GluCoach: Assist
Project Team
The project was created as an academic consumer health informatics (CHI) research project by a candidate group in the Masters of Science in Health Informatics & Analytics (MSHI&A) at Florida International University: Spencer Ash, Emma Blake, Oares Cabello, Kisha Hendricks, Ledurnis Lantigua, and Rick Morgan. Production assistance was provided by Veronica Fuentes.
The Problem Space
As life expectancy continues to increase, the prevalence of diabetes mellitus is increasing and there has been a noted epidemiologic shift in the disease occurrence towards individuals of greater age as discussed by Abdelhafiz, et al (Abdelhafiz, Chakravorty, Gupta, Haque, & Sinclair, 2014, p. 790). This is a category of inherited and/or acquired chronic diseases which result in either a diminished capacity for the pancreas to produce insulin or in the ineffectiveness of the insulin that is produced to regulate blood sugar.
Projections from the US Centers for Disease Control and Prevention (CDC) indicate that the prevalence of diabetes in the general population will nearly double by 2025 in large part due to the aging population of which approximately 25% of individuals ≥65 years of age currently have diabetes (Kirkman et al., 2012). Within this segment of the population, the incidence of diabetes is also projected to be 4.5-fold between 2005 and 2050 as compared to a 3-fold increase within the total population (Kirkman et. al, 2012). The burden of disease for diabetes is often described in terms of the effects on middle age working adults. However, in older individuals, diabetes is linked to higher mortality, reduced functional status, and additional substantial health burdens, as well as increased costs of health maintenance and increased risks of institutionalization which are compounded by demographic and clinical characteristic differences between these groups (Kirkman et. al, 2012). This significantly adds to the complexity of treatment recommendations for older individuals with diabetes and as noted by Abdelhafiz and Sinclair (2014). Management and care for older individuals with diabetes is further complicated by an increased likelihood for disabilities and additional chronic conditions which hindering activities of daily living, including self-care tasks such as glucose monitoring, changing insulin doses, or diet management (Abdelhafiz and Sinclair, 2014; Kirkman et. al, 2012).
There is a wide variance of abilities in this population from health care consumers who live very independently to those who may be fully dependent upon caregivers 24 hours a day. This non-heterogeneity characteristic of this population leads to challenges when it comes to diabetes care in this population. Additional considerations complicating the care of these patients include other comorbidities such as: geriatric syndromes, cognitive dysfunction, polypharmacy, physical impairments, unique nutritional issues, individual learning needs, and access to emotional, caregiver, and financial support. It is reasonable to expect the prevalence of multiple chronic diseases and other comorbidities in the aging populations to only increase, as supported by numerous studies and consistent with the data published (LeRouge et. al, 2011). It has been shown the incidence of diabetes increases with age (~12% in people aged 65–70 and ~15% in those over 80 years old). This results in older adults acquiring or presenting with new-onset signs and symptoms of diabetes (Abdelhafiz and Sinclair, 2014; Kirkman et. al, 2012) which adds to those who have already been living with this chronic illness. According to a population report published by the United States Census Bureau in 2014, “by 2030, more than 20 percent of U.S. residents are projected to be aged 65 and over, compared with 13 percent in 2010 and 9.8 percent in 1970 (U.S. Census Bureau, 2014).”
The influx of aging individuals will place significant pressure on healthcare systems and the utilization of services and resources. Because of this, we are interested in generating a more thorough understanding of this segment of the population with a particular focus on the subgroup of individuals with long-standing diabetes with a reduced capacity for self-care and poor adherence to a prescribed diabetic care plan. Within this context, we hope to explore the potential for consumer health informatics (CHI) technologies to assist in the proactive management of the healthcare needs of this subsegment of the population, to empower individuals living with diabetes to manage their care, to reduce the overall personal expense related to self-care activities, and to achieve positive outcomes that result in increased quality of life and extended longevity.
Projections from the US Centers for Disease Control and Prevention (CDC) indicate that the prevalence of diabetes in the general population will nearly double by 2025 in large part due to the aging population of which approximately 25% of individuals ≥65 years of age currently have diabetes (Kirkman et al., 2012). Within this segment of the population, the incidence of diabetes is also projected to be 4.5-fold between 2005 and 2050 as compared to a 3-fold increase within the total population (Kirkman et. al, 2012). The burden of disease for diabetes is often described in terms of the effects on middle age working adults. However, in older individuals, diabetes is linked to higher mortality, reduced functional status, and additional substantial health burdens, as well as increased costs of health maintenance and increased risks of institutionalization which are compounded by demographic and clinical characteristic differences between these groups (Kirkman et. al, 2012). This significantly adds to the complexity of treatment recommendations for older individuals with diabetes and as noted by Abdelhafiz and Sinclair (2014). Management and care for older individuals with diabetes is further complicated by an increased likelihood for disabilities and additional chronic conditions which hindering activities of daily living, including self-care tasks such as glucose monitoring, changing insulin doses, or diet management (Abdelhafiz and Sinclair, 2014; Kirkman et. al, 2012).
There is a wide variance of abilities in this population from health care consumers who live very independently to those who may be fully dependent upon caregivers 24 hours a day. This non-heterogeneity characteristic of this population leads to challenges when it comes to diabetes care in this population. Additional considerations complicating the care of these patients include other comorbidities such as: geriatric syndromes, cognitive dysfunction, polypharmacy, physical impairments, unique nutritional issues, individual learning needs, and access to emotional, caregiver, and financial support. It is reasonable to expect the prevalence of multiple chronic diseases and other comorbidities in the aging populations to only increase, as supported by numerous studies and consistent with the data published (LeRouge et. al, 2011). It has been shown the incidence of diabetes increases with age (~12% in people aged 65–70 and ~15% in those over 80 years old). This results in older adults acquiring or presenting with new-onset signs and symptoms of diabetes (Abdelhafiz and Sinclair, 2014; Kirkman et. al, 2012) which adds to those who have already been living with this chronic illness. According to a population report published by the United States Census Bureau in 2014, “by 2030, more than 20 percent of U.S. residents are projected to be aged 65 and over, compared with 13 percent in 2010 and 9.8 percent in 1970 (U.S. Census Bureau, 2014).”
