Introduction

Ensuring effective primary health care (PHC) for refugees is an increasing concern globally. In 2011, developed countries worldwide received 441,448 applications for refugee status, a 20% increase from the previous year [1]. A refugee is a person outside the country of his or her nationality who, “owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of particular social group or political opinion, is outside the country of his nationality and is unable or owing to such fear unwilling to avail himself of the protection of that country” [2]. The refugee community in Australia is culturally and linguistically diverse, originating from more than 40 different countries [1]. These refugees are racially, culturally and linguistically diverse and often have suffered extreme mental and physical trauma [3], coming from countries in situations of long-term war or conflict [4].

Refugees often have complex or multiple health care needs as a consequence of inequities in the social determinants of health: experiences of persecution, torture and other forms of trauma, deprivation, unhealthy environmental conditions and disrupted access to health care [5]. PHC services need to be able to meet these challenges both on refugees’ arrival and in their transition to long-term care [3], providing culturally appropriate and timely care [6].

Refugees are more likely to have increased morbidity, poor health habits and a decreased life expectancy [4].They may have difficulty navigating the new education, housing, social support services and health systems in their country of resettlement [7]. Limited local language proficiency has an impact on health [8] and on the quality and accessibility of care [9]. It also influences access to the resources required for health, such as education, employment and social support [8]. When several service providers are involved, poor integration reduces their ability to deliver care effectively and efficiently [10]. There is inadequate community support for refugees in moving between services and sectors [11, 12].

Enabling refugees to access timely, high-quality health care is crucial to their successful settlement and integration, as optimal health and well-being provide a stronger basis for them to adapt and thrive in their new country [11, 13]. Good physical and mental health are vital for refugees to deal effectively with the challenges of settling in a new country and to participate fully in economic, social and cultural life [12]. Although other population groups face access barriers, the diverse and complex health and well-being needs of people from refugee backgrounds require specific attention [14–16].

Diverse initiatives have been instituted by governments and by non-government organisations to address the needs of refugee populations in countries of resettlement. This review focuses on PHC service delivery models for refugee populations and the impacts of components of these models on access, coordination and quality of care in countries of resettlement, with the aim of informing refugee health policy and its implementation.

Method

A narrative systematic review was conducted. Medline, CINAHL, EMBASE, Cochrane Library, Scopus, Australian Public Affairs Information Service – Health, Health and Society Database, Multicultural Australian and Immigration Studies and Google Scholar were searched from February to March, 2012 for articles that included “refugee”, “primary health care” and “model of care” (or associated terms) in the title, keywords or abstract (Appendix 1). In Medline, the following search terms were used in the titles, abstract and keywords: (“Primary Health Care” OR “General Practice” OR “Comprehensive Health Care” OR “Physician, Family” OR “Family Practice” OR “Family Medicine” OR “Community Health Services”) AND (“Refugee” OR “Refugees” OR “Transients and Migrants” OR “Emigration and Immigration” OR “Asylum Seeker”) AND (“Model of Care” OR “Long-Term Care” OR “Models, Organisational” OR “Continuity of Patient Care” OR “Delivery of Health Care” OR “Patient Care Team”). The “explode” option was used to increase the depth of the search.

This was followed by a limited snowballing exercise, searching the reference lists of all included studies. Further studies and studies from the grey literature were identified with the assistance of the stakeholder advisory committee and international advisers and networks in Australia and overseas. These studies included reports on websites of key government (USA, Australia and Canada), international bodies (World Health Organization, World Bank, International Organization for Migration, United Nations High Commissioner for Refugees) and non-government organisations (The Victorian Foundation for Survivors of Torture, Boston Center for Refugee Health & Human Rights, Canadian Collaboration for Immigrant and Refugee Health, Ontario Multicultural Health Applied Research Network, and the Welcoming Communities Initiative). Systematic reviews fulfilling the selection criteria were also included in the review. The PRISMA checklist was followed for reporting [17].

