Questionnaire.

Prior to each FGD, refugee participants were asked to complete a short questionnaire with the help of the research team. This method triangulation increased the confirmability and dependability of the findings, thus its rigor [36]. The moderator assisted illiterate participants in completing the questionnaire.

The questionnaire (S1 Appendix) included sociodemographic information (age, education, casual labor, household conditions), and the Arabic-translated, previously validated version of the Household Food Insecurity Access Scale (HFIAS) [37,38]. The sociodemographic questionnaire, including the HFIAS scale, were used in previous studies conducted in Lebanon with Syrian refugees and Lebanese households showing good and reliable results [26,39,40]. The HFIAS reflects the food insecurity experience of the participants and their family members in the past four weeks, and it is comprised of nine questions on the occurrence and frequency of food insecurity experience during that period. To calculate the HFIAS score per household, a standard scoring procedure was used whereby 0, 1, 2, 3 points were allocated for “non-occurrence”, “rare”, “sometimes”, and “often”, respectively. The total HFIAS score ranged between 0 and 27, with higher scores indicating higher levels of household food insecurity. Scores were used to categorize households into four levels of food insecurity (food secure, mildly, moderately or severely food insecure) depending on the number of positive responses to questions related to severe conditions, and as per the HFIAS measurement and indicator guide [25]. For the purpose of analysis in this paper, the food secure and mildly food insecure households were merged. In addition, the short questionnaire asked respondents about the frequency of their household’s consumption of eight different food groups over the previous seven days to calculate their food consumption scores (FCS). The FCS is a composite score based on the dietary diversity, food frequency, and relative nutritional importance of different food groups with higher scores referring to higher dietary diversity and frequency of consumption among households [41]. The FCS was calculated for each household and later categorized into three groups (poor, borderline and high), as per the standard thresholds [27]. Briefly, the FCS is an index that was developed by the WFP in 1996 and it has been incorporated since then into surveys to assess the food security and dietary diversity of vulnerable communities and households [42]. In Lebanon, the FCS represents a key component of the WFP annual surveys conducted to assess the food security status of Syrian refugees [3,10,43].

Focus groups.

Liamputtong’s traditional approach [44] for conducting focus groups was adopted in the present study. For cultural appropriateness, gender congruence was observed, i.e. men moderators facilitated the men’s FGDs and women moderators for women’s groups. Each FGD consisted of 4 to 10 participants and was conducted in a communal room inside the ITS for a duration of one hour and a half using a FGD guide (S2 Appendix). At the start of FGDs, moderators introduced themselves and explained their roles that consisted of guiding the discussion while the observers examined the group dynamics and took notes. Concerning credibility, the moderators and interviewers conducting the FGDs and in-depth interviews shared the same first language as the participants. The decision to conclude data collection was done once we reached data saturation by group (registered with assistance, registered non-assistance, and unrecorded) and by gender within each group.

In-depth interviews.

Informed oral consent was secured from all key informants to ensure voluntary participation and confidentiality. The permission to audiotape the discussion was secured from participants prior to the start of the interview. Those who consented were interviewed by the moderators (D.N, H.I or F.AH) in a private setting at each of the designated key informant’s offices or at the Global Health Institute (GHI) offices in Beirut, except for one interview that was conducted via SKYPE to meet the interviewee’s preference. The interviews were conducted by the moderators using an interview guide (S3 Appendix). The interviewer(s) introduced themselves and re-iterated the purpose of the study in addition to emphasizing the need to understand the effect of registration status on the livelihoods of Syrian refugees. The non-participant observer recorded notes and intervened when in need to clarify certain terms or legal implications.

Qualitative analysis.

Audio recordings were transcribed verbatim concurrently with the data collection. The research team avoided translation while transcribing to observe fidelity of meanings [45]. Instead, data were analyzed in Arabic, and only the quotes that were used in the manuscript were translated from colloquial Arabic to English. After the completion of data collection, an inductive thematic analysis using Braun and Clarke approach was used for both interviews and FGD [46]. After the analysis was completed, all quotes were translated into English.