The influx of aging individuals will place significant pressure on healthcare systems and the utilization of services and resources. Because of this, we are interested in generating a more thorough understanding of this segment of the population with a particular focus on the subgroup of individuals with long-standing diabetes with a reduced capacity for self-care and poor adherence to a prescribed diabetic care plan. Within this context, we hope to explore the potential for consumer health informatics (CHI) technologies to assist in the proactive management of the healthcare needs of this subsegment of the population, to empower individuals living with diabetes to manage their care, to reduce the overall personal expense related to self-care activities, and to achieve positive outcomes that result in increased quality of life and extended longevity.
Gap Focus and Method to Address
After conducting an ecosystem analysis of the current market space we were unable to discover any consumer health technologies that satisfactorily met the needs and condition management goals of our targeted subpopulation: older adults age 55–75 with Type-1 diabetes and difficulties with adhering to diabetes protocols. As a result, the project team sought to reimagine diabetes management for the targeted subpopulation and devised the concept for GluCoach Assist, a device agnostic and multi-modal diabetes management solution designed for universal access and enhanced usability specifically for older adults with Type-1 diabetes.
From our analysis of the current consumer health ecosystem, it seems as though the user-centered design has not been adopted to appeal to our target subpopulation at large. Although some technologies have embraced user-centric design principles, none of which that we analyzed appeared to be specifically designed for the consumer within the demographics of older adults age 55–75 with Type 1 diabetes and poor compliance with diabetes management protocols. There are currently many CHT devices, applications, and platforms in the existing market space, however, they appear to be designed broadly for a generalized audience. Within our specific subpopulation of older adults with Type-1 diabetes and challenges with condition management, this could prevent a consumer from embracing CHT applications, if they feel that there is no solution designed specifically for their needs as they must often adapt solutions that utilize devices which are either designed for a broad audience or that are focused on other areas of concern such as Type 2 diabetes.
The GluCoach Assist platform has been intentionally created to be an inclusive diabetes management system supported by advanced machine learning to understand the unique needs of each individual and to provide a highly individualized experience that is accessible across a wide range of devices — from mobile and wearable devices to voice-based platforms such as Amazon Alexa and Google Home. Through GluCoach Assist, we aim to provide comprehensive diabetes wellness coaching to individuals within the targeted subpopulation in ways that fit seamlessly into each person’s lifestyle, regardless of the level of expertise with diabetes condition management or their familiarity with technology. With the GluCoach Assist concept, the overall goal is to facilitate diabetes self-management through personalized coaching, designed to support an individual’s goals and to affect behavior change through an accessible model supported by the latest technological innovations.
The GluCoach Assist platform has been intentionally created to be an inclusive diabetes management system supported by advanced machine learning to understand the unique needs of each individual and to provide a highly individualized experience that is accessible across a wide range of devices — from mobile and wearable devices to voice-based platforms such as Amazon Alexa and Google Home. Through GluCoach Assist, we aim to provide comprehensive diabetes wellness coaching to individuals within the targeted subpopulation in ways that fit seamlessly into each person’s lifestyle, regardless of the level of expertise with diabetes condition management or their familiarity with technology. With the GluCoach Assist concept, the overall goal is to facilitate diabetes self-management through personalized coaching, designed to support an individual’s goals and to affect behavior change through an accessible model supported by the latest technological innovations.
Content and Features
The underlying platform that supports the GluCoach Assist ecosystem is designed primarily to resemble an interactive personal health record (IPHR) and functions in a manner similar to a clinical application for diabetes condition management, diabetes education, and overall health promotion that can be personalized to achieve an individual’s goals for managing their diabetic conditions through a comprehensive intelligent diabetes coach powered by advanced artificial intelligence. The Glucoach Assist ecosystem allows for the individual consumer to utilize the platform in the most accessible way possible by allowing them to select the modality that best fits into their lifestyle, whether it be through a mobile device, a wearable, or even a voice-based platform.
GluCoach Assist is intended to be a single solution to meet all of the diabetes management needs for the individual and the features that are included in the platform consist of:
Glucose Level Monitoring
Blood sugar (glucose) levels can be manually added by the user or automatically synced to the platform from remote glucose monitoring systems. The application is intended to support various independent and standalone glucose monitoring systems as well as connections via. AppleHealth Kit and GoogleFit which can connect directly to the GluCoach Assist system. Connections such as the ones described above provide the user with the ability to seamlessly integrate data from multiple glucose monitoring sources so that the GluCoach system can continuously track and present glucose monitoring information through intuitive visualizations to the consumer without any action required on behalf of the consumer to enter data (ie. no manual data entry required). As a result information from insulin pumps can provide 24/7 updates on basal rates and basal insulin flow seamlessly into the GluCoach platform, which ensures timely interventions and behavioral recommendations in near real time.
Medication Management and Scheduling
With the GluCoach system, consumers are able to receive reminders from their personalized GluCoach when it is time to take their medications. Additionally, medications can be logged as they are taken and insulin pump basal rates can be integrated into these timelines to provide up-to-date metrics that inform coaching recommendations based on the consumer’s behavior.
