Area-Based Child Protection Risk Analysis: Ukraine

Area-Based Child Protection Risk Matrix for Ukraine

Armed conflict, mass displacement, and deepening poverty have created a child protection crisis in Ukraine that differs sharply from one oblast to the next. This analysis uses 12 open-source 2024 publications and datasets — household surveys, displacement tracking, needs assessments, protection monitoring, and coordination records — to build an oblast-level composite risk profile.

According to these sources, roughly 3.5 million children — 70% of all children in Ukraine — now lack access to basic goods and services, including adequate food, shelter, and functioning water supply. The UNICEF SitAn 2024 confirms a 35% birthrate decline and millions of women and children having fled the country. Frontline oblasts — Donetska, Kharkivska, Khersonska, Zaporizka, Sumska, and Luhanska — score highest on the composite risk index, driven by direct conflict exposure (576 education facilities damaged/destroyed in 2024 alone, compared to approximately 288 in 2023 per Education Cluster monitoring data), infrastructure destruction, and collapsing service delivery. UNICEF reached 62,091 children with individualised case management and provided Mental Health and Psychosocial Support (MHPSS) to 757,807 children, caregivers, and frontline workers in 2024.

The risk matrix uses eight scoring dimensions — conflict exposure, socio-economic vulnerability, displacement burden, service access gaps, child-specific deprivation, Explosive Remnants of War (ERW)/mine contamination, education disruption, and GBV/violence prevalence — using a weighted composite formula. The IOM DTM Round 18 (Oct 2024) confirms 3.6 million internally displaced people (IDPs) (24% children), with Dnipropetrovska (14%) and Kharkivska (12%) as the main hosting oblasts. The REACH MSNA 2024 found that 38% of assessed households nationally have severe or above sectoral needs, with the Humanitarian Needs and Response Plan (HNRP) serving as the coordination framework, with displaced, returnee, and disability-affected households reporting significantly higher unmet needs.

War-Driven Poverty Surge and Household Fragmentation

The full-scale invasion has transformed the makeup of Ukrainian households. Average household size fell from 2.57 persons in 2021 to 2.29 in 2023, a result of displacement, family separation, and emigration. The number of households with children dropped by 34% — from 5.5 million to 3.65 million. Among remaining households with children, single-child families now dominate at 68.7%, up from a pre-war pattern where multi-child households were more common.

The economic toll is visible in every indicator. The share of families with children living below the actual subsistence minimum rose by 1.7 times — from 20.6% to 35.5%. Those unable to afford adequate nutrition tripled from 3.4% to 12.2%. Real incomes for households with children nominally grew to UAH 19,101 per person (from 18,386 in 2021), but this hides a widening gap: households with three or more children saw real income fall, and frontline-region households faced the sharpest drops.

Multidimensional Child Poverty and Deprivation

OECD multidimensional poverty measures show three crises stacked on top of each other. 40.3% of children live in monetary poverty (below the actual subsistence minimum), 70.4% experience material deprivation (3+ out of 17 identified deprivations), and 40.1% are in households that self-assess their consumption capacity as insufficient. The intersection of all three dimensions captures 22.1% of all children — the most vulnerable group.

Material deprivation rates climb with household size: families with three or more children face a 80.8% deprivation rate, compared to 70.7% for single-child families. Households where all adults are of retirement age or have disabilities experience near-total child deprivation at 97.1%. The urban-rural split is wide — 73.9% in rural areas versus 67.6% in cities. IDP households with children stand out on several measures: 35.8% contain children (vs 26.9% nationally), and only 58.2% own their housing (vs 89% nationally).

CP AoR Case Data and Protection Concerns

The UNICEF 2024 Annual Report confirms 62,091 children received individualised CP case management support and MHPSS was provided to 757,807 children, caregivers, and frontline workers. The CPIMS+ system recorded 19,274 cases by November 2024, with psychosocial distress as the dominant concern (7,690 cases), followed by children affected by explosive ordnance (5,454), denial of access to services and resources (2,991), inadequate care arrangements (2,822), neglect (2,076), and lack of birth registration or legal identity (1,512), among other categories. An estimated 1.5 million children are at risk of depression, anxiety, and PTSD, with 84% of key informants identifying psychosocial well-being as the top child risk (Protection Cluster PAU, July 2024).

