Received: 25 November 2021 Revised: 9 March 2022 Accepted: 10 March 2022

DOI: 10.1002/ajhb.23747

ORIGINAL ARTICLE

Transient refugees' social support, mental health, and

physiological markers: Evidence from Serbian asylum
centers

Jelena Jankovic-Rankovic1 | Rahul C. Oka1 | Jerrold S. Meyer2 |

J. Josh Snodgrass3 | Geeta N. Eick3 | Lee T. Gettler1,4

1
Department of Anthropology, University
of Notre Dame, Notre Dame,
Abstract
Indiana, USA Objectives: Refugees seeking safety across international borders are often
2
Psychological and Brain Sciences, exposed to a wide breadth of psychosocially stressful experiences that may frac-
University of Massachusetts, Amherst,
ture existing sources of social support and impair the generation of new social
Massachusetts, USA
3
Department of Anthropology, University
relationships, with implications for their long-term health and resilience. Using
of Oregon, Eugene, Oregon, USA data from recently settled refugees in two asylum centers in Serbia, we examined
4
Eck Institute for Global Health, the associations between social support, mental health, and physiological
University of Notre Dame, Notre Dame, markers.
Indiana, USA
Methods: In this mixed-method study of refugees (age 18–50 years, n = 76),
Correspondence we collected key socio-demographic information and conducted semi-
Jelena Jankovic-Rankovic, 260 Corbett
structured interviews about refugees' journey and stay in Serbia, trauma/loss,
Family Hall, Department of
Anthropology, University of Notre Dame, and their sources of social support. We also collected self-reported measures of
Notre Dame, IN 46556, USA. mental well-being as well as physiological markers relevant to repeated expo-
Email: [email protected]
sure to chronic psychosocial stress (fingernail cortisol and dried blood spots for
Funding information analysis of Epstein–Barr virus [EBV] antibody titers).
Institute for Scholarship in the Liberal Results: We found that refugees with longer journeys reported lower social sup-
Arts, University of Notre Dame; Kellogg
Institute for International Studies, port than those with shorter journeys. Refugees with lower social support
University of Notre Dame; Nanovic reported poorer mental well-being, greater PTSD-related symptoms, and higher
Institute for European Studies, University
recent perceived stress than those with higher social support. We also observed
of Notre Dame
that refugees with lower social support and higher recent stress, respectively,
tended to exhibit higher fingernail cortisol levels. However, we did not observe
comparable patterns linking EBV antibodies with psychosocial functioning.
Conclusion: Our cross-sectional findings are consistent with the notion that
social support is likely to be a critical component in effective interventions
aimed at mitigating the adverse health effects of relocation-related illnesses
and poor social functioning as they await resettlement.

1 | INTRODUCTION UNHCR (2020), 79.5 million forcibly displaced individ-

uals worldwide sought protection within or outside of
Millions of people are currently undergoing forcible their home country in 2019; 26 million of them were ref-
migration and displacement. According to ugees relocated across international borders

Am J Hum Biol. 2022;34:e23747. wileyonlinelibrary.com/journal/ajhb © 2022 Wiley Periodicals LLC. 1 of 16

https://doi.org/10.1002/ajhb.23747
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 of 16 JANKOVIC-RANKOVIC ET AL.

(UNHCR, 2020). Violent conflicts, economic deprivation, Although definitions vary, interpersonal social support is
and human rights violations are often among the core often conceptualized as the help and care afforded
drivers of contemporary forced human migration. For through close encouragement of social networks/
example, over a million people, mainly from Syria, Iraq, relationships (Berkman et al., 2000). The health-
and Afghanistan, have been forced over the past decade promoting effects of social support are typically investi-
to embark on dangerous migratory journeys to flee war gated in the context of perceived (an individual's possible
and violence and to pursue safety far from their home- access to social support) and/or received social support
lands, with many attempting to reach Europe (actual utilization of support resources; Uchino, 2009).
(Sandalio, 2018; UNHCR, 2019). Moreover, past research has suggested that social support
Refugees' pursuit of protection is often permeated by may buffer against the adverse effects of distressful and
multiple losses and traumatic experiences before they traumatic life events (DeVries et al., 2003) and trauma-
flee, during their journeys, and in post-migration settings related psychopathology (e.g., anxiety, depression, and
(Ben Farhat et al., 2018; Hynie, 2018). These circum- PTSD; Bogic et al., 2012; Charuvastra & Cloitre, 2008;
stances have also simultaneously engendered a funda- Davidson et al., 2008). One pathway through which this
mental break with the well-known conditions of their may occur is through the effect of support on psychologi-
daily lives, including separation from family members cal processes associated with appraisals, emotions, and
and a lack of social connections, such as sources of feelings of control (Uchino, 2009). These processes, in
social/emotional support (Pike & Crocker, 2020; Strang & turn, are likely to be linked to variation in health-
Quinn, 2019). This may represent an understudied con- relevant physiological responses, such as stress-
tributor to their mental health (Bogic et al., 2015). Studies responsive immune system and hormone mechanisms,
show that forced migration and repeated exposure to including the hypothalamic–pituitary–adrenal (HPA) axis
extreme violence and trauma are linked to relatively poor and its production of cortisol (Flinn et al., 2011; Miller
mental well-being among refugee and migrant et al., 2007).
populations (Panter-Brick et al., 2020; Porter & During their transitional migratory period, refugees
Haslam, 2005; Reed et al., 2012; Steel et al., 2009; often experience uncertainty and heightened risks for
Wilson & Thayer, 2018). In particular, they tend to suffer recurrent psychosocial stress that are commonly linked
from higher rates of depression, posttraumatic stress dis- to increased cortisol production (Johnson et al., 1992;
order - PTSD, and other anxiety disorders than the non- Koolhaas et al., 2011; Miller et al., 2007; Panter-Brick
war affected general population (Fazel et al., 2005; et al., 2020). Short-term upregulation of cortisol release is
Morina et al., 2018). beneficial for individuals coping with acute stress via cor-
Refugees also tend to struggle with cultivating social tisol's metabolic, cardiovascular, and neurobiological
support (Miller et al., 2002; Strang & Quinn, 2019), which effects (Dedovic et al., 2009; Dickerson & Kemeny, 2004;
can exacerbate their feelings of loneliness and isolation Flinn et al., 2011). However, repeated exposure to chronic
(Dolberg et al., 2016; Gottvall et al., 2019), with likely psychosocial stress and excess cortisol production can
links to physical and mental well-being (Hawkley & contribute to adverse health outcomes, including cardio-
Cacioppo, 2010). Prior research has investigated associa- vascular disease, metabolic dysfunction, and accelerated
tions between disrupted social support and refugees' cognitive decline (Dickerson & Kemeny, 2004;
mental well-being, primarily in long-settled refugee McEwen, 1998; Miller et al., 2007; Steptoe &
populations (Ahern et al., 2004; Cummings et al., 2011; Kivimäki, 2013). Thus, cortisol is often used as a bio-
Panter-Brick et al., 2020). Consequently, we know little marker of the physiological response to psychosocial
about how the dynamics of low or disrupted social sup- stressors (Ice, 2007; Nepomnaschy et al., 2012). Studies
port interrelate with mental health among refugees in have found that recently settled refugees tend to exhibit
more intermittent, transitional stages of the forced migra- higher cumulative indicators of cortisol production,
tion process. During this period, displaced individuals including in hair (Dajani et al., 2018; Mewes et al., 2017;
often find themselves in challenging socio-ecological con- Panter-Brick et al., 2020) and fingernails (Jankovic-
ditions. In such circumstances, social support can be Rankovic et al., 2020), based on recent/ongoing psycho-
challenging to generate and maintain but it may also be a social stress, fear, and/or uncertainty. Social support's
critical buffer against the effects of trauma and psychoso- protective properties may help attenuate these links;
cial stress. though these associations may vary across displaced
Beyond refugee contexts, social support is a widely populations and be highly context-dependent (García
studied predictor of individuals' well-being and is gener- et al., 2017). Indeed, there is increasing evidence regard-
ally considered a critical protective factor concerning ing associations between greater social support and better
both mental and physical health (Uchino, 2006). neuroendocrine responses (e.g., lower levels of HPA axis
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
JANKOVIC-RANKOVIC ET AL. 3 of 16