Studies were included in the review if they were about PHC in the major developed countries accepting refugees for long-term resettlement (Australia, USA, Canada, Sweden, Norway, New Zealand, Finland, Denmark, Netherlands and the UK) [18] and were published in English between 1990 and 2011. Inclusion criteria were that studies evaluated models of care that included specific aspects of care for refugee populations and the organisation and/or delivery of PHC. We did not predefine the way in which these models were evaluated and we accepted measures of the outcomes based on service provider or client assessment using either qualitative or quantitative methods. Table 1 provides the key definitions used in the review. Figure 1 shows a framework for thinking about how these relate to each.

Table 1 Key definitions used in the review
Full size table
Figure 1
figure 1

Framework describing impact of primary health care service delivery models (adapated from Hogg et al.) [[29]].

Full size image

The papers were screened initially by CJ and a 20% random sample of excluded studies was reviewed and checked by MFH. Consensus was gained for uncertain articles through discussion between MFH and CJ. Included studies were screened based on the title and abstract and the full paper verified through assessment of its contents based on the inclusion criteria. Data extraction, including a quality assessment using a published checklist [31] was carried out by CJ and a 20% sample was verified by MFH. The extracted data were entered in an MS-Excel spread-sheet and included variables such as country of resettlement and country of origin, years in the destination country, location (urban or rural), characteristics of participants including type of refugee, country of origin, years in country, age, gender and major health problems or concerns. Also recorded was information about the study design, methodology and quality, types of service, model of care and the impact of services including access, coordination, quality of care, costs involved and the health outcomes of interventions. Because the outcomes were heterogeneous a meta-narrative synthesis was undertaken [32].

Results

The database searches and grey literature produced 2,139 papers which were assessed for inclusion in the review. We found relatively few evaluated models of refugee health care in countries of resettlement. Most evaluations focused on patient satisfaction rather than other outcomes. After the screening and verification stages, data were extracted from 25 studies that described evaluated models of PHC for refugees. The major reasons for exclusion were that the paper did not report empirical research, was not from the predefined countries of resettlement, was not based on primary health care or lacked evaluation of the models. Figure 2 shows a PRISMA flowchart of the selection process. The models were then analysed according to their impact on access, coordination and quality of care (Table 2).

Figure 2
figure 2

PRISMA flow diagram on selection of papers for the review.

Full size image
Table 2 Characteristics of the models and their impacts
Full size table

Models of care

There were 10 overseas and 15 Australian models of PHC for refugees. The characteristics of the models were described under six categories: service context, clinical model, workforce, cost to clients, health services and non-health services (Table 3). Case management and care planning were common components of the clinical models evaluated. In Australia, the majority of models used case management to coordinate a range of health (especially mental health, specialised services and access to general practitioners) as well as non-health services [20, 33–40]. Clinical services were often provided by a multidisciplinary team [20, 33, 37, 40–42]. Several studies described the role of specialist refugee health nurses in assessment and care coordination [20, 33, 34, 36, 38, 39, 43, 44]. Those roles routinely included assessment of health and social needs, immunisation, and case management of referral to other services and ongoing liaison and transfer to GPs. Training (including training in cross-cultural communication) underpinned the capacity of many health staff to provide appropriate care [30,33,34]. The provision and use of interpreters and bilingual staff were key components of many models [33, 35–37, 39, 42, 45–48]. In Australia, interpreters and/or bilingual staff were used in most services. Outreach into the homes of refugees or the community with a comprehensive range of service was another commonly described approach [33–36, 40, 42, 44, 46, 49, 50]. These services were delivered by a range of health professionals including health visitors, students and ethnic health workers. In Australia, outreach was a common model provided in conjunction with case management by specialist refugee health nurses.

Table 3 Characteristics of models of PHC for refugees and their corresponding evaluated components
Full size table

Mental health care (including counselling, cognitive behavioural therapy and psychiatry) was the most frequent health service described [33–35, 37, 39, 40, 43, 45, 48, 49],[51, 52]. Screening and assessment were described in several studies [34, 39, 41, 46, 47, 53]. Child health was the focus in several models [37, 45, 54]. A number of studies described models involving the provision of both health and non-health services (e.g. legal, housing, education, transport) [20, 33–41, 43, 44, 50, 51]. In Australia, many of the models involved screening or assessment and health as well as non-health services (often using a case management approach and in conjunction with settlement services).

