Utilization of Mobile Mental Health Services among Syrian Refugees and Other Vulnerable Arab Populations—A Systematic Review

The global refugee crisis is at its most critical state in history; Syria alone has produced 12 million internally displaced persons, with another 5 million refugees seeking protection across the globe. Faced with the heavy burden of mental distress carried by a massive refugee influx, many host nations lack the service capacity to respond adequately. While mobile mental health (mMHealth) applications and platforms have the potential to augment screenings and interventions for vulnerable populations, an insufficient gender and cultural adaptation of technology may drastically hamper its uptake in Arab refugees. Reporting only papers originating from Middle Eastern and/or Arab nations or refugee host nations, this systematic review evaluates the available literature published between 2000 and 2019 on the usage acceptability of mMHealth in Syrian refugees and other vulnerable Arab populations. We conducted a systematic review in PubMed, PsychInfo, Association of Computing Machinery (ACM) and the Directory of Open Access Journals (DOAJ) using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to identify studies that addressed mMHealth implementation in these populations; of a total of 607 articles identified, only 10 (1.6%) available, unique articles met our search criteria. These studies discussed the feasibility and efficacy of mMHealth applications and the barriers to their uptake. The few existing studies show positive impacts of mMHealth on the access to services and on treatment outcomes but also reveal a paucity of literature on mMHealth for vulnerable Arab populations. These findings indicate a critical need for research on the barriers to mMHealth uptake, to bolster service capacity in the Arab Region and in the refugee diaspora of other, non-Arab host countries.


Background
The global refugee crisis is currently at its most critical point it has ever been in history. In the last three decades devastating wars and violent conflicts in the Middle East, Africa, Central America, Eastern Europe and Central Asia have created millions of refugees worldwide, ushering in a new age of globalized forced migration and drastically shifting diasporic landscapes. By the end of war-impacted Arab countries, telehealth seems to offer a more viable solution than clinic-based direct care. Though many efforts have been put forth, one strategy that has recently gained momentum is the creation of mobile health (mHealth) and mobile mental health (mMHealth) interventions to achieve healthcare objectives and transform healthcare delivery worldwide.
There is no standardized definition of mHealth; however, the WHO's Global Observatory for electronic Health (GOe) defines mHealth as medical and public health practice supported by mobile and wireless devices [7]. The recent rapid rise in mobile phone penetration in developing countries could be leveraged to mitigate the challenges of a high population growth (whether through natural increase or through immigration), a high burden of disease prevalence, insufficient health care workforce, of large rural and/or marginalized immigrant populations and of inadequate financial resources to support healthcare infrastructure and health information systems [8]. With both providers and consumers having a greater access to mobile phones, the costs of delivering e-healthcare drop. Naturally, mHealth is not without its share of criticism. Disparities in low-and middle-income countries could exacerbate new forms of digital healthcare exclusion [9]. A rapid uptake of mHealth might alter the practice of healthcare and patient-physician relationships with unintended ill consequences if a preparatory technological education and cultural adaptation has not been adequately put in place to ease an organic transition to e-health [10]. In the context of electronic health records, patient confidentiality and data protection are of special concern [11]. Nevertheless, since the benefits of mHealth uptake outweigh the vulnerabilities mentioned above, especially in countries pressured by a sudden massive influx of refugees, one would expect to see strong policies supporting a faster uptake of mMHealth. Given the widespread penetration of mobile phone networks, even in low-and-middle income nations, this idea carries significant feasibility [6].
Pilot studies and initial reviews from other mHealth areas have reported on successes and barriers to mHealth success. A review of health system utilization by Sarria-Santamerra et al. discusses the general lack of utilization of health services by immigrant populations in a variety of host nations and conclude that further studies are needed to evaluate patterns of health resource usage among these populations [12]. A study from Ankara, Turkey identified language as a major barrier for utilization of health services [13]. The Turkish Ministry of Health is now employing Syrian Health Workers to improve access to Syrians living in Turkey [14]. Additionally, another study by Batniji et al. describes political and governmental instability and accountability as major barriers to all health services in the Arab world [15]. This issue is further compounded by the massive influx of Syrian refugees in these countries. Despite significant barriers, there have been pilot successes in mHealth. A study by Saleh et al. analyzed an mHealth platform for non-communicable disease management among Syrian refugees in a Lebanese refugee camp and found a significant decrease in mean systolic and diastolic blood pressures and HbA1c levels [16].