responses to stress and/or faster recovery from the physi- home country, followed by approximately 28% of partici-
ological effects of stress; Eisenberger et al., 2007; pants who identified Iran as their homeland. In terms of
Seeman & McEwen, 1996; Sippel et al., 2015) and partnering status, 35 participants were single, 35 married,
immune system functioning (Dickerson & Zoccola, 2009; one separated, and four divorced. The mean number of
McClure et al., 2010). children within the families was 1.2 ± 1.4 SD. We report
Past research has also shown that chronic psychoso- further descriptive statistics in Table 1. In all cases, par-
cial stress may alter immune system function and ticipants were informed about the project and its purpose,
increase vulnerability to adverse health outcomes and our team's permits to work in the area. Informed ver-
(McDade, 2007; Panter-Brick et al., 2008). In particular, bal consent was obtained before any data were collected.
ongoing psychosocial distress and upregulation of the The study was approved by IRB at the University of Notre
HPA axis can negatively affect cellular immunity. Disrup- Dame (protocol # 17-05-3916) and permission was
tion of cellular immunity can trigger viral reactivation of granted by the Commissariat for Refugees and Migration
latent viruses, such as Epstein–Barr virus (EBV; Glaser & of the Republic of Serbia. The present analyses build on
Kiecolt-Glaser, 2005; Panter-Brick et al., 2020). Such reac- our prior work in this area and draws on related data,
tivation may result in increased production of EBV anti- wherefore our methods are similar to those we reported
bodies (McDade, 2002; Sorensen et al., 2009), which previously (Jankovic-Rankovic et al., 2020).
serves as an indicator of stress-related physiological
effects on immune function (McDade et al., 2000). How-
ever, greater social support may enhance an individual's 2.2 | Interview, survey data, and
coping abilities or lead to the appraisal of stimuli as less physiological markers
stressful, thereby ameliorating the adverse effects of HPA
axis overstimulation and ultimately the immune system's 2.2.1 | Semi-structured interviews
responses to challenging events, thereby decreasing
health risks (Uchino, 2006). All participants were met in asylum centers and engaged
In this article, we aim to help address existing gaps in in interviews, which lasted approximately an hour. They
the literature by examining associations between social first reported key socio-demographic information (see
support, mental health, and psychobiology among refu- Table 1) and responded also to questions pertaining to
gee populations in transitional migratory stages. Specifi- their journey and stay in Serbia, including duration of
cally, we draw on data collected from adults (N = 76; the journey, duration of refugee status, and trauma/loss
46% female) residing in Serbia's asylum centers. Based on they experienced at home and during their journey to
our prior work in this area, we predicted that refugees Serbia. Most participants were interviewed in their native
with longer journeys would have lower social support. language (e.g., Farsi, Pashto, Urdu) by the first author
We also predicted that refugees with lower social support with the assistance of a trained interpreter. Participants
would exhibit greater psychosocial stress and more fluent in English preferred to be interviewed in English.
PTSD-like symptoms. Finally, we tested whether refugees Participants were asked questions about their migratory
with lower social support had higher cortisol (measured journey (e.g., What was your journey like?), and the types
via fingernails) and EBV antibody levels (measured via of trauma (physical, psychological, or both) they faced
dried blood spots) than those with higher social support. during their flight to safety (e.g., Did you experience any
traumatic event on your journey?). Next, after completing
the Friendship scale (see below), participants were asked
2 | METHODS to elaborate on their sources of social support and ways
of handling their concerns (e.g., Do you have people you
2.1 | Study population can trust with your intimate thoughts and feelings? Do
you have people you can ask for advice from when you
This mixed-method study was conducted from 30 May to have a problem?).
30 June 2018 in two asylum centers in Serbia, asylum
center “Krnjaca,” located in Belgrade, Serbia, and the
other asylum center “Bogovadja,” situated approximately 2.2.2 | Survey measures of mental well-being
70 km away from Belgrade. A total of 76 participants and psychosocial stress
(41 males; 35 females) had full data and were included in
the present study. The mean age of the participants was We used the 22-item Impact of Event Scale-Revised (IES-
30.1 years ±7.7 SD (range = 18–50 years), and approxi- R; Weiss & Marmar, 1997) to measure subjective distress
mately 48% of participants identified Afghanistan as their caused by traumatic events (alpha = 0.85), the 15-item
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 of 16 JANKOVIC-RANKOVIC ET AL.

TABLE 1 Descriptive statistics (n = 76) Refugee Health Screener 15 (RHS-15; Hollifield

et al., 2013) survey to screen individuals for both trauma-
Mean SD
related mental and physical health problems
Sociodemographic data
(alpha = 0.82), the 10-item Cohen Perceived Stress Scale
Age 30.14 7.75 (PSS; Cohen et al., 1983) to measure more recent psycho-
Number of children 1.29 1.47 social stress (i.e., over the previous four weeks)
Journey length 9.96 10.83 (alpha = 0.68), and finally Friendship scale (FS;
Time in Serbia 14.96 11.03
Hawthorne, 2006) to assess aspects of both perceived
social isolation/loneliness and connectedness over the
Biomarker data
previous four weeks in the refugees' lives (alpha = 0.50).
Nail cortisol (pg/mg)a,b 5.53 20.83 See the Supplemental Methods for further information
c,d
EBV IgG (AU/mL) 160.83 266.56 on these instruments.
Health survey data
RHS-15 total scoree 40.18 12.33
IES-R total score f
56.86 15.64
2.2.3 | Fingernail cortisol measurement
PSS neg. total scoreg 14.39 4.86
We collected fingernail clippings from participants' fin-
h
FS total score 13.55 4.52 gers (all fingers, both hands) at a single time point. Corti-
Partnering statusi (%) sol accumulates in fingernails, which are composed
Partnered 46.67 primarily of keratin, through similar mechanisms as in
Not partnered 53.33 hair, which is a more commonly used medium to mea-
sure the body's longer-term production of cortisol
Country of origin
(e.g., over the previous month; de Berker et al., 2007;
Afghanistan 48.68
Russell et al., 2012). However, based on our team's prior
Iran 28.95 ethnographic research at this site, it would not have been
Syria 3.95 feasible to collect participants' hair because of their reli-
Pakistan 2.63 gious beliefs (e.g., women's reluctance to take off their
Iraq 2.63
hijab in public or in the presence of a male interpreter)
as well as routine cultural practices, such as men keeping
Other <2.63
their hair very short to the scalp. Cortisol measured in
Education fingernails likewise provides a long-term aggregate indi-
<High school 34.21 cator of the body's exposure to cortisol 3–5 months prior
High school 21.05 to sampling (Izawa et al., 2015, 2021; Warnock
>High school 44.74 et al., 2010) and was thus appropriate for our study popu-
lation and research goals.
Trauma
The samples were stored at room temperature in indi-
No trauma 21.05
vidual, ID-coded plastic bags both in the field and at the
Physical 14.47 University of Notre Dame. The fingernail samples were
Psychological 3.95 then shipped to laboratory in the Department of Psycho-
Both 60.53 logical and Brain Sciences at the University of
a
Massachusetts-Amherst, where they were processed
We present raw values for nail CORT but used log-transformed values in all
using the following procedures modified from Doan
the analyses.
et al. (2018). First, each fingernail sample was placed into
b
n = 73; one man and two women did not have CORT data.
c
We present raw values for EBV antibodies but used log-transformed values
a 2.0-ml hardened microcentrifuge tube and weighed.
in all the analyses. Samples were then washed twice for 1 min with iso-
d
n = 66; seven men and three women were EBV seronegative. propanol to eliminate external contamination. After dry-
e
RHS-15 Total Score = Refugee Health Screener 15. ing the washed nails, grinding was performed for 2 min
f
IES-R Total Score = Impact of Event Scale-Revised. using a Mini-BeadBeater-16 (BioSpec, Bartlesville, OK)
g
PSS Neg. Total Score = Cohen Perceived Stress Scale - negatively worded with a single 6.35 mm chrome steel bead. Ground nails
items. were extracted overnight (18–24 h) with 1.5 mL HPLC-
h
FS Total Score = Friendship Scale.
grade methanol, after which the tubes were centrifuged,
i
n = 75; one man did not report partnering status.
1.0 mL of the supernatant was transferred to a clean tube,
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
JANKOVIC-RANKOVIC ET AL. 5 of 16

and the methanol was evaporated. The dried extract was with the data. Next, using two observational techniques -
reconstituted in 0.25 mL of Arbor Assays assay buffer word repetition and the presence or absence of expres-
and then filtered using a Corning Costar Spin-X 0.45 μm sion, the first author laid the groundwork for the theory-
cellulose acetate filter (VWR, Philadelphia, PA) to driven thematic analysis with flexibility in the coding
remove any residual particulate matter. Finally, the rec- process and generating themes (Russell et al., 2016). Dur-
onstituted extracts were analyzed for cortisol in duplicate ing the data coding process, the first author aimed to
along with standards and quality controls using the identify the relations between codes and merged or sepa-
Arbor Assays (Ann Arbor, MI) DetectX enzyme immuno- rated some of them where needed. Lastly, basic themes
assay kit (#K003–H1). Intra- and inter-assay coefficients were grouped under the broader theme of sources and
of variation (CVs) for this assay were <10%. Assay output types of support.
was converted to pg cortisol per mg nail to correct for
variation in sample weight.
3.2 | Statistical analyses

