UNDERSTANDING THE MENTAL HEALTH CONSEQUENCES OF FAMILY SEPARATION FOR REFUGEES: IMPLICATIONS FOR POLICY AND PRACTICE

Abstract

Consistent evidence documents the negative impacts of family separation on refugee mental health and concerns for the welfare of distant family members and desire to reunite with family members as priorities for refugees post-migration. Less is known about refugees’ emic perspectives on their experiences of family separation. Using mixed methods data from a community-based mental health intervention study, we found that family separation was a major source of distress for refugees and that it was experienced in a range of ways: as fear for family still in harm’s way, as a feeling of helplessness, as cultural disruption, as the greatest source of distress since resettlement, and contributing to mixed emotions around resettlement. In addition to these qualitative findings, we used quantitative data to test the relative contribution of family separation to refugees’ depression/anxiety symptoms, PTSD symptoms, and psychological quality of life. Separation from a family member was significantly related to all three measures of mental health, and it explained significant additional variance in all three measures even after accounting for participants’ overall level of trauma exposure. Relative to 26 other types of trauma exposure, family separation was one of only two traumatic experiences that explained additional variance in all three measures of mental health. Given the current global refugee crisis and the need for policies to address this large and growing issue, this research highlights the importance of considering the ways in which family separation impacts refugee mental health and policies and practices that could help ameliorate this ongoing stressor.

Ongoing conflicts in southwest Asia and East Africa have resulted in the number of forcibly displaced people rising to an all-time high at the end of 2015, an estimated 65.3 million people worldwide (UNHCR, 2016). This represented an increase of 12.4 million individuals during the 12 months of 2015, which equates to almost 34,000 people per day forced to leave their homes due to conflict or persecution. The majority of these 65 million people – approximately 40.8 million – are internally displaced persons, driven from their homes but remaining within the borders of their country of origin, while 21.3 million are classified as refugees. Refugee status is designated by the United Nations High Commissioner for Refugees and applies to those who have a well-founded fear of persecution should they return to their country of origin. States that produce refugees are sites of profound social upheaval, civil war, even genocide. Thus, most refugees have experienced traumatic events, such as torture, sexual assault, family fragmentation, and the death of loved ones (Knipscheer, et al., 2015).

Given the difficult life experiences that many refugees have had to overcome, it is unsurprising that high rates of mental illness have been documented among refugee populations. Numerous studies have found that symptoms consistent with PTSD, major depression, suicidality, and anxiety are reported more commonly amongst refugee groups living in exile than amongst the background populations in these states (Mills et al., 2005; Pham, Vinck, Stover, 2009; Robjant, Hassan, Katona, 2009; CDC, 2013; Heeren, Mueller, Ehlert, Schnyder, Copiery, Maier, 2012; Fazel, Wheeler, Danesh, 2005; Porter & Haslam, 2005). However, rates of psychological distress among refugees are lower than might be expected given their levels of trauma exposure and extensive research has documented the importance of broadening the ways in which the impact of trauma are measured to include culturally-specific measures of distress (Miller, et al., 2006) and psychosocial adaptive systems (Silove, 1999).

While there has been a great deal of research on the role of pre-migration experiences in refugee mental health, only recently have researchers come to recognize how profoundly post-migration circumstance impacts refugee well-being (Miller & Rassmussen, 2010; Tempany, 2009). In fact, psychosocial stressors such as discrimination, lack of economic opportunity, and significant social isolation some refugees experience after resettlement may more strongly predict emotional distress than exposure to trauma before or during flight (Pernice & Brook, 1996; Porter & Haslam, 2005; Rasmussen et al., 2010). Furthermore, research shows that trauma exposure does not straightforwardly cause emotional distress or mental illness in all people the same way – rather, its impacts are variable, depending in large part upon conditions that prevail after the trauma has occurred (Gorst-Unsworth & Goldenberg, 1998; Carswell, Blackburn, Barker, 2011).

Among the range of post-migration factors that impact refugee mental health, familial fragmentation has emerged as an important one among a variety of refugee communities (Rousseau, Mekki-Berrada, Moreau, 2001). Ethnographic research demonstrates that conceptions of family vary across cultures, with more expansive conceptualizations of family common in non-Western cultures (Georgas, Berry, van de Vijver, Kagitcibasi, & Poortinga, 2006). Of note, Muslim families often conceive of core family members from up to three or four generations (Dhami & Sheikh, 2000). In recognition of the diverse meanings that family carries in different cultural settings, the United Nations High Commissioner on Refugees (UNHCR), in its 2008 Annual Tripartite Consultation on Settlement, attempted to extend to non-biological members of a family group the same rights and protections as biological members (McDonald-Wilmsen & Gifford, 2009). The U.S. Refugee Resettlement Program resettles nuclear family units—parents and children. Moreover, family reunification for refugees outside the Refugee Resettlement Program has been limited, as the U.S State Department now requires applicants to obtain DNA tests or official adoption papers to ensure the veracity of familial relationships of applicants of its P-3 visa program, which allows refugees to apply for resettlement of family members fleeing crisis (Worth, 2015).

