Understanding the relationship between social support and mental health of humanitarian migrants resettled in australia

Doma et al BMC Public Health (2022) 22 1739 https //doi org/10 1186/s12889 022 14082 z RESEARCH Understanding the relationship between social support and mental health of humanitarian migrants resettl[.]

Understanding the relationship

between social support and mental health

of humanitarian migrants resettled in Australia Hemavarni Doma1*, Thach Tran1, Pilar Rioseco2 and Jane Fisher1

Abstract

Background: Forced migration can lead to loss of social support and increased vulnerability to psychological distress

of displaced individuals The aims were to ascertain the associations of sociodemographic characteristics and social support received by resettled adult humanitarian migrants in Australia; determine the relationship between social support and mental health at different intervals following humanitarian migration; and examine the modification effects of gender, age and migration pathway on that relationship

Methods: A secondary analysis was conducted of data generated in Waves One (three to six months after

resettle-ment), Three (three years after resettlement) and Five (five years after resettlement) of the Building a New Life in Aus-tralia prospective cohort study The association between sociodemographic characteristics and mental health were examined at each timepoint using a multivariate regression model Exploratory factor analysis was used to develop a two-factor social support scale (emotional/instrumental and informational support) from a larger set of items col-lected in the BNLA Psychological distress was measured by the Kessler-6 scale Path analysis was used to analyse the relationships between social support and psychological distress among the three time points considering socio-demographic characteristics simultaneously

Results: A total of 2264 participants were included in the analyses Age, gender, birth region, migration pathway,

education level and English proficiency were significantly associated with both social support types Main source of income was only significantly associated with informational support Remoteness area was only significantly associ-ated with emotional/instrumental support As emotional/instrumental support increased by one standard deviation (SD) at Wave One, psychological distress at Wave Three decreased by 0.34 score [95% CI (− 0.61; − 0.08)] As informa-tional support at Wave Three increased by one SD, psychological distress at Wave Five decreased by 0.35 score [95%

CI (− 0.69; − 0.01)] The relationships between social support and psychological distress varied between genders, age groups and migration pathways

Conclusion: Findings demonstrate the importance of emotional/instrumental support and informational support

for the medium and long-term mental health of humanitarian migrants This study also highlights the important of extending current social support provisions and tailoring programs to enhance support received by humanitarian migrant subgroups years after resettlement to improve mental health

Keywords: Humanitarian migrants, Social support, Mental health, Resettlement, Refugees, Asylum seekers

© The Author(s) 2022 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which

permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line

to the material If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder To view a copy of this licence, visit http:// creat iveco mmons org/ licen ses/ by/4 0/ The Creative Commons Public Domain Dedication waiver ( http:// creat iveco mmons org/ publi cdoma in/ zero/1 0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Open Access

*Correspondence: hemavarni.doma@monash.edu

1 Global and Women’s Health, School of Public Health and Preventive

Medicine, Monash University, Melbourne, Victoria, Australia

Full list of author information is available at the end of the article

According to the United Nations High Commissioner for

Refugees (UNHCR), more than 80 million people are

cur-rently displaced globally [1] Among them, more than 26

million are refugees and more than 4 million are asylum

seekers [1] Over the past decade, Australia has resettled

more than 110,000 refugees, the third-highest number

globally behind two other high-income countries,

experiences can increase the vulnerability of refugees to

mental health problems and psychological distress [3 4]

Yet, whilst the role that trauma has on the mental health

6], the detrimental impact of loss of social connections

on mental health after resettlement [3 7] in high-income

countries like Australia is less well described

Social relationships, or networks, provide social

sup-port [8] Whilst there are several definitions, social

sup-port is commonly described as the functional aspect of

relationships where resources, assistance and aid are

exchanged or provided to an individual through family,

friends, community groups, and government services [9]

Forced displacement leads to disruption of social

connec-tions for humanitarian migrants and involves rebuilding

social networks in the host country [7 10] Among Latin

American and African refugees in Canada, 80% of

refu-gees experience continued separation from their family

members for an average of 3.5 years post-migration [11]

Such disruption can lead to a loss of social support

The context and structures that influence the

provi-sion of social support are essential when discussing social

example, in a group of Chinese and Somali refugees and

immigrants in Canada, a country with broadly similar

humanitarian settlement services as Australia [13],

Stew-art [14] found that lack of financial resources (e.g.,

mon-etary savings from their country of origin) and language

proficiency impeded social support provisions including

access to education training in the host country Having

a small or less-established ethnic group in the host

coun-try was also a barrier to access social services and

sup-port [14] Importantly, Simich [15] found that recreating

social ties and social support, especially with refugees

from their ethnic group, was crucial to the emotional

wellbeing of refugees

Social support has become widely considered an

essential protective factor for mental health [16, 17]

