Mobile primary health care clinics for Indigenous populations in Australia, Canada, New Zealand and the United States: a systematic scoping review

Background Mobile clinics have been used to deliver primary health care to populations that otherwise experience difficulty in accessing services. Indigenous populations in Australia, Canada, New Zealand, and the United States experience greater health inequities than non-Indigenous populations. There is increasing support for Indigenous-governed and culturally accessible primary health care services which meet the needs of Indigenous populations. There is some support for primary health care mobile clinics implemented specifically for Indigenous populations to improve health service accessibility. The purpose of this review is to scope the literature for evidence of mobile primary health care clinics implemented specifically for Indigenous populations in Australia, Canada, New Zealand, and the United States. Methods This review was undertaken using the Joanna Brigg Institute (JBI) scoping review methodology. Review objectives, inclusion criteria and methods were specified in advance and documented in a published protocol. The search included five academic databases and an extensive search of the grey literature. Results The search resulted in 1350 unique citations, with 91 of these citations retrieved from the grey literature and targeted organisational websites. Title, abstract and full-text screening was conducted independently by two reviewers, with 123 citations undergoing full text review. Of these, 39 citations discussing 25 mobile clinics, met the inclusion criteria. An additional 14 citations were snowballed from a review of the reference lists of included citations. Of these 25 mobile clinics, the majority were implemented in Australia (n = 14), followed by United States (n = 6) and Canada (n = 5). No primary health mobile clinics specifically for Indigenous people in New Zealand were retrieved. There was a pattern of declining locations serviced by mobile clinics with an increasing population. Furthermore, only 13 mobile clinics had some form of evaluation. Conclusions This review identifies geographical gaps in the implementation of primary health care mobile clinics for Indigenous populations in Australia, Canada, New Zealand, and the United States. There is a paucity of evaluations supporting the use of mobile clinics for Indigenous populations and a need for organisations implementing mobile clinics specifically for Indigenous populations to share their experiences. Engaging with the perspectives of Indigenous people accessing mobile clinic services is imperative to future evaluations. Registration The protocol for this review has been peer-reviewed and published in JBI Evidence Synthesis (doi: 10.11124/JBISRIR-D-19-00057). Supplementary Information Supplementary information accompanies this paper at10.1186/s12939-020-01306-0.

mortality rate compared to non-Indigenous populations [18], culturally safe models of health care are needed which improve the accessibility of primary health care services [19]. Evidence supports that a greater participation of Indigenous people in their health care leads to better health outcomes [20,21]. Therefore, Indigenousgoverned health care services are inherent to the provision of culturally accessible health care [22]. In Australia, over 140 Aboriginal Community-Controlled Health Services (ACCHOs) provide primary health care services to Aboriginal and Torres Strait Islander people [23]. Internationally, evidence supports the important contribution of Indigenous-governed health organisations in providing culturally safe and accessible primary health care for Indigenous populations [24][25][26][27].
Mobile clinics implemented specifically for Indigenous populations and governed by Indigenous health organisations, may be one way to improve the accessibility of culturally safe primary health care for Indigenous populations. It is known that mobile clinics are able to deliver health care to populations experiencing health inequity, particularly in countries where health care can be otherwise inaccessible due to transport, financial or cultural barriers [28][29][30]. In the United States, there has been an upward surge in the implementation of mobile clinics, particularly of mobile clinics delivering primary health care services [31,32]. The support for mobile clinics in providing flexible and safe health care to vulnerable people has gained traction with the recent COVID-19 pandemic [33]. In other countries, mobile clinics have also been implemented with the purpose of screening for communicable and non-communicable diseases [34][35][36] and providing disaster relief [37,38]. Some research supports the potential for mobile clinics to be a cost-effective model of health care and improve the management of chronic disease [29,39].
There is also some evidence of mobile clinics being implemented specifically for Indigenous populations, either by an Indigenous health organization [40] or for a specific disease (e.g. diabetes) [41] or treatment (e.g. dialysis) [42]. What is not known, is the available evidence regarding the use of primary health care mobile clinics implemented specifically for Indigenous populations in Australia, Canada, New Zealand, and the United States who share a similar history of colonization, discrimination and barriers to accessing primary health care services [7]. This was apparent when undertaking a preliminary search of the literature for evidence around the effectiveness of mobile clinics for Indigenous populations, as part of seeking funding for a mobile clinic to be implemented in an Australian ACCHO. Indeed, it was an absence of evidence that made it difficult to obtain funding for the mobile clinic, justifying the need for a systematic scoping review. It is known that there is a vast body of literature regarding mobile clinics in the United States, yet there is very little focus on Native American, Native Hawaiian, and Alaskan Native populations [32]. A systematic scoping review was conceptualised to synthesise the available evidence regarding the use of primary health care mobile clinics implemented specifically for Indigenous populations in order to identify gaps in the literature and inform future research evaluating mobile clinics for Indigenous populations. Specifically, the review question developed was: What is the evidence surrounding the use of mobile primary healthcare clinics implemented for Indigenous populations in Australia, Canada, New Zealand, and the United States?
Specific objectives were to: (1) scope the models of primary health care clinics for Indigenous populations (in Australia, Canada, New Zealand, and the United States) as described in the literature, (2) determine geographically where mobile primary health care clinics for Indigenous populations (in Australia, Canada, New Zealand, and the United States) have been implemented and, (3) examine the findings of any evaluations of mobile primary health care clinics for Indigenous populations (in Australia, Canada, New Zealand, and the United States) that have been published in the literature.

