Search

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.

Non-communicable diseases (NCDs) have been on the rise in low- and middle-income countries (LMICs) over the last few decades and represent a significant healthcare concern. Over 85% of "premature" deaths worldwide due to NCDs occur in the LMICs. NCDs are an economic burden on these countries, increasing their healthcare expenditure. However, targeting NCDs in LMICs is challenging due to evolving health systems and an emphasis on acute illness. The major issues include limitations with universal health coverage, regulations, funding, distribution and availability of the healthcare workforce, and availability of health data. Experts from across the health sector in LMICs formed a Think Tank to understand and examine the issues, and to offer potential opportunities that may address the rising burden of NCDs in these countries. This review presents the evidence and posits pragmatic solutions to combat NCDs.

Background Mobile Health (mHealth) is becoming an important tool to improve health outcomes in maternal, newborn and child health (MNCH). Studies of mHealth interventions, have demonstrated their effectiveness in improving uptake of recommended maternal services such as antenatal visits. However, evidence of impact on maternal health outcomes is still limited. Methods A pseudo-randomized controlled trial (single blind) was conducted to assess the impact of a voice-message based maternal intervention on maternal health knowledge, attitudes, practices and outcomes over time: Pregnancy (baseline/Time 1); Post-partum (Time 2) and when the infant turned one year old (Time 3). Women assigned to the mMitra intervention arm received gestational age- and stage-based educational voice messages via mobile phone in Hindi and Marathi, while those assigned to the control group did not. Both groups received standard care. Results Two thousand sixteen women were enrolled. Interviews were conducted with 1516 women in the intervention group and 500 women in the control group at baseline and post-partum. The intervention group performed significantly better than controls on four maternal health practice indicators: receiving the tetanus toxoid injection (OR: 1.6, 95% Confidence Interval (CI): 1.05–2.4, p = 0.028), consulting a doctor if spotting or bleeding (OR: 1.72, 95%CI: 1.07–2.75, p = 0.025), saving money for delivery expenses (OR: 1.79, 95%CI: 1.38–2.33, p = 0.0001), and delivering in hospital (OR: 2.5, 95%CI: 1.49–4.35, p = 0.001). The control group performed significantly better than the intervention group on two practice indicators: resting regularly during pregnancy (OR: 0.7, 95%CI: 0.54–0.88, p = 0.002) and having at-home deliveries attended by a skilled birth attendant (OR: 0.46, 95%CI: 0.23–0.91, p = 0.027). Both groups’ knowledge improved from Time 1 to Time 2. Only one knowledge indicator, on seeking medical care during pregnancy, was statistically increased in the intervention group compared to controls. Anemia status at or near the time of delivery was unable to be assessed due to missing data from maternal health cards. Conclusions This study provides evidence that in low-resource settings, mobile voice messages providing tailored and timed information about pregnancy can positively impact maternal health care practices proven to improve maternal health outcomes. Additional research is needed to assess whether voice messaging can motivate behavior change better than text messaging, particularly in low literacy settings. Trial registration The mMitra impact evaluation is registered with ISRCTN under Registration # 88968111, assigned on 6 September 2018 (See https://www.isrctn.com/ISRCTN88968111).

Digital technological tools offer the opportunity to design and disseminate targeted information to influence health behaviors and outcomes. However, extended engagement with health information is vital to promote sustained behavior change in health consumers. Our content-led mobile application for children's health, Saathealth, deployed in a low-resource setting, was used to develop and run machine learning algorithms to build health recommender systems using content and collaborative filtering techniques. We aimed to explore changes in engagement associated with the recommendation systems on the Saathealth app by assessing various aspects of user engagement: videos watched on the app, time spent on the app, sessions on the app, and correct quiz responses. We conducted two A/B experiments to compare the effect of (i) content filtering with no recommendations, and (ii) content filtering with collaborative filtering, on content consumption on the Saathealth app. In experiment 1, the content filtering recommender system was associated with a 25.00% higher median number of videos watched per user compared to when no recommendations were provided after 45 days (5.00 [interquartile range (IQR), 2.00–12.00] vs. 4.00 [IQR, 2.00–10.50], respectively). Content filtering also led to 53.80% more complete video watches and 13.96% higher proportions of correct quiz responses. When the content filtering recommender system was compared with the collaborative filtering one in experiment 2, users in the collaborative filtering arm watched 66.67% more videos, both at 45 days and 90 days. At 90 days, the median number of videos watched per user was 5.00 (IQR, 2.00–9.25) in the collaborative filtering arm and 3.00 (IQR, 2.00–6.00) in the content filtering arm. Collaborative filtering also led to 15.01% more time spent on the app and 59.05% higher complete video watches. We found that machine learning-driven health recommender systems may be effective tools to sustain user engagement with health content. These tools have the potential to address various global health challenges by improving health awareness and behaviors in low-resource settings.