Although family is constructed differently across cultural settings, preliminary research suggests that the effects of familial separation may be similar for refugees from different geocultural regions (Rousseau, Mekki-Berrada, & Moreau, 2001). Studies have repeatedly documented concerns for the welfare of distant family members and the desire to reunite with said family members as prominent priorities for refugees in the post-migration period (McDonald-Wilmsen & Gifford, 2016; Momartin, Steel, Coello, Aroche, Silove & Brooks, 2006; Rousseau, Mekki-Berrada, & Moreau, 2001). Although family separation clearly negatively impacts refugees’ mental health, there is limited research that explores refugees’ perspectives on their experiences of separation from family members, including the ways in which family separation impacts their mental health and how its impacts are manifested in their daily lives. It is also important to understand the relative importance of family separation on refugee mental health compared with other traumas and life stressors. Given the current global refugee crisis and the need for policies to address this large and growing issue, research on familial fragmentation and mental health that addresses these gaps is particularly important.

Method

The data we are reporting on in this article were collected as part of a randomized controlled trial, funded by the National Institute on Minority Health and Health Disparities (R01MD007712) and approved by the University of New Mexico Human Research Protections Office, that examines the efficacy of a community-based advocacy, learning, and social support intervention to improve the mental health of recently arrived adult refugees from Afghanistan, Iraq and the Great Lakes Region of Africa. The mixed methods study uses a convergent parallel design and includes quantitative and qualitative data collected from participants at four time points over a period of 14 months; the data are analyzed separately and integration occurs at the analytic level (Creswell & Plano Clark, 2011). Quantitative data is being analyzed to determine the impact of the intervention on reducing refugees’ psychological distress, increasing protective factors, and engaging and retaining refugee adults with PTSD in Narrative Exposure Therapy (NET; Schauer, Neuner, & Elbert, 2005). NET is an evidence-based, short-term intervention that has been implemented and tested with people from many different parts of the world, including the Middle East and Africa. NET includes aspects of trauma-focused cognitive-behavioral therapy and testimony therapy and thus involves exposure but within the context of an individual’s whole life. The narrative structure of NET is more culturally consistent with many non-Western cultures, and it enables the individual to address multiple traumatic events. Qualitative interviews are being used to explore participants’ experiences of resettlement, to explore experiences with and effects of the intervention, to inform interpretation of quantitative data, and to investigate unexpected impacts of the study on participants.

Participants

The overall study will be implemented with four cohorts. For each year of the study, all refugees in Albuquerque from the above countries who have been resettled in the United States within the last three years are invited to join the study. Participants are then randomized by family groups into control and intervention conditions. The control group is invited to attend a one-time stress management session, and the intervention group is invited to participate in the six-month intervention in which participants are paired with undergraduate students who are trained in a mutual learning and advocacy model with the aim of increasing access to resources and social support. Participants in both conditions who are experiencing symptoms associated with post-traumatic stress disorder are offered Narrative Exposure Therapy.

The current research involves quantitative and qualitative data collected from initial pre-interviews with refugee participants in the first three cohorts of the larger study (participants enrolled between October 2013–November 2015). Quantitative and qualitative data were collected from 165 participants, through interviews and self-administered questionnaires translated into participants’ languages. Participants came from three main regions: Afghanistan/Persian area (52), Iraq/Syria (77), and Great Lakes Region of Africa (36). Potential participants (all refugees living in Albuquerque who were from the three regions of interest and who had been in the United States fewer than three years) were contacted by bilingual/bicultural study staff who briefly explained the study over the phone and arranged in-person meetings with refugees who were interested, at which time the study was explained in more detail. Eighty-five percent of eligible refugees agreed to participate in the study. Participant ages ranged from 18 to 71, with a mean of 36.29 and standard deviation of 12.13. Eighty-six (52.0%) were women. In terms of time in the United States, the mean time was 7.81 months (SD = 7.22), with 40 participants here fewer than two months, 49 participants here between 2.0 and 4.9 months, 38 participants here between 5–12 months, and 38 participants here more than 12 months.

Data Collection

Interviews took place in participants’ homes and were conducted by a bicultural interpreter who spoke the participant’s native language and a native English speaker. Whenever possible (and almost all of the time), project staff and interviewees were matched by gender: men were interviewed by men and women by women.

The qualitative semi-structured component of each interview was comprised of open ended questions that covered a variety of topics: personal background, resettlement experience in the United States, challenges and benefits of living in the United States, impacts of resettlement on family members, social support, access to resources and perceived cultural difference and cultural maintenance, hopes and goals for the future. Questions were asked in English, then interpreted, participants responded, and responses were interpreted back into English. The entire qualitative portion was recorded with the participant’s consent. Average interview length was 50 minutes; the range was 12 to 143 minutes.

Qualitative Analyses and Results

The importance of familial separation emerged as numerous study team members who participated in initial qualitative analyses were struck by the frequency, poignancy, and intensity of refugees’ discussions of their separation from family members and its deep impact on their daily lives and well-being. These observations were the impetus for the analyses and results described here.

Qualitative Data Analysis

English portions of the qualitative interviews were transcribed by a professional transcriptionist. When possible, project staff speaking the participant’s language checked interpretation for accuracy. Transcripts were then formatted, anonymized and imported into Nvivo 10 (QSR International, 2012), a qualitative data management package. All text was then coded using a multi-phase process. First, transcripts were coded according to question, allowing for rapid analysis of our semi-structured data. Second, all text was coded thematically using a hierarchical coding tree by two coders working independently. The data analysis team conducted a series of meetings to define codes to ensure consistency in the way they were applied to the data. Our goals were not to have coders apply the codes in exactly the same manner, but to allow for inclusion of the perspectives of a diverse team of analysts, which include refugee members, to apply the codes in a way that ensures ongoing discussion of interpretation and meaning. Third, analytic memos on major themes were created to analyze content for patterns and anomalies across the data, akin to focused coding as described by Charmaz (2014). The first author created analytic memos examining the effects of family separation on participant health.