Methods
This systematic scoping review examines the evidence surrounding the use of mobile primary healthcare clinics implemented for Indigenous populations in Australia, Canada, New Zealand, and the United States [43]. This review was conducted in accordance with the Joanna Briggs Institute (JBI) Reviewer's Manual 2017: Methodology for JBI Scoping Reviews [44]. Search terms were developed using a PCC (Population, Concept, Context) mnemonic. The premise and methods of this review, have been published elsewhere [43]. The Preferred Reporting Items for Systematic Reviews and Meta-analysis extension for scoping reviews checklist (PRISMA-ScR) [45] was adhered to in the reporting of this review (Additional file 1_PRISMA-ScR checklist).

Search strategy
The JBI three step search process was utilized to develop the search strategy [44]. This involved a preliminary search undertaken in MEDLINE and CINAHL using keywords from the review question. A tailored search was then developed for each information source. For database search strategies, a combination of Boolean operators, truncations and Medical Subject Headings (MeSH) were used (Additional file 2_ Academic database search strategies). Librarian assistance was provided for the development of the Ovid MEDLINE search strategy. Support was also provided in translating the search strategies into other databases. The reference lists of included studies were then searched for additional studies.
Multiple platforms were used to search for unpublished studies and grey literature which included: Australian, Canadian, New Zealand, and the United States Indigenous-specific research websites, Indigenous organisational websites, health services and health research websites and open access websites, repositories and catalogues (Additional file 3_Grey Literature sources).

Inclusion criteria and exclusion criteria
Literature based on the following criteria was considered (Table 1. Inclusion and exclusion criteria).
No restrictions were placed on the quality or study design used. All types of literature, including media releases, webpages and news articles, were considered. Literature published since 1 January 2006 was considered in order to capture mobile clinics implemented since the 'United Nations Declaration on the Rights of Indigenous Peoples' (2007), where a greater international focus on the need to work in partnership with Indigenous populations to improve health outcomes, was established [46].
For consistency, the term 'Indigenous' has been used throughout this review to refer to all clans, tribes and communities of Indigenous populations within a global context. We acknowledge the diversity and uniqueness of all Indigenous tribes, clans and nations. No disrespect is intended by the use of this term.

Study selection and data extraction
Searches for published and unpublished literature were conducted by two researchers (HB and GE). Titles and abstracts retrieved were screened independently by two reviewers (HB and GE). Full text review and data extraction were then undertaken independently by the same two reviewers. For articles not meeting the inclusion criteria, reasons for exclusion were provided. The reference lists of included citations were then screened for additional citations in order to scope for all possible citations meeting the inclusion criteria.
The published data extraction table was used and modified to extract the longitude and latitude coordinates for locations serviced by the included mobile clinics from publicly available information [43]. The coordinates were then imported into ArcGIS Arc-Map 10.6.1 (ESRI, CA, USA), a Geographical Information System (GIS), and mapped as point locations. Using a spatial join, the coordinates were linked with an underlying geographical characteristic described either as the Remoteness Structure (Australia) [47], Population Centre and Rural Area Classification 2016 (Canada) [48], or Urban status (United States) [49] to determine the classification of locations serviced by included mobile clinics. It is important to note that each country included in this review has a different rural area classification system. In Australia, Remoteness Structure comprises five categories: Major Cities of Australia, Inner Regional Australia, Outer Regional Australia, Remote Australia, and Very Remote Australia [47]. These classifications offer complete coverage of the Australian continent. Population centers in Canada are described as Small (1000-29,999), Medium (30,999) or Large (100,000 and over) with all other areas not classified, indicating very low population densities [48]. The urban footprint in the United States (high population density and urban land use) are described as Urban Clusters (2500-49,999) and Urbanised areas (> 50, 000) [49]. Like Canada, all other areas are not classified. The spatial data used was based upon each modern state's most recent census -2016 for Australia and Canada (next census due 2021), and 2010 for the United States (next census due 2020).
Review findings were developed using a descriptive approach that addressed the review objectives, as per the Joanna Briggs Institute (JBI) Reviewer's Manual 2017: Methodology for JBI Scoping Reviews [44]. This involved examining the evidence that met the inclusion criteria, providing a summary of citations and synthesising extracted data where possible (e.g. geographical characteristics of location(s) where mobile clinics were implemented).