Meal Planning
Glucoach provides access to one of the world’s largest food and nutrition databases and consumers have the ability to scan barcodes of foods and upload photos of meals. Information from the scanned barcodes will automatically be added to the GluCoach system and photos that are manually added by the user will be analyzed by advanced convolutional neural networks (CNNs) which have the capability to pull nutritional content from the provided images. As a result, there will be no need for the user to manually track their consumption of foods using traditional food logs or to calculate their carbohydrate loads, which may have an effect on their blood glucose levels.
Exercise Routines and Activity Tracking
The GluCoach system automatically tracks the user’s activity in cases where they utilize the platform on their mobile or wearable devices. On these devices, the system has the capability to gather physical activity information from onboard pedometers and accelerometers and can also sync activity information from Apple Health, GoogleFit, FitBit, and other fitness related applications such as Strava.
Automated Decision Support
One of the more unique features of the GluCoach system is its advanced automated decision support capabilities that are supported by the most up to date clinical evidence bases and are able to react in real-time to the information provided to the system by the consumer or their linked services. GluCoach is capable of understanding the individual consumers’ blood glucose levels and can correlate the results to activities such as taking medications, physical activity, as well as diet. As a result, it is able to provide actionable information and suggestions to the consumer in ways that best support their personalized condition management goals and diabetes care plans. The decision support functionality is also capable of estimating future blood glucose levels based on predicted trends (extrapolated from past consumer behavior) and can provide actionable food and activity suggestions directly to the user far in advance of any adverse situations.
Data and Report Sharing
With the GluCoach system, information related to the consumer’s condition, such as glucose readings, activity levels, medication compliance, food intake, and even adverse events can be communicated to the consumer’s healthcare provider, caretakers, and other designated individuals (denoted as authorized delegates) as a mechanism to avoid adverse health events and to ensure compliance with diabetic care protocols as determined by the preferences of the consumer. Of course, the autonomy, privacy, and self-determination of the individual consumer are of the utmost concern and no information will be shared unless otherwise instructed based on the consumer selected preferences — except in the case of the direst circumstances in which emergency notifications may be sent.
Personalized Coaching and Health Goals
With personalized coaching being the core element of the GluCoach platform, the individual consumer would be able to set their health goals and actively work to achieve them with the assistance of the digital GluCoach assistant. The consumer may set their goals or the goals can be predetermined based on any imputed health plan provided by a healthcare practitioner or diabetes care team.
Additionally, the GluCoach assistant is able to answer diabetes and care related questions posed by the user through either a chat or voice interface which provides educational information and personalized support to the consumer. The database utilized by the GluCoach is derived from the same reference base that is available to Certified Diabetes Educators and is also supported by the most up to date clinical research systems and clinical evidence repositories. This capability allows the GluCoach system to provide real-time 24/7 support to the consumer and also allows the consumer to effectively provide for their own self-management of their diabetes without needing to rely on busy healthcare practitioners who often have limited availability to help the consumer. As an example, where a diabetic individual may only see their physician or diabetes management team a few times a year, the GluCoach is capable of providing 24/7 comprehensive monitoring, feedback, and behavioral health coaching to the consumer which fills in the gaps between doctor’s appointments. The personalized coaching offered by the GluCoach Assist system includes:
It is also important to note that diabetes management consists of much more than medications, diabetes protocols, and blood glucose testing. Recognizing this, the GluCoach is also able to affect the lifestyle behaviors of the consumer and to radically change their behavior based on scientifically proven lifestyle and behavioral theory that is prompted based on the consumer’s interactions with the system and the consumer-generated data that is provided through the platform.
GluCoach Assist is intended to be a single solution to meet all of the diabetes management needs for the individual and the features that are included in the platform consist of:
Glucose Level Monitoring
Blood sugar (glucose) levels can be manually added by the user or automatically synced to the platform from remote glucose monitoring systems. The application is intended to support various independent and standalone glucose monitoring systems as well as connections via. AppleHealth Kit and GoogleFit which can connect directly to the GluCoach Assist system. Connections such as the ones described above provide the user with the ability to seamlessly integrate data from multiple glucose monitoring sources so that the GluCoach system can continuously track and present glucose monitoring information through intuitive visualizations to the consumer without any action required on behalf of the consumer to enter data (ie. no manual data entry required). As a result information from insulin pumps can provide 24/7 updates on basal rates and basal insulin flow seamlessly into the GluCoach platform, which ensures timely interventions and behavioral recommendations in near real time.
Medication Management and Scheduling
With the GluCoach system, consumers are able to receive reminders from their personalized GluCoach when it is time to take their medications. Additionally, medications can be logged as they are taken and insulin pump basal rates can be integrated into these timelines to provide up-to-date metrics that inform coaching recommendations based on the consumer’s behavior.
Meal Planning
Glucoach provides access to one of the world’s largest food and nutrition databases and consumers have the ability to scan barcodes of foods and upload photos of meals. Information from the scanned barcodes will automatically be added to the GluCoach system and photos that are manually added by the user will be analyzed by advanced convolutional neural networks (CNNs) which have the capability to pull nutritional content from the provided images. As a result, there will be no need for the user to manually track their consumption of foods using traditional food logs or to calculate their carbohydrate loads, which may have an effect on their blood glucose levels.
Exercise Routines and Activity Tracking
The GluCoach system automatically tracks the user’s activity in cases where they utilize the platform on their mobile or wearable devices. On these devices, the system has the capability to gather physical activity information from onboard pedometers and accelerometers and can also sync activity information from Apple Health, GoogleFit, FitBit, and other fitness related applications such as Strava.
