Initiation of the research at RSP

The idea of an anthropological assessment had been discussed at MSF for some years before my arrival in Amman. As the RSP entered its tenth year, in 2017, the organization’s determination to make an assessment that went beyond an evaluation of its medical aspects intensified. The mutual interest of the field and desk managers and CRASH, MSF’s centre for research and reflection, drove the launch of a deep, multifaceted research project. The programme managers wanted a more operational focus in the research, while members of the CRASH, some of whom were involved in the RSP from its beginning, had a long-standing interest in its development.

Multiple concerns about patients’ wellbeing in the hospital and after they returned home required answers. How did hospital staff and patients relate to one another and how much of their interaction was conducive to the patients’ recovery process? Were there any gaps in providing a truly healing environment for the victims of war? The RSP had been predominantly focused on improved mobility and functionality of patients’ limbs. To what extent did these programme objectives improve patients’ lives after hospitalization? How was their physical, emotional, social, and economic wellbeing after they returned home?

The research came at a time when the concept of a patient-centred approach was flourishing at MSF, and the RSP had declared it one of its main strategies. However, “patient-centred care” is a widely used but poorly understood concept in medicine, including within MSF operations. Despite attempts to develop a universal definition of this term (Stewart, 2001), there is currently no general consensus on what patient-centredness might mean or how it should be practised. Some of the literature focuses on patients’ needs and values (Kvåle and Bondevik, 2008), while others emphasize the involvement of a patient’s family members (Epstein and Street, 2011). Still more explore interactions between patients and healthcare providers (Berwick, 2009; Bauman et al., 2003; Epstein and Street, 2011). In the absence of a clear definition, I decided to take a unique approach and to focus my research on the patients’ perspectives and to ascertain what was important to them. The underlying goal was for these assessments to become the core element driving operational change in the future.

Additional institutional enquiries in the research had a more strategic lens. Could the considerably high budget that MSF allocated to the RSP (10 million EUR per yearThis is one of the highest budgets for a single programme within MSF operations. It is comparable to budgets required for specialized surgical hospitals in Europe or the US. ) potentially be used more effectively in other humanitarian aid projects? The very sustainability of the RSP was in question. An interesting and challenging task awaited me!

Methodology

I conducted the research between September 2017 and December 2018. An exploration of views of both the healthcare providers and the patients was needed, and I designed a research plan as a twofold process: in the first part of the investigation I focused on relations and perceptions between hospital staff and patients; in the second part I assessed the outcomes of the RSP from the patients’ perspective. This process resulted in comprehensive original data from 173 formal interviews with patients and MSF staff, as well as extended field observations. The information was further enriched with the review of internal MSF documents.Triangulation was achieved using the mix of methods described above and debriefing the research assistants/interpreters after each interview.

Interviews with MSF personnel and observations in the hospital

My research started in Al Mowasah hospital, where over a six-month period I observed interactions between hospital staff and patients, and explored staff perceptions of patients. Upon arriving in Amman in April 2017, I was soon integrated into the hospital microcosm and was invited to participate in department meetings, celebrations, and events. Daily observations of hospital life also took place during medical rounds, surgical interventions, in the hospital cafeteria, and in the outdoor recreation space.

Furthermore, my office became a place where staff members would stop to visit, bring me homemade desserts, show me photos of their families, chat about their weekend activities, and so on. At times, they also asked for confidential advice on issues such as how to resolve a conflict with a colleague. These visits to my office included descriptions of professional aspirations, gossip about colleagues and patients, chats about difficulties in their countries, and worries about possible complications arising from events in the operation theatre. This insider’s position allowed for interaction with staff both formally and informally.

My interactions with patients in the hospital were mostly based on observations of their daily life and limited chats in Arabic (greetings, enquiries about their names and country of origin). Those who spoke some English would occasionally talk about their injuries, how they acquired them, and their progress after surgery. They also expressed curiosity about where I was from, what my country was like, and my family.

In the first period of study I conducted daily observations of hospital life, following doctors and surgeons on their medical rounds, spending hours in the operation theatre and physiotherapy sessions, in medical consultation rooms, and in social spaces such as the hospital cafeteria and the outdoor recreational space for patients. The purposes of these observations were to establish a base of knowledge of medical procedures and to observe how patients and their caregivers interacted with each other.