Objectives mHealth interventions for MNCH have been shown to improve uptake of antenatal and neonatal services in low- and middle-income countries (LMICs). However, little systematic analysis is available about their impact on infant health outcomes, such as reducing low birth weight or malnutrition among children under the age of five. The objective of this study is to determine if an age- and stage-based mobile phone voice messaging initiative for women, during pregnancy and up to 1 year after delivery, can reduce low birth weight and child malnutrition and improve women’s infant care knowledge and practices. Methods We conducted a pseudo-randomized controlled trial among pregnant women from urban slums and low-income areas in Mumbai, India. Pregnant women, 18 years and older, speaking Hindi or Marathi were enrolled and assigned to receive mMitra messages (intervention group N = 1516) or not (Control group N = 500). Women in the intervention group received mMitra voice messages two times per week throughout their pregnancy and until their infant turned 1 year of age. Infant’s birth weight, anthropometric data at 1 year of age, and status of immunization were obtained from Maternal Child Health (MCH) cards to assess impact on primary infant health outcomes. Women’s infant health care practices and knowledge were assessed through interviews administered immediately after women enrolled in the study (Time 1), after they delivered their babies (Time 2), and after their babies turned 1 year old (Time 3). 15 infant care practices self-reported by women (Time 3) and knowledge on ten infant care topics (Time 2) were also compared between intervention and control arms. Results We observed a trend for increased odds of a baby being born at or above the ideal birth weight of 2.5 kg in the intervention group compared to controls (odds ratio (OR) 1.334, 95% confidence interval (CI) 0.983–1.839, p = 0.064). The intervention group performed significantly better on two infant care practice indicators: giving the infant supplementary feeding at 6 months of age (OR 1.4, 95% CI 1.08–1.82, p = 0.009) and fully immunizing the infant as prescribed under the Government of India’s child immunization program (OR 1.531, 95% CI 1.141–2.055, p = 0.005). Women in the intervention group had increased odds of knowing that the baby should be given solid food by 6 months (OR 1.89, 95% CI 1.371–2.605, p 2.5 kg (OR 2.279, 95% CI 1.617–3.213, p Conclusions for Practice This study provides robust evidence that tailored mobile voice messages can significantly improve infant care practices and maternal knowledge that can positively impact infant child health. Furthermore, this is the first prospective study of a voice-based mHealth intervention to demonstrate a positive impact on infant birth weight, a health outcome of public health importance in many LMICs.

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.

The Indian health care system lacks the infrastructure to meet the health care demands of the country. Physician and nurse availability is 30 and 50% below WHO recommendations, respectively, and has led to a steep imbalance between the demand for health care and the infrastructure available to support it. Among other concerns, India still struggles with challenges like undernutrition, with 38% of children under the age of five being underweight. Despite these challenges, technological advancements, mobile phone ubiquity and rising patient awareness offers a huge opportunity for artificial intelligence to enable efficient healthcare delivery, by improved targeting of constrained resources. The Saathealth mobile app provides low-middle income parents of young children nflwith interactive children’s health, nutrition and development content in the form of an entertaining video series, a gamified quiz journey and targeted notifications. The app iteratively evolves the user journey based on dynamic data and predictive algorithms, empowering a shift from reactive to proactive care. Saathealth users have registered over 500,000 sessions and over 200 million seconds on-app engagement over a year, comparing favorably with engagement on other digital health interventions in underserved communities. We have used valuable app analytics data and insights from our 45,000 users to build scalable, predictive models that were validated for specific use cases. Using the Random Forest model with heterogeneous data allowed us to predict user churn with a 93% accuracy. Predicting user lifetimes on the mobile app for preliminary insights gave us an RMSE of 25.09 days and an R2 value of 0.91, reflecting closely correlated predictions. These predictive algorithms allow us to incentivize users with optimized offers and omni-channel nudges, to increase engagement with content as well as other targeted online and offline behaviors. The algorithms also optimize the effectiveness of our intervention by augmenting personalized experiences and directing limited health resources toward populations that are most resistant to digital first interventions. These and similar AI powered algorithms will allow us to lengthen and deepen the lifetime relationship with our health consumers, making more of them effective, proactive participants in improving children’s health, nutrition and early cognitive development.