Cultural Modulators of Distress and Treatment Acceptance
However, many factors and barriers are left to consider in terms of social acceptability, cultural and gender norms and infrastructure for mMHealth success. Currently, there is a paucity of literature regarding the development and implementation of mMHealth interventions for mental health among refugee populations, especially in predominately Arab nations. The effectiveness of mobile health technologies to produce desired outcomes (e.g., screening for mental distress) may be reduced when used with an application that is not specifically designed for mHealth (e.g., Skype or WhatsApp). Furthermore, when mMHealth applications fail to formulate their questions (e.g., screening questions) in alignment with Arab idioms of distress, the Western mental constructs conveyed by such questions may not be clearly understood by Arab laypersons whose cultural expression of suffering may be different. Lastly but not in the least, the mode in which mobile health application may convey their questions to users may ignore or even conflict with Arab social norms of gendered communication. For example, questions probing a female patient's sex drive would be seen as unacceptable or even violating family space and sense of honor. For example, in a study conducted in 2011 by Barber et al. with 68 Palestinians in the West Bank, East Jerusalem and the Gaza Strip, the participants articulated their suffering in the following Arabic terminology: broken, crushed ( muh . at . t . ima), shaken up ( mahzūza), destroyed, ( mudammira) and exhausted, tired ( ta'bāna) [17]. The aspects of the sufferer were described as the self or spirit ( an-nafs), morale ( al-ma'naūiyyat) and hopes or aspirations for the future ( amālak aū t . amūh . atak ban-nisba lil-mustaqbil) [17].
The Arab construct of "hozon" ( , sadness and difficulty in the face of an acute or sudden stressor) may be referred to by Syrian refugees as al-hayat sawda ( , 'a black life') or iswadat al dounia fi ouyouni ( , 'life has blackened in my eyes') or could be used to signal intense grief from losses and withdrawal from social life [17,18]. When intended to signify a chronic state of depression, "hozon" is replaced by laypersons and mental health practitioners alike with halat ikti'ab ( , 'condition of ikti'ab'), an agglutinant concept of brooding, darkening of mood, aches, a gloomy outlook, somatization and social withdrawal [19].
Moreover, many Arab refugees do not perceive their suffering as the "clinical picture" of a mental disorder and reject the pathologization and overmedicalization of their war-induced distress [20]. The meaning they attribute to their extreme grief is that of a commensurate yet natural response of a normal human being who has been forced to live under extreme conditions of duress. This may require at times a de-pathologized naming of human suffering, as Joseba Achotegui argues in his "Ulysses Syndrome" [21].
Not only are the idioms of distress different in the Arab world, the very construction of personhood (embedded versus autonomous) and the relating, positioning and communicating across sexes is vastly different from Western norms [22]. Consequently, for a successful uptake, Western information and communication technology (ICT) transfer, in general and mMHealth in particular, may need to adapt in alignment with the gender and cultural reality of the Arab consumer.
The problem is that historically there appears to be not only a lack of mMHealth technology applied to the Arab and Syrian refugee populations but also a lack of cultural adaptation of the innovation [23]. Many attempts to apply a technology acceptance model (TAM) have ignored the models' weaknesses in predicting the cultures and behaviors of individuals and organizations within the complex health domain [24]. Such lack of differentiation between technological and human factors limits mMHealth practical applicability [25]. The development of a cultural and gendered innovation in mobile mental health in this region may better reflect the complex interactions between refugees, health systems and technology and may actually yield a higher user uptake [26], This matter has not been dealt with as rigorously as it should have been in the mMHealth literature. We surmise that part of the problem is the lack of an established theory base and of an evidence-based, consistent, gendered and culturally competent conceptual framing and grounding of mMHealth development, diffusion and transfer.
The present systematic review intends to investigate the current literature available regarding mMHealth and to elucidate key findings of outcomes, successes and barriers. Generally, the authors intend to highlight the current academic landscape of the mMHealth concept to some of world's most vulnerable populations: Arab vulnerable populations and Arab refugees in the Arab region and in non-Arab host countries. Our three pronged systematic review sought to explore the existing research literature on (1) the current presence, uptake and outcomes of mMHealth applications among Arab populations and Arab refugees, (2) the mMHealth usage acceptability and barriers to use among this populations and (3) the mMHealth adaptation to Arab culture and gender norms.