In a hospital like Al Mowasah, the diverse backgrounds of patients and the hospital staff (Iraq, Syria, Yemen, Jordan, European countries, Peru, the US, and Canada) created an interesting ground for exploration. Would nationality matter when it came to perception? Would the hospital staff’s perceptions of patients be in line with views that patients have about their own health?

Such questions led me to interview MSF staff and examine their opinions. The in-depth interviews I conducted with ninety-nine members of staff The sample size in qualitative research is not pre-determined. The interviewing process is considered complete when a so-called data saturation point has been reached. Data saturation is achieved when no new information emerges from additional data collection (Richie and Lewis, 2003). represented almost half of all the employees (209) at the RSP. Staff who participated in the interviews had medical and non-medical profiles, and I included expatriates and non-expatriates. I interviewed representatives from all hospital departments, including administration and managementProgramme managers, researchers, HR and finance departments, communications department. (19), hospital supportLogistics department, laboratory staff, pharmacy staff. (20), paramedical Physiotherapy and psychosocial departments. (20), medicalDoctors, nurses, and medical-department supervisors. (27), and surgicalSurgeons, surgical assistants, operating-theatre nurses, anesthesiologists. (13).

The focus of the interviews was on participants’ views on their professional role in relation to patients, their personal experiences with the patients and their perceptions of different patient groups (adults vs. children, differences between Iraqi, Syrian, and Yemeni, male vs. female, etc.). Depending on the preference of participants, the interviews were in English or Arabic, with translation provided by my assistant.

Selection of participants from among the patients

After the period of assessment in the hospital, my focus shifted to the patients, who I also aimed to interview during home visits. The process started with purposeful selection of potential candidates from the patients’ hospital register. The criteria I followed when selecting appropriate candidates were as follows: first, I focused on adult patients from Syria who now live in Jordan, and Iraqi patients who had returned to Iraq; second, preference was given to those who had had multiple procedures and a longer stay in the hospital; third, the candidates had to have completed the programme and been discharged after 2012. I did not select candidates from earlier in the programme in order to reduce recall biases – errors occurring in selective, incomplete, or inaccurate reporting of past events (Richie and Lewis, 2003).

I also left out those patients who were discharged less than six months prior to the study, as they were more likely still to be undergoing re-adjustment at home; and I excluded persons younger than eighteen (on the day of the interview) as the impact of treatment is different for adults, and it was beyond the scope of my project to explore in detail the situation for young people. Yemeni patients too were excluded from the study even though they represented a large group of patients in the hospital register. The security situation in Yemen at the time of the fieldwork prevented me from travelling to that country and thus there was no way I could gather comparable data from them.

After the selection was completed, I contacted patients by phone with the assistance of interpreters. We found that about two thirds of the registered phone numbers no longer functioned. In a few cases, relatives reported that the patient had moved abroad. However, none of the patients we reached refused the interview. In all, we reached seventy-four patients from Syria and Iraq (fifty-four men, twenty women) who were able and willing to participate in the interview and who had undergone a range of surgical procedures: orthopaedic, plastic, and maxillofacial (see Appendix). We set the dates and venues for our meetings with them.

Fieldwork

I travelled with my interpreter around Jordan to meet the thirty-nine Syrian participants, who had originally come from Daraa, Homs, Damascus (including Eastern Ghouta), and Aleppo. All of them now lived in Jordan as refugees, but many had resettled from the refugee camps to private accommodation. The majority received us in rented apartments, mostly in suburbs of Amman, or in rented houses in rural areas in the north of Jordan. Their homes differed in the size and amount of furnishing, from very simple (lacking any source of heating in winter, for instance) to more sophisticated (equipped with flat-screen TVs, sofas, and carpets). One of the families we visited in southern Jordan resided in a tent that they moved north in summer and south in winter to avoid the high cost of heating.