Qualitative Results

Data from the semi-structured qualitative interviews demonstrated that a major source of distress for participants is separation from their family members that comes with settling in a new country. This distress is experienced and described in a range of ways: as fear for family still in harm’s way, as a feeling of powerlessness to assist distant family members, as the greatest source of distress since resettlement, as mixed emotions around the decision to resettle in the United States, as feeling constrained about what one can communicate to distant family, and as a cultural disruption.

Fear for physical safety of family still in conflict zone

Interviewees discussed fears for family members’ safety on many occasions. Participants often described the negative emotional impact that these fears have on their lives here in the United States. One 33-year old Iraqi woman, Halima (all names are pseudonyms to protect the identity of participants), noted:

I’m not okay, psychologically. I feel really bad all the time, depressed, because we left our family in Iraq. We always hear of explosions and bombings in Iraq. Where my family lives in Baghdad, it’s not a very safe place. I’m not really happy.

A 49-year old Iraqi man, Mushtaq, noted how his concerns for his brother have a pervasive impact on his thought processes here in the United States:

It’s still bad in Iraq now. I have a lot concerns about my brother, because he lives—my brother, he lives under the stress, and the bad security situation. Just I’m putting my mind [to] how [to] push my brother to the safer place. How can I push him to move out of Iraq to safe place?

Numerous participants downplayed the other stressors that exist in their lives in the United States, in deference to their concerns for distant family.

The concerns participants had for their family members back home were made more urgent and distressing when information about those family members was difficult to find. One 40-year old Congolese woman, Meissa, described how: “I don’t feel really good about it because when it comes to my two sons I still don’t know where they are right now, so it’s really heartbreaking for me. I don’t feel good about it.”

Meanwhile, other participants described how news about the conflict back home created an urgent need to verify the safety of family members. One Iraqi man, Aleem, related: “I wanted always to check for their health or being well because of the bombing in Iraq and unsafe situation so when I hear anything I would have to call and check for them.”

Among the Iraqi and Afghan participants, those who were employed by U.S. or Coalition forces often described how their work history led to additional uncertainty or danger for their loved ones back home. One Afghan man noted that:

Because they would face arrest and threat, because they are parents of a person who have been working with the coalition forces for five years. So they are, it’s important that they hide. They should keep it secret.

One Iraqi man provided a glimpse into how affected he was by this problem, how he thought about his family more often than he thought about himself:

Interviewer: How does being in touch with them affect your well-being?

Participant: Yeah, [being in touch with family back home] affect[s] me a lot and it helped me during my resettlement here but I’m always thinking about them more than myself because I’m worried about them.

Powerlessness to help distant family

Alongside fears for the safety of family members still in the conflict zone, another prominent theme was a feeling of powerlessness to help those family members. One Congolese woman, Ayana, stated that this feeling of powerlessness issued not only from the distance between her and her family back home, but from her impoverished circumstances here in the United States, “ … if I had like money I could send her to support her. So that, but now we’re crying together. We don’t have anything.” Moving to the United States had changed nothing for her family back home, and the participant also expressed pessimism about her future ability to help them materially. This seemed to make the separation from family more difficult to bear for this participant.

The notion that one derives personal fulfillment from helping family members – and is left disheartened when unable to do so – was present in other accounts as well. As Haleema, an Iraqi woman in her 60s discussed:

Since my daughters are over there [in Iraq] I cannot help them. Like I cannot sit with them, see them, so … if they just get here I can like help them with their kids, like watching their kids, maybe feel comfortable when I see them…[Iraqis] want to talk and to see each other so I will feel good if I do that.

The idea that family interaction will provide fulfillment for the participant was notable, beyond the obvious benefit it would provide for her daughters.

Dispersed family means unmet emotional needs

Another theme prominent in our interviews was how individuals’ socio-emotional needs were not met when their family is spread out or distant. Participants detailed the ways in which having their relatives far away negatively impacted them, as well as relaying how their family members were responding to the participant being newly distant, here in the United States.

One Iraqi woman in her late 30s, Basmah, shared how her settling here in the United States had diminished her contact with her family still living in Iraq, made her ‘emotionally tired.’ And without her family, “here, there is nothing.”

Basmah’s comment highlighted that visiting with family is a restorative emotional experience from which one can draw strength, a sentiment seen in other participants’ accounts as well. A Iraqi man, age 28, Mukhalad, made very similar comments:

Interviewer: It [your mood] went down?

Participant: Yeah, ‘cause of my family. In Syria, I wasn’t with my family all the time, but I had the chance to go see them whenever I want, but now I can’t, and this is the first time I’ve been away from my family, like really, really far from them.

Mukhalad’s comment demonstrated that he considered a drop in mood to be the natural consequence of not having regular contact with family.

One Iraqi man in his late 40s named Mushtaq, spoke about how he felt emotionally isolated without his family members: “It’s not like a mental symptom. It’s like mental concerns. Because there is nobody from your friends, or from your relatives around you, can’t listen to you, or you can’t participate your concerns with them.” For Mushtaq, the separation from family meant not having anyone to discuss one’s concerns with. Other participants made similar observations, namely, that family helps an individual to deal with adversity, and that not having family around made difficult experiences even harder. Peter, a 28-year old Congolese man, had similar sentiments about the adversity he faced soon after arriving to the United States:

Physically, it didn’t really affect me, but mentally it really affected, ‘cause I was hurt. Especially at that time, with all that was happening, I really thought about my mother a lot. I wished she was there for me. Going through all those problem[s] without her, it really hurt my heart.