2.2.4 | Dried blood spot collection and EBV All statistical analyses were conducted using Stata
measurement (v. 14.0). We evaluated statistical significance at p ≤ 0.05.
First, primarily for descriptive purposes, we ran Pearson's
We collected four blood spots using standardized filter bivariate correlations for key study continuous variables
paper (Schleicher and Schull #903, Keene, NH) following (see Table 2 for full results). We then used ordinary least
established collection and laboratory procedures squares (OLS) regression to test the relationship between
(McDade et al., 2000). Each participant's finger was refugees' length of journey and social support. Using
cleaned with alcohol and then pricked with a sterile, dis- Akaike Information Criterion (AIC) model selection pro-
posable microlancet. Blood spot samples were dried at cedures (see the Supplemental Methods), we then consid-
room temperature, placed in individual plastic bags, ered participants' sex, age, trauma experiences, and
refrigerated the same day, and stored at 20 C in Serbia; residence time in Serbia as potential covariates for the
they were shipped to the University of Notre Dame for model predicting social support from journey length (see
storage at 80 C. They were later shipped on dry ice to Table 3 for full results).
the University of Oregon for analysis of Epstein–Barr Next, we used multivariate analysis-of-variance
virus (EBV) antibody titer analyses. Specifically, follow- (MANOVA) to test whether social support predicted refu-
ing Eick et al. (2016), EBV immunoglobulin G (IgG) anti- gees' mental health outcomes for each scale (perceived
body levels in DBS were assayed at the Global Health stress; IES-R; RHS-15). We included the following
Biomarker Laboratory at the University of Oregon and covariates, which emerged from AIC model selection pro-
are expressed in the arbitrary units of absorbance units cedures: age, trauma experiences, and journey length.
per milliliter. Levels of IgG antibodies against the viral Following statistically significant MANOVA results, we
coat antigen (VCA) of Epstein–Barr virus were assessed then used the Stata ‘mvreg’ command, which uses OLS
in dried blood spots as described in Eick et al. (2016). All regression, to test for links between the predictors and
samples were run in duplicate, and samples with a the individual dependent variables. The full model results
CV > 15% were rerun. Final EBV IgG titers are reported for these analyses are presented in Table 4, with
in absorbance units (AU)/mL. Intra- and inter-assay CVs covariates included based on AIC model selection proce-
were 3.6% and 13.7%, respectively. The lower limit of dures (see the Supplemental Methods).
detection was 5.17 AU/mL. Seronegativity was defined as
a DBS EBV VCA IgG titer of <22 AU/mL (Eick
et al., 2016). We found ten participants to be EBV sero- 4 | RESULTS
negative. Because of a skewed distribution, the raw EBV
variable was natural log transformed for all analyses. 4.1 | Qualitative results

Participants in the two asylum centers spoke about trau-

3 | DATA ANALYSES matic experiences and perceived and/or received social
support from their family members, spouses, and/or fri-
3.1 | Qualitative analysis ends following displacement. Fifty-one participants
(67.1%) recounted numerous instances in which they
This analysis was guided by the literature on social sup- were reluctant to seek support from their social network,
port, and transcripts were first read to gain familiarity while twenty-five interlocutors (32.9%) expressed their
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 of 16 JANKOVIC-RANKOVIC ET AL.

TABLE 2 Pairwise correlations for key study variables

Variables (1) (2) (3) (4) (5) (6) (7) (8) (9)
(1) FS total score 1.000
(76)
(2) PSS neg. total score 0.29* 1.000
(76) (76)
(3) IES-R total score 0.29* 0.51*** 1.000
(76) (76) (76)
(4) RHS-15 total score 0.32** 0.57*** 0.71*** 1.000
(76) (76) (76) (76)
(5) Age 0.13 0.09 0.005 0.12 1.000
(76) (76) (76) (76) (76)
(6) Journey length 0.27* 0.31** 0.21† 0.26* 0.008 1.000
(76) (76) (76) (76) (76) (76)
(7) Time in Serbia 0.08 0.02 0.03 0.008 0.04 0.06 1.000
(76) (76) (76) (76) (76) (76) (76)
(8) Log nail cortisol 0.21† 0.22† 0.09 0.01 0.03 0.07 0.02 1.000
(73) (73) (73) (73) (73) (73) (73) (73)
†
(9) Log EBV 0.006 0.14 0.20 0.07 0.04 0.05 0.03 0.07 1.000
(66) (66) (66) (66) (66) (66) (66) (66) (66)

Note: Pearson's r is listed in the top row of each cell with the bivariate sample size in parentheses in the second row. †p < 0.1, *p < 0.05, **p < 0.01,
***p < 0.001.

content at having close confidants nearby. As we explain to Pakistan and then to Iran […] we did not
below, refugees routinely faced various social and psy- eat for 3–4 days or drink any water for two
chological hardships on their journey to Serbia that seem days while hiding in the forest […] I was with
to have affected their trust in people and eagerness to cre- my family in Greece, and we wanted to go to
ate and maintain profound relational exchanges with Austria. We paid smugglers to take us there.
close confidants and other refugees in asylum centers. My family was placed in one vessel, and I in a
Forced migration of people is routinely marked by different one. The police intercepted my boat,
recurrent trauma and psychosocial stress. Such experi- but not the one my family was in; they man-
ences are inextricable parts of the drive of refugees to aged to reach Germany. I was deported back
seek safety across international borders to escape debili- to Bulgaria. Being separated from my family
tating conditions in their homelands. Due to numerous affected my sanity… That's why I decided to
structural constraints and restrictive migration policies, continue my journey and come to Serbia
refugees are often compelled to navigate rugged terrains, (Asylum center “Krnjaca,” May 30, 2018).
physical barriers, police patrols, and smuggling networks
on their lengthy journey to reach protection across inter- Even as they negotiate their lives in this new liminal
national borders. Not all refugees succeed in their first space of an asylum center, most study participants con-
attempt to cross multiple borders and reach safety farther tinue to grapple with the hardships of journey-related
afield. Caught in a vicious cycle of numerous border- trauma and loss, as well as uncertainty and a new life in
crossing attempts, refugees are routinely subjected to var- the shadows on a daily basis. Such past and present trau-
ious torments, including separation from their loved matic experiences are likely to cause disruptions in their
ones, endless walking for days without food and/or system of social support. Specifically, forced
water, and physical/psychological abuse at the hands of displacement-related experiences have opened the door to
the border police and smugglers. One participant said: different forms of mistrust and constraints upon their
social and emotional interactions. Most interlocutors
I traveled with my family. We came from (67.1%) felt that they would be better off if they kept their
Afghanistan to Serbia on foot. First, we went emotions and concerns to themselves instead of sharing
 JANKOVIC-RANKOVIC ET AL.

TABLE 3 Model fit statistics (AIC) for key covariates predicting refugees' social support, mental well-being, and physiological markersa

Predicting social support from journey lengtha

Base model AIC: 442.3

Covariates AIC with the covariate added to the base model

Sex 443.4
Age 442.9
Trauma 444.3
Time in Serbia 443.5
Predicting outcomes from social supportb
FS base model for PSS neg. FS base model for RHS-15 FS base model for IES-R FS base model for CORT FS base model for EBV
AIC: 452.2 AIC: 592.2 AIC: 629.8 AIC: 226.9 AIC: 188.1
Covariates AIC with the covariate added to the base model
Sex 453.3 592.5 630.1 228.1 189.8
Age 452.7 591.7 631.7 228.9 190.0
Journey length 449.2 591.3 630.2 228.9 189.8
Trauma 453.6 589.3 629.2 228.4 190.1
Time in Serbia 454.2 594.1 631.8 228.8 190.0
a
The base model predicts social support (the FS scale) from journey length.
b
The base model predicts each outcome variable (i.e., the mental well-being surveys and physiological markers) from social support (the FS scale).
7 of 16

15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 of 16 JANKOVIC-RANKOVIC ET AL.

TABLE 4 MANOVA models predicting refugees' mental well-being and follow-up OLS regression models

Model 1 Model 2

Predictors Pillai's trace F Df Sig. Pillai's trace F Df Sig.

FS scores 0.1256 3.45 (1, 74) 0.021 0.1095 2.83 (4, 71) 0.048
Journey length 0.0501 1.21 (4, 71) 0.311
Trauma 0.0748 1.86 (4, 71) 0.144
Age 0.0732 1.82 (4, 71) 0.152
OLS regression models following the MANOVA results
(N = 76 for all models)
RHS-15

Model 1 Model 2

b SE p b SE p
FS scores 0.87 (0.29) 0.005 0.84 (0.30) 0.007
Age 0.35 (0.17) 0.042
Journey length 0.10 (0.13) 0.436
Trauma 7.72 3.48 0.030
Model R2 0.10 0.21
IES-R

Model 1 Model 2

b SE p b SE p
FS scores 1.01 (0.38) 0.010 0.96 (0.38) 0.013
Trauma 6.65 4.20 0.118
Model R2 0.08 0.11
PSS neg.

