Attacks on healthcare in the Syrian conflict

The various mechanisms used by humanitarian health responders, human rights agencies and legal actors in Syria report high incidence of AHCC and indicate similar trends. Actors involved in reporting AHCC in Syria emphasize the importance of multidisciplinary efforts with a strong focus on accountability in any reporting mechanism for AHCC.

Mechanisms for reporting AHCC in the Syrian conflict

Using a literature review, online searches and key informant interviews, this study has identified and analysed all available reporting mechanisms of AHCC in the Syrian conflict that functioned between March 2011 and June 2018. Based on key informant interviews, peaks of AHCC incidents in Syria happened in the summer of 2012 and again in July 2014 with the intensification of the armed conflict. Throughout 2015 and 2016, AHCC incidents took place regularly, with an unprecedented peak in the fourth quarter of 2016 concurrent with the Battle of Aleppo. In 2017, following the conclusion of a de-escalation agreement, there was an overall decrease in AHCC incidents despite the Khan Shaykhun chemical attack in April. A further peak resembling that of the Battle of Aleppo was observed when the Syrian government took control of opposition-held eastern Ghouta in the first quarter of 2018.

A variety of actors were involved in reporting AHCC, but with substantial differences in relation to the purpose of the reporting, the methodologies used and the geographical coverage. Table 1 lists all the identified reporting mechanisms, alongside basic features of each one.

Table 1. AHCC reporting mechanisms in Syria from 2011

Date	Mechanism	Institution	Geographical focus of AHCC reporting	Main source of data26
Mar. 2011–	PHR	PHR	All Syria – Global coverage	Secondary
Jun. 2011–	VDC	Violations Documentation Center	All Syria	Primary
Jun. 2011– (AHCC: Nov. 2016–)	SNHR	Syrian Network for Human Rights	All Syria	Primary
Feb. 2014–	SAMS	SAMS	Opposition- controlled areas	Primary
Apr. 2015–	SiND	Insecurity Insights	All Syria – Global database	Secondary
2015–17	Annual reporting	Safeguarding Health in Conflict	All Syria – Global database	Secondary/ Primary
Nov. 2015–Mar. 2018	MVH	WHO-led Health Cluster in Gaziantep, Turkey	Opposition- controlled areas	Primary
Mar. 2018–	SSA	WHO	All Syria	Primary and secondary
Not known	HCiD	ICRC	Syrian government- controlled areas	Primary
Not known	MRM	UNICEF	All Syria and Syrian government focus	Secondary

Three principal types of actor are involved in reporting AHCC in Syria: humanitarian health responders; human rights agencies; and legal actors. Data practices – i.e. the collection, verification and use of data – reflected the objectives and purpose of the governing agency. While reporting by health actors tends to be more representative and extensive in documenting all alleged incidents, reporting by legal actors used more rigorous, case-by-case approaches to establish proof and intentionality and to identify perpetrators. While many human rights organizations had general reporting on human rights violations, some, such as the Syrian Network for Human Rights (SNHR), created a special reporting tool for AHCC.

PHR’s reporting played a key role in bringing global attention to the issue of attacks on healthcare facilities, and encouraged local and international health actors to engage in reporting such attacks.

The first organization to report on AHCC in a systematic way was PHR, which did so from the onset of the conflict in April 2011. PHR’s reporting played a key role in bringing global attention to the issue of attacks on healthcare facilities, and encouraged local and international health actors to engage in reporting such attacks. In the first three years of the conflict, AHCC incidents were reported as human rights violations or IHL infractions by local organizations, such as the Violations Documentation Center (VDC) and the SNHR. From 2013, PHR began building the capacity of local health and human rights actors to help expand and improve the reporting and documentation of AHCC. SAMS, which established its reporting mechanism in 2014, was considered the leading local medical actor in reporting AHCC. The Syrian health directorates also played a key role in documenting and archiving AHCC incidents in each governorate and supported the health cluster efforts. While a wide range of agencies and organizations participated in the collection and publication of data on AHCC, efforts were not made to establish an overarching dataset combining all reports.