In a systematic review of 36 studies, social support was

consistently associated with protecting adults from

be a protective factor of mental health in humanitarian

where ongoing separation from family, social networks

and sources of social support was associated with increased psychological distress [19] Further, symptoms

of depression decreased in refugee groups as sources

of support from friends and family increased [16, 20] Conversely, refugees with weaker social networks and support reported more severe mental health problems [18] In Australia, in a group of 63 Sudanese refugees in Southeast Queensland, stronger support was a significant predictor of better mental health [3] Therefore, social support post-migration appears to play a role in shaping the experiences of refugees during resettlement and pro-tecting against adverse psychological distress

Although the influence of social support on mental health in humanitarian migrant communities has been documented, there remains a gap in the evidence about the relationship between specific types of social sup-port, time since resettlement and mental health among humanitarian migrants residing in a high-income coun-try Hence, among adult humanitarian migrants resettled

in Australia, the aim was to: (1) describe the specific types

of social support offered to humanitarian migrants; (2) describe the sociodemographic characteristics associated with receiving social support; (3) determine the relation-ship between types of social support and mental health at different times after resettlement; and (4) understand the effect modification of gender, age and migration pathway

on the relationship between social support and mental health

Methods

Setting

Australia is a high-income country, resettling humani-tarian migrants long-term, yearly However, social and health services for humanitarian migrants may not be immediately provided upon arrival [21]

In Australia, the Humanitarian Settlement Pro-gram (HSP) provides support services to humanitarian

Perma-nent protection affords humanitarian migrants the right

to work, study and permanently resettle in Australia

HSP provides support on services including connect-ing with community groups, access to housconnect-ing, English proficiency training, and Medicare [22] These services are provided by organisations such as Settlement Ser-vices International (SSI) and Adult Multicultural Edu-cation Services (AMES) Australia across 11 loEdu-cations

informational support services on finding employment,

includ-ing translation services, basic household goods packages and on arrival logistical needs such as transportation

from the airport and assistance in finding short-term

and long-term accommodation [23, 24] SSI also assists

in emotional support provisions by linking refugees

HSP is short term (six to 18 months) with the

expecta-tion that humanitarian migrants will eventually

transi-tion to services provided within the community and seek

support through other programs, including the

Settle-ment EngageSettle-ment and Transition Support (SETS)

that aims to support the specific needs of humanitarian

migrants [25]

Building a new life in Australia study

The present study is a secondary analysis of data

col-lected from the Building a New Life in Australia

(BNLA) study, a large-scale longitudinal cohort

study tracing the settlement journey of humanitarian

migrants in all Australian states Data from the BNLA

study are available to researchers

The BNLA study has been commissioned by the

Department of Social Services (DSS) and undertaken by

the Australian Institute of Family Studies (AIFS) Detailed

information about BNLA has been reported elsewhere

offshore humanitarian migrants, including refugees (Visa

Subclass 200), women-at-risk (Visa Subclass 204) and

permanent onshore humanitarian migrants on the

pro-tection visa (Visa Subclass 866) People granted

perma-nent protection visas between May and December 2013

(three to six months before the recruitment dates) were

eligible

First, AIFS randomly identified and selected eligible

primary visa applicants (PAs) aged 18 years or older from

11 sites in Australia across metropolitan and regional

areas using the Settlement Database, which provides

sta-tistical data on all permanent arrivals to Australia [27]

Recruitment site was selected by AIFS to ensure each site

optimally represented the diversity of humanitarian visa

subclasses, and rate of humanitarian migrant settlement

[26]

AIFS partnered with Colmar Brunton Social Research

(CBSR), and Multicultural Marketing and Management

(MMM) Both CBSR and MMM collected the data and

conducted the fieldwork for the BNLA study AIFS

sup-plied the contact details of the principal applicants to

CBSR who invited each of them to participate in the study

[26] Following initial contact, CBSR interviewers

tel-ephoned each potential participating principal applicant

to ascertain their interest in participating in the study and

schedule an interview [26] For each principal applicant

who agreed to participate, up to two secondary applicants

who were on the same visa as the principal applicant,

residing in the same household as the principal applicant, and 15 years or older were randomly selected and invited

to participate in the BNLA A total of 2399 people (prin-cipal applicants = 1509, secondary applicants = 890) were recruited