Automated Decision Support
One of the more unique features of the GluCoach system is its advanced automated decision support capabilities that are supported by the most up to date clinical evidence bases and are able to react in real-time to the information provided to the system by the consumer or their linked services. GluCoach is capable of understanding the individual consumers’ blood glucose levels and can correlate the results to activities such as taking medications, physical activity, as well as diet. As a result, it is able to provide actionable information and suggestions to the consumer in ways that best support their personalized condition management goals and diabetes care plans. The decision support functionality is also capable of estimating future blood glucose levels based on predicted trends (extrapolated from past consumer behavior) and can provide actionable food and activity suggestions directly to the user far in advance of any adverse situations.
Data and Report Sharing
With the GluCoach system, information related to the consumer’s condition, such as glucose readings, activity levels, medication compliance, food intake, and even adverse events can be communicated to the consumer’s healthcare provider, caretakers, and other designated individuals (denoted as authorized delegates) as a mechanism to avoid adverse health events and to ensure compliance with diabetic care protocols as determined by the preferences of the consumer. Of course, the autonomy, privacy, and self-determination of the individual consumer are of the utmost concern and no information will be shared unless otherwise instructed based on the consumer selected preferences — except in the case of the direst circumstances in which emergency notifications may be sent.
Personalized Coaching and Health Goals
With personalized coaching being the core element of the GluCoach platform, the individual consumer would be able to set their health goals and actively work to achieve them with the assistance of the digital GluCoach assistant. The consumer may set their goals or the goals can be predetermined based on any imputed health plan provided by a healthcare practitioner or diabetes care team.
Additionally, the GluCoach assistant is able to answer diabetes and care related questions posed by the user through either a chat or voice interface which provides educational information and personalized support to the consumer. The database utilized by the GluCoach is derived from the same reference base that is available to Certified Diabetes Educators and is also supported by the most up to date clinical research systems and clinical evidence repositories. This capability allows the GluCoach system to provide real-time 24/7 support to the consumer and also allows the consumer to effectively provide for their own self-management of their diabetes without needing to rely on busy healthcare practitioners who often have limited availability to help the consumer. As an example, where a diabetic individual may only see their physician or diabetes management team a few times a year, the GluCoach is capable of providing 24/7 comprehensive monitoring, feedback, and behavioral health coaching to the consumer which fills in the gaps between doctor’s appointments. The personalized coaching offered by the GluCoach Assist system includes:
- Medication compliance and medication reminders
- Blood sugar testing and reminders of when to measure blood glucose
- Physical activity tracking and tracking related to personalized health goals (such as glucose control and medication adherence)
- Monitoring carbohydrates and making other healthy diet choices
It is also important to note that diabetes management consists of much more than medications, diabetes protocols, and blood glucose testing. Recognizing this, the GluCoach is also able to affect the lifestyle behaviors of the consumer and to radically change their behavior based on scientifically proven lifestyle and behavioral theory that is prompted based on the consumer’s interactions with the system and the consumer-generated data that is provided through the platform.
Conceptual Designs and Prototypes
Mobile App Concept
View interactive prototype here: Interactive Prototype (https://invis.io/8VRIFPRQPSM#/357755983_Splash)
The mobile component of the GluCoach Assist platform relies on current smartphone technology and is accessible on both iOS (iOS 11.0 or later. Compatible with iPhone, iPad, and iPod touch and Android (version 5.0 and up) devices. Platform features employed by this variation of the system rely on:
- Bluetooth device integrations: The solution allows individuals to connect to blood glucose testing and medication delivery devices which provide the application with information related to constant glucose monitoring, automatic insulin pump data, and digital scales. Note that this functionality is limited only to devices that communicate via. Bluetooth unless other methods of integration are provided by hosted sites and/or applications which provide an application programming interface (API).
- API Integrations: Application Programming Interfaces are utilized to gather information about the user’s activities and data. Supported connections include fitness apps (such as FitBit, GoogleFit, Apple Health, and others), electronic medical records and patient health portals (for incorporating diabetes management plans), and remote monitoring devices (such as connected glucose monitors and insulin delivery systems). Additional examples of API integrations across applications (native and web) include: Dexcom, myfitnesspal, Accu-chek, Dario, Nike, Strava, misfit, Garmin, OneTouch, lark, iHealth, Carrotfit, Withings, among others.
- Device Camera: Access to the mobile devices camera is required for automatic food logging and nutrition calculations (ie. carbohydrate calculations), which are supported by convolutional neural networks (CNNs) built upon Google Vision in conjunction with a nutritional database and supported by the Cloud Vision API.
- Nutritional Database: The nutritional database used by GluCoach is supported by Nutrino Health’s FoodPrint database as well as the nutrition database of the U.S. Department of Agriculture. This allows users to scan food label barcodes, manually input food sources, and/or utilize GluCoach’s CNN for determining nutritional intake which can then be tracked and factored into coaching recommendations.
- Neural Networks and Convolutional Neural Networks (CNNs): The GluCoach Assist platform relies heavily upon artificial intelligence (AI) and machine learning for understanding user behavior, analyzing data trends, and providing personalized coaching recommendations. A specific subset of Convolutional Neural Nets is also used to determine nutrient content from images of foods provided by the user to aid in calculating carbohydrate and other nutrient intakes. The machine learning utilized by GluCoach is dependent on self-learning models which use backpropagation to learn about user behavior and to make adjustments in recommendations provided to the user. The artificial intelligence platform supporting GluCoach is supported by the use of Tensorflow Processing Units (TPUs) which are AI accelerator application-specific integrated circuits (ASIC) developed by Google specifically for neural network based machine learning
- Voice Assistant: The mobile variation of the Glucoach system also features an integrated voice assistant “Diabetes Management Coach,” which in this case is a stand-alone voice assistant (not that the voice assistant is also supported on a variety of voice assistant platforms such as Amazon Echo and Google Home enabled devices. In an attempt to provide universal access to the voice assistant, users may also interact with the voice coach through a text messaging chat function.
