and delayed malaria treatment-seeking cannot be established. Second, as a facility-based study, the findings may not be generalizable to caregivers who did not seek care at health facilities, introducing the possibility of selection bias. Third, although efforts were made to minimize recall bias by restricting the recall period to six months, caregivers may not have accurately remembered the timing of symptom onset or previous exposures. In particular, previous waiting time at health facilities was based on caregivers’ recall and was collected using broad predefined categories (< 30, 30–60, and > 60 min) rather than an exact duration. Although this approach was intended to reduce recall burden, some measurement and recall error may remain. Fourth, information on behavioral factors was collected through face-to-face interviews and may therefore be subject to social desirability bias, with caregivers potentially underreporting practices such as self-medication or over reporting desirable health-seeking behaviors. Fifth, the temporal relationship between self-medication and delayed formal treatment-seeking could not be established because both were assessed retrospectively. The strong association observed between these variables may partly reflect their close relationship or shared underlying barriers to accessing formal healthcare. Finally, although multivariable analysis was performed to control for measured confounders, residual confounding from unmeasured variables, including detailed household socioeconomic status, cannot be excluded. Conclusion This study identified key determinants of delayed malaria treatment seeking-among under-five children, including children whose caregivers had no formal education, poor perception of malaria, practiced self-medication, caregivers who did not reported previous history of malaria, and living more than five kilometers from health facilities. These factors collectively significantly exacerbate delayed treatment seeking of malaria in under-five children in the study area. Recommendations Based on the findings of this study, the Tigray Regional Health Bureau, Samre District Health Office, and local health facilities should strengthen post-conflict malaria control efforts by improving community health education to enhance caregivers’ awareness and perception of malaria, promote prompt treatment-seeking, and discourage self-medication. Efforts should also focus on improving access to timely malaria diagnosis and treatment, particularly for communities located far from health facilities, through the restoration and strengthening of primary healthcare services, including health posts and health centers. Ensuring the continuous availability of malaria diagnostic tests and antimalarial medicines at nearby health facilities is essential to reduce delays in treatment. In addition, community outreach and health extension programs should be strengthened to improve access to malaria services and support early treatment-seeking among caregivers of under-five children in post-conflict settings.