In total, UNICEF’s 2024 humanitarian response reached 9.8 million people including 2.5 million children across all sectors (health, nutrition, education, WASH, child protection, and social protection). The HNRP identified 2.9 million children (of 14.6 million people in need), while UNICEF’s HAC appeal required $633.6 million to address needs of 6.5 million people. Nearly 6 million Ukrainian refugees (88% women and children) are hosted across Europe, with a 70% family separation rate exposing children to trafficking, exploitation, and abuse risks. The IOM DTM Round 18 (Oct 2024) counted 3.6 million IDPs domestically (57% female, 24% children), with Dnipropetrovska (14%) and Kharkivska (12%) as principal hosting oblasts. At least 2,502 children have been verified as killed or injured since February 2022 (669 killed, 1,833 injured; OHCHR verified through December 2024). Actual numbers are likely significantly higher. The gap between need and response is wide: 62,091 children reached by case management represents just 2.1% of the 2.9 million children in need identified under the HNRP. Ukraine’s pre-war residential care system housed approximately 100,000 children; deinstitutionalisation efforts have stalled under conflict, and children in institutions in frontline areas face evacuation, separation, and care-continuity problems daily. Family Tracing and Reunification (FTR) capacity is far below what the scale of family separation demands.

Area-Based Child Protection Risk Matrix

The map below shows the composite child protection risk classification for each of Ukraine's 25 assessed oblasts (government-controlled areas). Hover over an oblast to see its severity level and composite score. Click for a breakdown of all eight dimension scores.

Child Protection Risk Code Definitions

The five-level risk code system aligns with the IASC/JIAF 2.0 humanitarian severity scale and the Global Protection Cluster PiN/Severity methodology. Risk codes CP-1 through CP-5 map to composite score thresholds from the eight-dimension weighted model (see Methodology above). The table below sets out the operational criteria and recommended response type for each level, based on CP AoR coordination data, HSESS findings, and the REACH MSNA 2024.

Gaps in Social Services for Children

Only 11.1% of households with socially vulnerable persons receive social services, leaving 88.9% without coverage — of whom 31.7% report an active need for services they cannot access. Among households with children, the gap between need and receipt is worst for IDP families (64.8% changed location of service access due to war) and families with a child with a disability (only 7.4% reported needing services while 15.9% pay for private alternatives).

Social assistance reaches some groups better than others. Of the major social support programmes, assistance to low-income families is the most effectively targeted (78.8% of recipients are poor), but its coverage is low: only 1.4% of all Ukrainian households receive it. IDP assistance reaches 4% of households nationally but accounts for 29.8% of recipients’ total income — indicating high dependence. Critically, child-specific benefits such as pregnancy/childbirth support (29.4% of HH with children income), single-mother allowances (31.3%), and guardianship assistance (24%) make up a large share of income for the poorest households.

IDP Households and Child Protection Implications

IDP households look different from the general population, and those differences raise child protection risks. They are larger (average 2.5 vs 2.3 persons), more likely to contain children (35.8% vs 26.9%), and overwhelmingly reliant on rented housing (58.2% rent vs 9.2% nationally). Only 24.2% of IDP households with children own their dwelling. Overcrowding is widespread: the Eurostat-method overcrowding rate for households with children nationally stands at 86.6%, with IDP families in hostels (5%), modular towns (0.6%), and communal apartments (4.4%) facing the most precarious conditions.

Displacement data shows that while western oblasts absorbed the first wave (Zakarpatska showing nearly 100% Feb 2014–Jan 2022 movers), eastern regions like Mykolaivska and Kyivska received waves throughout all conflict phases. Households with at least one IDP member report much higher need: 29.2% of IDP households with children received NGO assistance (vs 5.3% for one-child families nationally), and 42.4% of IDP families with children assessed this support as something they “could not live without.”

CP AoR Operational Coordination (National & Regional Meeting Minutes)

The CP AoR National and Regional Meeting Minutes (January–November 2024) underpin the coordination-level data behind this risk matrix. Records from 11 National CP AoR meetings and over 40 Regional CP AoR meetings across seven sub-national coordination hubs (Kharkiv, Dnipro, Donetsk, Kherson, Mykolaiv, Sumy, and Odesa, plus Lviv and Chernihiv) track how the response has developed month by month.

Response Scale and Service Coverage

By the mid-year review (July 2024), the CP AoR reported 213 partners providing assistance to 743,000 children and 299,000 parents/guardians across 1,141 communities. The web-based Child Protection Referral Pathways system mapped 1,303 organisations (humanitarian and state social service), 982 mobile teams, and 2,376 referral focal points across 25 regions, 126 districts, 350 hromadas, and 1,536 municipalities. By year-end, 181 projects had been launched (90 completing in 2024, 91 carrying into 2025).