Each FGD participant was given an ID during coding. The first letters described the gender of the participants, with ‘M’ denoting men and ‘W’ for women. The second letter referred to the registration status, where ‘R’ was for registered and ‘NR’ for the unregistered. For registered participants, ‘A’ was for those receiving assistance and ‘NA’ for those not receiving assistance. The first number was the number of the focus group, and the second was for the number of the participant preceded by a P (e.g. P2). For FGDs, and when we were unable to track each participant’s statements, the participant identifier was excluded. For example, M.R.NA.1 was a man, registered, not receiving assistance from the FGD, with the exact participant unidentified. As for codes from the interviews, the source of the quote was indicated by the number of the interview and the affiliation of the key informant, for example, iNGO referring to an interview with a representative from an international NGO.

Data was thematically analyzed along six phases. In phase 1, two members of the research team read and re-read each transcript to get acquainted with the information. In phase 2, descriptive coding line-by-line for each transcript was conducted. In phase 3, an initial list of all codes was generated by the two team members (D.N. and A.H.). In phase 4, the research team discussed the relationships between codes. A log of potential themes and sub-themes was developed, including a list of definitions and quotes to illustrate each theme and sub-theme. In phase 5, the list of candidate themes and sub-themes was further refined based on consensus among all research team members and to highlight the existing relationships between them. As for phase 6, the findings were presented in a narrative form, and a synthesis of the results was included. These findings were supported with quotes from interviewees and beneficiaries relating to identified themes and sub-themes.

Quantitative analysis.

Data from the short questionnaire completed with participants at the beginning of the focus groups were entered and analyzed using Statistical Package for the Social Sciences (SPSS) version 22 software [47]. Analysis of variance (ANOVA) with Tukey’s post-hoc analysis and chi-square analysis were conducted for continuous and categorical variables, respectively. Results were reported as means and standard deviations (SD) for continuous variables and as frequencies and percentages [n (%)] for categorical variables. Statistical significance was reported at p-value ≤5%.

Integrated analysis.

Analysis of integrated findings between FGD and short questionnaire were integrated by contiguous narrative approach where the quantitative findings were presented first followed by the qualitative findings. Findings were further discussed for concordance, discordance or expansion. Integration of results was assessed by the research team [48]. Findings were reported based on Good Reporting of A Mixed Methods Study (GRAMMS) criteria (S4 Appendix) [49]. Data was analyzed manually.

Dependence on cash assistance for survival.

All registered refugees who were receiving assistance reported food and/or MPC assistance as their main and essential sources of livelihoods, “God help those not receiving assistance” (W.R.A.2: P: 10), despite being insufficient and unsustainable. Registered refugees not receiving assistance and unregistered refugees also highlighted the importance of assistance, and believed their situations would improve if they were to receive any type of assistance by UNHCR. “If I can get anything, if I can get only $260 (referring to the cash assistance), I will be able to get by”; “The registered have everything secured for them, every month they receive fuel, packages, cash, everything.” (M.NR.4). Key informants confirmed refugees’ dependence on assistance for their livelihoods and survival. However, the majority of these informants clarified that the assessments of refugees’ needs, and accordingly assistance allocation, under most projects were not based just on registration status, but also on their level of vulnerability. Key informants also echoed the fears of refugees that the current programs providing assistance to refugees were short-term, and that long-term, sustainable projects are not feasible in the Lebanese context due to donor or governmental restrictions: “You cannot do anything long-term because long-term means permanent, permanent means that [refugees] are going to stay here and we go into a political conflict” (iNGO—Int. 3).

Informal employment and exploitative work conditions.

Employment was regularly identified as an important source of livelihood among all the participants, especially among those not receiving assistance (whether registered without assistance or unregistered). In fact, all of the participants were informally employed (i.e. without work permits), mainly in construction, agriculture, and tobacco picking, with an average daily salary of $4. There were a couple of cases of women cleaning houses and one case of garbage collector, but these were only reported amongst unregistered refugees. Regardless of registration status, work opportunities were highly irregular, often provided and controlled by the gatekeeper, known as “shaweesh”, and seasonal, mostly restricted to the agricultural season in the summer. These findings were consistent with those reported by the NGO representatives, who echoed the engagement of refugees, regardless of their registration status, in informal employment without legal work permits as their source of livelihoods.

