Behavioral intervention for weight loss
Nudging healthy behavior change
Build a mobile, weight loss tool that will intervene on and nudge middle-aged women with lower income to pursue healthy eating and moving behavior
APP-Me, a mobile digital tool that reminds users of cues pertinent to eating and sedentary behavior and motivates them to achieve their goals
APP-Me is currently being tested in a federally-funded clinical trial — on top of a traditional coach-based model of weight loss support that is operated within safety-net clinical sites
Obesity rates are high among middle-aged adults generally and among poorer and minority adults particularly. Behavioral interventions for weight loss have been less effective in lower income women. These poorer outcomes may in part be related to these women having more frequent exposures to social and physical situations that are obesogenic, i.e., eating and sedentary cues or situations. Working with obese, lower income black and white women, Addressing People and Place Microenvironments (APP-Me) was designed to create awareness of self-behavior at times and places of frequent eating and sedentary behavior.
Middle-aged women (35-64yr) from lower income group
How it works
APP-Me is an automated short messaging system (SMS) that pushes awareness messages to its users at the right place and time to bring participants’ attention back to healthy eating and physical activity. Awareness messages are text, image, audio, or a combination, and are delivered to users’ mobile devices with the intent of creating awareness at the times and places of frequent eating or sedentary behavior. Understanding and building personalized behavioral pattern for users helps achieve this.
Personalized behavioral patterns are initially built using a mobile ecological momentary assessment (EMA) system that is capable of performing experience sampling. EMA question formats include eating, drinking, social interaction, physical activity, and location questions. EMA assessments occur in the moment and reduce recall bias. To achieve maximum timely awareness and context learning, APP-Me also includes an onboarding process where users set approximate times appropriate to messages. This process includes users setting approximate times when they woke up, ate, slept, and wished to not be disturbed. Onboarding information is used to guide or narrow the time window during which EMA messages can be sent. User responses to EMA messages over 4 weeks are used to further narrow the time windows during which appropriate awareness messages can be sent to users.
Awareness messaging begins after a month-long EMA assessment. APP-Me includes pushing personalized, automated awareness messages that are required to be likely personal, in order to bring users’ attention to healthy behavior change. Hence, APP-Me allows users to create messages. APP-Me also includes a message library built by our team with help from sample of target users, health coaches, and healthcare providers. Messages are coded by broad themes (e.g., activity, food, neutral) and subthemes (e.g., gardening, inspiration, grocery shopping, eating out, walking, family, religion, celebrities).
APP-Me includes a user profile where users’ set or modify preferences for themes and subthemes. Users’ rating of messages for their selected themes or subthemes, when received, are used to refine further, appropriate messages that can be pushed for each user.
In sum, novel aspects of APP-Me include the involvement of a low income population, the use of data on time and place of obesogenic behavior, and message delivery time-tailored to an individual’s behavioral patterns.
Semi-structured interviews and affinity diagram
Ideation with game play
Co-design of messages and interface
Think-aloud evaluation in controlled settings
Evaluation in the field
Response with low cost and effort is possible when user responds with selection as opposed to typing. Thus, APP-Me supports a form of communication where users respond by selecting a response option
3-4 weeks of experience sampling with a response rate of over 50% for a user is all that is needed before sending awareness messages. However, right place and time can change based on seasons. Hence, there is a need for repetitive experience sampling at regular intervals (e.g., once in 3 months)
We learnt that social connection can significantly impact choices. A user being held accountable for signing up for a goal and being congratulated for achieving the goal by a loved one can go a long way. Users suggested the need for messages created by a their social connection
User feedback, such as “I like or dislike this message” is an absolute necessity for system to learn and refine context-sensitive messaging further. There is however a need for detailed feedback. E.g., a user can dislike a message if the message arrived at the right time, but possibly at incorrect location
Strategized and led product design research methods
Mentored and collaborated with design team to design probes and prototypes
Compiled and handed-off product requirements and prototypes to developers