Evacuation and Family Separation

Regional meeting minutes document the scale of ongoing mandatory evacuations. In Donetska oblast, 1,900 children remained in forced evacuation territories as of September 2024, with 10 settlements under active mandatory evacuation orders covering 336 children. In Sumska oblast, mandatory evacuation from 7 communities (101 settlements) affected 475 children, with specific concentrations in Bilopillia (145 children), Vorozhba (200), and Uhroidy (45). In Kharkivska oblast, 4,178 orphans and children deprived of parental care were registered, with 98% placed in family-based care; evacuation from Borivska (119 children) and Kupiansk district (269 children) was largely completed by November, with transit centres operational in Izyum and Kharkiv City. Dnipropetrovska reported 1,500 children remaining in danger in the Pokrovsk direction, with a transit shelter accommodating 60 mothers with children.

Sub-National Partner Presence

Partner density varies by oblast, broadly tracking the risk levels. Kharkivska had the highest concentration with 49 active CP partners and 16 mobile teams, followed by Dnipropetrovska (35 reporting partners) and Mykolaivska (42 partners). In contrast, Donetska and Chernihiv oblasts operated with 12–15 partners each, and the lowest coverage areas — Rivnenska, Luhanska, and Volynska — had minimal partner presence. The CP AoR minutes consistently note that rural coverage remained critically low at 5–12% in Donetska and Chernihiv versus 72% urban nationally, with transport access the primary barrier.

MSNA CP Findings (May–July 2024)

The REACH MSNA 2024, presented to the National CP AoR in September 2024, surveyed 10,434 households across 24 oblasts and 105 raions. Key CP-relevant findings: 32% of children (aged 5+) exhibited psychosocial distress signs, rising to 45% in Mykolaivska and 42% in Khersonska; 44% of adults reported psychosocial difficulties; and 42% of households reported barriers to MHPSS access. Households with children reported higher protection needs than those without, particularly in northern oblasts (Sumska, Chernihivska). In education, 88% of children were enrolled but 75% learning online, with offline schooling highest in Mykolaivska (24%), Sumska (24%), and Dnipropetrovska (19%). A 20% decline in learning outcomes was documented through the Rapid Gender Analysis.

2025 Planning Trajectory

The HNRP 2025 planning cycle, documented in the October and November National CP AoR meetings, recalibrated the response to $128.8 million (down from $139.6M in 2024), targeting 2 million people (60% children, 40% caregivers). The CP-specific PIN was refined to 865,000 IDP children (down from 1 million) plus 2.4 million children and caregivers in non-displaced conflict-affected populations. The 5% target cut and 10% budget reduction point to a tighter focus on frontline oblasts, not a drop in assessed need.

Additional Risk Dimensions from 2024 Publications

Three additional dimensions add detail to the risk scores: explosive remnants of war (ERW) contamination, education disruption, and GBV/violence prevalence. Each draws on dedicated 2024 open-source publications.

ERW/Mine Contamination (Dimension D6)

Ukraine’s National Mine Action Authority reports that 156,000 km² of territory has been exposed to conflict and requires survey. An estimated 30% of the country may be contaminated with mines or UXO. In 2024, over 600 child casualties were recorded from explosive weapons, with the highest concentration in Kharkivska (21 victims in Q3 2024 alone). Landmines have been found in 11 of 27 regions. The 5.4 million people requiring Mine Action assistance include a disproportionate share of children, who face lasting injuries including limb amputations and loss of eyesight. Risk education reached 1.08 million children and 200,325 caregivers through mobile teams in 2024.

Education Disruption (Dimension D7)

Attacks on education facilities doubled in 2024: 576 facilities (schools, kindergartens, universities) were damaged or destroyed compared to 256 in 2023. The REACH MSNA 2024 confirms that 75% of children engage primarily in remote learning, with top challenges including lack of internet connectivity, disrupted learning from air raid alerts, and lack of peer contact. The Education Cluster documented acute responses in Kharkivska, Sumska, and Donetska oblasts (May–July 2024), where cities have resorted to underground schools in metro systems. The State Service for Education Quality assessment in 2024 showed declining test results for 8th graders and students in remote/hybrid formats.

GBV/Violence Prevalence (Dimension D8)

The UNFPA GBV assessment (Feb–Apr 2024) confirms an estimated 2.5 million vulnerable people at high risk of GBV. Government-reported rates of sexual violence and domestic violence affecting children increased significantly between 2022 and 2023. In 2024, areas retaken by Ukraine and frontline territories saw a doubling of intimate partner violence and domestic violence cases. REACH’s CP Assessment found that in Dnipropetrovska, 25% of children reported incidents of violence, with girls aged 15–17 at 30% versus 20% of boys. By May 2024, 82 GBV partners reported services to 261,500 people (15% children).