Model 1 Model 2

b SE p b SE p
FS scores 0.31 (0.11) 0.010 0.24 (0.12) 0.050
Journey length 0.11 (0.05) 0.028
Model R2 0.08 0.10

them with close confidants. They also expressed their dis- centers. They reported relying on their own personal
trust in people. One interlocutor described it as following: capacity rather than seeking help from others in dealing
with problems. They typically resorted to activities such
It is hard to trust people and relate to them as walking, self-analysis, and thinking. Participants also
when I experienced bad things in my life and expressed their unwillingness to engage in or uphold
on the journey to Serbia. I am here with my social interactions with other refugees while still in tran-
wife and kids, but I do not share my prob- sit. Two interlocutors explained:
lems and feelings with them. I often feel
alone; it is hard to find people outside your I do not have anyone to share my problems
family you can trust and share your prob- and concerns with… I do not trust anyone. I
lems or concerns with (Asylum center would love to have someone to share my
“Krnjaca,” June 8, 2018). thoughts and feelings with, but I have not
found such a person so far. When I have prob-
As a consequence of their journey-related experi- lems, I go somewhere outside the asylum cen-
ences, refugees' mistrust continues to persist and pervade ter, sit, and think through those issues alone
their interactions with other people in the asylum (Asylum center “Krnjaca,” June 8, 2018).
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
JANKOVIC-RANKOVIC ET AL. 9 of 16

I do not talk with my family about my prob- actual utilization of social support afforded the partici-
lems and concerns. Sometimes I feel I am pants a sense of security, relief, and ease. They felt they
alone and do not have any real friends here. did not have to go through all problems alone. Trust in a
If I have a problem, I take a walk and think family member, partner, and/or a friend to share most
about possible solutions (Asylum center intimate thoughts when they felt sad, lonely, and disap-
“Krnjaca,” June 10, 2018). pointed comforted the participants and provided them
with the opportunity to mentally unload. Two partici-
In the state of shattered social support systems and pants stated the following:
impaired social relationships, refugees may be at risk of
experiencing social isolation and loneliness, even when I talk to my wife about everything. She is my
they have their families and friends nearby. biggest support and a person of trust. I also
The intricate intersection of border crossing experi- talk to my family back in my home country
ences, daily struggles, and challenges in maintaining (Asylum center “Bogovadja,” June 13, 2018).
social support appears to have ripple effects on refu-
gees and their families. Social relationships and inter- I have friends with whom I talk about my
actions are critical to human functioning and well- problems and resolutions. I can also share
being. In the context of disrupted social support, a my concerns and feelings with them. I have
warm hug, a good listener, even a handshake might be been here in Serbia for two years now, and
a profound act of human connection and utter care. during the time, I have managed to find a
When these intimate and meaningful interactions turn few people I can trust and talk with about
into nonexistent or dysfunctional ones, they may add anything (Asylum center “Bogovadja,” June
another layer to the refugees' ongoing stress and fur- 17, 2018).
ther intensify the decline of social contact. According
to most of the participants (67.1%), numerous social Perceiving and receiving support was imperative to
and psychological hardships appear to have taken a toll some participants. They were more likely to feel confi-
on their readiness to engage in profound relational dent and reassured, which further enhanced their trust in
exchanges or even share their concerns with close con- the people with whom they unreservedly shared their
fidants. Specifically, a desire to protect their loved ones thoughts, emotions, and concerns without embarrass-
from any potential problems, reluctance to engage in ment or discomfort. Participants also expressed satisfac-
new social interactions, and readiness to tackle their tion with having family and friends who helped them
problems alone are some of the core remarks that have face their problems, provided place to talk about any-
informed the participants' narratives. Such disconnec- thing, and above all, who understood their past and pre-
tions from their familial and social networks may exac- sent struggles. Refugees' lived realities have been and still
erbate their feelings of loneliness and social isolation are fraught with worries and trauma. Existing meaning-
and critically contribute to poor health outcomes. One ful affective social ties and support are likely to buffer the
participant explained: effects of uncertainty and psychosocial stress during the
emotionally and mentally demanding periods between
My problem is that I keep everything to displacement and resettlement.
myself. I can talk to my wife, but, at the While a small number of participants expressed their
same time, I do not want to put a huge bur- content with having family and friends nearby, most of
den on her. When I feel that way, I usually them, as we noted, said that they preferred not to rely
go outside the center or occupy myself with even on the people they considered a reliable source of
some physical work to distract my thoughts protection, particularly during challenging times. They
or to find a solution (Asylum center chose to face problems and cope with them in seclusion
“Bogovadja,” June 12, 2018). while living in uncertain conditions. Refugees' narratives
in which they recounted physical and psychosocial diffi-
In contrast, 25 interlocutors (23.9%) expressed their culties en route apparently continue to affect their social-
contentment with having their spouses, friends, and fam- ity and systems of support and to reinforce mistrust in
ily members nearby. These are their go-to people of trust, others while living in asylum centers. Moreover, refugees
particularly when they are experiencing challenging and tended to refrain from establishing and preserving social
difficult times. Having close confidants in an asylum cen- relationships, emphasizing their and other people's inten-
ter reassured the participants that they were not lonely tion to continue the journey to the desired destination. In
and/or friendless. Moreover, potential access to and/or the context of such impaired social functioning, refugees
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 of 16 JANKOVIC-RANKOVIC ET AL.

often lack appropriate support, coping responses, and We then used MANOVA models to examine the link
care, which may be critical in overcoming the adverse between social support and outcome variables associated
impacts of trauma and stress. with refugees' mental health (perceived stress; IES-R;
RHS-15; Figure 1). In Model 1 predicting refugees' mental
health, individuals with lower social support exhibited
5 | QUANTITATIVE RESULTS poorer mental well-being (all p's < 0.05; see Table 4).
Addition of trauma, age, and journey length as covariates
5.1 | Bivariate correlations in Model 2 revealed that refugees who reported trauma
on their mobility journeys reported poorer physical
Bivariate correlations showed that the refugees' social sup- health and mental well-being (RHS-15 scores; p < 0.05),
port was negatively correlated with their mental health while refugees with longer journeys had higher recent
(RHS-15, r = 0.32, p = 0.005; IES-R, r = 0.29, perceived stress (PSS negative sub-scale scores; p < 0.05;
p = 0.010; and, PSS negative sub-scale, r = 0.29, see Table 4 for full results). In OLS regression models of
p = 0.010). Although neither correlation was statistically refugees' mental health (RHS-15 scores), post-traumatic
significant, refugees' social support was also negatively cor- stress-related symptomology (IES-R scores), and recent
related with fingernail cortisol level (r = 0.21, p = 0.072), perceived stress (PSS negative sub-scale), refugees with
while their recent perceived stress was positively correlated lower social support reported poorer mental well-being
with their fingernail cortisol level (r = 0.22, p = 0.051; see in MANOVA (Model 1; all p-values<0.05; see Table 4).
Table 2 for full results). With the addition of trauma, age, and journey length in
Model 2, refugees with longer and more traumatic jour-
neys exhibited poorer mental and physical health than
5.2 | Refugees' well-being and social other participants (RHS-15 scores; all p values <0.05). In
support Model 2, the finding for journey length remained signifi-
cant for PSS negative sub-scale scores (p < 0.05), while
In predicting the refugees' social support, we found that experiences of trauma were no longer significantly linked
individuals with longer journeys reported lower social to refugees' IES-R scores (p > 0.05; see Table 4 for full
support (b; 95% CI: 0.11; 0.20, 0.02; p: 0.017; Model results).
R2: 0.07). Based on model selection procedures, sex, age, Finally, in predicting the refugees' cortisol, we found
trauma, and residence time in Serbia did not improve that individuals who reported lower social support
model fit (Table 3) and thus we did not include them as tended to exhibit higher cortisol levels, though our find-
covariates. ing did not reach statistical significance (p = 0.072).

