Exploring relationships between conflict intensity, forced displacement, and healthcare attacks: a retrospective analysis from Syria, 2016-2022.

Introduction: Attacks on healthcare have been committed throughout the Syrian conflict in violation of International Humanitarian Law (IHL), contributing to the devastation of the country's healthcare system. The conflict has also forcibly displaced over half of Syria's pre-conflict population, 7.2 million of whom are internally displaced. In this retrospective analysis, we aim to assess the relationships between (1) healthcare attacks and general conflict and (2) healthcare attacks and forced displacement between 2016 and 2022.

Methods: Data on healthcare attacks, conflict events, and displacement were extracted from the Syrian American Medical Society (SAMS), Uppsala Conflict Data Project, and OCHA Türkiye, respectively. The analysis addresses three research questions: the associations between (1) healthcare attacks and conflict events, (2) healthcare attacks and conflict events in the week after an attack on a healthcare facility, and (3) healthcare attacks and forced displacement. For each, we used generalized additive models with a negative binomial distribution that also accounted for spatial and temporal factors.

Results: SAMS recorded a total of 541 attack events, comprising 650 attack rounds over 235 facilities between 2016 and 2022. Conflict events were significantly associated with healthcare attacks in the same week (IRR: 1.14, 95% CI 1.12-1.17), and healthcare attacks in one week were associated with a maximum of 1.44 greater risk (95% CI 1.08-1.91) of conflict events in the following week, even when accounting for general conflict levels in the previous weeks. Healthcare attacks were also significantly associated with increased displacement up to three months following the attacks.

Discussion: We find that healthcare facilities are not avoided during conflict (as obliged under IHL), and that healthcare attacks significantly precede an escalation of general conflict in the same area. Healthcare attacks are also significantly associated with displacement for months following the attacks, even when accounting for conflict levels. Based on these findings, we present a framework outlining one pathway through which healthcare attacks may contribute to larger conflict tactics. Our findings highlight the critical role of healthcare infrastructure in conflict and reaffirm calls to hold perpetrators of these attacks accountable.