- Automated Text Messaging and Email Communications: In addition to in-app notifications and reminders, the platform is also able to generate text notifications (SMS) and email communications to the user, their health care proxies/delegates, and healthcare providers. This form of communication relies on permissions from the user and allows for automated messages to be communicated to the various audiences so that they may be aware of the recommendations being provided to the user, the user’s blood glucose readings, compliance with medication protocols, food intake, and other behavioral indicators such as physical activity levels. Emergency notifications may also be sent through the GluCoach system in cases where medications have been missed or where blood glucose levels are indicated to be at dangerous levels.
Wearable Concept
The GluCoach system is also available on wearable devices supported by WearOS, Apple Watch, and other platforms which allow 3rd party applications, such as FitBit. In this particular modality, the wearable GluCoach system must be paired with the mobile application and provides the functionality of viewing blood glucose, activity, and food intake data. Depending on the technological specifications of the device, the wearable application also supports voice interactions with the GluCoach voice assistant (eg. Apple Watch and WearOS). No manual input is supported through this modality aside from interactions associated through the voice assistant coach.
Voice Concept
View a video of live interaction with the voice prototype (on Google Assistant) here: Video of live interaction with GluCoach prototype (https://drive.google.com/a/fiu.edu/file/d/1d4-pgKd5u8aFubE8_lKPRSM-L5ewMTnI/view?usp=sharing)
One of the core functionalities of GluCoach Assist is the voice assistant coach, which has been engineered as a standalone service integrated into the mobile and wearable variations of the platform as well as an independent service provided on both the Google Assistant and Amazon Echo platforms (both of which contribute and pull information from the central GluCoach application system through API connections). With the voice assistant coach (the “GluCoach”), users are able to interact with the GluCoach assistant and it becomes possible to interact with the GluCoach system without any manual input (eg. data inputting or manual field entry) from the user. The GluCoach voice assistant is hosted in the cloud and is supported either independently by GluCoach or on the associated services such as Amazon’s Echo platform as a standalone skill or Google’s voice platform as an independent voice agent (an example of an interaction flow with the GluCoach Agent on the Google Voice platform can be seen here: http://bit.ly/2U9AxnS). The voice assistant coach’s infrastructure is built upon the Google Cloud platform which allows for the ability to scale to millions of users while also leveraging Google’s machine learning engines and products such as Google Cloud Speech-to-Text and Google’s machine Vision. Software development kits (SDKs) are also used to enable to a voice assistant to engage with users wherever they are and across all devices, whether they are users are on-the-go or at home, through wearables, phones, cars, speakers and other smart devices.
Data Collection and Sharing
In terms of data collection, the GluCoach system relies significantly on automated data sources (such as glucose meters and insulin pumps) as well as user-generated and/or provided data. Data that is shared will be limited to text (SMS), email communications, and other avenues of sharing as established in information sharing protocols through services such as the GluCoach voice platforms, EHR integration, and/or patient portal, PHR (including IPHR), or other agreements.
In terms of data sharing, no information provided by the user or captured by automated processes will be shared without direct and explicit permissions from the user. Any data shared will be provided either to the directed health delegates or designated healthcare provider, diabetes management team, diabetes counselor, or healthcare organization/system. Information that is shared with entities other than the individual will be limited to behavioral patterns (medication and insulin compliance), nutritional intake (carbohydrate loads), physical activity, blood glucose, and A1C results. Primarily, consumer data will only be shared with these entities in accordance with management plan agreements or in cases where abnormal or dangerous activities or interpretive scores (eg. elevated blood glucose levels) are indicated to require interventions from external sources.
- Demographic Data: The GluCoach relies on personal information that users provide when establishing their GluCoach profiles either through manual methods or as data permissions granted when establishing a profile through their Google profile accounts. The type of information utilized by GluCoach in terms of demographic information is primarily used to generate coaching recommendations and may include: the consumer’s name, gender, sex, date of birth, address, phone number, and email address.
- User information generated from connected devices, applications, and clinical systems: The GluCoach system will obtain information from any connected source, which includes connected devices and applications as well as informational platforms such as patient portals and electronic health systems. The types of data gathered from these sources may include glucose readings, insulin delivery metrics and volumes, diabetes management plans, clinical lab results, medication records, physical activities, sleep patterns, and quality, nutritional intake and other data deemed essential for managing diabetes and/or associated medical conditions.
- Patient Reported Data: The GluCoach system may obtain data directly from the individual consumer from manual or voice inputs in the form of self-reported glucose and A1C results, nutritional intake, physical activity, demographic information, medications, diabetes management plans, sleep patterns and quality, and voice inputs (used to enhance natural language processing capabilities).
In terms of data sharing, no information provided by the user or captured by automated processes will be shared without direct and explicit permissions from the user. Any data shared will be provided either to the directed health delegates or designated healthcare provider, diabetes management team, diabetes counselor, or healthcare organization/system. Information that is shared with entities other than the individual will be limited to behavioral patterns (medication and insulin compliance), nutritional intake (carbohydrate loads), physical activity, blood glucose, and A1C results. Primarily, consumer data will only be shared with these entities in accordance with management plan agreements or in cases where abnormal or dangerous activities or interpretive scores (eg. elevated blood glucose levels) are indicated to require interventions from external sources.