F I G U R E 1 Line of best fit plots and 95% CI with scatter plots overlaid for refugees' mental well-being and physical health (RHS score),
PTSD-related symptoms (IES-R score), and recent psychosocial stress (PSS neg. Score) based on their reported social support (friendship
scale score). The y axis in each panel is centered near the mean and ranges from approximately 2 SD to +2 SD for RHS, IES-R, and PSS
neg scores, respectively. See Table 2 for the bivariate correlations between the featured variables in the figure and Table 4 for the full model
results
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
JANKOVIC-RANKOVIC ET AL. 11 of 16

Meanwhile, refugees' social support was not statistically on their protracted journeys (Arsenijevic et al., 2017;
significantly linked to refugees' EBV antibody titers Sandalio, 2018).
(p > 0.05). Based on model selection procedures, Our findings are in line with prior research showing
sociodemographic covariates, trauma, and residence time that refugees' exposure to stress and trauma are linked to
in Serbia did not improve model fit for physiological lower social support, greater feelings of loneliness, and
markers (Table 3). poor social interactions once they have reached security
across borders (Liamputtong & Kurban, 2018). In past
studies, lower social support was linked to poorer mental
6 | DISCUSSION well-being (Hynie, 2018; Pike & Crocker, 2020; Porter &
Haslam, 2005). In particular, a study among Syrian refu-
In this study of recently settled refugees, we tested for gees resettled in Sweden found that participants who
relationships between their social support and the length reported higher social support were less likely to suffer
of their journey from their home countries to Serbia and from PTSD symptomology than their peers with lower
how social support was linked to mental well-being and social support (Gottvall et al., 2019). Similarly, another
physiological markers relevant to psychosocial stress. The study among Bosnian refugees in the U.S. found that
small body of prior work that has examined links post-migration social isolation was significantly associ-
between recently settled refugees' social support and ated with PTSD symptomology in the community group
health suggests that enhanced social support may have compared to participants in the clinical group (Miller
protective effects on refugees' health and buffer against et al., 2002).
their likely loneliness and isolation (Correa-Velez Furthermore, we found that recently settled refugees
et al., 2020; Stewart, 2014). These past findings help with lower social support reported poorer mental and
underscore the importance of our findings here, which physical well-being, greater PTSD-related symptoms, and
indicate that migration-related trauma and stress higher recent perceived stress than those with higher
adversely affect refugees' systems of support. Further- social support. Similarly, Schweitzer et al.' (2006) study of
more, we discuss short- and long-term implications for recently settled Sudanese refugees in Australia found that
the well-being and resilience of refugee populations. We loss of social support contributed to higher levels of psy-
contextualize our findings regarding forced migration chosocial stress and poorer mental health outcomes.
and social support vis-à-vis concepts pertaining to Meanwhile, among Q'eqchi' refugee women in Southern
chronic psychosocial stress and neuroendocrine Mexico, those with weakened social support networks
responses (Bartram, 2015; Kawachi & Berkman, 2001; due to forced displacement, violence, and encampment
Miller et al., 2007; Panter-Brick et al., 2020). reported higher stress and symptoms of trauma than
women with stronger networks (Warner, 2007). Although
further research is needed, the congruity of our findings
6.1 | Mental health suggest that social support is a robust correlate of refu-
gees' feelings of better mental and physical health across
We found that refugees who had experienced longer jour- cultural boundaries and settlement sites, which is unsur-
neys reported lower social support than those with prising given the adversity and challenges refugees face
shorter journeys, on average. Although our study is cross- (Mawani, 2014; Stewart, 2014). Our findings add to this
sectional, this aligns with the idea that traumatic and literature by indicating that the protective properties of
stressful experiences may fracture existing sources of social support are likely critical for alleviating and buffer-
social support and impair the generation of new social ing against the adverse mental health effects of
relationships. The lives of many forcibly displaced people displacement-related psychosocial stress and trauma,
are characterized by recurrent stress and trauma at all although longitudinal and intervention-based study
stages of the migration process (Arsenijevic et al., 2017; designs are required to properly disentangle temporal
Zimmerman et al., 2011). The closure of borders and gov- and potentially causal relationships between refugee
ernments' abilities to police, control, discourage, and cur- social support and well-being.
tail refugee movements across international borders The qualitative dimensions of our study are likewise
(Gerard & Pickering, 2013; Weber, 2017) have impeded consistent with the intersection between refugees'
safe and legal migration of most forcibly displaced indi- journey-related distress and social relationship difficulties
viduals. Consequently, many refugees must embark on in Serbia. Specifically, the participants' personal and
dangerous and risky routes to seek protection outside culture-specific accounts concerning social support rev-
their homeland, thereby exposing themselves to various ealed that most recently settled refugees had experienced
physical hardships, psychosocial stress, and uncertainty numerous social and psychological hardships during
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
12 of 16 JANKOVIC-RANKOVIC ET AL.

their migratory journey to Serbia, including physical suggest that establishing peer support and community-
trauma, violence, uncertainty, family separation, and fear based activities may help refugees regain trust in others
that affected their engagement in profound relationships while encouraging and empowering them to engage
and willingness to share their concerns with close confi- socially since the effects of impaired social support or
dants. Once in an asylum center, they characterized those relationships over a more extended period may lead to
distressing experiences as constraints on their familial chronic illnesses that can negatively impact their subse-
and social networks while simultaneously enhancing quent integration in post-migration settings (Ahmad
their desire to protect their loved ones from personal et al., 2020). We hope that this study lays a foundation
problems and additional stress, frustration, and despair. for future research of social support and its influence on
In our study, most participants in the Serbian asylum health, particularly among displaced populations in more
centers came from countries where culture is collective intermittent transitional stages of migration, as a deeper
in its orientation. Cultures with collectivist worldviews understanding of this relationship may be critical in
tend to prioritize the benefit of the community and social designing effective interventions to mitigate the adverse
harmony that promote family and group well-being com- health effects of confined social settings on refugees.
pared to more individualistic cultures in which value is
placed on independence, autonomy, and self-assertion
(Buse et al., 2013; Mojaverian & Kim, 2013). Therefore, 6.2 | Cortisol, EBV, and social support
refugees' propensities towards seclusion and loneliness
weakened their own social ties because they deliberately Complementing our findings regarding social support
curtailed their relations with family and friends in the and health, we found that refugees who reported lower
asylum center, yet this may have helped safeguard the social support tended to exhibit higher fingernail cortisol
welfare of their loved ones and promoted social cohesion levels. While the association did not reach statistical sig-
within the family. In their qualitative interviews, some nificance, we observed patterns in the same direction for
participants did express the desire to withhold their own associations between low social support levels and higher
struggles with poor mental well-being to buffer their cortisol observed previously (Cacioppo et al., 2000;
loved ones. Past studies in this domain revealed that Matheson et al., 2008) with an effect size (r = 0.21) con-
some refugees feared future loss and consequently sistent with those in prior studies of social support, psy-
showed a tendency to isolate and refrain from investing chosocial stress, and neuroendocrine responses (Chida &
affection in social interactions, which might further dam- Steptoe, 2009; Rosal et al., 2004). We also found that
age their relationships with their close confidants recently settled refugees, who reported higher recent psy-
(Blackwell, 1993; Blair, 2001). Aligning with this perspec- chosocial stress, had higher cortisol, aligning with our
tive, our findings further highlight the potential positive past results and other findings (Dajani et al., 2018;
effects that establishing support programs in transitional Jankovic-Rankovic et al., 2020; Mewes et al., 2017).
settlements may have on improving refugees' existing Though this latter finding was not the primary focus of
relationships and expanding their opportunities to meet the study, the patterns of our results for cortisol, social
and socialize with new people, potentially reducing their support, and recent psychosocial stress hint at the need
social isolation. to understand further these bio-psychosocial processes at
Lastly, refugees' narratives revealed that the transi- transitional migratory stages, given the well-established
tional stage in their migration process dramatically links between chronically elevated cortisol and poorer
affected their social interactions with other refugees health outcomes (Herane-Vives et al., 2018; Staufenbiel
residing in asylum centers. As they contemplate their et al., 2013).
next steps in their migratory journeys, refugees' ability We note that we did not observe the same patterns
and willingness to build strong social ties and support for social support and the stress-relevant physiological
decline because they often redirect their energy and time measure of EBV antibodies. Some past studies have
to finding ways and critical resources necessary to con- shown positive correlations between higher EBV anti-
tinue their journey. Furthermore, because of past trau- bodies and lower levels of social support. However, those
matic experiences, refugees' mistrust of other people studies have often been conducted in Euro-American
likely contributes to their reluctance to establish and populations (e.g., focusing on adult caregivers or univer-
maintain social relationships while in transit. Thus, sity students) (Glaser et al., 1985; Kiecolt-Glaser
weakened social relationships, lack of trust, and patterns et al., 1991). In those settings, individuals are typically
of social and emotional disengagement may render every- exposed to substantially different psychosocial and
day life more difficult and have adverse effects on refu- ecoimmunological contexts than the refugees we focused
gees' mental health (Biehl, 2015). Our qualitative findings on here, who are primarily individuals of Middle Eastern
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
JANKOVIC-RANKOVIC ET AL. 13 of 16