Non-communicable diseases (NCDs) have been on the rise in low- and middle-income countries (LMICs) over the last few decades and represent a significant healthcare concern. Over 85% of “premature” deaths worldwide due to NCDs occur in the LMICs. NCDs are an economic burden on these countries, increasing their healthcare expenditure. However, targeting NCDs in LMICs is challenging due to evolving health systems and an emphasis on acute illness. The major issues include limitations with universal health coverage, regulations, funding, distribution and availability of the healthcare workforce, and availability of health data. Experts from across the health sector in LMICs formed a Think Tank to understand and examine the issues, and to offer potential opportunities that may address the rising burden of NCDs in these countries. This review presents the evidence and posits pragmatic solutions to combat NCDs., Video abstract Point your SmartPhone at the code above. If you have a QR code reader the video abstract will appear. Or use: https://youtu.be/eB0JOLAYTPs

Additional file 1 Supplement 1. Time 1 (Baseline) Survey. Questionnaire FIRST INTERVIEW: Impact Evaluation of mMitra.

Mobile health (mHealth) offers new opportunities to improve access to health services and health information. It also presents new challenges in evaluating its impact, particularly in linking the use of a technology intervention that aims to improve health behaviors with the health outcomes that are impacted by changed behaviors. The availability of data from a multitude of sources (paper-based and electronic) provides the conditions to facilitate making stronger connections between self-reported data and clinical outcomes. This commentary shares lessons and important considerations based on the experience of applying new research frameworks and incorporating maternal and child health records data into a pseudo-randomized controlled trial to evaluate the impact of mMitra, a stage-based voice messaging program to improve maternal, newborn, and child health outcomes in urban slums in India.

This paper reports findings from a study aimed to understand how urban poor pre- and postnatal women in India use ICT tools like mobile phones for information seeking. The study was divided in 2 phases. In phase 1, we conducted semi-structured interviews with 66 pre- and postnatal women from urban poor slums located in 4 cities in India. In phase 2, we conducted a survey with 102 pre- and postnatal women from urban poor slum located in 9 cities in India. The aim was to understand care giving during and after pregnancy, nutrition, terminology used during pregnancy, and interpretation of visual material. This paper summarizes the findings, which include the importance of the use of appropriate terminology to design content in a multilingual country, identifying triggers for information seeking behaviour, and approaches to technology acceptance for information seeking. Based on these findings, we make suggestions for design of future ICT-based interventions in resource constraint settings.

BACKGROUND While mobile technologies have unlocked opportunities to address health care access and affordability gaps, billions of emerging mobile users are yet to reap the benefits of connected health. OBJECTIVE We studied data enabled mobile phone ownership and behaviors amongst low income young Indian families. Our goal was to study the barriers to usage in this rapidly growing segment and how these barriers can be overcome to design meaningful health interventions. METHODS Starting with secondary research, our team conducted primary research interviews in a group of 250 young families (pregnancy or an infant in the family) within six diverse low income communities in Mumbai and Delhi. The surveyed population represents the global segment of the next billion emerging mobile users. The hypotheses generated from our research were used for rapid prototype development, testing and iteration over two rounds with similar users in Mumbai. Prototype development was informed by learnings from commercial apps used by emerging mobile users that address patchy connectivity, low data speeds and cost conscious consumers. The prototype testing included qualitative and quantitative methods. RESULTS Our primary research showed rapidly increasing and recent ownership of internet enabled phones (83%) in our surveyed population. 56% of users had acquired their smartphone in the last year and only 10% had owned a smartphone for more than 2 years. Phone ownership had a strong gender bias with only 39% women owned phones compared with 85% in men. Smartphone ownership was linked to monthly house income, rising from 35% ownership in households earning less than 20,000 INR to 90% in households earning 20,000-40,000 INR. Phone ownership across the same income segments increased dramatically in digitally ready neighborhoods that had a superior infrastructure and connectivity. The main reasons for not using the internet were lack of knowledge (80%), cost (57%), lack of local language content (39%) and lack of motivation (35%). Emerging mobile users were drawn to the internet primarily for entertainment (95%), social networking (82%), and online gaming (74%). Only 18% of the population was using internet to seek health care related information, and when they did, digital health information was reported as least reliable, with close to 90% of respondents ranking them lowest on the reliability scale. Our users were more likely to respond to the test mobile app when it was data light, transparent, and allowed for offline viewing and sharing. Behaviorally, reducing cognitive load significantly allowed users to make confident decisions. This meant creating linear user journeys, minimizing screen choices and nudging users towards specific messages using a combination of rewards and loss aversion techniques. Even though literate, our users responded overwhelmingly to visual and video content than to written text. Finally, activating the user journey on the app was much more powerful when done through a trusted community resource. CONCLUSIONS Slimming down existing health apps won’t address the unique needs of emerging mobile users. Designing connected health interventions for the next billion mobile users requires multi-disciplinary teams to design content experiences and user journeys rooted in local ecosystems and tailored to the users’ unique sensibilities.