Peter’s comment suggested that family plays an important role in buffering a person against emotional shocks.

A variation on this theme emerged from a number of interviews – namely, that proximity of family allows individuals to be their best selves and to excel in life. Conversely, the absence of family served to slow or prevent growth or progress. A young Iraqi man, Amar, who was in his mid-20s, said the presence of family members was necessary for his own progress:

Interviewer: What are your hopes and ambitions for your family?

Participant: I wish they would come to the United States, come to this place so we’re all together again and they can help me to improve in my life basically, yeah.

A Congolese man, Zachariah, who was 40 years old, said being in touch with his family was “what adds beauty” to his life in the United States, and “If I’m not in contact with them I won’t be happy.” These passages demonstrate the various emotional responses that participants experienced during family separation; participants often characterized this separation as their single greatest source of distress.

Separation as greatest source of distress

In a number of passages, participants stated that the distance from their family members, and the emotional distress this engendered, was either their foremost concern in the United States, or the most distressing aspect of their life post-resettlement. For Jameel, a 63-year old Iraqi man, being distant from his sons still in Iraq, was the only grievance he had about the United States: “So the main problem is the sons but otherwise I love U.S. I feel very safe, very happy, like all the benefits, everything is good.” In a separate interview, his wife Asal, an Iraqi woman in her late 50s, likewise characterized this distance from children as her chief concern:

Okay, so the most thing make me sad [is] my son in Iraq, I wish that I can make him come here like in any way ‘cause he’s [the youngest] and he was the closest [son] to us, like every time I talk to him he tells me, “I feel alone in the whole world, like when you just left.” So I’m so sad for him. I want him to come here so it’s the most thing that makes me sad.

Later on, speaking more directly about her health, Asal described how the presence of her sons here in the United States was the only missing precondition for her to have good health:

Interviewer: Okay. So you talked a little bit about some health difficulties you’re facing. What do you think you need in order to heal or get better?

Participant: My sons. Yeah.

Interviewer: Okay, how do you think that will make you better?

Participant: When they come here first of all I wouldn’t be alone so that’s going to help a lot. Second, I feel that they can help me, like help me for work, help me for not to be alone, to be a big family, to be happy, and … sometimes when I talk to my sons’ kids, like when I just hear their voices I think to go back, like just for them. So I just want to be with them, to be happy.

The participant believed that her sense of loneliness would be resolved by her sons’ arrival, that they could help her carry out her daily duties, and then described the duty she felt to care for her grandchildren, which at times gave her the urge to return to Iraq, in spite of the risks.

A 19-year old Iraqi woman, Delal, had similar feelings about her distance from her parents, which she characterized as the factor that most impacted her health here in the United States: “My parents is the most thing that makes me feel really bad because I [am] so worried about them, I miss them, I feel sad, yeah, so it’s mostly this [that has affected my health during resettlement].”

Felicia, the Congolese woman quoted earlier, had a similarly comprehensive description of how her family arriving to the United States would benefit her health:

Interviewer: So you’ve mentioned that you have headaches, and sometimes you think a lot, and the stress, what do you think you need in order to heal or get better?

Participant: I need like someone to talk to. I need like to see my family. Yeah. …

Interviewer: Like how do you think it would help you?

Participant: To erase my stress. Yeah.

Interviewer: And if you were able to talk to your family how would that help?

Participant: So much, it going to help me so much because my family, they’re part of me.

The presence of one’s family is felt by participants to be a necessary condition for maintenance or recovery of good health. Given the scope of the distress that family separation caused participants, participants often experienced mixed emotions over their decision to resettle in the United States.

Feeling conflicted about decision to resettle in the United States

Because distance from family was often an overriding concern for study participants, one that sometimes dwarfed their relative satisfaction with other dimensions of life, participants also stated that separation from family contributed to emotional conflict related to their presence here in the United States. As one 44-year old Iraqi woman, Walaa, related,

[I]t’s really hard for me just to leave them there behind, but at the same time, I’m thinking of my kids and their safety, which is my priority … There, no, I have to be worried because they might face bombs and kidnapping and that is not safe at all for them. Here, they are really fine and safe, and that will make me feel really—it’s a relief and make me feel comfortable. What can I do? It just have a price.

The “price,” for Walaa seemed to be a trade-off between her children living with safety and security here in the United States, which had obvious and overriding benefit in her mind, but which came at the cost of distance from other family members. “What can I do?” she asked, as if to say that there was no perfect course of action in her situation, but that she considered the needs of her children first.

An Iraqi woman in her mid-30s, Halima, described how, for her, the safety of her immediate family was likewise paramount:

My family in Iraq, they are sad that we left. They’re not really happy, but we are very happy because now we’re safe. It’s so different in here. We were living in a war zone, so this is really safe. We’re happy.

Another Iraqi woman, Azhar, who was 20 years old, described how the dissonant feelings that participants experience are matched by a peculiar blend of emotions in their family members back home:

So they were very sad that we are leaving because, you know, it’s far away from here. We are going to miss them a lot. But at the same time, they encouraged us to come here because, you know … when we were in Iraq during the war, my dad got kidnapped, my brother get to be disabled because of the war there, so they were praying for us, encouraging us to go and say please, be safe, we want you to live. “Go and try to find a better life, a safer life there”…. So it’s kind of having two things like they are sad we are not there but at the same time they are comfortable, they are happy that we are safe here now.

Thus, for Azhar, this “success” was a mixed experience, tempered by sadness and longing.