origin (McDade et al., 2016; Panter-Brick et al., 2020). focused on EBV antibodies from dried blood spots as an
Our result suggest that more research is needed among indicator of cell-mediated immunocompetence sensitive
diverse refugee populations to better understand under to chronic psychosocial stress, rather than a broader
what conditions psychological factors and social support panel of immune markers, such as a range of pro- and
may interrelate with immune markers, such as EBV anti- anti-inflammatory cytokines (McDade, 2007; Miller
body levels. et al., 2007). Thus, in our future work at this site and else-
where in the region, our goal is to expand the breadth of
our immune marker analyses to look at C-reactive pro-
7 | LIMITATIONS tein (CRP) and IgG and IgE antibodies (Urlacher
et al., 2018).
There are several limitations to this study that merit dis-
cussion. First, its cross-sectional design precludes us from
assessing directional inferences between variables in our 8 | CONCLUSION
key results. For example, we predicted that lower social
support would be associated with the refugees' poorer Social support is well-characterized as a critical factor
mental health profiles, consistent with the idea that and essential component in maintaining an individual's
social support may buffer against trauma-related psycho- health and well-being (Uchino, 2006). In the present
pathology and enhance coping abilities and resilience to study, we build upon this research to show that recently
stress. However, it is also possible that poor mental settled refugees, who reported lower social support, had
health can impair an individual's ability to cultivate sup- poorer mental well-being and higher post-traumatic and
port and social relationships (Bogic et al., 2012; Davidson recent perceived stress scores than other participants. We
et al., 2008). Longitudinal and intervention-based study also found that refugees that had experienced longer
designs would help disentangle key health-related drivers journeys reported lower social support. Generally, our
and social processes during these challenging periods fol- correlative findings are consistent with the notion that
lowing forced migration. enhanced social support is likely a necessary means and
We also measured cortisol from fingernails, a rela- resource for the recovery, resilience, and health of
tively new approach to studying the body's cumulative recently settled refugees as they await permanent
exposure to the hormone. Prolonged exposure to resettlement. Additionally, given that refugee populations
demanding psychosocial conditions may lead to maladap- tend to experience ongoing uncertainty and stress (Ben
tive neuroendocrine responses. Notably, they can disrupt Farhat et al., 2018) and suffer from higher rates of mental
or dysregulate the production of cortisol throughout the disorders (Fazel et al., 2005; Morina et al., 2018), we join
day, affecting its diurnal curve, which has been linked to calls to consider the refugees' social support and services
an array of health outcomes (Adam et al., 2017; Fries a public health policy agenda (Gottvall et al., 2019).
et al., 2009). In our study, it was not feasible to collect
and store repeated samples of blood or saliva throughout CONFLICT OF INTEREST
the day to capture those diurnal dynamics in this study. The authors declare that they have no conflict of interest.
We acknowledge that a longer-term, cumulative indicator
of the body's exposure to cortisol, such as in nails, cannot DA TA AVAI LA BI LI TY S T ATE ME NT
address dynamic diurnal aspects of HPA functioning and The data that support the findings of this study are avail-
health. That said, our study aimed at measuring corre- able from the corresponding author upon reasonable
lates of chronic psychosocial stress and trauma, rendering request.
cortisol measured in fingernails – an aggregate indicator
of the prolonged body's exposure to cortisol 3–5 months ORCID
prior to sampling (Izawa et al., 2021; Warnock Jelena Jankovic-Rankovic https://orcid.org/0000-0002-
et al., 2010) - an appropriate approach for our study pop- 5381-0467
ulation and research goals. Geeta N. Eick https://orcid.org/0000-0001-7512-3265
Lastly, we measured EBV antibodies, an indirect cell- Lee T. Gettler https://orcid.org/0000-0001-7325-8852
mediated marker of chronic stress and trauma (McDade
et al., 2000). Research has shown a link between chronic RE FER EN CES
psychosocial stress and increased EBV antibody level Adam, K. E., Quinn, E. M., Tavernier, R., McQuillan, T. M.,
(Sorensen et al., 2009; Worthman & Panter-Brick, 2008). Dahlke, A. K., & Gilbert, E. K. (2017). Diurnal cortisol slopes
Since we wanted to keep our protocol as minimally- and mental and physical health outcomes: A systematic review
invasive as possible and due to funding constraints, we and meta-analysis. Psychoneuroendocrinology, 83, 25–41.
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
14 of 16 JANKOVIC-RANKOVIC ET AL.

Ahern, J., Galea, S., Fernandez, W. G., Koci, B., Waldman, R., & Cummings, S., Sull, L., Davis, C., & Worley, N. (2011). Correlates of
Vlahov, D. (2004). Gender, social support, and posttraumatic depression among older Kurdish refugees. Social Work, 56,
stress in postwar Kosovo. The Journal of Nervous and Mental 159–168.
Disease, 192(11), 762–770. Dajani, R., Hadfield, K., van Uum, S., Greff, M., & Panter-Brick, C.
Ahmad, F., Othman, N., & Lou, W. (2020). Posttraumatic stress dis- (2018). Hair cortisol concentrations in war affected adolescents:
order, social support and coping among afghan refugees in A prospective intervention trial. Psychoneuroendocrinology, 89,
Canada. Community Mental Health Journal, 56(4), 597–605. 138–146.
Arsenijevic, J., Schillberg, E., Ponthieu, A., Malvisi, L., … Davidson, G. R., Murray, K. E., & Schweitzer, R. (2008). Review of
Zachariah, R. (2017). A crisis of protection and safe passage: refugee mental health and wellbeing: Australian perspectives.
Violence experienced by migrants/refugees travelling along the Australian Psychologist, 43(3), 160–174.
Western Balkan corridor to northern Europe. Conflict and de Berker, D. A., Andre, J., & Baran, R. (2007). Nail biology and nail
Health, 11(1), 1–9. science. International Journal of Cosmetic Science, 29, 241–275.
Bartram, D. (2015). Forced migration and “rejected alternatives”: A Dedovic, K., Duchesne, A., Andrews, J., Engert, V., &
conceptual refinement. Journal of Immigrant and Refugee Stud- Pruessner, J. C. (2009). The brain and the stress axis: The neu-
ies, 13(4), 439–456. ral correlates of cortisol regulation in response to stress.
Ben Farhat, J., Blanchet, K., Bjertrup, P. J., … Cohuet, S. (2018). NeuroImage, 47(3), 864–871.
Syrian refugees in Greece: Experience with violence, mental DeVries, A. C., Glasper, E. R., & Detillion, C. E. (2003). Social mod-
health status, and access to information during the journey and ulation of stress responses. Physiology & Behavior, 79, 399–407.
while in Greece. BMC Medicine, 16(1), 1–12. Dickerson, S. S., & Kemeny, M. E. (2004). Acute stressors and corti-
Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From sol responses: A theoretical integration and synthesis of labora-
social integration to health: Durkheim in the new millennium. tory research. Psychological Bulletin, 130(3), 355–391.
Social Science & Medicine, 51(6), 843–857. Dickerson, S. S., & Zoccola, P. M. (2009). Toward a biology of social
Biehl, K. (2015). Governing through uncertainty experiences of support. In S. J. Lopez & C. R. Snyder (Eds.), Oxford library of
being a refugee in Turkey as a country for temporary asylum. psychology. Oxford handbook of positive psychology (pp. 519–
Social Analysis, 59(1), 57–75. 526). Oxford University Press.
Blackwell, R. D. (1993). Disruption and reconstitution of family, Doan, S. N., Deyoung, G., Fuller-Rowell, T. E., Liu, C., & Meyer, J.
network, and community systems following torture, organized (2018). Investigating relations among stress, sleep and nail cor-
violence, and exile. In J. P. Wilson & B. Raphael (Eds.), Interna- tisol and DHEA. Stress, 21(2), 188–193.
tional handbook of traumatic stress syndromes (pp. 733–741). Dolberg, P., Shiovitz-Ezra, S., & Ayalon, L. (2016). Migration and
Plenum Press. changes in loneliness over a 4-year period: The case of older
Blair, R. G. (2001). Mental health needs among Cambodian refu- former Soviet Union immigrants in Israel. European Journal of
gees in Utah. International Social Work, 44(2), 179–196. Ageing, 13(4), 287–297.
Bogic, M., Ajdukovic, D., Bremner, S., & Franciskovic, T. (2012). Eick, G., Urlacher, S. S., McDade, T. W., Kowal, P., &
Factors associated with mental disorders in long-settled war Snodgrass, J. J. (2016). Validation of an optimized ELISA for
refugees: Refugees from the former Yugoslavia in Germany, quantitative assessment of Epstein-Barr virus antibodies from
Italy and the UK. British Journal of Psychiatry, 200(3), 216–223. dried blood spots. Biodemography and Social Biology, 62(2),
Bogic, M., Njoku, A., & Priebe, S. (2015). Long-term mental health 222–233. https://doi.org/10.1080/19485565.2016.1169396
of war-refugees: A systematic literature review. BMC Interna- Eisenberger, N. I., Taylor, S. E., Gable, S. L., Hilmert, C. J., &
tional Health and Human Rights, 15(1), 29. Lieberman, M. D. (2007). Neural pathways link social support
Buse, N. A., Bernacchio, C., & Burker, E. J. (2013). Cultural varia- to attenuated neuroendocrine stress responses. NeuroImage
tion in resilience as a response to traumatic experience. Journal (Orlando, Fla.), 35(4), 1601–1612. https://doi.org/10.1016/j.
of Rehabilitation, 79(2), 15. neuroimage.2007.01.038
Cacioppo, J. T., Ernst, J. M., Burleson, M. H., … Berntson, G. G. Fazel, M., Wheeler, J., & Danesh, J. (2005). Prevalence of serious
(2000). Lonely traits and concomitant physiological processes: mental disorder in 7000 refugees resettled in Western countries:
The Macarthur social neuroscience studies. International Jour- A systematic review. Lancet, 365, 1309–1314.
nal of Psychophysiology, 35, 143–154. Flinn, M. V., Nepomnaschy, P. A., Muehlenbein, M. P., & Ponzi, D.
Charuvastra, A., & Cloitre, M. (2008). Social bonds and posttraumatic (2011). Evolutionary functions of early social modulation of
stress disorder. Annual Review of Psychology, 59, 301–328. hypothalamic-pituitary-adrenal Axis development in humans.
Chida, Y., & Steptoe, A. (2009). Cortisol awakening response and Neuroscience and Biobehavioral Reviews, 35(7), 1611–1629.
psychosocial factors: A systematic review and meta-analysis. Fries, E., Dettenborn, L., & Kirschbaum, C. (2009). The cortisol
Biological Psychology, 80(3), 265–278. awakening response (CAR): Facts and future directions. Inter-
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global mea- national Journal of Psychophysiology, 72(1), 67–73.
sure of perceived stress. Journal of Health and Social Behavior, García, A. R., Gurven, M., & Blackwell, A. D. (2017). A matter of
24(4), 385–396. perception: Perceived socio-economic status and cortisol on the
Correa-Velez, I., Green, A., Murray, K., Schweitzer, R. D., Island of Utila, Honduras. American Journal of Human Biology,
Vromans, L., Lenette, C., & Brough, M. (2020). Social context 29(5), 1–17. https://doi.org/10.1002/ajhb.23031
matters: Predictors of quality of life among recently arrived ref- Gerard, A., & Pickering, S. (2013). Gender, securitization and tran-
ugee women-at-risk living in Australia. Journal of Immigrant sit: Refugee women and the journey to the EU. Journal of Refu-
and Refugee Studies, 18(4), 498–514. gee Studies, 27(3), 338–359.
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
JANKOVIC-RANKOVIC ET AL. 15 of 16