Key Child Protection Events and Response Timeline

Policy and Programming Recommendations

Analytical Framework and Composite Risk Scoring

Purpose and Scope

This risk matrix assigns an oblast-level severity classification across Ukraine, intended to guide humanitarian programming, resource allocation, and inter-cluster coordination. The analysis covers 25 oblasts (Ukraine’s principal administrative divisions, equivalent to regions; including Kyiv City) within government-controlled areas (GCA) only. Temporarily occupied territories of Crimea and non-GCA portions of Donetska, Luhanska, Zaporizka, and Khersonska oblasts are excluded due to access constraints and absence of reliable data. The scope follows the OCHA Humanitarian Programme Cycle (HPC) 2024/2025 and the Joint Intersectoral Analysis Framework (JIAF 2.0) five-phase severity scale.

Severity Classification Framework

The five-level child protection risk code system (CP-1 through CP-5) is aligned with the IASC/JIAF humanitarian severity scale (Phase 1 Minimal through Phase 5 Catastrophic) and the Global Protection Cluster Methodology for Calculating Protection Severity. The classification combines CP AoR operational severity with socio-economic and protection indicators across eight weighted dimensions. Each oblast receives a normalised score of 1–5 per dimension. These are combined into a single weighted composite score, which is then mapped to the severity phase thresholds:

Composite Scoring Formula

The composite risk score C_i for each oblast i is calculated as a weighted arithmetic mean of eight normalised dimension scores:

Dimension Definitions, Indicators, and Weights

Normalisation Process

Each dimension score is normalised to a 1–5 integer scale at the oblast level. The normalisation follows a min-max rescaling approach: for each indicator, the minimum value across the 25 assessed oblasts is anchored to 1 and the maximum to 5, with intermediate values assigned proportionally. Where multiple indicators feed a single dimension, they are averaged within the dimension before rescaling. This follows the JIAF 2.0 severity scoring methodology, where threshold indicators are used to assign severity levels and the highest severity across contributing indicators determines the dimensional score.

Weight Justification

Dimension weights reflect how strongly each factor predicts child protection outcomes in Ukraine’s conflict setting. The two highest-weighted dimensions — D1 Conflict Exposure (0.20) and D5 Child-Specific Deprivation (0.15) — carry a combined weight of 0.35 because they are the main drivers of CP severity in active conflict (CPMS 2019; Global Protection Cluster PiN methodology). D2 Socio-Economic Vulnerability (0.15) is weighted equally to D5 given the documented 1.7-fold poverty surge since 2022 and its direct causal link to child neglect, exploitation, and family separation. The three newer dimensions (D6–D8) each carry 0.10, reflecting their importance but also the thinner sub-national data available for them so far.

Data Sources and Vintage

The analysis draws on 12 open-source 2024 publications and datasets. The principal sources and their temporal coverage are:

• HSESS (Household Socio-Economic Status Survey): UCSR/UNICEF, December 2023–February 2024, 8,023 households, CARI methodology, national coverage with oblast stratification
• UNICEF SitAn 2024 (Situation Analysis of Children in Ukraine): Published 2024, national scope, secondary data synthesis
• REACH/IMPACT MSNA 2024 (Multi-Sector Needs Assessment): Household-level survey, national and oblast coverage, JIAF 2.0-aligned severity indicators
• IOM DTM Rounds 16–18 (Displacement Tracking Matrix): April–October 2024, 23 oblasts, raion-level displacement data
• Protection Cluster PAU (Protection Analysis Update): July 2024, national coverage with sub-national severity classification
• CP AoR / CPIMS+ (Child Protection Information Management System): Operational data through November 2024, 1,220 communities, 19,274 individual cases
• OCHA HNRP 2024 (Humanitarian Needs and Response Plan): December 2023, 14.6 million People in Need (PIN), 2.9 million children
• National Mine Action Authority / ACAPS: 2024 thematic reports on ERW contamination extent and child casualties
• Education Cluster: 2024 monitoring data on attacks, 7 priority oblasts, State Service for Education Quality assessments
• UNFPA GBV Assessment: February–April 2024, national coverage, GBV incident trends
• ADM2 CSV Datasets: Administrative unit-level facility and service access data, 140 districts
• UNICEF HAC Annual Report 2024: Year-end response data, outcome indicators, financial requirements

Limitations and Caveats