Another Iraqi man, Zaid, age 50, hinted at the “mixed” emotional experience of leaving family behind to come to the United States, and how this caused “emotional issues:”

One feet here, one feet back home. Do you understand me? Then I feel like always I am in the middle, or I can tell you: my body here, but my soul there. That’s what. Then that’s me, cause me like emotional issue[s], mental things together.

A related but distinct theme of emotional conflict was relayed by participants who said that while their families were happy they were in the United States, ironically, they were not happy being here. A 22-year old Afghan man, Farid, stated: “They are so happy for me. But I’m not so happy.”

Still another example of emotional dissonance was found in the comments of some Afghan participants. A 40-year old Afghan woman named Diba described how the expectations and preferences of her family members back in Afghanistan contributed to her family’s uncertainty when immigrating to the United States:

We didn’t tell them until we [arrived] here. Then we informed the families in Afghanistan that [we] were in U.S. Some of them were happy, some of them not, some have bad mentality about U.S. They don’t know what’s going on here.

When Diba said that her relatives “don’t know what’s going on here,” she seemed to suggest that her relatives did not know about the aspects of life in the United States that she and her family had come to value. The discordance that participants experienced when confronted with the opinions and experiences of distant family members contributed to another theme found in our interviews, that participants felt constrained about what they could communicate to their distant family members.

Feeling constrained in what one can communicate to distant family

During interviews, participants frequently relayed their misgivings about communicating to distant family members the full details of their resettlement here in the United States. One Iraqi man, Saad, aged 40, discussed how he withheld details of his difficulties in order that he might prevent his family from feeling frustration:

Honestly, when I am talking to them and when I keep contact with them I have to say that everything is perfect, and I have to—because I don’t want to push them for frustration, in order not to be frustrated over there, because they are planning to come here. I don’t have to say that [resettlement agency] treated me just like this way, as I mentioned before. I have to keep that by myself in order to prevent them from the frustration.

Hearing of the difficult times he had experienced in the United States, he believed, would be very distressing to them.

Similarly Mushin, a 43-year old Iraqi man shared how he avoided communicating with his family through video calls, so that his family would not be alarmed by his appearance, which he considered to be an indicator of his unhealthy state. When asked how it impacted his health when he talked to his family, he responded, “Actually I feel sad because they used to see me in a better situation, healthy. And if they see me like that right now I will feel very sad and uncomfortable, so that will affect my situation, my health.”

An Iraqi woman in her late 30s, Shaymaa, shared that she was reluctant to contact her family because she did not want to tell them how difficult life had proven for her in the United States:

Actually, I try to not contact them very much because I don’t want to tell them the issue, the truth. Because our people, they know America is very great place and all the people here are very happy and rich, but this is not the truth. Then I cannot call them and tell them the[se] things ha[ve] not happened to me. Then I wait ‘til maybe after one year, or I don’t know when, but just ‘til I can find a good job, like my job as a teacher or teacher assistant. … I cannot tell them now what’s happening right now, that’s difficult.

Like Saad, Shaymaa had made the judgment that sharing her difficulties with her distant family members would not serve her interests or theirs.

Separation as cultural disruption

While some participants described the toll that distance from family took on them in emotional terms, others spoke of challenges that it presented to cultural or religious practice. Zaid mentioned that in Iraqi culture, the word “family” means extended family rather than being limited to nuclear family. An Iraqi woman, Areej, who was in her early 30s, voiced a similar idea when she said that the most difficult thing about coming to the United States was leaving their families, because they are “a part of us.”

Mukhalad attributed straying from strict religious practice to not having family members around him:

Like I told you, my family’s not here, so there is nothing to keep me in base with that, with my religion, with my spiritual things. It’s very hard to do it at home, so yeah. I’ve been starting to lose track of that.

His matter of fact response suggested the ways in which absence of family members could result in disconnection with well-established spiritual practices. One Afghan man, Atash, who was in his mid-40s, spoke about the universality of familial separation for refugees, and offered a unique description of how he imagined the family: “[It’s] a big concern for all the refugees. The family is like a body. If you lose a body part, you are disabled.”

Meanwhile, Zachariah, a Congolese man, described his wish to see his parents immigrate to the United States in terms of the cultural continuity they would be able to provide for his children:

I really need my parents here. I really want to see my family here because in the country back home they are not okay… And I need my children also to learn about the culture where I come from. I think that legacy can only be given from my parents.

Zachariah later repeated his wish to have his parents immigrate to the United States as his “number one” worry. When asked why this was important, he again emphasized the cultural background that they would serve to connect him and his family with, because connection to one’s culture had ontological value in itself: “It is important because I believe somebody who does not know he or she comes from is a lost person. He loses focus in life.” A 19-year old Afghan woman named Elham similarly described her wish to see her family in cultural terms:

I miss my family. There is no one around me to talk with me and guide me in my life in the U.S. The culture is very different here and I need someone to communicate [with me] and guide me.

The participant framed her desire for family members close by as “guidance,” implying that she felt disoriented, even adrift, without them.

Our qualitative results show that in response to open-ended questions about refugee resettlement, health, and well-being, family separation was named repeatedly as the most distressing aspect of many refugees’ resettlement experiences. Participants discussed family separation as having a large impact on their mental and physical health, and mentioned it in ways that suggest it also affected their feelings of self-efficacy and agency with respect to post-migration life.