Furthermore, FGD participants reported numerous cases of work exploitation, mostly pertaining to lack of payment by their employers. One participant said: “The employer tells you to go away and I’ll come pay you later, but he doesn’t come; I mean here we are deprived of our rights and there’s nothing we can do, we have no support or back-up” (M.R.NA.1). Key informants echoed the poor and exploitative work conditions of refugees, whether through being underpaid by their employers, controlled by the gatekeepers “shaweesh”, or abused verbally and physically: “The employer rarely deals with these people. It’s usually the Shaweesh that beats them with a stick sometimes, if he doesn’t beat them, there’s the verbal abuse that they suffer all day long, and the emotional effect.” (iNGO—Int. 3). Some key informants reported that Syrian women resort to socially less acceptable jobs, such as sex work and begging: “So it is more of a coping mechanism for women who are willingly doing that, but for younger girls it is sometimes a coping mechanism and sometimes exploitation against their will” (Local NGO—Int. 9). Nevertheless, sex work and begging were not reported by refugees during the FGD.

Child labor and early marriage.

Children as young as 11 years old were essential to the livelihoods of their families, regardless of registration status. “The children help, they help the mom in work” (W.R.NA.2); “12 year olds can work, but if they are 7 years old or younger they can’t work in anything” (W.NR.2). According to key informants, child labor is widespread and normalized in Lebanon, further increasing rates of children who were out of school. One of the main reasons mentioned by key informants was that children under 15 years of age are not required to have legal papers in Lebanon, and as such they can easily cross checkpoints and are better positioned to be working than their parents.” One key informant also mentioned the preference of employers in hiring children in certain cases: “As of 6 years old they send them to work, especially I don’t know if you know about Marijuana but you have to weed it by hand, there are no pesticides, they prefer small size people, because they don’t damage the crop and they are near the ground to do it with their bare hands or with a small shovel…, imagine the smell they are smelling all day, I wonder if they get dizzy or sick” (iNGO—Int.3).

Regardless of registration status, early marriage was another subtheme that emerged from FGD, and it was considered a culturally acceptable practice among many Syrian refugees. According to the key informants, early marriage was a strategy utilized to alleviate the financial burdens faced by families, as marrying girls earlier decreases family expenses. Moreover, few unregistered men in the FGD expressed their need for a young girl to help them around the household: “There was no one to help my mother, so I got married I had to get married. My mother is an older woman, she has high blood pressure, diabetes and poor eyesight, you know what I mean… I want someone to work instead of my mom, I have 2–3 younger sisters, yes I want a girl to help my mother, I married her and now she helps her” (M.NR.4).

Accruing debt and social exchange.

All FGD participants, regardless of whether they received assistance or not, agreed that accruing debt was another source to support their livelihoods. “I swear we are all in debt, there isn’t one household who is not in debt” (M.R.A.1) said one registered man who was receiving assistance. Syrian refugees reported being in debt to shopkeepers, pharmacists, neighbors, property owners or “shaweesh” (ITS gatekeepers), and the level of debt of refugees accumulated more in winter due to lack of seasonal employment. In addition, the magnitude of debt was significantly higher amongst unregistered refugees: “We request debts to survive, up to 1000$, what can we do, we need to live regardless of excessive debt” (M.NR.1). Registered refugees with assistance who were working were able to repay some of their debts, whereas it was very challenging for their unregistered counterparts to pay off their dues, thus accumulating larger amounts of debts.

Dependence on debt was also echoed by key informants as a negative livelihood strategy and coping mechanism, particularly amongst refugees who were unregistered: “It is true that debt is common regardless of registration status, but for the others (referring to those receiving assistance) every month they pay back some of their debt, the problem is for those unregistered with the UNHCR or registered and not benefitting from assistance, their debt is very big, large, very, and this affects a lot of things such as sending their children to work, sending the wives for extra work, the type of work becomes more difficult because the number of unregistered refugees is increasing, and as such work opportunities are decreasing so they go to work on the roads in prostitution for example and other jobs. The debt is a big problem because all Syrians are now in debt” (Local NGO—Int. 9).