Glaser, R., & Kiecolt-Glaser, J. K. (2005). Stress-induced immune eastern refugees living in Melbourne, Australia. Children and
dysfunction: Implications for health. Nature Reviews. Immunol- Youth Services Review, 85, 99–106.
ogy, 5(3), 243–251. Matheson, K., Jorden, S., & Anisman, H. (2008). Relations between
Glaser, R., Kiecolt-Glaser, J. K., Speicher, C. E., & Holliday, J. E. trauma experiences and psychological, physical and neuroen-
(1985). Stress, loneliness, and changes in herpesvirus latency. docrine functioning among Somali refugees: Mediating role of
Journal of Behavioral Medicine, 8, 249–260. coping with acculturation stressors. Journal of Immigrant and
Gottvall, M., Vaez, M., & Saboonchi, F. (2019). Social support atten- Minority Health, 10(4), 291–304.
uates the link between torture exposure and post-traumatic Mawani, N. M. (2014). Social determinants of refugee mental
stress disorder among male and female Syrian refugees in health. In L. Simich & L. Andermann (Eds.), Refuge and resil-
Sweden. BMC International Health and Human Rights, 19, 28. ience: Promoting resilience and mental health among resettled
Hawkley, L. C., & Cacioppo, J. T. (2010). Loneliness matters: A the- refugees and forced migrants (pp. 27–50). Springer.
oretical and empirical review of consequences and mecha- McClure, H. H., Martinez, C. R., Jr., Snodgrass, J. J., … McDade, T. W.
nisms. Annals of Behavioral Medicine, 40(2), 218–227. (2010). Discrimination-related stress, blood pressure and Epstein-
Hawthorne, G. (2006). Measuring social isolation in old adults: Barr virus antibodies among Latin American immigrants in Ore-
Development and initial validation of the friendship scale. gon, US. Journal of Biosocial Science, 42, 433–461.
Social Indicators Research, 77, 521–548. McDade, T. W. (2002). Status incongruity in Samoan youth: A bio-
Herane-Vives, A., Fischer, S., De Angel, V., Wise, T., … Cleare, A. J. cultural analysis of culture change, stress, and immune func-
(2018). Elevated fingernail cortisol levels in major depressive tion. Medical Anthropology Quarterly, 16(2), 123–150.
episodes. Psychoneuroendocrinology, 88, 17–23. McDade, T. W. (2007). Measuring immune function: Markers of
Hollifield, M., Verbillis-Kolp, S., Farmer, B., Toolson, E. C., cell-mediated immunity and inflammation in dried blood spots.
Woldehaimanot, T., Yamazaki, J., Holland, A., St Clair, J., & In G. H. Ice & G. D. James (Eds.), Measuring stress in humans:
Soohoo, J. (2013). The refugee health Screener-15 (RHS-15): A practical guide for the field (pp. 181–207). Cambridge Univer-
Development and validation of an instrument for anxiety, sity Press.
depression, and ptsd in refugees. General Hospital Psychiatry, McDade, T. W., Georgiev, A. V., & Kuzawa, C. W. (2016). Trade-offs
35(2), 202–209. between acquired and innate immune defenses in humans.
Hynie, M. (2018). The social determinants of refugee mental health Evolution, Medicine, and Public Health, 2016(1), 1–16.
in the post-migration context: A critical review. Canadian Jour- McDade, T. W., Stallings, J. F., Angold, A., Costello, E. J.,
nal of Psychiatry, 63(5), 297–303. Burleson, M., Cacioppo, J. T., Glaser, R., & Worthman, C. M.
Ice, G. H. (2007). Measuring emotional and behavioral response. In (2000). Epstein-Barr virus antibodies in whole blood spots: A
G. H. Ice & G. D. James (Eds.), Measuring stress in humans: A practi- minimally invasive method for assessing an aspect of cell-
cal guide for the field (pp. 60–93). Cambridge University Press. mediated immunity. Psychosomatic Medicine, 62(4), 560–567.
Izawa, S., Keiichi, M., Tsuchiya, M., Mitani, T., Midorikawa, T., McEwen, B. S. (1998). Protective and damaging effects of stress
Fuchu, T., Komatsu, T., & Togo, F. (2015). Cortisol level mea- mediators. The New England Journal of Medicine, 338(3), 171–
surements in fingernails as a retrospective index of hormone 179. https://doi.org/10.1056/NEJM199801153380307
production. Psychoneuroendocrinology, 54, 24–30. Mewes, R., Reich, H., Skoluda, N., Seele, F., & Nater, U. M. (2017).
Izawa, S., Sugaya, N., Ogawa, N., Shirotsuki, K., & Nomura, S. Elevated hair cortisol concentrations in recently fled asylum
(2021). A validation study on fingernail cortisol: Correlations seekers in comparison to permanently settled immigrants and
with one-month cortisol levels estimated by hair and saliva non-immigrants. Translational Psychiatry, 7(3), E1051.
samples. Stress, 24(6), 734–741. Miller, E. G., Chen, E., & Zhou, S. E. (2007). If it goes up, must it
Jankovic-Rankovic, J., Oka, R. C., Meyers, J. S., & Gettler, L. T. come down? Chronic stress and the hypothalamic- pituitary-
(2020). Forced migration experiences, mental well-being, and adrenocortical axis in humans. Psychological Bulletin, 133(1),
nail cortisol among recently settled refugees in Serbia. Social 25–45.
Science & Medicine, 258, 113070–113070. Miller, K. E., Weine, S. M., Ramic, A., Brkic, N., Bjedic, Z. D.,
Johnson, E. O., Kamilaris, T. C., Chrousos, G. P., & Gold, P. W. Smajkic, A., Boskailo, E., & Worthington, G. (2002). The rela-
(1992). Mechanisms of stress: A dynamic overview of hormonal tive contribution of war experiences and exile-related stressors
and behavioral homeostasis. Neuroscience & Biobehavioral to levels of psychological distress among Bosnian refugees.
Reviews, 16, 115–130. Journal of Traumatic Stress, 15(5), 377–387.
Kawachi, I., & Berkman, F. L. (2001). Social ties and mental health. Mojaverian, T., & Kim, H. S. (2013). Interpreting a helping hand:
Journal of Urban Health, 78(3), 458–467. Cultural variation in the effectiveness of solicited and
Kiecolt-Glaser, J. K., Dura, J. R., Speicher, C. E., Trask, O. J., & unsolicited social support. Personality and Social Psychology
Glaser, R. G. (1991). Spousal caregivers of dementia victims: Bulletin, 39(1), 88–99.
Longitudinal changes in immunity and health. Psychosomatic Morina, N., Akhtar, A., Barth, J., & Schnyder, U. (2018). Psychiatric
Medicine, 53, 345–362. disorders in refugees and internally displaced persons after forced
Koolhaas, J. M., Bartolomucci, A., Buwalda, B., De Boer, S. F., displacement: A systematic review. Frontiers. Psychiatry, 9, 433.
Flügge, G., Korte, S. M., Meerlo, P., Murison, R., Olivier, B., Nepomnaschy, P. A., Lee, T. C. K., Zeng, L., & Dean, C. B. (2012).
Palanza, P., Richter-Levin, G., Sgoifo, A., Steimer, T., Stiedl, O., Who is stressed? Comparing cortisol levels between individuals.
Van Dijk, G., Wöhr, M., & Fuchs, E. (2011). Stress revisited: A American Journal of Human Biology, 24, 515–525.
critical evaluation of the stress concept. Neuroscience and Biobe- Panter-Brick, C., Eggerman, M., Mojadidi, A., & McDade, T. W.
havioral Reviews, 35(5), 1291–1301. (2008). Social stressors, mental health, and physiological stress
Liamputtong, P., & Kurban, H. (2018). Health, social integration in an urban elite of young afghans in Kabul. American Journal
and social support: The lived experiences of young middle- of Human Biology, 20(6), 627–641.
 15206300, 2022, 7, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajhb.23747 by UFPR - Universidade Federal do Parana, Wiley Online Library on [31/05/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
16 of 16 JANKOVIC-RANKOVIC ET AL.