Initially, I expected to spend much of my fieldwork in the refugee camps in Jordan, but only three participants still resided in Za’tari refugee camp, living in metal containers. The Za’tari camp is organized in lines of these containers, which are separated by mostly unpaved streets. Small shops with basic items and NGO offices are located between residential containers. The entire area of the Za’tari camp is fenced and only accessible at entry checkpoints. Each family unit that we visited had one or two containers arranged next to each other. There was a tiny kitchen, a small bathroom, and family living space. Each family unit had a water tank. Electricity was provided for limited hours a day, but one family had a small generator to cover the electricity gaps.

The second part of my fieldwork took me to Iraq. When I first arrived in Iraq I was struck by the reminders of war that could be seen on every corner, now mingled with ordinary life. I was impressed by the nicely manicured bushes along the highway, a smell of cardamom spiced coffee spreading through the hustle and bustle of the streets, and shopping malls filled with upper-class Baghdadi residents who were shopping, chatting in the cafés, eating delicious meals, or smoking sweet-smelling shishas. But military checkpoints, high walls, observation towers, and body searches at the entrances of buildings were routine, standing in stark contrast to ordinary day-to-day Middle Eastern life – a clear reminder that existence in Iraq is far from peaceful.

Compared to Jordan, fieldwork in Iraq was more challenging. We conducted interviews during Ramadan (May, June 2018), when fasting made patients reluctant to travel a great distance. In addition, security rules were stricter in this period, which was right after national elections, and hence our movements were limited. For each participant who agreed to be interviewed, we needed to carry out additional security assessments to confirm that their residence was relatively safe. Sometimes participants did not feel comfortable sharing the location of their residence, and they requested that the interview take place at MSF premises. Eventually, we were only able to visit eleven in their homes and we met the other twenty-four either in a MSF-run hospital in Baghdad or at the Fallujah hospital in Anbar.Transport reimbursement was provided to participants who travelled to meet the research team.

We conducted fieldwork in Baghdad and Al-Anbar provinces. Some participants from other provinces (two from Babylon and two from Diyala) travelled to meet us in Baghdad.Babylon was relatively stable at that time, but Diyala still had many security incidents. In the month of June 2018, it was the worst affected governorate, with fifty-two civilian casualties (sixteen killed, thirty-six injured) (UNAMI, 2018a; 2018b).The home visits we carried out in Baghdad were in the districts/neighbourhoods of Dara, Adhamiyah, Sleikh, and Saydia. These neighbourhoods are predominantly occupied by Sunni families, and the residents are middle to upper-middle class. The neighbourhoods we visited varied from those that looked obviously more upscale, characterized by large houses and wide, clean streets, to those with houses under construction, rubbish on the streets, and plots of agricultural land stretching along the railways. Regardless of the area, all of the houses were well maintained and equipped. These areas of Baghdad were controlled either by the Iraqi army, the federal police, or Kurdish security forces. According to the MSF security in charge, the level of incidents (explosions, kidnaping) was relatively low in these districts (two to three per month) during the period of fieldwork.

The areas of Baghdad where we were denied access due to security restrictions (New Baghdad, Shuala, Sadar City, Abu Ghreb, Shaab, Mahmudia, Rifak, and Alan) are mostly occupied by Shia or mixed communities. Mostly lower-economic-status to middle-class families reside in these areas. Despite security forces operating there (federal police or the Iraqi army), the MSF security advisor commented that the areas were still controlled by the sectarian militia and may have Islamic State of Iraq and the Levant (ISIS) sleeper cells. Security incidents were still frequent (daily explosions, shooting, kidnapping), and, thus, we stayed away from these neighbourhoods.

The area we visited in Anbar had a rural, green feel to it, with date-palm plantations and some patches of agricultural land stretching between them. Anbar was drastically affected during the US invasion and ISIS control. There were still security incidents reported daily, particularly from the Western Anbar (Iraqi News, 2018). Evidence of the recent mass destruction was obvious and was an ever present part of our drives through the area. Half-demolished buildings were everywhere, and bullet holes were visible on many of the intact buildings. Roads were in relatively good condition but were characterized by frequent military checkpoints, where we needed to exit the car and show our IDs before we were allowed to pass. On the main highway every day, I observed passing military convoys armed to the teeth.