Quantitative Analyses and Results

After exploring the in-depth qualitative data on refugees’ experiences of familial fragmentation and separation and the ways in which it impacted their mental health and resettlement experiences, we decided to augment these analyses with an examination of our quantitative data. The mixed method approach of the overall study was chosen to provide opportunities for such triangulation of findings (Creswell & Plano Clark, 2011). In this case, we sought to explore whether our quantitative data confirmed our qualitative findings related to the negative impact of family separation on refugee participants’ mental health, and furthermore, to examine the strength of the relationship of family separation to participants’ mental health, relative to other traumatic stressors.

Quantitative Measures and Data Analysis

This study included three measures of psychological well-being: the Hopkins Symptom Checklist–25 measure of emotional distress (Derogatis, 1974), the PTSD Checklist – Civilian Version (Weathers, et al., 2013), and the WHO Psychological Quality of Life scale (WHOQOL Group, 1998). All measures were translated by two people and independently back-translated by two others. Discrepancies in translation were resolved through discussion among translators and the fourth author. Although not included here, the overall study also involved the development and testing of culturally-specific measures of psychological distress for each refugee group.

Hopkins Symptom Checklist-25 (HSC-25)

The HSC-25 is a self-report measure of anxiety and depression symptoms that has been used successfully and repeatedly with populations throughout the world, including refugees (e.g., CepedaBenito & Gleaves, 2000; Hermansson, Timpka, & Thyberg, 2002; Ichikawa, Nakahara, & Wakai, 2006; Mollica, Wyshak, de Marneffe, Khuon, & Lavelle, 1987; Nickerson et al., 2011; Smith-Fawzi et al., 1997) and was rated by Hollifield and colleagues (2002) in their review of mental health measures for refugees as one of only two instruments to meet all five of their established criteria. Response choices for each item are on a Likert-type scale ranging from 1 (not at all) to 4 (extremely). The HSC-25 produces three scores: total score (mean of all 25 items), depression score (mean of the 15 depression items), and anxiety score (mean of the 10 anxiety items). It has been consistently shown in multiple populations that the total score is highly correlated with severe emotional distress of unspecified diagnosis, and that the depression score is correlated with major depression as defined by the Diagnostic and Statistical Manual of the American Psychiatric Association, IV Version (DSM-IV). We used the total score (calculated as the mean of all items) in the current study. Participants’ scores ranged from 1.00 to 3.67, with M = 1.64, SD = 0.64. Typically, a score of 1.75 or above suggests clinically significant distress; 29.4% of our participants had scores above this cutoff. Cronbach’s alpha in our study was .96, with a range of item-total correlations from .66 – .79.

PTSD Symptom Checklist – Civilian Version (PCL-C)

The PCL-C is a self-report measure of 17 PTSD symptoms that reflect the DSM-IV criteria/symptoms. An advantage of the PCL-C is that it asks about symptoms in relation to generic “stressful experiences,” which allows for inclusion of the multiple traumas refugees may have experienced because symptom endorsements are not attributed to a specific event (although we also used a measure of trauma exposure, described subsequently, to verify that participants had experienced at least one trauma that would warrant completion of the PCL-C). The PCL-C is designed to produce a total symptom severity score (range = 17–85), which is obtained by summing the scores from each of the 17 items that have response options ranging from 1 (not at all) to 5 (extremely). The PCL-C can also be used to make a “presumptive” determination of whether respondents meet the DSM-IV criteria for PTSD, but we did not attempt to make diagnoses for our study, both because of our focus on symptom change rather than diagnosis and because we did not include a structured clinical interview. Cut-point scores ranging from 30–50 have been recommended for the PCL-C, depending upon the population and setting of administration (Blanchard, Jones-Alexander, Buckley, & Forneris, 1996; Freedy, et al., 2010). Based on review of relevant research, our clinical team chose to use a cut-point score of 40 for this study. Our participants’ scores ranged from 17.00 to 81.00, with M = 33.91, SD = 16.20, alpha = .95, range of item-total correlations = .45–.80, and 27.8% of participants having a PTSD score above the clinical cut-point score of 40.

World Health Organization Quality of Life Scale – Brief Version (WHOQOL-BREF)

The WHOQOL-BREF is a 26-itme quality of life measure that was collaboratively developed by researchers at 15 international WHO field centers, with the goal of creating a quality of life assessment that would be applicable cross-culturally and that would assess respondents’ quality of life from their own perspective (The WHOQOL Group, 1998). The full version of the WHOQOL has 100 items. The WHOQOL has four domains: physical health, psychological well-being, social relationships, and environment. For this study, we used three items from the psychological well-being subscale of the WHOQOL-BREF, which includes items that measure positive feelings, spirituality/religion/personal beliefs, and thinking, learning, memory and concentration. Response choices for each item are on a Likert-type scale ranging from 1 (not at all) to 5 (an extreme amount). Total subscale score is calculated as a sum of all items (possible range of 3 to 15), with higher scores indicating more positive quality of life. Participants’ scores ranged from 3.00 to 15.00, with M = 9.78, SD = 2.88, alpha = .80, range of item-total correlations = .36–.47. The WHOQOL-BREF is not designed to have specific cut-off scores, but rather is used to obtain a subjective assessment of respondents’ quality of life.

Trauma exposure and family separation

Trauma exposure was assessed by a 27-item checklist that was created for our study in consultation with several clinical researchers who focus on trauma exposure in refugee populations and included items from Weine and colleagues (1995), Foa and colleagues (1993), and the Harvard Trauma Questionnaire (Mollica, 1992). Participants’ scores ranged from 0 to 25 (indicating that some participants did not endorse any of the 27 items of trauma exposure while some participants reported experiencing 25 of the 27 types of trauma); M = 8.45, SD = 6.19, alpha = .91, item-total correlations = .24–.59. One trauma exposure item in the 27-item checklist assessed whether the participant had experienced “separation from a family member or loved one”; this item was endorsed by 93 (56.4%) of the participants. This item was removed from the overall trauma score in order to examine its separate effects; for the remaining 26-item scale, M = 7.72, SD = 5.81.