Results
Database searches yielded 1672 citations. An additional 91 citations were retrieved from an extensive search of the grey literature and targeted organisational websites. A total of 1350 unique title and abstracts were screened, after duplicates were removed. The full texts of 123 citations were screened in accordance with the review criteria, identifying 39 relevant citations ( Fig. 1 -PRISMA Flow Diagram). An additional 14 citations were snowballed from 39 included citations, resulting in a total of 53 included citations discussing 25 mobile clinics.
Reasons for excluding citations were provided (Additional file 4_Excluded studies) and included: not an Indigenous-specific mobile clinic (n = 39), no mobile clinic (n = 25), not a primary health care mobile clinic (n = 12), sub-studies already included in search (n = 3), sub-studies did not meet the inclusion criteria (n = 3) and audio-recording not available (n = 2).

Finding 1: geographical distribution of mobile clinics for Indigenous populations
Of the 25 mobile clinics included (many servicing multiple locations), most were implemented in Australia (n = 14), followed by the United States (n = 6) and Canada (n = 5). No primary health care clinics implemented specifically for Māori populations in New Zealand, were retrieved from the search ( Table 2). In Australia, the majority of locations serviced by mobile clinics were located in Very Remote Australia (n = 44; Table 3; Fig. 2). This was compared to Inner and Outer Regional Australia, which both had a similar amount of locations represented (n = 15 and n = 17 respectively). The remoteness classification with the least amount of locations was Major Cities of Australia (n = 2).
In Canada, most locations serviced by a mobile clinic were outside the formal classification of population centres (n = 142; Table 3; Fig. 3). There was a declining presence of mobile clinics with the increasing size of population centres. This was similar to the United States where two thirds of mobile clinic activity was in areas classified as being outside Urbanised Areas or Urbanised Clusters (n = 24, Table 3; Fig. 3). Locations with a mobile clinic presence were more numerous in Urbanised Clusters (n = 11) compared to Urbanised Areas (n = 1).

Finding 2: primary health mobile clinic models for Indigenous populations
Of the mobile clinics included in the search (n = 25), the types of primary health care services and targeted populations varied (Table 3). These included delivering a broad range of general primary health care services (n = 13), providing disease specific services (e.g. diabetes management, screening and education n = 6, renal disease and other chronic disease screening n = 1, breast cancer screening n = 1, ear disease screening n = 3) and opportunistic health services and health promotion (n = 1) to Indigenous populations.
Most of the mobile clinics were implemented for Indigenous populations across the lifespan (n = 15), with fewer implemented for a specific age, gender group or population with chronic disease (infants, children or young people aged less than 18 years n = 4, people with diabetes n = 4, women n = 1, adults n = 1). There was evidence of Indigenous organisations governing and/or implementing 14 of the 25 mobile clinics (56%), with the remainder implemented in partnership with a non-Indigenous organisation or institution (n = 10). No information was provided about the involvement of Indigenous people in the implementation of one mobile clinic [67].
Information about the funding source(s) was retrieved for 19 of the 25 (76%) mobile clinics. Various sources were used to fund the mobile clinics which  Of the 25 included mobile clinics, 13 (52%) had evidence of some form of evaluation (Table 4). Of these 13 mobile clinics, most of the evaluation findings were disseminated in the non peer-reviewed literature or grey literature (n = 7 mobile clinics), with fewer evaluation findings disseminated in the peer-reviewed literature (n = 6 mobile clinics).
Of the evaluated mobile clinics, various approaches to undertaking an evaluation were used. Some evaluations produced multiple citations for a single mobile clinic  (Table 4). Most of the evaluations used quantitative methods of evaluation (n = 11) including descriptive statistics (e.g. of clinical indicators, patient demographics, service data), surveys and longitudinal data. One of these included a cost-effectiveness analysis [52]. Two evaluations used a mixed methods approach consisting of both quantitative and qualitative methods of evaluation. Of the two mobile clinics evaluated using mixed methods (e.g. including qualitative methods of data collection such as interviews and focus group sessions), one evaluation did not provide qualitative data [63], whereas the other provided rich qualitative findings with evidence of engaging with the perspectives and voices of Indigenous people [65,66]. Evaluations were heterogeneous in terms of evaluation methods and outcomes, making it difficult to compare findings. However, the participant sample included in evaluations was those receiving the services of the respective mobile clinic with a client or patient record (Table 4).

