Octennial Review (2010-2018) of Literature on M-Learning for Promoting Distributed-Based Medical Education in Sub-Saharan Africa

Medical education in Africa is in desperate need of reforms, evident in widespread diseases, and an inability to mobilise and train the required medical workforce to deal with these health issues. However, the exponential rise in the use of mobile technologies due to the spread of the Internet and increased telecommunication networks offer an opportunity for the transformation of medical education and practice through the deployment of mobile devices as a medium for learning and conveying health care services to the remote and resource-constrained locations of Sub-Saharan Africa (SSA). This paper reviewed articles on the affordances of m-learning for distributed medical education in SSA published between 2010-2018. Results from 18 articles identified in the review revealed a slow-paced ascendancy of practice and research in the field; it further exposed competing priorities, infrastructural deficit, and chronic workforce shortages as the bane of m-learning implementation in the subregion. This paper makes recommendations that will enhance the growth of mobile-based distance medical education and practice in SSA.


Introduction
Globally, the advent of the Internet has led to a progressive transformation in all facets of human activities. Amongst these, is the impact of Internet-driven technologies that have reshaped the educational landscape. It has become apparent that the traditional form of teaching and learning confined to the classroom is no longer efficient due to limited out of class collaboration and inadequate infrastructure (Africa-America Institute (AAI), 2015; Organisation for Economic Co-operation and Development (OECD), 2009). However, electronic learning (e-learning) has the affordances to accommodate more people, especially those with limited access to regular schooling (Yunusa & Dalhatu, 2014) as an alternative medium to education. The increasing rate of the remote and mobile workforce has resulted in the emergence of the use of mobile technologies for communication, collaboration, and exchange of occupational knowledge practices and training. Consequently, this has evolved into various forms of technology-enhanced learning solutions across human endeavour and the ubiquity of mobile devices for mobile learning is now potentially more revolutionary than e-learning (Okai-Ugbaje, Ardzejewska, & Ahmed, 2017).
Keegan (as cited in Ally, 2009) predicted that the future of mobile learning is "wireless," and today virtually all human activities are gradually transforming into online using several mobile devices and platforms. Sub-Saharan Africa (SSA) has the most significant number of adolescents and young adults in the world. The International Telecommunication Union (ITU) report (2016) on mobile cellular penetration indicate a mobile penetration rate of 65% in SSA. Relatedly, Masita-Mwangi, Mwakaba, Ronoh-Boreh and Impio (2012) note that 43% of the population in SSA is below the age of 15 years making the region the youngest region in the world. Consequent upon which many African countries are making efforts in widening access and creating opportunities for the expansion of the Higher Education institutions to accommodate these young adults. However, these efforts are ineffective due to the rising number of students with an average of 50% more students per professor in African universities compared to the global average AAI (as cited in Bervell & Umar, 2017).
One key aspect of education in sub-Saharan Africa that needs a technological intervention due to the problem above is medical education. The use of mobile learning offers opportunities for training more medical staff and promoting distance-based medical practices in Sub-Saharan Africa. Recently, authors such as Chaya, Pilot, and Urassa (2018) ;Lazarus, Sookrajh, and Satyapal (2017a) and Witt et al. (2016) had echoed the need for reforms in medical education and practice in SSA when they suggested the mobile learning paradigm. Recognising the need for mobile learning, these African medical practitioners have suggested that the adoption of mobile technologies and online courses will enable class sizes to increase dramatically in a short space of time and could potentially be significant in mobilising a medical workforce that is struggling with traditional modalities that cannot produce the desired results. The growth of these technologies across the world and Africa, in particular, has prompted some studies on the use of mobile devices in medical education in SSA (Adebara, Adebara, Olaide, Emmanuel, & Olanrewaju, 2017;Chaya, Pilot, & Urassa, 2018;Frehywot et al., 2013;Ibrahim, Salisu, Popoola, & Ibrahim 2014;Lazarus, Sookrajh, & Satyapal, 2017b;Masika et al., 2015;Witt et al., 2016b). Koole (2009) defines mobile learning as "a process resulting from the convergence of mobile technologies, human learning capacities, and social interaction" (p.25). Accordingly, this definition placed m-learning within four central constructs namely, technological tools, context and social interactions buttressed by Krull and Duart (2017) and Naismith, Lonsdale, Vavoula, and Sharples in promoting medical education and practice in Sub-Saharan Africa. The World Health Organization (WHO) (2011) conducted a global survey on the explosion of m-health and telemedicine activities around the world. Even though the report provided some extracts on efforts at using mobile technologies to deliver health services and information to remotely located and resource-constrained communities, the outcome of this report indicated that out of the 114 countries surveyed, SSA had the least positive indices on initiatives in m-learning in medical education (WHO, 2011). Against this background, this paper reviewed published articles based on studies carried out in the subregion between 2010-2018, bringing to fore the trends, milestones, and barriers to effective implementation of mobile technology-based learning solutions in medical education and contributes a reference that will guide policy and practice in the subregion.
Forthrightly, the study sought to answer the following questions:

