JMIR Publications

ABSTRACT

Background:

Cardiovascular diseases (CVDs) are increasingly affecting younger populations, particularly in African Americans in the Southern USA. Comorbidities and environmental factors such as trauma, stress, depression, diet, pollution, racism, poverty, and structural violence all combine to exacerbate These factors manifest early in life, and as such monitoring and interventions need to be implemented long before individuals present with symptoms, and they enter the chronic phase of the disease. However, younger populations are notoriously hard to recruit and retain, often because of lack of concern for long term health. For African Americans, low levels of trust in historically discriminatory systems or institutions as well as cultural differences with health providers and/or researchers serve as additional barriers to optimal health.

Objective:

The study aimed to evaluate the design requirements of urban AA adolescents to develop an mHealth framework to reduce CVD risk factors by monitoring nutrition, sleep, physical and mental health.

Methods:

Urban African Americans, ages 18-29 years old participated in user-centered design sessions. This HealthTech sessions were guided a modified version of is the design thinking approach. The first three phases (empathize, define, and ideate) are underpinned by leveraging constructs of behavior change theories, specifically, the Health Belief Model (HBM) and the Social Cognitive Theory of Mass Communication (SCTMC). The images were analyzed using NVivo 12, a qualitative analysis software. Using a grounded theory approach, an open-coding method was applied to a subset of data, approximately 20% or 5 complete prototypes to identify themes. To ensure intercoder reliability, two research team members analyzed the same subset of data.

Results:

An analysis of the emerging design requirement were customization; incentive motivation; social engagement; awareness/education/recommendations; behavior tracking; location services; access to health professionals; data user agreements; and health assessment.

Conclusions:

This led to the design of a cross-platform app to collect standardized health surveys, narratives, geolocated pollution, weather and food desert exposure data, physical activity, social network, and physiology through point of care devices. A HIPAA-compliant cloud infrastructure was developed to collect, process and review data, as well as generate alerts, to allow automated signal processing and machine learning on the data to produce key alerts. Integration with wearables and electronic medical records (via FHIR) was also implemented. The framework we have designed provides a comprehensive health and exposure monitoring system which allows for a broad range of compliance, from passive background monitoring to active self-report. The system is scalable through the cloud infrastructure and extensible. Through an open-source BSD license, the system can be leveraged by entrepreneurs and researchers alike to generate high quality data for predictive health. Clinical Trial: N/A