In addition to accumulation of debt, unregistered refugees expressed reliance on more positive mechanisms such as social exchange within Syrian refugee families and among friends and neighbors to share food and/or assistance. For example, few unregistered refugees and their registered parents, or elderly unregistered parents and their registered children reported often sharing assistance: “I have children, I let my mother and father bring me food assistance cards and diapers for the children” (M.NR.4). However, some unregistered participants living away from their families believed that only other unregistered neighbors were willing to share their food and help each other because they understood each other’s difficult situations “Today my situation is under zero for example? I have 2 cups of sugar, Fatima (unregistered neighbor) comes to borrow a cup of sugar, I will keep a cup and give her the other” (W.NR.2). Registered refugee participants rarely mentioned sharing of assistance between community members and only a few of the international and local NGO representatives referred to the social exchange of assistance between refugees.

High food insecurity and low dietary diversity.

All refugee participants, regardless of assistance and registration, reported difficulties in securing food for the family and poor dietary diversity. Occasionally, refugee women or children working in agriculture were able to take home leftover vegetables. To overcome food shortages, refugees resorted to different coping strategies such as buying cheaper, but low-quality, foods and over time decreasing even the quantity of food consumed. Food diversity was worse among refugees without assistance (registered or unregistered), whereby the latter would frequently skip meals and were more likely to suffer severe food insecurity or even occasionally eat expired or rotten foods. “There are times where we have no lunch and no dinner, there are other times where we only eat bread and tea” (M.NR.1). At times of prolonged food shortages, adults would skip meals to prioritize their children needs. “For example if the kids don’t feel full, we say let the kids eat until they’re full, we can handle hunger but the kids can’t so we give them the food, they are the priority” (M.NR.4).

Interviews with key informants confirmed these findings. When discussing skipping meals as a coping mechanism, one local NGO representative said, “It is standard now eating 2 meals a day. It’s become a daily reality [for refugees], their behavior has changed and it’s not a temporary coping mechanism anymore.” (Local NGO—Int. 6).

Overcrowding and poor living conditions.

Overcrowding was a common concern among participants regardless of registration status. “We are 7 people in this tent, in this room, in this 1 room, we all sleep in it” (W.R.A.2). In addition, participants reported overall poor and unsanitary living conditions and that the availability of proper waste disposal systems were inconsistent between ITS. In order to cope with these challenges, many participants had to dig unsanitary holes in the ground that they would discharge manually. Participants also reported a high incidence of gastrointestinal infections amongst children attributed also to poor water quality. “It is possible that the landlord will evict us due to the holes we dig as bathroom replacements, because we cannot dig a new hole every other day, I mean we suffer from this, it is our biggest problem, it causes sickness” (M.R.A.2). Refugees also experienced extremes in temperatures due to poor housing, as their shelters were very hot in the summer, and cold in winter. The cold weather was commonly dealt with by burning fuel, plastic or any available item such as clothes, resulting in respiratory complications amongst children.

Key informants confirmed many of the experiences reported by the refugees in terms of the poor and unsanitary living conditions that were causing health issues and breakouts “We stopped our desludging for like a month and the situation is deteriorating in some sites; health breakouts like scabies and other health issues at the community level. Management of grey water is not a priority although we are now witnessing some sites that are highly under risk because of that and it’s causing like a plague of rodents, rats, and no activities are being done for this regard.” (iNGO—Int.5).

Another challenge that a few refugee participants expressed is their constant fear of eviction from their tents due to their inability to cover rent for the land property owners: “You don’t know what day you will get evicted, in the extreme cold, in winter he (referring to the landlord) would evict us” (W.R.NA.1). A few unregistered refugees lived on their employers’ land or warehouses where they had to work anytime in return for housing and faced eviction threats if they did not work upon their employer’s request. One of the registered women, who was not receiving assistance and was living in an employer’s warehouse, explained: “It’s mandatory, my husband used to work for him, you know what I mean? I mean instead of payment he gives us shelter, but now there is no work, he says I want to evict you” (W.R.NA.1). As a coping mechanism, both registered and unregistered participants shared their tents with relatives and friends when they first arrive to Lebanon. It is worth noting that the sharing of housing was often a transition or temporary situation amongst registered refugees. “Of course, for example if someone from your family comes from Syria where is he going to stay? He needs to wait until the UNHCR gives him assistance so of course he is going to live with his family in the same tent until the UNHCR gives him assistance” (M.R.NA.1). On the other hand, unregistered refugees reported the sharing of housing with other refugees as a more permanent living arrangement: “If I go back to my tent I would die of cold, that’s why I am living at my parents” (M.NR.3: P5).