Evidence Review
Johns Hopkins University researcher Ahmed Hassoon made headlines this year when he published his success in changing the behavior of overweight and obese cancer survivors utilizing patient coaching via SMS texting and a voice-controlled artificial intelligent application hosted by Amazon’s Echo device (more commonly referred to as, Alexa). Hassoon reflected on the significance of his findings in a magazine published by Johns Hopkins, “It’s a very cost-effective means of reaching out and delivering personalized advice and monitoring compared to conventional ways,” he says. “We think it saves tens of thousands of dollars without compromising the personal connection to patients” (Kimmel in the Community, 2019, p. 13). In our investigation, we reached out to Dr. Hassoon to try to determine if his research may be translational and predict parallel success for GluCoach.
We discovered Dr. Hassoon’s research team has been working on developing his own application designed as a “personal coach” application for diabetes patients. He has been working on developing a model since 2017 which would be compliant with the Americans with Disabilities Act (1990), as well as regulations pertaining to patient privacy and data security. Upon reviewing his PATH study design for the cancer patients, it looks like some of the privacy and security regulations were avoided by using a secured server operated by Johns Hopkins University (Hassoon et al., 2018). In an email, Dr. Hassoon offered the following statement regarding the work to develop an application or platform hosted by devices, such as Amazon’s Echo:
We discovered Dr. Hassoon’s research team has been working on developing his own application designed as a “personal coach” application for diabetes patients. He has been working on developing a model since 2017 which would be compliant with the Americans with Disabilities Act (1990), as well as regulations pertaining to patient privacy and data security. Upon reviewing his PATH study design for the cancer patients, it looks like some of the privacy and security regulations were avoided by using a secured server operated by Johns Hopkins University (Hassoon et al., 2018). In an email, Dr. Hassoon offered the following statement regarding the work to develop an application or platform hosted by devices, such as Amazon’s Echo:
There are some issues that you need to be aware off. But the top one is patient privacy and security of data. Issues related to reliability of the algorithm and the continue maintenance and improvement of the voice design… I would not focus on the technology at all. The technology is the easiest part. The hardest part is legal, privacy, conversation design, data infrastructure and cost, hacks, compliancy, and patient’s engagement. These things take long time to resolve and usually cost the most. The technology part is the cheapest part.
(A. Hassoon, personal communication, April 11, 2019)
Adult learning allows individuals to cope with the changes in health, namely the decreased physical abilities, increased medical conditions, and the associated changes in lifestyle and social conditions (Hill & Ziegahn, 2010). Modern andragogy is primarily based on the work of Dr. Malcolm Knowles and centered on self-directed learning. Knowles is best known for establishing 4 Principles of Andragogy and 5 Assumptions of the Adult Learner illustrated in Appendix A (Knowles, 1984). These are important because other adult learning theories will build upon these concepts. We also have noticed that in our target population, these learning principles and assumptions seem to be most relevant to the adoption of Consumer Health Technology among Baby Boomers, yet the designs of these applications with consumer health education seem to be based in pedagogy and not andragogy. Designers should consider creating around the user’s desire to learn to create greater adoption of the Baby Boomer consumer.
Although meta-analysis and scholarly systematic reviews of evidence exist in this field, many are studying data collected from small sample sets. Few investigations study our target population. No consumer health technology is designed and on the market that is comparable to the level of service provided by GluCoach.
It is apparent that the Consumer Health Technology field is not utilizing designs centered for older adults. Gao (2017) recognized there is a “failure” in the consumer health technology design to meet the needs of older, diabetic patients. As the Baby Boomers continue to age, they will continue to increase demands on already overburdened healthcare delivery systems. This is a huge opportunity for developers to create technology to assist in the health and wellness of Baby Boomers. There is a very real economic value that would translate into healthcare savings if the technology is developed to keep this population healthier, and thus, decreasing demands on healthcare delivery systems.
Unfortunately, only a handful of mobile diabetes management designed for older adults exists in published literature. The Health Data Science Lab of the University of Waterloo (Kim & Lee, 2017) found upon reviewing 51 research articles reviewing mobile health apps, only 3 studies targeted older adults. Researchers at the University of North Carolina (Chung, Griffin, Selezneva, & Gotz, 2018) studied 309 apps, but they only noted 1 was focused on older adults. Arnhold (2014) reviewed 656 mobile applications for diabetics. His team analyzed the usability and found it to be “moderate to good” for apps offering a narrow range of functions but “considerably worse” for apps offering more functions. This means the quality of user experience for Baby Boomers decreases, as the application offers more features. This is a design flaw and evidence of lack of involvement of the user in the design of these applications.
Although older adults are a population that may benefit greatly from the health consumer technology, it is apparent there is not enough research and design being invested in meeting the needs of this population. After reviewing the current published literature, our target population seems to be most concerned with cost, security, and privacy. Application design needs to incorporate adult learning theory to be successful, but they seem to fail to capture motivations of the Baby Boomer population set. Current evidence points to Baby Boomers adopting this technology if they can see a direct correlation with being enabled to age at home. The generational cohort is one that has not followed previous generations, and thus, it is due the most thorough of investigations. Their identity has constantly evolved and changed through their lifespan.
Baby Boomers are more likely to use an application if it is easy to use and individualized to aesthetically appeal to their age. Application designers should take caution to not design applications that feel too clinical or favor functionality over pleasing aesthetics.There has not been a consumer health technology landmark doctrine to prove that the use of mobile applications will improve lab values or other health indicator measures. The literature is often reporting small statistical significances in reported findings. This suggests that more scientific research should be performed with greater reach to continue to capture meaningful data. However, there does seem to be some qualitative consensus, consumer health technology applications seem to improve the user’s perceived comfort and self-confidence. Most of the authors in the published literature also provide conclusions or discussion that show support consumer health technology has the potential to be beneficial in the patient’s management of diabetes.