Methodology
This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol by Moher, Liberati, Tetzlaff, Altman, and Prisma Group. (2009). The PRISMA protocol is a step-by-step process to search, collect, analyse, synthesise, and report findings from the literature (see Figure 1). The first step was to search databases for studies on Mobile learning and Medical Education research in SSA. Search terms such as "m-learning," "m-health," "e-health," "mobile technologies," "medical education," "telemedicine," "adoption," "implementation," "mobile learning," and "Africa," were explored in prominent databases such as Google Scholar, Science Direct, Scopus, IEEE, PubMed, Microsoft Academic, and Institutional database. Reference pages of retrieved articles were screened for relevant studies for the review materials. The obtained articles were then sorted based on inclusion, exclusion, and eligibility criteria, and set out for the review (see Table 1). While, Table 2 depicts the summary of the reviewed studies.  2. Design of the study: The review focused on three broad study designs namely; qualitative, quantitative, and mixed method (Creswell, 2013).
3. Theoretical framework: The review focused on the theories adopted or adapted for the studies on m-learning in medical education in sub-Saharan Africa.
5. Subjects of the study: Subjects were; instructors, students (Interns), nurses, other health professionals or both.
6. Statistical tools: Instruments employed for data analysis were; qualitative thematic/narrative (Creswell,2013) Table 3 presents the results on the geographical spread of articles on m-learning in medical education in Sub-Saharan Africa:   Figure 2.

The Methodology Employed in Studies of M-Learning in Medical Education in SSA
Regarding research question three, the methodologies or approaches employed included; research design, data collection, subjects, sample size, and statistical tools used for the analysis.

Study Design and Instruments
The aspects of the methodology assessed were the research design and instruments adopted by the various studies, as presented in Table 4.

Subjects and Sample Size
The subjects selected for the studies and their corresponding sample sizes are featured in Table 5.

Statistical Tools Employed for the Analysis
The quality and reliability of research findings lie with the analytical tools used for the analysis of the study. In that respect, this study sought to find out and aggregate the analytical tools used in the studies on m-learning in medical education in SSA as depicted in Table 6.

100%
As shown in Table 6, eight studies (44.4%) used descriptive statistics, i.e. simple frequencies and percentages to analyse their data. Five studies used thematic and narrative analysis for their data analysis, especially for qualitative data. Four studies employed predictive and group difference analysis tools to analyse their data (22.2%). The least used statistical tool was a correlation which was featured in only one study (5.7%).

Effective Factors/Milestones Determining M-Learning in Medical Education in SSA
In response to research question four of this review, the study catalogued various factors reported across the reviewed studies as determinants or milestones of m-learning use or adoption. Results from the analysis produced ten elements and their frequencies of occurrence as represented in Figure 3.

Challenges Involved in M-Learning Implementation in Medical Education in SSA
The outcome of problems in the reviewed studies is depicted in Figure 4.

Determinants of m-learning in Medical Education in SSA
values  reported competing priorities in governance as a determinant factor with 12 studies (66.7%) reporting a lack of technical skills. The least challenges in the reviewed studies include: battery life 2 (11.1%) and small screen 2 (11.1%).