Limited access to healthcare.

There was a consensus among refugees that registration with UNHCR determined access to healthcare: “those who have a code [registered] get helped with the payment, they are much better off” (M.NR.3). However, out-of-pocket payments for hospital admissions and transportation fees were among the main barriers to access health care in Lebanon among all refugee participants. Refugees reported that they were required to pay upfront to be provided care at hospitals: “the hospital needs $3000 only as an insurance, but she is not allowed to enter; even if you tell them she will die, they still say she can’t enter!!” (M.R.A.3). Although healthcare subsidies were provided for registered refuges, out-of-pocket costs for antenatal care and deliveries were barriers to all refugees: “It’s expensive, especially for Syrians, you always have to pay out of pocket” (W.NA.R.3). One registered woman narrated her experience resorting to a home delivery due to the high out-of-pocket payment for hospital delivery.

Many registered and unregistered participants stated that discrimination and poor treatment in hospitals by healthcare professionals were also a common experience: “When I went to get the operation, they were yelling at me, and when I was done, they lifted me and then threw me on the bed, and even under anesthesia you feel the pain, you feel the pain of being thrown on the bed” (W.R.A.1. P5).

Due to financial constraints, registered and unregistered refugee women reported prioritization of medical conditions. For example, medical issues that were not perceived as crucial and urgent, such as reproductive tract infections, were almost always ignored without resorting to medical attention and instead were treated with home remedies “I swear the most important issue we have been facing is reproductive tract infections, and I can only rely on God because the cheapest medication for infection costs $20, I swear it’s very expensive so we just stay sick” (W.R.NA.2. P.6). This was echoed in interviews with key informants: “Not going to the necessary checkups, not taking medications all together because they cannot afford it. What is really sad is that sometimes people are sick, but they don’t know because they don’t get checked up.” (Local NGO—Int. 6).

Another coping mechanism for overcoming financial barriers to health services was seeking alternative informal sources of healthcare, such as pharmacists for medical advice, as it was much cheaper and easily accessible. “There’s no one other than pharmacies” (W.R.NA.P9). Many participants also opted to acquire their medications from Syria where prices are much lower, either done firsthand or through an intermediary. Some participants mentioned that individuals occasionally went to Syria to get medical care and then came back into Lebanon illegally. “We asked for assistance to cover the surgery, but we don’t have papers, we couldn’t do it, he had to go to Syria, and he has to serve in the army, so they took him and he had to stay” (W.NR.3.P4). These alternative sources of healthcare were also reported by local and international NGOs: “Some Syrians sometimes prefer to go to Syria to put their selves at risk to do some surgeries and come back. For example, there are some women who are going for C-sections and they are putting themselves at risk because some of them are undocumented” (iNGO—Int. 8).

A few participants mentioned switching IDs between registered and unregistered participants in order to access medical services, “For example my son today is not registered, and it is necessary to get him to the hospital, I’m going to have to admit him under my nephew’s name, he’s the same age and registered. It’s a rare case but these are our circumstances, it’s better than having my son die, I’m going to have to do this, this is not acceptable for the hospital nor for us but we are forced to do this what can we do” (M.NR.2). Only one key informant mentioned this coping mechanism: “We deal with cases of registration you sometimes you will hear about cases where women using the ID of her sister to give birth in the hospital” (iNGO—Int. 7).

Integrated analysis.

Integration of qualitative and quantitative data, albeit being modest, revealed mostly concordance specifically on food insecurity challenges. According to the questionnaire results, the majority of Syrian refugees were experiencing moderate to severe food insecurity and had poor or borderline food consumption, with the highest percentages observed among those who were unregistered followed by those who were registered without assistance. Similarly, FGD findings echoed the challenges faced by Syrian refugees in securing food for their families due to the high food prices in the market, particularly among those who were unregistered/unrecorded.

For employment, discordance was noted between the FGD and questionnaire findings. While FGDs and interview results reported almost universal informal employment, results from the questionnaire showed that only 25% were involved in labor. This discrepancy can be interpreted either due to the seasonality and irregularity of work opportunities, especially that the data was collected in the winter, which is considered a low agricultural season, or perhaps due to the fear of refugees from disclosing this information in the questionnaire as their employment was informal and illegal (i.e. without work permit).