Discussion
To our knowledge, this is the first systematic scoping review examining primary health care mobile clinics implemented for Indigenous populations in Australia, Canada, New Zealand, and the United States. This review locates evidence of mobile clinics that have been implemented specifically for Indigenous populations (with the exception of New Zealand), and highlights the potential for mobile clinics to improve the accessibility of primary health care services. These findings are a valuable contribution to the growing body of international literature around the use of mobile clinics [28,29,32,33,36,38]. Before discussing the implications of these findings, it is important to reiterate that Indigenous populations are diverse, have different languages, cultural identities, customs, lore and spiritual beliefs [16]. However, Indigenous populations in Australia, Canada, New Zealand, and the United States share the experience of colonization and require culturally safe health care embedded in the principles of self-determination [7,16,17,46]. Likewise, there are key differences between the health care systems of Australia, Canada, New Zealand, and the United States, which may account for variations in the implementation of mobile clinics specifically for Indigenous populations. Australia, Canada, and New Zealand have universal access to health care for all populations [100][101][102] which differs from the partially-funded health care system in the United States [103]. There are also complexities around the policies of each modern state regarding the funding of Indigenous-governed health services and programs [104]. In the United States, funding is allocated through the Indian Health Service (IHS), with a key criticism being the failure to provide sufficient resources to meet the health care needs (particularly primary health care needs) of a growing Native American, Native Alaskan, and Native Hawaiian population [17,105]. In Australia and Canada, Indigenous health organisations (e.g. ACCHOs in Australia and onreserve First Nations health services in Canada) receive some funding from governments to provide primary health care services to Indigenous populations, yet inequities exist in the distribution of funding (e.g. lack of funding for Métis People) and power imbalances between government and Indigenous health-organisations [17,27,104]. The funding structure in New Zealand differs again, with a more integrated approach of health service delivery through mainstream health services or private agencies and greater participation of Māori People in the process of informing the policy of District Health Boards (DHB) [17,106]. The need to reform health care systems for the provision of equitable and culturally safe health care for Indigenous populations, has been widely discussed in the peer-reviewed literature [27,104,105].
There are also variations as to how population density is described in Australia, Canada, and the United States, which also has implications for interpreting the findings of this review (see Table 3). Australia's Remoteness Structure [47] has a complete coverage of the continent, whereas Canada and the United States classify their urban areas by population size [48,49]. Although there are other geographical methods for classifying population density (e.g. in Australia, Modified Monash Model [107]), this review has included classification methods used by decision-makers in each country at the time of analysis. For example, the Australian Government's Rural Health Multidisciplinary Training (RHMT) Program [108] utilizes the Remoteness Structure [47] to guide investment to improve the recruitment and retention of health professionals in rural and remote Australia. Likewise, the Population Centre and Rural Area Classification 2016 (Canada) [48] and Urban status (United States) [49] are both based on the most recent census for each respective country and are used in government decision-making. Acknowledging these variations, this review identifies a pattern of increasing presence of mobile clinics in areas with lower population densities (see Table 3). Geographical gaps in service provision are evident (Figs. 2 and 3), indicating that the implementation of mobile clinics for Indigenous populations is not widespread. There are also variations in the models of primary health care mobile clinics implemented for Indigenous populations. Most of the mobile clinics retrieved by this review targeted Indigenous populations across the lifespan, indicating a holistic family-centered model of primary health care, which is a preferred characteristic of Indigenous primary health care services [109]. Some mobile clinics targeted specific chronic diseases prevalent in Indigenous populations (e.g. diabetes) [110] and prevention of chronic disease for specific populations (e.g. otitis media in Aboriginal and Torres Strait Islander children) [111]. Although there was some evidence of Indigenous organisational governance or involvement in the implementation of most mobile clinics, it was difficult to ascertain the degree of Indigenous community ownership. This is a key issue which has been discussed in another review examining chronic disease programs implemented for Aboriginal and Torres Strait Islander populations [112], and in the international literature examining health services and programs for Indigenous populations [27,113,114]. Indigenous community ownership of mobile clinics is imperative to ensuring culture, self-determination, and community participation are embedded in the delivery of primary health care services [109].