Background: While mobile technologies have unlocked opportunities to address health care access and affordability gaps, billions of emerging mobile users are yet to reap the benefits of connected health. Objective: We studied data enabled mobile phone ownership and behaviors amongst low income young Indian families. Our goal was to study the barriers to usage in this rapidly growing segment and how these barriers can be overcome to design meaningful health interventions. Methods: Starting with secondary research, our team conducted primary research interviews in a group of 250 young families (pregnancy or an infant in the family) within six diverse low income communities in Mumbai and Delhi. The surveyed population represents the global segment of the next billion emerging mobile users. The hypotheses generated from our research were used for rapid prototype development, testing and iteration over two rounds with similar users in Mumbai. Prototype development was informed by learnings from commercial apps used by emerging mobile users that address patchy connectivity, low data speeds and cost conscious consumers. The prototype testing included qualitative and quantitative methods. Results: Our primary research showed rapidly increasing and recent ownership of internet enabled phones (83%) in our surveyed population. 56% of users had acquired their smartphone in the last year and only 10% had owned a smartphone for more than 2 years. Phone ownership had a strong gender bias with only 39% women owned phones compared with 85% in men. Smartphone ownership was linked to monthly house income, rising from 35% ownership in households earning less than 20,000 INR to 90% in households earning 20,000-40,000 INR. Phone ownership across the same income segments increased dramatically in digitally ready neighborhoods that had a superior infrastructure and connectivity. The main reasons for not using the internet were lack of knowledge (80%), cost (57%), lack of local language content (39%) and lack of motivation (35%). Emerging mobile users were drawn to the internet primarily for entertainment (95%), social networking (82%), and online gaming (74%). Only 18% of the population was using internet to seek health care related information, and when they did, digital health information was reported as least reliable, with close to 90% of respondents ranking them lowest on the reliability scale. Our users were more likely to respond to the test mobile app when it was data light, transparent, and allowed for offline viewing and sharing. Behaviorally, reducing cognitive load significantly allowed users to make confident decisions. This meant creating linear user journeys, minimizing screen choices and nudging users towards specific messages using a combination of rewards and loss aversion techniques. Even though literate, our users responded overwhelmingly to visual and video content than to written text. Finally, activating the user journey on the app was much more powerful when done through a trusted community resource. Conclusions: Slimming down existing health apps won’t address the unique needs of emerging mobile users. Designing connected health interventions for the next billion mobile users requires multi-disciplinary teams to design content experiences and user journeys rooted in local ecosystems and tailored to the users’ unique sensibilities.

We describe a qualitative user study that we conducted with 64 people living with HIV/AIDS (PLHA) in India recruited from private sector clinics. Our aim was to investigate information gaps, problems, and opportunities for design of relevant technology solutions to support HIV treatment. Our methodology included clinic visits, observations, discussion with doctors and counsellors, contextual interviews with PLHA, diary studies, technology tryouts, and home visits. Analysis identified user statements, observations, breakdowns, insights, and design ideas. We consolidated our findings across users with an affinity. We found that despite several efforts, PLHA have limited access to authentic information. Some know facts and procedures, but lack conceptual understanding of HIV. Challenges include low education, no access to technology, lack of socialisation, less time with doctors and counsellors, high power-distance between PLHA and doctors and counsellors, and information overload. Information solutions based on mobile phones can lead to better communication and improve treatment adherence and effectiveness if they are based on the following: repetition, visualisation, organisation, localisation, and personalisation of information, improved socialisation, and complementing current efforts in clinics.