In Anbar we conducted home visits in Ramadi town, its rural suburbs, and Falujah town. Here the houses varied widely from very comfortable upper middle-class houses to improvised shelters covered with iron sheets, with cupboards instead of windowpanes, and a lack of furnishings. One participant resided illegally in an abandoned government apartment. We conducted one of the interviews in the MSF car, due to the participant’s embarrassment at the condition of his significantly damaged house. He resided with his relatives. We were restricted from visiting participants in Garnea, which was still considered unstable, so we arranged to interview them in the Fallujah hospital. My impression from fieldwork is that Anbar maintains a better sense of social cohesion than Baghdad, but the long-term destruction of the physical environment is more evident there.

Home visits and the interviews with MSF patients

My aim was, as far as possible, to interview our participants at home in the hopes of making them feel at ease and to observe their living situation and their interactions with family members and friends, important elements that bear on their daily reality and reveal the truth of their life with disability. The central focus of the interviews was participants’ descriptions of the events that led to the injuries, their experiences of treatment at MSF and with other health providers, and their perception of their quality of life and wellbeing after the treatment at the RSP. The interpreters who accompanied me in the fieldwork assisted me with direct interpretation from English to Arabic and vice versa.

I recorded all interviews on a digital voice recorder, which in total exceeded eighty-six hours of interviews, averaging an hour and ten minutes per participant. My field team and I spent at least an additional hour and a half per participant in introductions, informal chatting, reassurance, responding to questions, and social interactions with family members. I also carried out supplementary observations (participant observationsParticipant observation is a core method in anthropology that offers researchers the opportunity to gain additional insights through experiencing and observing phenomena under study (Richie and Lewis, 2003).) – for example, of patients’ physical and social environments – and noted them during home visits. To determine socio-economic profile of participants (details in Appendix) I completed a short socio-economic questionnaire for each participant at the end of the interview.

We were received with generous expressions of hospitality during our home visits. After initial greetings from family members, we were invited into the salon, typically an entry room of the house/apartment, furnished with ground mats and pillows, where visitors are usually seated. In Jordan we were frequently served tea or Syrian-style coffee. On occasion, a traditional breakfast (bread, olive oil, olives, cheese, stuffed aubergine, and sweet pastries) was brought to us after the interview. Iraqi participants often invited us to stay for iftar, an evening meal that breaks the fast during Ramadan. In Anbar, as a part of the expected dress code, I wore a hijab (a head scarf), and this was greatly appreciated, particularly after I clarified that I am not a Muslim.

Often, family members (spouses, parents, and children) were present throughout the interview; occasionally, they would enter the room during the home visits. In order to respect confidentiality, we confirmed with participants that this arrangement was acceptable to them, and they usually responded along the lines of “After what we went through together there are no secrets between us.” The family members present often acted as the patient’s caregiver, and they added important information to the patient’s narrative. Such information might otherwise have been missing, because participants were often unconscious right after the injury. Caregivers provided insight into changes in the participants’ behaviour over time. Family members also accompanied those participants who came to meet us at MSF premises.

I was concerned that the interviews focused on memories related to war would cause emotional pain to the participants. Before we started the interviews, I reassured them that they were under no obligation to respond to my questions and should share only the details they felt comfortable sharing. I put forward my best effort to be as sensitive as possible, observing participants’ reactions to my questions. If I observed any sign of distress, I asked if they wished to continue. When talking about the war context and the circumstances under which the injury occurred, my questions were never specific to the events but, rather, open-ended, such as “And then? And what happened after?” With this strategy I aimed to place participants in control of the narrative and the details they wanted to remember and share.

Some participants communicated very clearly that “they did not mind my questions, but they did not want to remember related traumatic events.” I respected this, and on such occasions shifted to another line of questioning. In contrast, some insisted on describing, quite graphically, the events of violence. Additionally, some showed me videos taken during the event that caused the injury. I still wonder if these video examples were used to communicate that this was the participant’s destiny. Or whether they were an attempt to make me a witness to their suffering, which extended beyond their bodies to those who were killed in front of them.