The BNLA comprises five waves of data collected annu-ally from 2013 to 2018 Data from waves two and four were collected via a questionnaire administered through

a computer-assisted telephone interview (CATI) with

an interviewer and interpreter present if required by the

were collected during home visits by CBSR fieldworkers

adminis-tered either via computer-assisted self-interviews (CASI), which used a computer tablet with audio and flashlight function to enable participants to listen to the ques-tions or computer-assisted personal interviews (CAPI), which enabled participants to complete the survey with

an interviewer present [26] Participants were given the option to choose their mode of interview When neither method was feasible, an accredited interpreter was pre-sent over the phone or in person alongside an interviewer

to pose questions and record answers [26]

The questionnaire was translated into multiple lan-guages (e.g., Arabic, Persian, Dari) and designed based on the work of the BNLA advisory group comprising experts

in different areas such as survey methodology, longitudi-nal studies, and refugee and migrant studies

Participants

This secondary analysis included all primary applicants and secondary applicants aged 18 years or older who pro-vided data for the BNLA 15- to 17-year-old participants were excluded from this study because the mental health

of adolescents and adolescent social support services may differ from adults

Data sources

This secondary analysis used data collected in Waves One (baseline, three to six months after resettlement), Three (three years after resettlement) and Five (five years after resettlement)

Social support

into four main types Emotional support is defined as expressions of care, comfort and empathy in social inter-actions; instrumental support as tangible, task-oriented, and material assistance; informational support as the provision of suggestions, advice and new information; and appraisal support as communicating information relevant to self-evaluation such as constructive

differentiated into four types, some social ties may

pro-vide one or more types of support [8]

We created a 10-item scale to measure social support

provided to humanitarian migrants using questions in

across the BNLA questionnaires were selected to be

included in the scale according to the theory of social

[28] and Barrera [29] Exploratory factor analysis was

conducted on the selected items Two subscales

(fac-tors) were identified which corresponded to three social

support types: emotional, instrumental and

informa-tional support Emoinforma-tional and instrumental support were

measured with the first factor, and informational support

with the second factor Appraisal support was not

meas-ured as it was not identified as a factor from the selected

items of the BNLA

The first subscale, emotional/instrumental support,

consists of three items assessing the support and

com-fort provided by a community to assist with resettlement

The scores were summed and standardised (mean = 0

and SD = 1) to create a total emotional/instrumental

sup-port score where a high score indicates higher emotional/

instrumental support (Supplementary file 2) The

inter-nal consistency of the scale was tested using Cronbach’s

alpha coefficient where a coefficient > 0.8 indicates high

internal reliability For the emotional/instrumental

sup-port subscale, the internal consistency was α = 0.83 at

Wave One and α = 0.86 at Wave Three

The second subscale, informational support, consists of

seven items assessing whether information, suggestions

and advice on services essential to integrate and function

in society have been received The scores were summed

and standardised to create a total informational support

score where a high score indicated higher informational

support subscale, the internal consistency was α = 0.91 at

Wave One and α = 0.92 at Wave Three

Psychological distress

Psychological distress symptoms were assessed using

the Kessler-6 scale (K6) that included six items

describ-ing depression and anxiety symptoms [30, 31] The items

are scored on a five-point scale: 1 (none of the time), 2

(a little of the time), 3 (some of the time), 4 (most of the

time), and 5 (all of the time) The scores were summed

to create a total scale score, with a higher score

indicat-ing more symptoms of psychological distress In this

study, the internal consistency for the K6 was α =0.89 at

Wave One, α =0.90 at Wave Three and α = 0.92 at Wave

Five The scale has also been translated and validated

across different languages including in Arabic where the

K6 demonstrated high internal consistency (Cronbach’s

α =0.81) and high convergent validity with two other scales: the Generalised Anxiety Disorder (GAD-7) and Somatic Symptoms Scale (SSS-8) [32]

Socio‑demographic characteristics

At baseline, socio-demographic characteristics were col-lected using study-specific questions on age, gender, marital status, country of birth, remoteness area, educa-tion level, and main sources of income

English proficiency in the BNLA was determined by four items: how well do you (1) understand spoken lish, (2) speak English, (3) read English, (4) write Eng-lish The items were scored on a four-point scale: 1 (very well), 2 (well), 3 (not well), and 4 (not at all) The items were reverse-scored and recoded into 0 (not at all), 1 (not well), 2 (well), and 3 (very well) and summed to create a total score where a high score indicates a higher profi-ciency The internal consistency for this scale is α =0.96

in this study

Migration pathways were assessed as to whether humanitarian migrants arrived in Australia via the onshore or offshore pathway The offshore pathway is for those granted permanent protection visas (i.e., Visa Sub-classes 200, 201, 202, 203 and 204) before arriving in

pathway is for those granted a permanent protection visa (i.e., Visa Subclass 866) after arrival in Australia and would be termed asylum seekers [33]