Free transport to health centres was provided by many Australian models [35–38, 41, 44]. Low- or no-cost care for patients was provided in a number of overseas and Australian models through use of government insurance-only payments or volunteers [35, 41, 44, 46, 51]. Negotiation of pro bono medical services and volunteers was a feature of many Australian and US models [33, 35, 41, 46].

Impact on access

Sixteen studies evaluated the impact of the models of care on access to health care [6, 20, 33–36, 38, 39, 43, 44, 46],[47, 51–54]. The models broadly addressed several components: affordability, appropriateness and acceptability of primary care, and a variety of health and non-health services. Strategies used to enhance access included multidisciplinary staff (medical, nursing, allied health, non-health workers), use of interpreters and bilingual staff, no-cost or low-cost services to consumers, outreach services (many in refugees’ homes), free transport for appointments, longer consultation hours, patient advocacy and use of gender-sensitive providers. All these strategies had a positive impact on client satisfaction and increased utilisation of services. Other strategies to contain costs were the use of volunteers and minimising fees so that they were covered by government health insurance. Staff advocating on behalf of patients increased their access to housing, social security payments and medical services. Gender-sensitive providers helped overcome cultural barriers that prevented women from accessing health services. Teaching patients how to navigate the health care system led to better understanding and increased utilisation of health services. Bilingual and culturally appropriate information provision in written and video form enhanced health literacy and access to care. However, lack of interpreters in needed languages, unmet health needs, and shortage of doctors willing to accept fees limited to government insurance levels remained major barriers.

Impact on coordination

Six studies [20, 34, 38, 40, 43, 44] evaluated the impact of models of refugee care on coordination of health care services. A number of studies described models involving the provision of both health and non-health services (including legal assistance, housing, education and transport). Coordination between the different health care services and services responding to the social needs of clients was most frequently addressed through case management conducted by a refugee health nurse or other health professional and often involved visiting refugee clients in their home in the community. Team coordination, especially across agencies, was also used. These interventions were associated with improved communication and coordination between service providers, as well as improved access to preventive health services. Four of the six case management models reported improved outcomes. Individual case managers provided easier transition between the primary health care and hospitals [20], good coordination among stakeholders [43] and improved team work where workers alerted each other of patient issues [40]. Please refer to Table 2 for details. Two Australian studies without improved outcomes reported that coordination was compromised by limited access to services by some groups of patients and by the capacity of staff to meet the needs of patients [35, 38]. The study using a multidisciplinary team approach to coordination reported that services were able to meet clients’ needs [36].

Impact on quality of care

Seventeen studies evaluated the impact of models on quality of care [6, 34–39, 41–43, 45, 48–50, 52, 54, 55]. The most frequently cited interventions were training providers in culturally sensitive care, appropriate use of interpreters, and bilingual staff. These resulted in improved client satisfaction, increased reporting of physical and psychological symptoms by the patients, improved referrals, improved physical and mental health, and increased access to health services. There were marked reduction in risk related to medical interpreters’ service [42], and increased access to culturally appropriate services [43]. Nonetheless, many patients continued to experience barriers, including persisting cultural and language barriers [36, 38, 39]. There were also difficulties reported in putting cultural sensitivity training knowledge into practice [38].

Discussion

The aim of this review was to identify evaluated models of PHC for refugees in countries of resettlement and to evaluate their impact on access, coordination of care and quality of care. Relatively few studies met the inclusion criteria. There were many similarities among the evaluated models, with some variation according to the context and resources available. Integration between the different health care services and services responding to the social needs of clients was most frequently addressed by a case management approach conducted by a refugee health nurse or other health professional and often involving home visiting refugee clients in their home in the community. Interpreters, bilingual staff, and training of staff in cross-cultural management were also used to facilitate access to and quality of health and social care.