Materials and Methods
To analyze the existing literature regarding mMHealth applications in Syrian refugees and Arab populations, we conducted a literature review in the following four databases: PubMed/Medline, PsychInfo, Association of Computing Machinery (ACM) and the Directory of Open Access Journals (DOAJ). To ensure validity and standardization of our methods, we utilized the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [27]. The context of our research questions is provided in Table 1, utilizing a format provided by PRISMA [27]. This literature review was conducted from March 2019 to November 2019. Qualitative studies (Studies that provided data and statistics) 3.
Outcomes based studies (Studies that utilized data to discuss differences in outcomes between mMHealth and other treatments)

1.
What is the current state of mMHealth utilization towards Syrian Refugees and Arab populations? 2.
What types of mMHealth solutions have been utilized? Are they being utilized for monitoring, reporting, tracking or alleviating mental health burden? Any other uses? 3.
If mMHealth applications are being used, what has been their efficacy? Are these solutions effective? 4.
What are the barriers to mMHealth utilization in the study populations? * PICOS = Participants, Interventions, Comparisons, Outcomes, Study Design. Adopted from the PRISMA Guidelines [27].

Information Source and Search Strategy
The review considered English language quantitative and qualitative primary studies on mMHealth published between 2000 and 2019 in the peer-reviewed literature from four databases: PubMed, PsychInfo, ACM and DOAJ. The retrieved journal articles were downloaded into Endnote X8 citation management software and duplicates were removed. The authors were divided among two primary research teams, each with the same tasks: (1) Screen the four databases with our standard search methodology and (2) Evaluate the quality of the articles. After the primary researchers of each team screened the titles and abstracts, the remaining members of the research team reviewed the results based on the inclusion and exclusion criteria. Two quality appraisal meetings with both teams were held to discuss all uncertainty regarding articles for inclusion or exclusion articles for inclusion. The remaining relevant articles were organized into three major categories-excluded literature, background literature and possibly relevant literature.
We utilized a search algorithm using Boolean methodology to create the ideal search query for all 4 databases. The search algorithm encompassed all possible terminology for mHealth, as well as a series of factors that qualified the search. The exact search queries for each journal are tabulated in Table 2.

Inclusion/Exclusion Criteria
Articles were included only if they (1) originated from Middle Eastern and/or Arab nations or refugee host nations; (2) focused on a study population of Arabic speakers and/or Syrian refugees; and (3) primarily discussed mMHealth applications and/or feasibility of this technology within the study population. The review excluded studies published before 2000, grey literature and non-peer-reviewed articles. Only articles that met all criteria were included for further evaluation. The included articles were assessed for key findings related to mMHealth within the study population of interest.

Data Extraction
Each research team performed a criteria-based data extraction (citation, participant characteristics, methods, results, key conclusions, recommendations); afterwards the results were cross-checked by two researchers for any possible data errors. Figure 1 shows the PRISMA process undertaken for this review.

Inclusion/Exclusion Criteria
Articles were included only if they (1) originated from Middle Eastern and/or Arab nations or refugee host nations; (2) focused on a study population of Arabic speakers and/or Syrian refugees; and (3) primarily discussed mMHealth applications and/or feasibility of this technology within the study population. The review excluded studies published before 2000, grey literature and non-peerreviewed articles. Only articles that met all criteria were included for further evaluation. The included articles were assessed for key findings related to mMHealth within the study population of interest.

Data Extraction
Each research team performed a criteria-based data extraction (citation, participant characteristics, methods, results, key conclusions, recommendations); afterwards the results were cross-checked by two researchers for any possible data errors. Figure 1 shows the PRISMA process undertaken for this review.

Inter-rater Reliability
The selection phase of this review was rigorously based on inter-rater agreement or disagreement between the two research teams. Cohen's kappa coefficients (k) were calculated to quantify inter-rater agreement. We used the IBM Statistical Package for the Social Sciences (SPSS)

Inter-Rater Reliability
The selection phase of this review was rigorously based on inter-rater agreement or disagreement between the two research teams. Cohen's kappa coefficients (k) were calculated to quantify inter-rater agreement. We used the IBM Statistical Package for the Social Sciences (SPSS) version 22.0, IBM, Armonk, NY, USA, to determine the statistical significance of agreement between the research teams. The primary response categories were "Yes" or "No/Not applicable." "Yes" referred to agreement on selecting an article for inclusion, while "No/Not applicable" meant disagreement on article inclusion. The two primary categories of agreement were "Initial Screening" and "Quality Evaluation." The category "Initial Screening" was then divided among agreement/disagreements of inclusions based on the following-title, abstract, full article and type of article (reviews, outcomes studies, qualitative studies, randomized control trials, etc.) and additionally the level of comprehensiveness of the search. The category "Quality Evaluation" was then divided among agreements/disagreements regarding quality disputes of the following-topic, study design and quality of the research journal.