Statistical analysis

Analyses were conducted in three stages In stage one, the associations between the sociodemographic charac-teristics and each social support type at every time point were examined using a multivariate regression model, controlling for sociodemographic characteristics

In stage two, a path model was used to analyse the relationship between social support and

adapted from the model proposed by Heaney and Israel [8] and Watkins et al [34], and draws upon the

This model composes of the directional pathways and correlations between emotional/instrumental support subscale, informational support subscale, psycho-logical distress symptoms, and baseline socio-demo-graphic characteristics at each time point (Waves One, Three, Five) All of the directional pathways and

All of the path coefficients were interpreted as linear regression coefficients as all endogenous variables, which were caused by one or more variables in the

model, were in continuous scales The fit of the path

model was evaluated using the following criteria: Root

Mean Square Error of Approximation (RMSEA) < 0.05,

Comparative Fit Index (CFI) ≥ 0.90, and Tucker-Lewis

Index (TLI) ≥ 0.90, which indicate a good fit [35]

In stage three, the path model developed in stage two

was re-run for subgroups by gender (male and female),

age groups (18 to 29 years old, 30 to 44 years old and

45 to 75 years old), and migration pathways [offshore

pathway (refugees) and onshore pathway (asylum

seek-ers)] to determine if the effect of social support on

psychological distress was modified by gender, age and

migration pathway

The estimation method used for the path models was

the maximum likelihood for missing values (MLMV)

which accounted for missing data by adjusting the

like-lihood function to capture information on variables

that are observed by cases [36] All analyses were

con-ducted using Stata Version 16

Ethics

Ethics approval for the BNLA study was obtained from the Australian Institute of Family Studies Human Research Ethics Committee (13/03) The Monash Univer-sity Research Ethics Committee granted ethics exemp-tion to use the data

Results

Participant characteristics

Among 2399 participants of the BNLA, 2264 (1509 PAs and 738 SAs) were eligible for this study We excluded

135 people who were < 18 years old All 2264 participants provided data three to six months after resettlement (Wave One) Among those, 1779 (1155 PAs and 624 SAs) were followed-up three years after resettlement (Wave Three) and 1765 (1144 PAs and 621 SAs) five years after resettlement (Wave Five) Hence, the loss to follow-up rate from three to six  months to five  years after reset-tlement is 22.1%, with 77.9% of respondents from Wave One retained

Fig 1 Conceptual framework of the correlation between emotional/instrumental support, informational support and psychological distress

The mean age of participants included in this study was

36.6 years (Table 1) More than half of the participants were

men Humanitarian migrants arrived in Australia from

five regions, most were from the Middle East

Approxi-mately 60% of participants were married Most

partici-pants arrived via Australia’s offshore settlement pathway,

received government payments and lived in major cities

Younger humanitarian migrants were more likely to

have received both types of social support than those

women had received more emotional/instrumental

support and less informational support at each Wave

Humanitarian migrants who were married had been

given more emotional/instrumental support three years

after resettlement than those who were not married

(Table 2)

The association between birth region and emotional/ instrumental support was most significant five years after resettlement, where humanitarian migrants born

in the Middle East, South-East Asia and Central Asia received less of this support type than those born in

pro-vided with less emotional/instrumental support than those born in Africa across all Waves Whilst those born

in the Middle East and South-East Asia appeared to receive less emotional/instrumental support than those born in Africa, the difference was not statistically signifi-cant three to six months and three years after resettle-ment The association became significant five years after resettlement

Five years after resettlement, all humanitarian migrants born in regions other than Africa received less informa-tional support compared to those born in Africa Those born in South-East Asia received less informational sup-port than those born in Africa across all Waves (Table 3) Humanitarian migrants more proficient in English receive more emotional/instrumental and informational support at each Wave For educational level pre-arrival

to Australia, compared to those with six or fewer years

of schooling, humanitarian migrants with 12 or more years of schooling received less emotional/instrumental support than those with six or fewer years of schooling three to six months and five years after resettlement but received more informational support at all time points after resettlement (Tables 2 and 3)

Three to six months after resettlement, humanitarian migrants residing in regional Australia received more emotional/instrumental support compared to those residing in major cities (Table 2) Humanitarian migrants with primary income of government payments received less informational support three to six months and five years after resettlement than those reliant on their own salary or from another source (Table 3)