An analysis was conducted of the methodologies used across the above-mentioned mechanisms, in order to assess which held most relevance for comparative study. Five mechanisms emerged as the best candidates. The prioritization was performed by means of an unstructured process, based on numerous factors: these included specificity both to the context and to AHCC, wide geographical coverage, representativeness, completeness and the strictness of verification processes. The five mechanisms thus identified for further comparative study were:

MVH
PHR
SAMS reporting
SiND
SNHR reporting

A comparison of AHCC reporting mechanisms in Syria

A comparison was conducted using governorate-level monthly data from the mechanisms listed above. All datasets were publicly available with the exception of the SAMS dataset, which was obtained through a data-sharing agreement. The data then were cleaned and merged within a single set. A basic descriptive analysis was conducted, followed by a multivariate test to establish continuity and discrepancies between the constituent datasets. Due to the heterogeneity of these datasets, statistical approaches have been accompanied by a visual presentation of the study’s findings for the sake of clarity and comparison. The software used for the quantitative analysis was Stata 15.

Table 2 summarizes the main descriptive analysis. By running a multivariate test on the means, some weak evidence emerged to suggest that the data varied between different mechanisms (P value = 0.0663). This difference could be attributed to various factors. These factors will be discussed in the following sections.

Table 2. Descriptive analysis of country-level results

Period	Reporting method	Number of months (observations) included	Total number of incidents	Min.–max.	Mean	Interquartile range
Mar. 2011–Dec. 2017	PHR	81	492	0–16	6.01	0–15
Feb. 2014–Dec. 2017	SAMS	48	463	0–41	9.65	0–27
Apr. 2015–Oct. 2017	SiND	31	434	0–51	14	1–33
Jan. 2016–Dec. 2017	MVH	24	267	0–30	11.29	3–22
Nov. 2016–Dec. 2017	SNHR	15	377	6–56	25.13	11–43

Despite the differences between these reporting mechanisms, there remains an approximate symmetry across them. As Figure 1 shows, all mechanisms indicate synchronized peaks of AHCC incidents, especially during military events that have attracted wide media attention such as the invasion of the eastern section of Aleppo city by Syrian government forces in 2016. Also, the periods with minimum incidents tend to be concurrent across all mechanisms, synchronized with the implementation of cessation of hostilities agreements, for example in March 2016 and the summer of 2017. A study published by the open-access journal PLOS Medicine in 2018 compared individual incidents of AHCC in Syria as documented in two reporting mechanisms (SAMS and PHR) and found that there was some overlap.27 Mechanisms that use primary data sources tend to capture more incidents than those that rely solely on secondary sources. This suggests that secondary data, such as media reports, miss a proportion of AHCC incidents.

Figure 1. Frequency of reported AHCC incidents in Syria (2011–18), by the five leading reporting mechanisms

— Source: https://syriamap.phr.org; primary data collected by SAMS staff; http://insecurityinsight.org/services/the-data-base; monthly reports of the MVH mechanism produced by the WHO office in Gaziantep, Turkey, via https://www.humanitarianresponse.info; monthly reports of the SNHR accessed via https://sn4hr.org.

A similar, and even more striking, symmetry is noticeable in the reporting of the numbers of medical staff killed in AHCC events (Figure 2), with synchronized peaks and minimums across all reporting mechanisms. Yet the peaks here are sharper according to the reporting mechanisms using secondary data, such as SiND, compared with those using primary data, such as SAMS. This might indicate that secondary sources tend to report higher numbers of deaths among medical staff compared with primary sources, which can better verify this type of data.

In contrast to the pattern observed concerning the frequency of AHCC incidents, the earlier phases of the conflict show higher frequencies of killings of medical personnel than the later phases. The reason behind this contradiction might be the introduction of various protection measures, ranging from collective measures such as site selections and fortification of health facilities, alarm networks and evacuation plans, to individual measures related to medical workers’ behaviour before, during and after attacks.