Review Results
A total of 607 articles were identified using the Boolean search queries on PubMed, PsychInfo, ACM and DOAJ databases. Of the 607 articles, only 10 (1.6%) unique articles met all inclusion criteria. Of the 10 included articles, 7 studies discuss the utilization of novel mMHealth applications, 5 discuss barriers, 5 discuss attitudes and acceptance of mMHealth, 1 is a randomized crossover study and 1 is a randomized control trial analyzing efficacy of an mMHealth prototype. Seven unique mMHealth prototypes were discussed and included 3 web-based interventions, 2 mobile applications, 1 video/audio transfer tool and 1 tele-conferencing tool. The articles report 64-88.2% of their study populations express positive interest in mMHealth applications and positive attitudes toward mMHealth utilization. All 7 individual mMHealth technologies were efficacious as screening tools and psychiatric treatment and the majority of participants and providers had positive attitudes and displayed acceptance of mMHealth. Additionally, the studies report 64-93.4% of participants owning a mobile phone and being connected to a mobile network. The randomized control trial shows significant efficacy of a mMHealth application prototype on symptom relief of post-traumatic stress disorder. The randomized crossover trial showed good efficacy and feasibility of a mobile application. Barriers discussed primarily included primarily cultural, financial, technical, infrastructural and data privacy limitations and lack of mental health awareness in the study populations. Results and key findings are tabulated in Table 3 [6,28-36].   They preferred telepsychiatry in their mother tongue rather than interpreter-assisted care.
• Limited infrastructure regarding setup of telepsychiatry areas and exam rooms

Inter-rater Reliability
The k calculated for each agreement/disagreement category and sub-divisions were consistently high in the initial screening and quality appraisal phases of this review. These results statistically, indicate consistent agreement among the authors and teams regarding the inclusions/exclusion of studies. The values of Cohen's Kappa showed an acceptable range of 0.83-0.95. In an overall assessment of inter-rater reliability, the screening and evaluation process showed significant agreements (p < 0.001) regarding the selection of articles with the given inclusion and exclusion criteria. These results are tabulated in Table 4. Table 4. Inter-rater Reliability (K Table).