Panter-Brick, C., Wiley, K., Sancilio, A., Dajani, R., & Hadfield, K. Stewart, J. M. (2014). Social support in refugee resettlement. In L.
(2020). C-reactive protein, Epstein-Barr virus, and cortisol tra- Simich & L. Andermann (Eds.), Refuge and resilience: Promot-
jectories in refugee and non-refugee youth: Links with stress, ing resilience and mental health among resettled refugees and
mental health, and cognitive function during a randomized forced migrants (pp. 91–108). Springer.
controlled trial. Brain, Behavior, and Immunity, 87, 207–217. Strang, B. A., & Quinn, N. (2019). Integration or isolation? Refu-
Pike, I. L., & Crocker, R. M. (2020). “My own corner of loneliness:” gees' social connections and wellbeing. Journal of Refugee Stud-
social isolation and place among Mexican immigrants in Ari- ies, 34(1), 328–353.
zona and Turkana pastoralists of Kenya. Transcultural Psychia- Uchino, B. N. (2006). Social support and health: A review of physio-
try, 57(5), 661–672. https://doi.org/10.1177/1363461520938286 logical processes potentially underlying links to disease out-
Porter, M., & Haslam, N. (2005). Predisplacement and post- comes. Journal of Behavioral Medicine, 29(4), 377–387.
displacement factors associated with mental health of refugees Uchino, B. N. (2009). Understanding the links between social sup-
and internally displaced persons: A meta-analysis. Journal of port and physical health: A life-span perspective with emphasis
the American Medical Association, 294(5), 602–612. on the separability of perceived and received support. Perspec-
Reed, R. V., Fazel, M., Jones, L., Panter-Brick, C., & Stein, A. tives on Psychological Science, 4(3), 236–255.
(2012). Mental health of displaced and refugee children UNHCR. 2019. Desperate journeys: Refugees and migrants arriving in
resettled in low-income and middle-income countries: Risk and Europe and at Europe's borders. UNHCR. Available at: https://
protective factors. The Lancet, 379(9812), 250–265. www.unhcr.org/desperatejourneys/. Accessed: September 1, 2021.
Rosal, M. C., King, J., Ma, Y., & Reed, G. W. (2004). Stress, social UNHCR 2020. Global trends: Forced displacement in 2019. Available
support, and cortisol: Inverse associations? Behavioral Medicine, at: https://www.unhcr.org/5ee200e37.pdf. Accessed: September
30(1), 11–21. 1, 2021.
Russell, E., Koren, G., Rieder, M., & Van Uum, S. (2012). Hair cortisol Urlacher, S. S., Ellison, P. T., Sugiyama, L. S., Pontzer, H., Eick, G.,
as a biological marker of chronic stress: Current status, future … Snodgrass, J. J. (2018). Tradeoffs between immune function
directions and unanswered questions. Psychoneuroendocrinology, and childhood growth among Amazonian forager-horticul-
37, 589–601. https://doi.org/10.1016/j.psyneuen.2011.09.009 turalists. PNAS, 115(17), 3914–3921.
Russell, H. B., Wutich, A., & Ryan, G. W. (2016). Analyzing qualita- Warner, R. F. (2007). Social support and distress among
tive data: Systematic approaches. SAGE Publications. Q'eqchi'Refugee women in Maya Tec'un, Mexico. Medical
Sandalio, R. (2018). Life after trauma: The mental-health needs of Anthropology Quarterly, 21(2), 193–217.
asylum seekers in Europe. MPI. Available at: https://www. Warnock, F., McElwee, K., Seo, R. J., McIsaac, S., … Young, A. H.
migrationpolicy.org/article/life-after-trauma-mental-health- (2010). Measuring cortisol and Dhea in fingernails: A pilot
needs-asylum-seekerseurope. Accessed: August 1, 2021. study. Neuropsychiatric Disease and Treatment, 6(1), 1–7.
Schweitzer, R., Melville, F., Steel, Z., & Lacherez, P. (2006). Weber, B. (2017). The EU-Turkey refugee deal and the not quite
Trauma, post-migration living difficulties, and social support as closed Balkan route. SOE. Available at: http://library.fes.de/
predictors of psychological adjustment in resettled Sudanese pdf-files/bueros/sarajevo/13436.pdf. Accessed: August 1, 2021.
refugees. The Australian and New Zealand Journal of Psychia- Weiss, D. S., & Marmar, C. R. (1997). The impact of event scale-
try, 40(2), 179–187. revised. In J. P. Wilson & T. M. Keane (Eds.), Assessing psycho-
Seeman, T. E., & McEwen, S. B. (1996). Impact of social environ- logical trauma and PTSD (pp. 399–411). Guilford Press.
ment characteristics on neuroendocrine regulation. Psychoso- Wilson, M., & Thayer, Z. (2018). Impact of acculturation on depres-
matic Medicine, 58, 459–471. sion, perceived stress and self-esteem in young middle eastern
Sippel, L. M., Pietrzak, R. H., Charney, D. S., Mayes, L. C., & American adults. Annals of Human Biology, 45(4), 346–353.
Southwick, S. M. (2015). How does social support enhance https://doi.org/10.1080/03014460.2018.1484160
resilience in the trauma-exposed individual? Ecology and Soci- Worthman, C. M., & Panter-Brick, C. (2008). Homeless street children
ety, 20(4), 10. https://doi.org/10.5751/ES-07832-200410 in Nepal: Use of allostatic load to access the burden of childhood
Sorensen, M. V., Snodgrass, J. J., Leonard, W. R., McDade, T. W., … adversity. Development and Psychopathology, 20(1), 233–255.
Alekseev, V. P. (2009). Lifestyle incongruity, stress and immune Zimmerman, C., Kiss, L., & Hossain, M. (2011). Migration and
function in indigenous Siberians: The health impacts of rapid health: A framework for 21st century policy-making. PLoS Med-
social and economic change. American Journal of Physical icine, 8(5), e1001034.
Anthropology, 138, 62–69.
Staufenbiel, S. M., Penninx, B. W., Spijker, A. T., Elzinga, B. M., & SU PP O R TI N G I N F O RMA TI O N
Van Rossum, E. F. (2013). Hair cortisol, stress exposure, and Additional supporting information may be found in the
mental health in humans: A systematic review. online version of the article at the publisher's website.
Psychoneuroendocrinology, 38(8), 1220–1235.
Steel, Z., Chey, T., Silove, D., Marnane, C., Bryant, R. A., & Van
Ommeren, M. (2009). Association of torture and other poten- How to cite this article: Jankovic-Rankovic, J.,
tially traumatic events with mental health outcomes among Oka, R. C., Meyer, J. S., Snodgrass, J. J., Eick, G.
populations exposed to mass conflict and displacement: A sys- N., & Gettler, L. T. (2022). Transient refugees'
tematic review and meta-analysis. Journal of the American
social support, mental health, and physiological
Medical Association, 302(5), 537–549.
Steptoe, A., & Kivimäki, M. (2013). Stress and cardiovascular dis-
markers: Evidence from Serbian asylum centers.
ease: An update on current knowledge. Annual Review of Public American Journal of Human Biology, 34(7), e23747.
Health, 34(1), 337–354. https://doi.org/10.1146/annurev- https://doi.org/10.1002/ajhb.23747
publhealth-031912-114452