Artificial intelligence for strengthening healthcare systems in low- and middle-income countries: a systematic scoping review.

In low- and middle-income countries (LMICs), AI has been promoted as a potential means of strengthening healthcare systems by a growing number of publications. We aimed to evaluate the scope and nature of AI technologies in the specific context of LMICs. In this systematic scoping review, we used a broad variety of AI and healthcare search terms. Our literature search included records published between 1st January 2009 and 30th September 2021 from the Scopus, EMBASE, MEDLINE, Global Health and APA PsycInfo databases, and grey literature from a Google Scholar search. We included studies that reported a quantitative and/or qualitative evaluation of a real-world application of AI in an LMIC health context. A total of 10 references evaluating the application of AI in an LMIC were included. Applications varied widely, including: clinical decision support systems, treatment planning and triage assistants and health chatbots. Only half of the papers reported which algorithms and datasets were used in order to train the AI. A number of challenges of using AI tools were reported, including issues with reliability, mixed impacts on workflows, poor user friendliness and lack of adeptness with local contexts. Many barriers exists that prevent the successful development and adoption of well-performing, context-specific AI tools, such as limited data availability, trust and evidence of cost-effectiveness in LMICs. Additional evaluations of the use of AI in healthcare in LMICs are needed in order to identify their effectiveness and reliability in real-world settings and to generate understanding for best practices for future implementations.

The effect of a short, animated story-based video on COVID-19 vaccine hesitancy: A study protocol for an online randomized controlled trial.

Introduction: Exposure to a high volume of vaccine misinformation on social media can have a negative effect on vaccine confidence and rates. To counteract misinformation, we designed a collage of three short, animated story-based (SAS) videos to convey scientifically informed and accessible information about COVID-19 vaccine applicable to a social media context. Methods and analysis: We will conduct an online randomized controlled trial primarily to: (1) determine the effectiveness of SAS videos in improving COVID-19 vaccine knowledge; (2) evaluate the effectiveness of SAS videos in increasing behavioral intent for COVID-19 vaccination; and (3) quantify people's interest in watching SAS videos about the COVID-19 vaccine. We also aim to identify barriers and facilitators to COIVD-19 vaccinations that have been shown to minimize vaccine hesitancy between vaccinated and unvaccinated populations. Using a web-based recruitment platform, a total of 10,000 adults from the United States will be recruited and randomly assigned to (1) a SAS video collage arm, (2) an attention placebo control video arm, or (3) no intervention arm (1:1:1). Furthermore, we will measure behavioral intent to obtain information on vaccination regarding COVID-19. At the end of the trial, participants randomized to arm 2 and arm 3 will be given the option of watching one of the intervention videos voluntarily to assess participant engagement with SAS videos. Finally, we will assess individual factors associated with vaccine hesitancy - hope, optimism, COVID-19 perceived risks and benefits, self-efficacy, perceived social norms, and trust - and compare vaccinated and unvaccinated participants across the three arms. Discussions: Evidence-based information from official channels can be complex and inaccessible to the general public, whereas false information on social media is frequently shared in brief postings, images, or videos that can easily reach the general public, thereby rapidly disseminating (mis-)information. To avoid the spread of misinformation, social media may be used to deliver evidence-based and emotionally compelling information in a readily accessible format in order to pre-empt misinformation. Our findings may help inform future SAS efforts addressing COVID-19 and other important public health challenges. Ethics and dissemination: The study was approved by the Heidelberg University Hospital's Ethics Committee (S-163/2022). The trial was registered with German Clinical Trials Register (www.drks.de) on 5 January 2022: number DRKS00027938. Findings of the study will be published in peer-reviewed scientific publications and possibly presented at scientific conferences.

Telemedicine in Low- and Middle-Income Countries During the COVID-19 Pandemic: A Scoping Review.

Background: COVID-19 has impacted the capacity of healthcare systems worldwide, particularly in low- and middle-income countries (LMICs), which are already under strain due to population growth and insufficient resources. Since the COVID-19 pandemic's emergence, there has been an urgent need for a rapid and adequate reaction to the pandemic's disruption of healthcare systems. To this end, telemedicine has been shown in prior research to be a feasible approach. The overarching objective of this scoping review was to determine the extent and acceptance of telemedicine in healthcare in low- and middle-income countries (LMICs) during the COVID-19 pandemic. Methods: This scoping review followed PRISMA guidelines and Arksey and O'Malley's five-stage framework to identify available evidence. We systematically searched four academic databases for peer-reviewed literature published between January 2020 and April 2021: Medline, PubMed, Web of Science, and Scopus, as well as Google Scholar as a source for grey literature. Results: The search identified 54 articles with 45,843 participants, including 6,966 healthcare professionals and 36,877 healthcare users. We identified a range of reasons for introducing telemedicine in LMICs during COVID-19, most notably to maintain non-emergency healthcare, enhance access to healthcare providers, and reduce the risk of infection among health users and providers. Overall, healthcare providers and users have shown a high level of acceptance for telemedicine services. During the COVID-19 pandemic, telemedicine provided access to healthcare in the majority of included articles. Nonetheless, some challenges to accepting telemedicine as a method of healthcare delivery have been reported, including technological, regulatory, and economical challenges. Conclusion: Telemedicine was found to improve access to high-quality healthcare and decrease infection risk in LMICs during COVID-19. In general, infrastructure and regulatory barriers found to be the most significant barriers to wider telemedicine use, and should be considered when implementing telemedicine more broadly. There appears to be a need to prioritize patient data safety, as many healthcare practitioners utilized commercial apps and services as telemedicine systems. Additionally, it appears as though there is a need to increase capacity, skill, and transparency, as well as to educate patients about telemedicine.