Discussion, Recommendations, and Conclusion
The aggregate of the reviewed studies on m-learning in medical education in Sub-Saharan Africa indicated that western and southern Africa contributed more studies in m-learning in medical education. The outcome was because four studies were from Nigeria, and one was from Ghana, while Botswana and South Africa had three studies and two studies respectively. These subregions; southern and western Africa, are home to two of the most economically advanced countries in Africa as corroborated by Okai-Ugbaje, Ardzejewska, and Ahmed (2017) and Bervell and Umar (2017). The authors reported that world bank classification of countries in the region by income placed South Africa as an upper middle-income country while Ghana, Nigeria, Kenya, Tanzania, Uganda, and Botswana as lower-middle-income countries with potentials for further growth (Bervell & Umar, 2017). Tanzania. However, to have an effective continuous professional development program that will help in mitigating the impact of chronic diseases such as tuberculosis and HIV, there must be sustainable financial resources to provide the infrastructure that will lead to increased capacity and skills of the healthcare workers (Feldacker et al., 2017).
The results as presented in this study may be considered abysmal against the background that sub-Saharan Africa comprises of 48 countries out of which only nine states produced the number of reviewed articles (18). This supposedly indicates the low level of m-learning use in medical education in the subregion and holds promise for further research in that domain.
On models and designs adopted in research on m-learning, the review found out that there were no research models or conceptual models used in all the studies included in the report. Models are important because they provide powerful ways to address key information system research problems such as understanding information technology usage (Chin & Todd, 1995). Also, Borner (as cited in, Baltimore, Charo, & Kevles, 2016) asserts that models of technology and innovation help to inform policy decisions in education, health care, and other sectors which empower experts to make informed recommendations and predictions. Against this backdrop, model-based research will be constructive since m-learning is an attempt to improve the teaching and learning process in medical education, an approach that involves the use of mobile technologies for instructional, collaboration, and remote access to materials for just-in-time learning (Frehywot et al., 2013). However most of the studies in this review dwelled on practical solutions such as testing the effectiveness of smartphone (Android-based) applications with point-of-care programs for seeking medical information (Chang et al., 2012) and expert system developed to ease communication between doctors and health personnel across remote locations (Olajubu et al. 2014).
Regarding research design and instruments used in m-learning in medical education research in SSA, most of the reviewed studies applied the quantitative research design. Although the quantitative design is a standard measure and an excellent way of finalising results in the scientific field, its limitation is not accommodating findings that could augment the analysis Creswell (as cited in Bervell and Umar, 2017). However, a qualitative design was used in three of the studies, while a mixed method design was featured in five of the reviews. The authors are of the view that mixed method research approaches offer better ways to explain and understand the complexities of the organisational, social, and scientific phenomenon.
The study also revealed the use of all health personnel within the health care system as the most recurring subjects in the studies. This is understandably so as the interrelationship that exists between these actors in the implementation of any technology intervention in the health sector is critical to its success.
The dominant statistical tool adopted for data analysis in the studies on m-learning in medical education in SSA was descriptive statistics. The thematic narrative was the next most used statistical tool, and regression and Chi-square analysis appeared in only four studies. As outlined by Trochim (2006), descriptive statistics are used to describe the essential features of the data in a survey by porviding simple summaries about the sample and the measures. Using simple graphical analysis, they form the basis of virtually every quantitative analysis of data (Trochim, 2006). Perhaps, it is the most prevalent in health and medical research because it addresses visible and practical scientific evidence regarding what is or what the data shows or represents. However, relying only on descriptive statistics eliminates important relationships within mobile learning and even further makes it difficult in predicting the occurrence of other variable based on their predictors (Miles, Huberman, & Saldaña, 2014).
The significant determinants of m-learning in medical education implementation revealed were infrastructure and faculty limitations; competing priorities; budget constraints; system functionality and relevance; sustainability; socio-economic issues (affordability); promotion of health literacy; ethical issues related to privacy; emotional support and belongingness; knowledge and attitude; and mobile technology literacy and technical issues. Some of the studies reported on efforts to satisfy these factors or overcome the constraints. Gupta, Marsden, Oluka, Sharma, and Lucas (2017) suggested that to overcome issues of sub-optimal Internet connectivity (IT infrastructure) and network performance, intervention models like fair usage policy; network monitoring; effective enforcement of access regulations; and the use of user authentication systems should be adopted. For these interventions to succeed, there must be international solidarity and public/private partnerships to tackle the problems of shortage as recommended by Chenault (2011).
Recognising the critical role of evaluation on the success of policy implementation and change management across contexts, the WHO Global Observatory (2011)  this has a substantial effect on the acceptance of the technology and vice versa (Adebara et al., 2017;Ibrahim et al., 2014;Obi et al., 2018). However, the attitude of medical students and physicians or doctors is influenced by some specific factors such as; awareness, skills, training, and availability of the relevant infrastructure (Adebara et al., 2017;Ibrahim et al., 2014;Obi et al., 2018). With this, the success or otherwise of m-learning in medical education implementation in SSA depends on how to overcome the myriads of problems identified by the studies in the review.
Despite the challenges to the integration of mobile learning in health education in SSA, authors such as Saharan Africa, to do a comprehensive evaluation of the existing system. A thorough analysis of the research on ICT and technology-based solutions will inform decisions that will galvanise effort towards improved access and health care delivery at all levels (especially in resource-constrained communities).
In furtherance of these objectives, Masika et al. (2015) recommends the public-private-partnership model, which considers the scarcity of resources in most SSA countries.