Quantitative Data Analysis

To explore whether the negative impact of family separation on refugee participants’ mental health could be seen in the quantitative data, we first used t-tests to compare individuals who reported family separation with those who did not on each of the three measures of psychological well-being. We then performed a multivariate analysis of covariance (MANCOVA) on the set of three mental health variables by family separation, adjusting for the potentially confounding effects of overall trauma and months in the United States. To reduce the impact of missing data, we imputed three scores affecting two participants, using expectation maximization (Enders, 2010).

Quantitative Results

Table 1 contains means, standard deviations, and t-tests comparing refugees who had experienced family separation with those who had not on the three measures of psychological well-being plus overall trauma exposure and weeks in the United States. Refugees who reported family separation had significantly higher levels on the Hopkins measure of anxiety and depression and the PTSD checklist and significantly lower levels of psychological quality of life. They also reported higher levels of overall trauma. There were no differences on length of time in the United States.

Table 1.

Psychological Well-being, Overall Trauma Exposure, and Months in the United States by Family Separation

Dependent Variables	Family Separation (n=93)		No Family Separation (n=72)		Comparative Statistic
	Mean	SD	Mean	SD	t(163)	p
Hopkins Symptom Checklist	1.85	0.66	1.37	0.49	5.31	<.001
PTSD Checklist	40.12	17.04	25.52	10.30	6.81	<.001
WHO Psychological Quality of Life	9.04	2.78	10.74	2.73	3.91	<.001
Overall Trauma Exposure (26-item)	9.90	5.33	4.90	5.20	6.03	<.001
Months in United States	8.53	7.94	6.88	6.09	1.51	0.13

Table 2 presents the results of MANCOVA by family separation on the three psychological well-being measures, adjusting for overall trauma and time in the United States. As expected, the effect of overall trauma exposure was significant in the multivariate test and all univariate tests, accounting for 3% to 25% of the variance in the individual psychological well-being measures. After accounting for the effects of overall trauma, family separation showed significant effects in both the multivariate test as well as the univariate tests for all three psychological well-being measures. Family separation accounted for an additional 4% of the variance in the Hopkins measure of anxiety and depression symptoms, 7% of the variance in PTSD symptoms, and 5% of the variance in psychological quality of life.

Table 2.

MANCOVA on 3 Measures of Psychological Well-being by Family Separation, adjusting for Overall Trauma and Months in United States

	Multivariate Tests			Univariate Tests for each Dependent Variable
				Hopkins Symptom Checklist			PTSD Checklist			WHO Psychological Quality of Life
Effects	F(3, 159)	p	η2	F(1, 161)	p	η2	F(1, 161)	p	η2	F(1, 161)	p	η2
Family Separation	4.51	0.01	0.08	7.04	0.01	0.04	11.89	<.001	0.07	7.83	0.01	0.05
Overall Trauma Exposure (covariate)	19.60	<.001	0.27	29.90	<.001	0.16	53.40	<.001	0.25	4.17	0.04	0.03
Months in United States (covariate)	1.09	0.36	0.02	0.01	0.93	0.00	0.21	0.64	0.00	1.31	0.25	0.02

To assess whether other specific types of trauma also explained additional variance after controlling for overall trauma exposure, each of the other 26 items of the trauma exposure measure was tested. Only one other item – having experienced physical assault, beating, or torture – explained additional variance in all three measures of psychological well-being.

Discussion

In-depth qualitative interviews with 165 recently resettled refugees from Afghanistan, Iraq, and the Great Lakes Region of Africa revealed that family separation was one of the most salient and distressing aspects of their resettlement experiences. Quantitative data from the same refugee participants confirmed the strong negative association between family separation and mental health, and further highlighted that family separation explained additional variance in depression/anxiety symptoms, PTSD symptoms, and psychological quality of life after controlling for overall trauma exposure. This is particularly striking because only one of the other 26 types of trauma exposure measured explained additional variance in all three measures of mental health. These findings emphasize the importance of conducting mixed methods research with refugee populations to enable us to more comprehensively understand refugee experiences and also suggest numerous implications for policy and practice.

Refugee participants repeatedly discussed separation from family members as a major stressor and family reunification as one of the primary needs they had to improve their quality of life in the United States. While much research on refugee well-being has focused on pre-migration trauma, increasingly researchers have sought to examine resettlement stressors and the role they play in mental health and social integration (e.g., Carswell, Blackburn, & Barker, 2011; Miller & Rasmussen, 2010). Family separation can be construed as both a pre-migration trauma and as an on-going post-migration stressor; it is a temporal process that does not end, and indeed may be exacerbated by resettlement in the United States as geographical distances between family members increase. Rousseau and colleagues (2001) note that: “For refugees who have had traumatic experiences, extended separation from family members may serve as a continuing link to an unbearable past” (p. 41). Thus, familial separation is a phenomenon that connects pre- and post-migration experiences. This understanding is consistent with transnational theorizing of migration, which has upended the notion of a distinct separation between pre- and post-migration life (Basch, Glick Schiller, Svanton Blanc, 1994).