Most exciting for the GluCoach is the literature that exists representing increased accuracies and reduced calculation errors through the utilization of diabetes management applications using cameras/photographic input (Domhardt et al., 2015)(Zhang, Yu, Siddiquie, Divakaran, & Sawhney, 2015)(Anthimopoulos et al., 2015). Researchers at the National Institute for Health Innovation (Gemming, Utter, & Mhurchu, 2015) in New Zealand pointed to the benefit of the food photography leading to increased patient reported dietary with records of the actual food intake. This is exciting because GluCoach is designed to interface with mobile device cameras for both nutritional calculations and food logging.
With the current market not providing any product to meet the needs and desires of the Baby Boomer with Type 1 diabetes, GluCoach is primed to go to successfully go to the market and help increase compliance and patient adherence to self-management. The key to our success is in our multi-faceted design that is purposeful to be easy to use and appeal to varying preferences for learning and usability of the Baby Boomer consumer.
Although meta-analysis and scholarly systematic reviews of evidence exist in this field, many are studying data collected from small sample sets. Few investigations study our target population. No consumer health technology is designed and on the market that is comparable to the level of service provided by GluCoach.
It is apparent that the Consumer Health Technology field is not utilizing designs centered for older adults. Gao (2017) recognized there is a “failure” in the consumer health technology design to meet the needs of older, diabetic patients. As the Baby Boomers continue to age, they will continue to increase demands on already overburdened healthcare delivery systems. This is a huge opportunity for developers to create technology to assist in the health and wellness of Baby Boomers. There is a very real economic value that would translate into healthcare savings if the technology is developed to keep this population healthier, and thus, decreasing demands on healthcare delivery systems.
Unfortunately, only a handful of mobile diabetes management designed for older adults exists in published literature. The Health Data Science Lab of the University of Waterloo (Kim & Lee, 2017) found upon reviewing 51 research articles reviewing mobile health apps, only 3 studies targeted older adults. Researchers at the University of North Carolina (Chung, Griffin, Selezneva, & Gotz, 2018) studied 309 apps, but they only noted 1 was focused on older adults. Arnhold (2014) reviewed 656 mobile applications for diabetics. His team analyzed the usability and found it to be “moderate to good” for apps offering a narrow range of functions but “considerably worse” for apps offering more functions. This means the quality of user experience for Baby Boomers decreases, as the application offers more features. This is a design flaw and evidence of lack of involvement of the user in the design of these applications.
Although older adults are a population that may benefit greatly from the health consumer technology, it is apparent there is not enough research and design being invested in meeting the needs of this population. After reviewing the current published literature, our target population seems to be most concerned with cost, security, and privacy. Application design needs to incorporate adult learning theory to be successful, but they seem to fail to capture motivations of the Baby Boomer population set. Current evidence points to Baby Boomers adopting this technology if they can see a direct correlation with being enabled to age at home. The generational cohort is one that has not followed previous generations, and thus, it is due the most thorough of investigations. Their identity has constantly evolved and changed through their lifespan.
Baby Boomers are more likely to use an application if it is easy to use and individualized to aesthetically appeal to their age. Application designers should take caution to not design applications that feel too clinical or favor functionality over pleasing aesthetics.There has not been a consumer health technology landmark doctrine to prove that the use of mobile applications will improve lab values or other health indicator measures. The literature is often reporting small statistical significances in reported findings. This suggests that more scientific research should be performed with greater reach to continue to capture meaningful data. However, there does seem to be some qualitative consensus, consumer health technology applications seem to improve the user’s perceived comfort and self-confidence. Most of the authors in the published literature also provide conclusions or discussion that show support consumer health technology has the potential to be beneficial in the patient’s management of diabetes.
Most exciting for the GluCoach is the literature that exists representing increased accuracies and reduced calculation errors through the utilization of diabetes management applications using cameras/photographic input (Domhardt et al., 2015)(Zhang, Yu, Siddiquie, Divakaran, & Sawhney, 2015)(Anthimopoulos et al., 2015). Researchers at the National Institute for Health Innovation (Gemming, Utter, & Mhurchu, 2015) in New Zealand pointed to the benefit of the food photography leading to increased patient reported dietary with records of the actual food intake. This is exciting because GluCoach is designed to interface with mobile device cameras for both nutritional calculations and food logging.
With the current market not providing any product to meet the needs and desires of the Baby Boomer with Type 1 diabetes, GluCoach is primed to go to successfully go to the market and help increase compliance and patient adherence to self-management. The key to our success is in our multi-faceted design that is purposeful to be easy to use and appeal to varying preferences for learning and usability of the Baby Boomer consumer.
Care Continuum
GluCoach Assist has been designed through a user-centered design process which considered the consumer as a central element to the platform’s functionality. With this in mind, the design considers a diversity of lifestyles within the targeted subpopulation to aid in improving their health. Like any new device or technology, the consumer, in this case, older adults with type-1 diabetes, needs to be engaged and willing to interact with the services included within the GluCoach Assist platform. As stated earlier, this is a single solution designed to meet and adapt to the needs of this particular population. This is accomplished by catering to different levels of technology familiarity and consumer needs as supported by the evidence presented in the various components of our research series. Arguably, this technological solution would best fit the younger segment of the targeted sub-population, given they have more exposure to technology and to which, adaptability to a new device comes easier. GluCoach Assist will be a seamless transition for them since they already depend on their smartphones to connect to emails and social media as well as various other forms of digital interactions. Build in medication reminders can alleviate the challenges faced by diabetics regardless of age, by staying connected with the GluCoach the individual is taking charge of their health.
Working individuals within this subpopulation and those with a busy schedule could potentially be the best target population for GluCoach Assist. Aside from timely reminders, the program also offers historical tracking and report creation and other services, such as meal planning. By eliminating the guesswork of counting carbohydrates while trying to eat healthier and within a budget. Making healthier choices will also be easier when dining out with friends and family with the support of the GluCoach Assistant/Health Coach. Tracking activities will also assist in making lifestyle changes to improve diabetes management with continuous use and dedication.