Refugees need to be able to access the same primary care services as the local population. Thus all primary care services need to be prepared to deliver health care to refugees in their local area and some services should develop models of care specifically addressing the needs of refugees because of the demographics of their local communities. Clients of refugee-specific services need to be able to transition into ongoing mainstream PHC [56]. Yet their transition is influenced by factors including lack of knowledge about available services and/or how they work, language barriers, and lack of appropriate services. Refugees may also be reluctant to use existing services because of fear, distrust, negative experiences and lack of confidence, sociocultural barriers and political, economic and administrative constraints on access to the health services [57, 58]. The models in our review aligned broadly with the elements of accessibility proposed by Penchansky and Thomas [59]:

  • Increasing awareness and health literacy in using health services with interventions involving media and health education.

  • Outreach to facilitate registration or clinic attendance.

  • Improving acceptability and appropriateness through the use of interpreters and bilingual workers.

  • Coordinating service networks (often facilitated by refugee health nurses) to improve access to range of services and to transport.

  • Reducing cost to clients by use of pro bono providers or not using co-payments.

Coordination of care has been discussed in the literature describing models of care for refugees. For example, there has been debate about integration between government and non-government services, the need for refugee-specific health services and mainstream services (which may include medical specialists) and the balance of emphasis on initial assessment compared with providing ongoing long-term care [60, 61]. In this review we found that coordination of care was largely focused on integrating care across the number of health and non-health services that might be involved. The two main coordination models were case management and team coordination, and these were associated with improved communication and coordination between service providers.

Good patient–provider communication is of paramount importance to quality of care. Patient dissatisfaction arises more frequently from poor communication than from medical errors [62], and language barriers degrade the quality of care, resulting in poorer health outcomes [63, 64]. Culture frames the experience and expression of emotional distress and social problems [65]. Language barriers and the cultural complexity of assessing symptoms have been shown to prevent adequate diagnosis and treatment [65]. Several studies in our sample evaluated impacts on quality of care. These service models included use of interpreters, bilingual staff, cross-cultural training of staff and specialised refugee health nurses, and engagement with the community. They were associated with improvements in staff confidence, detection of problems at assessment, clients’ assessment of the quality of communication and interpersonal care. These measures are broadly consistent with international policy [57].

The review had some important limitations. Few studies in the international databases described evaluated models. Few studies described coordination tools or protocols and regional coordination. Many of the studies were identified from website search and from our key informants in Australia, the UK, Canada and New Zealand. Because the search of the grey literature was less systematic and reliant on key informants, it is possible that other international grey literature was missed. There were high levels of heterogeneity in the impacts and outcomes evaluated. This made meta-analysis impossible. Thus a qualitative approach was used to analyse and compare the studies. There was little information on the cost of services or models. This meant that we were unable to undertake any comparative cost analysis of models. There is a need for more rigorous evaluations, especially focused on the impact of innovative models on access and quality of care especially for women.

We found no evaluated model of PHC that specifically focused on women or men; hence it was not possible to conduct a gender analysis. Similarly, previous studies of migrant utilisation of health services have not included a focus on women [66].

Despite the limitations of the available evidence, there are some implications for policy and practice. Many of the models emerged in response to the needs of local communities. Therefore the effective components identified in this review need to be flexibly applied in the local context. For example, case management is commonly used as a component of models and appears to be broadly successful in improving access and coordination. This makes sense where there are relatively few refugees requiring case management and where the focus is on coordination between services and integration of the refugee into long-term care. It is, however, potentially expensive and the case coordinator needs to have some specialised training. Workforce planning and development in this field is thus important, especially for nurses. At present, the education and training available for the workforce in many countries remains limited [57]. Although specialised services and providers may be useful, especially for on-arrival assessment, other forms of service delivery based on and integrated with mainstream PHC services are needed.