Figure 2. Frequency of reported killings of medical staff in AHCC incidents in Syria (2011–18), according to the five leading reporting mechanisms

The geographical coverage of each mechanism depends largely on their primary source of information. For example, as SAMS operated primarily in northwest Syria (in particular, in Idlib governorate), it is more likely to report incidents in these areas than in eastern governorates such as Raqqa and Deir ez-Zor. In contrast, SiND, which relies mainly on secondary data, reports more rarely on Idlib governorate and much more frequently on Raqqa governorate, which was the principal seat of power for ISIS during the studied period.

Figure 3. Percentage of AHCC incidents per governorate per mechanism, 2016–17, according to four reporting mechanisms28

Regarding the geographical distribution of AHCC incidents, all reporting mechanisms indicate that most incidents took place in the areas that are severely affected by conflict. In 2016–17 the highest numbers of incidents were reported in the governorates of Aleppo and Idlib. However, if the period under consideration is extended to include 2018, Rif Dimashq governorate would show a similarly high cumulative frequency of AHCC incidents, with more than 40 such incidents having been recorded during the Syrian government’s invasion of eastern Ghouta in the first quarter of that year. According to PHR, the vast majority of AHCC incidents during the conflict happened in opposition-controlled areas, with more than 90 per cent of these incidents being committed by the Syrian government and its allies.29

Figure 4. Frequency of AHCC incidents per governorate per mechanism, 2016–17, by four reporting mechanisms30

The AHCC reporting mechanisms analysed in this study vary in terms of several conceptual and methodological elements. The elements identified in this study, each of which are discussed in section 5, are: taxonomy (i.e. the terminology and definitions used); the source of the data collected (i.e. primary/secondary) and the verification process; and the purpose of reporting.

Who should be responsible for reporting?

During the workshop conducted in Gaziantep, Turkey in June 2018, for the purpose of this research, 30 participants – mainly health workers in local networks and NGOs, together with three WHO health cluster staffers – were asked to indicate their preference for who, or what type of agency, should be involved in AHCC reporting in opposition-controlled areas in Syria. This was to better understand the overall sentiment related to the relevance of reporting AHCC and the purpose of collecting the data. Participants ranked the relevance of each role with respect to reporting AHCC from zero to five, with zero being least important and five most important. Results are shown in Table 3.

Table 3. Priority actors in reporting AHCC in opposition-controlled areas in Syria, as ranked by workshop participants

As Table 3 shows, the participants broadly prioritized the role of legal bodies in reporting AHCC. Most of them expressed the view that reporting AHCC should involve actors who have the mandate and influence to bring perpetrators to justice. While participants emphasized in the accompanying discussion the importance of involving local actors and NGOs in reporting AHCC, they did not see these actors as having any influence on accountability.

These findings support the argument that reporting AHCC should be a multidisciplinary effort, involving actors from different sectors. They also emphasize the perceived importance of international bodies and demonstrate a strong desire for accountability. Given that the sample of participants in this workshop was small, non-randomized and not representative, these findings could be further assessed through context-specific stakeholder mapping and followed up using quantitative and qualitative methods to identify priority actors.

A primary source is defined as immediate, first-hand accounts of a topic from people who had a direct connection with it. A secondary source is defined as a reply based on other organizations’ reporting, news and media or social media.

Haar, R. J., Risko, C. B., Singh, S., Rayes, D., Albaik, A. et al. (2018), ‘Determining the scope of attacks on health in four governorates of Syria in 2016: Results of a field surveillance program’, PLOS Medicine, 24 April 2018, https://journals.plos.org/plosmedicine/article/comments?id=10.1371/journal.pmed.1002559.

SNHR does not provide data disaggregated by governorate.

PHR (2021), ‘Physicians for Human Rights’ Findings of Attacks on Health Care in Syria’.

SNHR does not provide data disaggregated by governorate.

Topic themes

Defence and security

Economics and trade

Environment

Health

Institutions

Major powers

Politics and law

Society

Technology

Regions

Africa

Americas

Asia-Pacific

Europe

Middle East and North Africa

Russia and Eurasia

Events

Publications

Become a member

The Queen Elizabeth II Academy for Leadership and the Next Generation

About us

Dr Abdulkarim Ekzayez

04 Analysis of AHCC reporting in the Syrian conflict