Surprisingly, despite many of participants still experiencing a high level of emotional distress during the interview, many of them communicated gratitude that “somebody took time and was willing to listen to them,” and their mood seem to lift by the end of the process. Their eagerness was obvious in some of their comments. For instance, “I am happy that you came here to my home and asked me questions. Even if you want to interview me every day, I am ready” (RSP9, Syrian, male (M)). Or another participant, who was a political prisoner commented: “I’m sorry I bothered you with my story while you have nothing to do with this information. I just wanted to tell you my story; I honestly don’t know why [smiles]” (RSP46, Iraqi, M).

Analysis of the interviews

After we completed the interviewing process the audio recorded material was prepared for analysis. I used a thematic-analysis approach (Richie and Lewis, 2003), applying the following steps. My field assistants transcribed hours of audio records into Word documents. After the transcription process was finished, I read the transcripts to develop a coding system. The transcripts were coded using MAXQDA software. After coding was complete, segments of the data were organized into tables (matrices) and further examined to identify patterns (themes in the thematic-analysis approach).

Figure 3. Patient at home. Male relatives gathered around their injured family member.

Figure 4. Researcher conducting interviews. The researcher was usually placed opposite the participant and recorded the interview on a digital voice recorder.

When analysing data from the interviews with the hospital staff I noted any similarities or differences in the views of the participants from different hospital departments or different geographical regions. In analysis of the patients’ interviews I made a comparison between participants living with similar disabilities in different contexts. For example, I compared Syrian refugees in Jordan with residents in Iraq and/or the outcomes for male and female participants. Each theme was described in detail and illuminated by using participants’ direct quotes. I analysed information collected by questionnaire in Excel to determine the average socio-economic status of participants (Appendix).

Processes to ensure good quality of data

Having only a one-off opportunity to record each interview, it was essential to ensure good quality of data. This required accurate translation and transcription, and as both processes were performed by research assistants, I paid special attention to their selection. I carefully evaluated each applicant’s research background, their fluency in both Arabic and English, and previous experience in humanitarian settings. During the selection process potential candidates underwent written tests of their translation and transcription skills. The tests were designed to assess the candidate’s written and oral translations and computer-based transcriptions. Texts in English and in Arabic (one page in each language) were translated. Subsequently, their work was compared to originals translated by a professional interpreter to assess the accuracy of translation and efficiency of time management.

Oral interpretations were tested by providing candidates with short sections of written text (in English and Arabic) for them to translate into either Arabic or English. When translation was given in Arabic, a third person translated it back to me in English. This was, again, compared with the original written text. Transcriptions were tested by giving candidates instructions and audio recordings of two short dialogues in English to be transcribed. The accuracy of the transcripts (how many words/phrases were missing, any changes in meanings) and time management were assessed and scored.

The interpreters who were ultimately selected were further trained during a two-day workshop that I organized and ran. We covered specific behavioural guidelines for fieldwork, research ethics, and the main principles of qualitative research. Explanations, illustrative videos, and written materials were provided.

Before starting the fieldwork and interviewing patients we pre-tested an interviewing guide and socio-economic questionnaire on volunteers (patients) in the hospital. Through this process, the research assistants practised translations while interviewing and had the opportunity to check if any words used by Syrian volunteers appeared difficult to understand. In such cases, those words were double-checked with a Syrian member of staff (we took this approach to all interviews). Volunteers gave us feedback on the clarity of the questions and their sequence. The topic guide, which was ultimately used for interviews, was then revised to include the suggested improvements.

When a research assistant started transcribing the interviews, I often randomly compared parts of the transcriptions with the audio recordings. If there was a difference, we discussed why this had occurred and how to correct for it in the future. I repeated this process of cross-checking until we achieved good-quality transcriptions. To achieve a consistent level of translation and to double-check the on-site translations, some interviews were transcribed by an interpreter who had not been involved in the interview process.

The entire research project received ethics clearance from ethics committees at Comité de protection des personnes Sud-Ouest et outre-mer III, Bordeaux, France, and from Al Mowasah hospital, in Amman.

By implementing these rigorous steps, I was able to collect, document, and analyse a trove of rich information portraying the life of war victims and the staff providing their care. We will now continue and look into the daily life that unfolds inside the MSF hospital and that carries with it unique emotional bonds between the people behind the hospital walls.