Importantly, our results further elucidate the ways in which familial separation spans pre- and post-migration experience through participants’ descriptions of their ongoing fears for family still in harm’s way, their feelings of powerlessness to assist their distant family members, their separation from family as the greatest source of distress since resettlement, their experiences of mixed emotions around the decision to resettle in the United States, their constrained feelings about being able to seek social support from distant family members because of not wanting to burden family members living in danger with their difficulties, and the cultural disruption that resulted from being separated from family.

Our qualitative data show that refugees experience separation from family in a number of ways, and describe the emotional impacts of this separation in diverse language. Common to these accounts, however, is the notion that this separation is a noxious experience, a stressor. As discussed previously, research has shown that ongoing stressors play a critical role in mediating the impacts of past traumas on emotional well-being in refugees (Steel et al., 1999). For professionals working with refugees and for policymakers who make decisions that impact refugees’ ability to reunify with family members, the need to neutralize ongoing stressors should be paramount. However, as Rousseau and colleagues (2001) speculate, countries of permanent settlement may tend to focus on past traumas in studies of refugee mental health:

Perhaps the fact that there is greater academic interest in trauma than in separation reflects the political dimension of these phenomena: armed conflict and war trauma are seen as the violence of others (Kleinman & Kleinman, 1997), whereas an examination of prolonged separations highlights Western administrative violence (Rousseau, Mekki-Berrada & Moreau, p. 56).

Rousseau and colleagues suggest that the extended separations from family that refugees experience – often as relatives’ refugee or immigration cases pend according to an unknown timeline – are examples of “Western administrative violence” that presumably continue to cause harm to refugees and should be acknowledged and addressed when considering how to improve the well-being of refugees.

Indeed, a common feature of many of the separations that our interviewees related was an unknown duration. Participants variously articulated hopes, dreams, and, occasionally, plans for the reunion with family, but seldom stated concrete dates or timelines. It is this uncertainty around the reunion with family that makes these separations examples of “ambiguous loss,” a kind of relational disorder described by Boss (2010) in which a loved one is either physically present but psychologically absent, or psychologically present but physically absent. Examples relevant to this study included kidnapping, disappearance, or migration – all situations in which loss was unclear, confusing, incomprehensible, and, critically, defied closure (Boss, 2010). In such losses, there is “no verification, no closure, no rituals for support, and thus no resolution of grief” (Boss, 2010, p. 144).

In the case of third country resettlement, one could argue that safety concerns are dramatically lessened for at least some members of the family, which certainly creates an avenue for increased hope for the future and family reunification. Still, it is important to note that Laban and colleagues (2004) found that protracted administrative limbo was an important risk factor for psychopathology in Iraqi asylum seekers, with those who had spent longer in the process more likely to have higher levels of psychiatric distress. Hauff and Vaglum’s (1995) research also suggests that longer separations may contribute to more pathology: “Refugees may feel that they never will see their families again, but it is difficult to mourn since the people they miss are in fact still alive. Exile normally implies an unfavourable situation for mourning, due to social deprivation and isolation and a feeling of a lack of safety” (p. 366).

Limitations

Although our study makes important contributions to understanding the impact of familial separation on mental health of refugees, one notable limitation is a lack of inclusion of measures that systematically assessed the frequency and types of contact that participants had with family members from whom they were separated. Over the last two decades, the global dissemination of communication technology has contributed to the ability of refugees to be in contact with family members to an unprecedented degree. In the open-ended interviews, many participants indicated that they communicated with family members in their home countries or other places abroad often daily or at least weekly. They communicated by phone, but more often by free internet phone and video calling services, including Skype, Facebook and Viber. In addition, cable, satellite and internet access to news in their native languages meant that participants were continuously aware of news from their home countries and other places where their family members lived, which were often places of continued conflict and danger. Participants indicated this this level of contact and awareness of potential dangers to family members had impacts on their health and well-being. While many participants discussed the positive effects of speaking with family members, as seen in our results, many other refugees felt they could not be honest about the difficulties they faced in the United States for fear of distressing family members further. These contrasting experiences related to contact with family members who were far away but could be communicated with much more easily than in the past is an important area for future research. Although there is evidence that being with family members can have protective effects for people who have experienced trauma (e.g., Tsoi, Gabriel & Felice, 1986), less is known about connecting with family members through virtual means.

Conclusion

Historically, family reunification has been a cornerstone of U.S. immigration and refugee policy. Our research shows that family separation not only has serious health impacts for individuals, but that it also may have important effects on refugee integration in U.S. society in the long-term. Further, recent government policy changes and actions, including an Executive Order that cuts the number of refugees to be resettled in the United States by more than half in the current year and bans immigration from seven countries, as well as policies requiring DNA testing for reunification of refugee family members, and more generally, the increase in deportations of unauthorized Mexican immigrants, many from mixed status families, indicate an erosion of this important aspect of U.S. immigration and refugee policy. Given our findings of the importance of family for refugee mental health, policies and practices that facilitate family reunification may be important for promoting refugee and immigrant well-being. Our findings are consistent with other studies (e.g., Carswell, 2011) that have pointed to the need for greater responsiveness on the part of government agencies to the various needs of refugees, including prioritizing family reunification.

Public Policy Relevance Statement.

Although family reunification has been a consistent consideration within U.S. immigration and refugee policy in the past, recent government actions and policies suggest that this has changed, and that, in fact, current U.S. policies and practices have actually increased refugees’ and immigrants’ risks for being separated from their families. This study finds that family separation has dire mental health consequences for individuals and may impede newcomer integration in U.S. society. Thus, our findings highlight the importance of local, state, federal and international immigration and refugee policies returning to a prioritization of family reunification.