Real-life Evaluation of an Interactive Versus Noninteractive e-Learning Module on Chronic Obstructive Pulmonary Disease for Medical Licentiate Students in Zambia: Web-Based, Mixed Methods Randomized Controlled Trial.

BACKGROUND: e-Learning for health professionals in many low- and middle-income countries (LMICs) is still in its infancy, but with the advent of COVID-19, a significant expansion of digital learning has occurred. Asynchronous e-learning can be grouped into interactive (user-influenceable content) and noninteractive (static material) e-learning. Studies conducted in high-income countries suggest that interactive e-learning is more effective than noninteractive e-learning in increasing learner satisfaction and knowledge; however, there is a gap in our understanding of whether this also holds true in LMICs. OBJECTIVE: This study aims to validate the hypothesis above in a resource-constrained and real-life setting to understand e-learning quality and delivery by comparing interactive and noninteractive e-learning user satisfaction, usability, and knowledge gain in a new medical university in Zambia. METHODS: We conducted a web-based, mixed methods randomized controlled trial at the Levy Mwanawasa Medical University (LMMU) in Lusaka, Zambia, between April and July 2021. We recruited medical licentiate students (second, third, and fourth study years) via email. Participants were randomized to undergo asynchronous e-learning with an interactive or noninteractive module for chronic obstructive pulmonary disease and informally blinded to their group allocation. The interactive module included interactive interfaces, quizzes, and a virtual patient, whereas the noninteractive module consisted of PowerPoint slides. Both modules covered the same content scope. The primary outcome was learner satisfaction. The secondary outcomes were usability, short- and long-term knowledge gain, and barriers to e-learning. The mixed methods study followed an explanatory sequential design in which rating conferences delivered further insights into quantitative findings, which were evaluated through web-based questionnaires. RESULTS: Initially, 94 participants were enrolled in the study, of whom 41 (44%; 18 intervention participants and 23 control participants) remained in the study and were analyzed. There were no significant differences in satisfaction (intervention: median 33.5, first quartile 31.3, second quartile 35; control: median 33, first quartile 30, second quartile 37.5; P=.66), usability, or knowledge gain between the intervention and control groups. Challenges in accessing both e-learning modules led to many dropouts. Qualitative data suggested that the content of the interactive module was more challenging to access because of technical difficulties and individual factors (eg, limited experience with interactive e-learning). CONCLUSIONS: We did not observe an increase in user satisfaction with interactive e-learning. However, this finding may not be generalizable to other low-resource settings because the post hoc power was low, and the e-learning system at LMMU has not yet reached its full potential. Consequently, technical and individual barriers to accessing e-learning may have affected the results, mainly because the interactive module was considered more difficult to access and use. Nevertheless, qualitative data showed high motivation and interest in e-learning. Future studies should minimize technical barriers to e-learning to further evaluate interactive e-learning in LMICs.

The Impact of Wearable Technologies in Health Research: Scoping Review.

BACKGROUND: Wearable devices hold great promise, particularly for data generation for cutting-edge health research, and their demand has risen substantially in recent years. However, there is a shortage of aggregated insights into how wearables have been used in health research. OBJECTIVE: In this review, we aim to broadly overview and categorize the current research conducted with affordable wearable devices for health research. METHODS: We performed a scoping review to understand the use of affordable, consumer-grade wearables for health research from a population health perspective using the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) framework. A total of 7499 articles were found in 4 medical databases (PubMed, Ovid, Web of Science, and CINAHL). Studies were eligible if they used noninvasive wearables: worn on the wrist, arm, hip, and chest; measured vital signs; and analyzed the collected data quantitatively. We excluded studies that did not use wearables for outcome assessment and prototype studies, devices that cost >€500 (US $570), or obtrusive smart clothing. RESULTS: We included 179 studies using 189 wearable devices covering 10,835,733 participants. Most studies were observational (128/179, 71.5%), conducted in 2020 (56/179, 31.3%) and in North America (94/179, 52.5%), and 93% (10,104,217/10,835,733) of the participants were part of global health studies. The most popular wearables were fitness trackers (86/189, 45.5%) and accelerometer wearables, which primarily measure movement (49/189, 25.9%). Typical measurements included steps (95/179, 53.1%), heart rate (HR; 55/179, 30.7%), and sleep duration (51/179, 28.5%). Other devices measured blood pressure (3/179, 1.7%), skin temperature (3/179, 1.7%), oximetry (3/179, 1.7%), or respiratory rate (2/179, 1.1%). The wearables were mostly worn on the wrist (138/189, 73%) and cost <€200 (US $228; 120/189, 63.5%). The aims and approaches of all 179 studies revealed six prominent uses for wearables, comprising correlations-wearable and other physiological data (40/179, 22.3%), method evaluations (with subgroups; 40/179, 22.3%), population-based research (31/179, 17.3%), experimental outcome assessment (30/179, 16.8%), prognostic forecasting (28/179, 15.6%), and explorative analysis of big data sets (10/179, 5.6%). The most frequent strengths of affordable wearables were validation, accuracy, and clinical certification (104/179, 58.1%). CONCLUSIONS: Wearables showed an increasingly diverse field of application such as COVID-19 prediction, fertility tracking, heat-related illness, drug effects, and psychological interventions; they also included underrepresented populations, such as individuals with rare diseases. There is a lack of research on wearable devices in low-resource contexts. Fueled by the COVID-19 pandemic, we see a shift toward more large-sized, web-based studies where wearables increased insights into the developing pandemic, including forecasting models and the effects of the pandemic. Some studies have indicated that big data extracted from wearables may potentially transform the understanding of population health dynamics and the ability to forecast health trends.