RESUMO
Populations in resource-limited communities have low health awareness, low financial literacy levels, and inadequate access to primary healthcare, leading to low adoption of preventive health behaviours, low healthcare-seeking behaviours, and poor health outcomes. Healthcare providers have limited reach and insights, limiting their ability to design relevant products for resource limited settings. Our primary preventive health intervention, called the Saathealth family health interventions, is a scaled digital offering that aims to improve knowledge levels on various health topics, nudge positive behaviour changes, and drive improved health outcomes. This case study presents our learnings and best practices in scaling these digital health interventions in resource-limited settings and maximising their impact. We scaled the Saathealth interventions to cumulatively reach >10 million users across India using a multi-pronged approach: (1) ensuring localization and cultural relevance of the health content delivered through user research; (2) disseminating content using omni-channel approaches, which involved using diverse content types and multiple digital platforms; (3) using iterative product features such as gamification and artificial intelligence-based (AI-based) predictive models; (4) using real-time analytics to adapt the user's digital experience by using interactive content to drive them towards products and services and (5) experiments with sustainability models to yield some early successes. The Saathealth family health mobile app had >25,000 downloads and the intervention reached >873,000 users in India every month through the mobile app, Facebook, and Instagram combined, from the time period of February 2022 to January 2023. We repeatedly observed videos and quizzes to be the most popular content types across all digital channels being used. Our AI-based predictive models helped improve user retention and content consumption, contributing to the sustainability of the mobile apps. In addition to reaching a high number of users across India, our scaling strategies contributed to deepened engagement and improved health-seeking behaviour. We hope these strategies help guide the sustainable and impactful scaling of mobile health interventions in other resource-limited settings.
RESUMO
BACKGROUND: Digital phenotyping is the real-time collection of individual-level active and passive data from users in naturalistic and free-living settings via personal digital devices, such as mobile phones and wearable devices. Given the novelty of research in this field, there is heterogeneity in the clinical use cases, types of data collected, modes of data collection, data analysis methods, and outcomes measured. OBJECTIVE: The primary aim of this scoping review was to map the published research on digital phenotyping and to outline study characteristics, data collection and analysis methods, machine learning approaches, and future implications. METHODS: We utilized an a priori approach for the literature search and data extraction and charting process, guided by the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-analyses Extension for Scoping Reviews). We identified relevant studies published in 2020, 2021, and 2022 on PubMed and Google Scholar using search terms related to digital phenotyping. The titles, abstracts, and keywords were screened during the first stage of the screening process, and the second stage involved screening the full texts of the shortlisted articles. We extracted and charted the descriptive characteristics of the final studies, which were countries of origin, study design, clinical areas, active and/or passive data collected, modes of data collection, data analysis approaches, and limitations. RESULTS: A total of 454 articles on PubMed and Google Scholar were identified through search terms associated with digital phenotyping, and 46 articles were deemed eligible for inclusion in this scoping review. Most studies evaluated wearable data and originated from North America. The most dominant study design was observational, followed by randomized trials, and most studies focused on psychiatric disorders, mental health disorders, and neurological diseases. A total of 7 studies used machine learning approaches for data analysis, with random forest, logistic regression, and support vector machines being the most common. CONCLUSIONS: Our review provides foundational as well as application-oriented approaches toward digital phenotyping in health. Future work should focus on more prospective, longitudinal studies that include larger data sets from diverse populations, address privacy and ethical concerns around data collection from consumer technologies, and build "digital phenotypes" to personalize digital health interventions and treatment plans.