The use of interpreters and bilingual workers is well-documented as essential in facilitating access to care and delivering quality care. Interpreter services can be on-site or through a telephone service, especially for more routine care [67]. Use of informal interpreters such as family members can undermine the quality of care [63]. Thus it is important to make interpreter services more available to assist in PHC for refugees, both in person and on the phone.

Conclusion

Refugees experience unique health problems which can be met by PHC services. Relatively few studies have evaluated models of PHC delivery to the refugee community. Much of the literature currently focuses on describing the health problems or access barriers experienced by refugees. This study goes further in identifying the specific strategies required to provide accessible and well-coordinated care for refugees, such as case management, use of specialised staff, interpreters and outreach. These have financial and workforce implications and comparable evaluation data needs to be collected to allow comparison of the cost and effectiveness of different models.

Appendix 1

Search terms

1. Medline: 771 search result

PHC:

Primary health care mp. Or exp Primary Health Care/ or general practice mp. Or exp General Practice/ or exp Community Health Services/ or exp Comprehensive Health Care/ or exp Physicians, Family/ or exp Family Practice/ or family medicine mp.

Refugee:

Refugee mp. Or exp Refugees/ or exp “Transients and Migrants”/ or exp “Emigration and Immigration”/ or asylum seeker.mp.

Model of care:

Exp “Continuity of Patient Care”/ or exp “Delivery of Health Care”/ or exp Patient Care Team/ or model of care.mp. or exp Long-Term Care/ or exp Models, Organisational/

2. CINAHL: 561 search results (limiting search to 1990 - 2011)

PHC:

(MH “Community Health Centers”) OR (MH “Community Mental Health Services+”) OR (MH “Community Mental Health Nursing”) OR (MH “Health Information Networks”) OR (MH “Community Health Nursing+”) OR (MH “Community Health Services+”) OR (MH “Community Health Workers”) OR (MH “Community Networks”)

Refugee:

(MH “Transients and Migrants”) OR “transients and migrants” OR (MH “Refugees”) OR “refugee” OR “asylum seeker” OR (MH “Emigration and Immigration”) OR (MH “Immigrants, Illegal”) OR (MH “Immigrants+”)

Model of care:

(MH “Models, Psychological+”) OR (MH “Models, Educational”) OR (MH “Models, Structural+”) OR (MH “Multidisciplinary Care Team+”) OR (MH “Gender Specific Care”) OR (MH “Health Care Costs+”) OR (MH “Health Services Needs and Demand”) OR (MH “Health Care Delivery+”) OR (MH “Health Care Delivery, Integrated”) OR (MH “Nursing Care Plans+”) OR (MH “Nursing Care Delivery Systems+”) OR (MH “Outcomes (Health Care)+”) OR “model of care”

3. Embase: 578 search results

PHC:

exp primary health care/ or exp general practice/ or exp community health nursing/ or exp general practitioner/ or exp community care/ or exp family medicine/

Refugee:

asylum seeker*.mp. or “transients and migrants”.mp. or “emigration and immigration”.mp.

or exp refugee/ or refugee*.mp.

Model of care:

exp patient care/ or exp model/ or exp “organization and management”/ or model of care.mp. or exp health care delivery/ or exp medical care/

4. Cochrane library: 14 search results

primary care* or family medicine* or general practice* or community health*

model of care

5. Scopus: 62 results

“Primary care” or “primary health care”

Refugee* or “asylum seeker*”

6. APAIS health: 8 search results (limiting search to 1990-2011)

“primary care” or “primary health care” or “family medicine” or “general practice” or “community health” or “nursing servic*e” or “allied service*”

Refugee* or “asylum seeker*”

7. Health and society database: 31 search results

“primary care” or “primary health care” or “family medicine” or “general practice” or “community health” or “nursing service*” or “allied service*”

“asylum seeker’ or “refugee” or “refugees”

8. MAIS (Multicultural Australian and Immigration Studies): 46 search results (limiting search to 1990-2011)

“primary care” or “primary health care” or “family medicine” or “general practice” or “community health” or “nursing service*” or “allied service*”

“asylum seeker” or “refugee” or “refugees”