Recommendations for Future Research
Based on the findings of the review the following recommendations are proposed: 1. Given that our study revealed simple descriptive statistics as the most used analysis, we suggest that future studies focus on studies in m-learning adoption and acceptance that may reveal the use of advanced statistical tools in medical and healthcare education research in SSA.
2. There is a need for more research in medical education and practice primarily in the use of mobile technologies as only nine countries out of 48 African sub-Sahara nations met the inclusion criteria. Also, future studies should address the limitation of our study by extending the scope to include other African countries and beyond.

Governments across the sub-Sahara region and institutional leadership in Higher Education
Institutions should realign their priorities for human capital development by allocating more funds for critical infrastructural provision (that addresses poor bandwidth, sub-optimal Internet access etc.) and training and retraining of the workforce to foster the growth and development of m-learning in medical education and practice in the region. (2014)'s postulated baseline requirements for the adoption of any form of e-Learning technology are worth noting here, and include:

Oyo and Kalema
• The formation and funding of a hub for the coordination of the e-learning platform (be it m-learning in medical education, or in any other technology-mediated platform); • Development of contents, curricula, and program accreditation for m-learning; • The content delivery mechanism for m-learning in ODL; • Provision of access to computers, the Internet, and mobile technologies and applications; and • Adequate funding of mobile learning projects.

Conclusion
This paper reviewed articles on mobile learning research trends and challenges to its implementation in medical education in SSA between 2010 to 2018. It identified the research designs, subjects, sample size, and instruments and statistical tools employed for the analysis in the studies. Additionally, the paper focused on the factors determining m-learning adoption and associated barriers to medical education in SSA. The study provided an overview of the state of the literature in that domain and sought to guide the direction of future studies, policy, and practice. Mtebe and Raphael (2018) opined that the proliferation of mobile phones in SSA and the emergence of fibre optical marine cable and adoption of various electronic learning systems would continue to increase, making the need for effective strategies to leverage these technologies very critical. However, we conclude this discussion anchoring on Traxler (2009)'s view on mobile learning: "mobile learning is uniquely suited to support context-specific and immediate learning, it is a major opportunity for distance learning since mobile technologies can situate learners and connect learners" (p.18). Mobile learning is indeed an opportunity for addressing the challenges of educational, social, political, and economic development of all resource constrained economies in general and of SSA in particular.

Limitations
1. The review concentrated on articles written in the English language on mobile learning, mobile health, and Telemedicine based on research in medical education and practice in SSA without considering studies done in other African languages.
2. The review focused on m-learning without considering other aspects of technology intervention in Healthcare systems in Africa.