The impact of armed conflict on vaccination coverage: a systematic review of empirical evidence from 1985 to 2025.

Abstract

Background: Armed conflict disrupts health systems and undermines routine immunization, contributing to excess morbidity and mortality. This systematic review examines empirical evidence on the impact of armed conflict on vaccination services and coverage, identifying patterns of disruption across geographic settings and conflict types.

Methods: This study followed PRISMA guidelines and was registered in PROSPERO (CRD420251064804). We searched seven databases for peer-reviewed and grey literature (1985-2025) reporting quantitative comparisons of vaccination coverage before and after conflict onset, or between conflict-affected and unaffected populations. Screening and data extraction followed standardized systematic review protocols, with dual validation of a subset of studies. Due to methodological heterogeneity across studies, a meta-analysis was not conducted.

Results: Of 8,043 citations screened, 33 met the inclusion criteria. Most focused on child immunization in settings across the Eastern Mediterranean (15, 45%) and African (12, 36%) regions. Data sources included household surveys (22, 67%) and health system records (8, 24%). Conflict exposure was most commonly measured using battle-related deaths (15, 45%). Analyses employing individual-level data were most common (10, 30%), followed by subnational administrative data (9, 27%). Nearly all studies (31, 94%) were observational or quasi-experimental. In 28 (85%) studies, conflict was associated with reduced vaccination coverage, sometimes exceeding 20% points for vaccines such as BCG, DTP, and polio. Declines were most pronounced in settings with civil war and moderate to high conflict intensity. Two studies reported localized increases in vaccination coverage, possibly due to targeted humanitarian interventions. Effect estimates were larger in studies using national or administrative-level data compared to those using household-level data, underscoring methodological variation as a key contributor to heterogeneity in reported impacts.

Conclusions: Armed conflict is consistently associated with substantial declines in childhood vaccination coverage, most pronounced in civil war and military occupation settings and across conflicts with moderate-to-high annual BRDs. Regional disruptions were especially severe in the Eastern Mediterranean and sub-Saharan African regions. We found substantial variation in estimated effect sizes across analytic units (individual, household, region, country), suggesting that more aggregated data may better capture the broader impact of conflict on vaccination rates. Future research should incorporate standardized conflict and vaccination metrics to improve the generalizability of findings.