Fig. 2 Location of mobile clinics implemented for Indigenous populations in Australia
A paucity of published and publicly available evaluations of primary health care mobile clinics implemented specifically for Indigenous populations is also highlighted. This is despite a growing body of literature evaluating mobile clinics implemented for general populations and those atrisk for developing chronic disease, particularly in the United States [28, 31-33, 36, 39]. Although there is heterogeneity in the approaches used to evaluate mobile clinics implemented for Indigenous populations, there is some evidence that supports the potential for mobile clinics to increase attendance rates to services [54,62,69,72] and improve clinical indicators (e.g. BMI, HbA1C) of targeted chronic diseases (e.g. diabetes) in Indigenous people accessing mobile clinic services [41,79]. However, evaluation methods have relied heavily on the analysis of patient records and service data (see Table 4). The perspectives and insights of Indigenous people accessing mobile clinic services is largely absent. Findings support the need for high quality evaluations of Indigenous health programs which integrate qualitative evidence regarding the views and perspectives of Indigenous people [115]. An absence of qualitative data around the effectiveness of mobile clinics makes it difficult to know whether mobile clinics have potential to improve the cultural accessibility of primary health care services for Indigenous populations. This is a gap in existing knowledge which requires further research.
It is also difficult to examine how sustainable primary health care mobile clinics are when implemented for Indigenous populations. It is noted that the five diabetes mobile clinics retrieved from Canada were funded under the Aboriginal Diabetes Initiative (ADI), yet it is difficult to identify from the available literature as to whether all of these mobile clinics have been sustained over time under the original funding arrangement [79]. This highlights a key issue mediating the sustainability of mobile clinics in general, being the reliance on multiple funding sources (e.g. government and philanthropic) and/or short funding cycles [33]. There is also limited cost-effectiveness data around the use of mobile clinics for Indigenous populations [52]. Future research should include economic evaluations, coupled with an evaluation of the effectiveness and cultural acceptability of mobile clinics for Indigenous populations. This is imperative to informing the allocation of resources by decision-makers (e.g. governments and Indigenoushealth organisations) to mobile clinics.  In partnership (1) Feasibility study [51] (2) Cost-effectiveness analysis [52] (3) Pre and post-intervention analysis of hospital ENT service utilization [53] (4) Retrospective review of service activity from 2009 to 2011 [55] (5) Retrospective review of service activity from 2009 to 2014 [54] (    (1) Descriptive statistics of services delivered [73] (2) Descriptive statistics of attendance rates [74] (3) Descriptive statistics in annual report [75] (   [82] (2) Telephone survey [83] (3) Longitudinal analysis [84] (

Limitations
Every effort has been made to search academic databases and grey literature sources for primary health care mobile clinics that have been implemented for Indigenous populations in Australia, Canada, New Zealand, and the United States. In Australia, it is known that a significant proportion of health research involving Aboriginal and Torres Strait Islander populations is published in the grey literature [116]. A thorough search of grey literature information sources across key websites has been undertaken through the independent searching of two researchers and follow up of organisations, authors and researchers for additional information. Therefore, a limitation of this review is the manual processes required to undertake this search and the acknowledgement that there is the potential for some mobile clinics to be missed due to this.

Conclusions
This review identifies geographical gaps and a paucity of evidence around the implementation of primary health care mobile clinics for Indigenous populations in Australia, Canada, New Zealand and the United States. The findings support the need to undertake rigorous mixed methods evaluations of primary health care mobile clinics implemented specifically for Indigenous populations. Through the involvement of Indigenous people in the evaluation process, greater insights will be obtained as to the potential for mobile clinics to improve access to culturally safe and holistic primary health care services. It is important for organisations implementing primary health mobile clinics for Indigenous populations, to share their experiences by making evaluations publicly available, ideally through the peer-reviewed literature. This is essential in developing evidence around innovative models of health care that have the potential to improve health outcomes for Indigenous people globally. Dissemination of evaluation evidence concerning mobile clinics will also be invaluable to decision-makers, including Indigenous health organisations, who are considering allocating resources to a primary health care mobile clinic.