Discussion
The few studies that were available discussed implications, efficacy and barriers to mMHealth applications within the target population. Though only seven articles included in this review referred to mobile phone applications, their data clearly suggest that mMHealth applications carry high feasibility and potential, as many patients in the target areas of the Middle East own a mobile phone and have access to a cellular network [12]. Additionally, the cross-sectional studies suggest that the majority of Arab patients and Syrian refugees perceive mMHealth as positive and welcome the prospect of using mobile technology to monitor and bolster their mental health, among other use cases of this novel technology [4, 28,29]. In terms of statistical efficacy of mMHealth vs conventional treatments, not many studies were available. However, the single randomized control trial included in the present study found a statistically significant reduction in post-traumatic stress syndrome symptoms in Arab patients from war-torn backgrounds [34]. Additionally, the remainder of the articles showed efficacy of the technologies as screening tools for this population [6,[28][29][30][31][32][33][34][35][36]. It is imperative that mMHealth innovations are compatible with local needs of intended adopters and implementing institutions. This is evidenced by the barriers identified in five of the articles including inadequate technological capacity, insufficient or lack of cultural adaptation and acceptance and poor credibility as barriers to mHealth implementation [28,29,31,34,35]. Given the extent of the current refugee crisis and the incidence and prevalence of mental health disorders among these vulnerable populations, it is imperative that further studies are conducted to evaluate the efficacy of this potentially groundbreaking technology. With regard to the success of mMHealth uptake among vulnerable Arab populations, many barriers exist spanning stigmatization, technological literacy, technological access, general distrust towards healthcare providers and political conflicts [14,28,30,35]. In Arab communities, there is a strong social stigma associated with mental illness as men who seek care are considered weak and women associate it with helplessness and being 'cursed' [12,37,38]. Another barrier is the preferred provider-patient relationship in this population, in which Arabs expect an authoritative style of communication where providers give concrete and specific solutions rather than long-term, "collaborative," dynamic psychotherapy [37].
Additionally, technological barriers can significantly hinder mHealth uptake, including lack of access to Wi-Fi networks. A survey conducted of Palestinians in the West Bank showed an estimated 80% of the population have regular access to a smartphone, with 70% having a pay-as-you-go plan, whereas in Egypt only 45% of the Syrian refugees have access to a smartphone [12]. Another study from Syrian refugees in Egypt found 70% of refugees reported the cost of mobile access to be a barrier [12,37]. Additional barriers reported from the study were-lack of credibility of services (57%), acceptability of the application (47%) and technical literacy (25%) [12].
Mobile mental health is a promising way to reach a large portion of this vulnerable population; since most refugees have smartphones, the accessibility of a mobile app technology is feasible for usage. Additionally, mental health services in general that have been culturally, religiously and linguistically adapted to refugees in the Levantine region (Syria, Lebanon, Palestine) have been met with overwhelmingly positive results [13,15,16,37]. Therefore, the ideal mMHealth platform would incorporate linguistic, cultural and religious adaptations.
mHealth is a fairly novel concept, thus the literature is generally lacking in evidence of success globally and in all forms of healthcare from mental health to communicable diseases to trauma systems registries. Despite its novelty, mHealth has had some inspiring successes that may serve as precedents for governments and NGOs around the globe. A review article by Abaza and Marschollek published in 2017 identified 255 distinct articles addressing different applications of mHealth technology [38]. These studies spanned the globe and covered all continents except Antarctica and addressed mHealth applications for chronic diseases, transplantation, dermatology, dentistry, health promotion, maternal and child health and so forth [38]. Though many of the included articles had limitations of size, the conclusions of the study were positive for the development of both SMS and application based mHealth solutions for a diverse array of healthcare conditions [38].
Despite challenges and barriers, the prospect of mHealth carries significant potential. According to the WHO's GOe, the unprecedented spread of mobile and electronic technologies has given over 85% of the world's population-over 5 billion individuals-coverage with a commercial wireless signal [7]. The development of mobile phone networks in low-and-middle income countries has even superseded infrastructure development of roads, electricity and traditional internet deployment [7]. As technological sophistication continues to advance, speeds of data transmission and availability of lower-cost mobile devices will transform data exchange globally [7]. Since 2011, more progress has been made globally in eHealth. The key findings of the 2016 WHO GOe Report show that 58% of responding WHO Member States have an eHealth strategy, 90% of countries with an eHealth strategy report that they have special funding available for it, 50% of WHO member countries have government-supported health internet sites that offer information in multiple languages and 75% of WHO member countries have institutions that offered training in information communication technologies (ICT) for health professionals [8]. Given these global conditions, it appears that the integration and development of mHealth platforms could potentially bolster capacity, accessibility and delivery of care systems to vulnerable individuals even in the most remotes areas of the world.
We acknowledge our study had limitations. We utilized only four databases (PubMed, PsychInfo, ACM Portal and DOAJ) and therefore we might have missed relevant literature in other repositories. Additionally, our algorithmic search method may have not included literature available outside of the keyword inputs. However, based on the sensitivity of the four databases used for this systematic review and on the appropriate inter-rater reliability of our search methods, evidenced by high Cohen's kappa values, we are confident that the results of this study are significant. A third limitation of our systematic review lies in the circumstances that prevented the conduction a single meta-analysisinsufficient data availability, lack of qualitative data that would better answer review questions and significant differences in the few identified studies that prevented the combination of data.
Nevertheless, it is exactly this paucity of data in the field of Arab mMHealth technology that exemplifies the current challenges and directions for further research.

Conclusions
The mental health needs of a massive refugee influx often exceed a host nation's mental health system capacity to respond adequately. mMHealth technology diffusion and transfer could greatly complement clinic-based care while also bolstering system capacity. A clear understanding of barriers to mMHealth implementation holds the potential of advancing feasible solutions to unexpectedly high demands. The present study affirms the paucity of literature and lack of evidence available regarding the uptake of mMHealth interventions among Syrian refugees and other vulnerable Arab populations. Whether in the Arab Region or in the diaspora of non-Arab host countries, the study also identifies some of the current challenges to mMHealth implementation. Most importantly, the study findings suggest that some the main obstacles to mMHealth usage acceptability in Arab populations, may be the lack of an established theory base congruent with the Arab socio-cultural norms and the lack of a realistic conceptual framing and grounding of the mMHealth uptake problem. Recommendations include a paradigm shift from the current Western models of eHealth acceptance and diffusion, to an evidence-based Arab gendered and culturally competent conceptual framework capable of informing more user-acceptable adaptations of mMHealth.