Humanitarian migrants who came to Australia via the offshore pathway (refugees) received more emotional/ instrumental support compared to those who came

to Australia via the onshore pathway (asylum seek-ers) three and five years after resettlement (Table 2) Humanitarian migrants arriving as refugees received less informational support than those arriving as asylum seekers three to six months and three years after reset-tlement (Table 3)

Relationship between social support and mental health

social support types and mental health at differ-ent time points had a good fit (CFI = 0.98, TLI = 0.90, RMSEA = 0.03)

Table 1 Demographic characteristics at baseline (n = 2264)

Statistics

Gender, n (%)

Marital status, n (%)

Birth region, n (%)

Education level prior to arrival, n (%)

Remoteness area, n (%)

Migration pathway, n (%)

Main source of income, n (%)

Respondent’s own salary or spouse’s/partner’s/parent’s

Mode of interview, n (%)

Computer-assisted self-interview 1582 (69.9%)

Computer-assisted personal interview with interviewer 633 (28%)

Computer-assisted personal interview with interpreter 49 (2.1%)

The emotional/instrumental support received by

humanitarian migrants three to six months after

reset-tlement affected psychological distress three years after

resettlement As the emotional/instrumental support

increased by one SD three to six months after

resettle-ment, psychological distress (K6 score) three years after

resettlement decreased by 0.34 score The information

support received three years after resettlement influenced

psychological distress three years after resettlement

As informational support three years after resettlement

increased by one SD, the psychological distress score five

years after resettlement decreased by 0.35 (Table 4)

Effect modification of gender, age and migration pathway

on the relationship between social support and mental health

The relationships between social support and psycho-logical distress were slightly different between men and women Informational support influenced the severity

of psychological distress score in both groups but at different time points Psychological distress five years after resettlement decreased as the informational sup-port received by men increased three years after reset-tlement In contrast, psychological distress five years after resettlement decreased as the informational

Table 2 Association between demographic characteristics (baseline) and emotional/instrumental support at each time point

† Statistical significance set at p < 0.05

a Multiple regression coefficient

Age (in 10 years) −0.002 (− 0.010 to 0.061) 0.667 − 0.215 (− 0.302 to − 0.128) 0.000 − 0.123 (− 0.216 to − 0.030) 0.009 Gender

Women 0.257 (0.063 to 0.451) 0.009 0.235 (0.019 to 0.452) 0.033 0.300 (0.069 to 0.530) 0.011 Marital status

Yes 0.112 (−0.089 to 0.314) 0.273 0.309 (0.081 to 0.538) 0.008 0.090 (−0.154 to 0.334) 0.471 Birth region

Middle East −0.106 (− 0.497 to 0.285) 0.596 0.182 (−0.319 to 0.683) 0.476 −0.653 (−1.203 to − 0.102) 0.020 South-East Asia −0.406 (− 0.943 to 0.132) 0.139 − 0.182 (− 0.839 to 0.474) 0.586 − 0.930 (− 1.671 to − 0.189) 0.014 Southern Asia 0.297 (− 0.193 to 0.788) 0.235 0.017 (−0.575 to 0.610) 0.954 0.127 (−0.537 to 0.792) 0.707 Central Asia −0.990 (−1.413 to − 0.566) 0.000 − 0.895 (− 1.427 to − 0.363) 0.001 −0.898 (−1.485 to − 0.311) 0.003 English proficiency

(score) 0.132 (0.095 to 0.169) 0.000 0.0676 (0.027 to 0.109) 0.001 0.091 (0.047 to 0.136) 0.000 Education (pre-arrival)

6 or fewer years of

7 to 11 years of

schooling −0.288 (−0.548 to −0.027) 0.031 0.061 (−0.229 to 0.352) 0.679 −0.057 (−0.363 to 0.245) 0.705

12 or more years of

schooling −0.508 (−0.790 to − 0.225) 0.000 −0.076 (− 0.389 to 0.237) 0.634 − 0.404 (− 0.740 to − 0.069) 0.018 Remoteness area

Regional Australia 0.547 (0.203 to 0.890) 0.002 0.208 (−0.167 to 0.584) 0.277 −0.181 (−0.580 to 0.218) 0.374 Migration pathway

Offshore 0.153 (−0.121 to 0.427) 0.274 0.904 (0.575 to 1.234) 0.000 0.549 (0.185 to 0.914) 0.003 Main source of income

Own or

spouse/par-ent’s salary, savings Ref

Government

pay-ments 0.020 (−0.313 to 0.353) 0.906 0.135 (−0.246 to 0.515) 0.488 0.033 (−0.257 to 0.323) 0.824