Those that are just entering into retirement age may face the challenge of learning new technology, making it less appealing, which has informed our design decisions to make the system nearly universal when it comes to technological and interaction preferences. Within the retired population, we also see those that aged with technology and could consider the benefit of having helpful information that is just a click, tap, or conversation away. However, since most older adults own a smartphone, adding GluCoach into their digital interactions be seen as an added benefit. Learning how to use it and seeing the benefits will depend on them. In the end, being able to have easy to read glucose levels and a list of medications at your disposal can decrease the chances of errors as shown in studies conducted by the healthcare design agency Worrell (Engaging Patients with the Power of Voice). Keeping track of activities will be easier as the program measures all activities regardless of level, making it a suitable device to have on vacations and daily routine. Setting health goals is one of the many ways GluCoach can improve overall health for this population.
Note that it is important to keep in mind that the GluCoach Assist platform does not replace the care provided by physicians or diabetes management teams by any means. However, it could potentially improve a patient’s overall health by making them aware of blood sugar levels and how to improve them with diet and activity, as well as providing critical interactions outside of clinical encounters. The aging population of older adults can also incorporate GluCoach Assist into their daily routine and stay compliant with their medications and having their levels available on their smartphone to share with physicians or nurses.
Healthcare providers also play an important role in the use of the GluCoach system since it keeps the patient accountable for health maintenance and potential improvement. Not only do physicians can have access to data collected (if this option is selected by the consumer), but they can also set health goals for the patient such as: eating better, working on lowering blood levels, or goals focused on staying compliant with medications. Family members can also be involved in the care of those using the device, by being keeping track of data collected and making sure medications are not taken. Additionally, family members and caretakers may be able to help to set up the system and establish individual goals for the individual consumer, where once it is running it gives peace of mind to those individuals by tracking and keeping the important details related to the diabetes care of their loved ones.
Working individuals within this subpopulation and those with a busy schedule could potentially be the best target population for GluCoach Assist. Aside from timely reminders, the program also offers historical tracking and report creation and other services, such as meal planning. By eliminating the guesswork of counting carbohydrates while trying to eat healthier and within a budget. Making healthier choices will also be easier when dining out with friends and family with the support of the GluCoach Assistant/Health Coach. Tracking activities will also assist in making lifestyle changes to improve diabetes management with continuous use and dedication.
Those that are just entering into retirement age may face the challenge of learning new technology, making it less appealing, which has informed our design decisions to make the system nearly universal when it comes to technological and interaction preferences. Within the retired population, we also see those that aged with technology and could consider the benefit of having helpful information that is just a click, tap, or conversation away. However, since most older adults own a smartphone, adding GluCoach into their digital interactions be seen as an added benefit. Learning how to use it and seeing the benefits will depend on them. In the end, being able to have easy to read glucose levels and a list of medications at your disposal can decrease the chances of errors as shown in studies conducted by the healthcare design agency Worrell (Engaging Patients with the Power of Voice). Keeping track of activities will be easier as the program measures all activities regardless of level, making it a suitable device to have on vacations and daily routine. Setting health goals is one of the many ways GluCoach can improve overall health for this population.
Note that it is important to keep in mind that the GluCoach Assist platform does not replace the care provided by physicians or diabetes management teams by any means. However, it could potentially improve a patient’s overall health by making them aware of blood sugar levels and how to improve them with diet and activity, as well as providing critical interactions outside of clinical encounters. The aging population of older adults can also incorporate GluCoach Assist into their daily routine and stay compliant with their medications and having their levels available on their smartphone to share with physicians or nurses.
Healthcare providers also play an important role in the use of the GluCoach system since it keeps the patient accountable for health maintenance and potential improvement. Not only do physicians can have access to data collected (if this option is selected by the consumer), but they can also set health goals for the patient such as: eating better, working on lowering blood levels, or goals focused on staying compliant with medications. Family members can also be involved in the care of those using the device, by being keeping track of data collected and making sure medications are not taken. Additionally, family members and caretakers may be able to help to set up the system and establish individual goals for the individual consumer, where once it is running it gives peace of mind to those individuals by tracking and keeping the important details related to the diabetes care of their loved ones.
Policy, Legal Aspects, and Standard
In the development of this tool, we are aware of possible liability exposure related to the use and development. As such we have also considered the possible need for insurance products such as Commercial General Liability Insurance (CGL), Technology Errors and Omissions Coverage (tech E&O) and cyber, and D&O to provide defense and indemnity protection against possible claims.
There is a Digital Health Innovation Action Plan generated by the FDA to safeguard high-quality, safe and effective digital health products to the consumer. According to the FDA “This plan lays out the Center for Devices and Radiological Health’s (CDRH) vision for fostering digital health innovation while continuing to protect and promote the public health.”
To ensure we are in compliance with FDA policies and standards we carefully examined and followed the guidance proved by the FDA. The21stCentury Cures Act (12/13/2016) amended the definition of “device” in the Food, Drug and Cosmetic Act to exclude certain software functions. GluCoach is a Non-device function so it is currently not regulated by the FDA. Therefore GluCoach is not subjected to FDA rules and regulation.
The team prudently studied the Resource Guide for Implementing the Health Insurance Portability and Accountability Act (HIPAA) Security Rule document (2008) presented by the U.S. Department of Commerce. Particularly in section 2.1. and 2.1.1. on page 6 of the document. GluCoach does not fall under the covered entities handling EPHI and as such would not be subjected to HIPAA’s security rules. GluCoach will not be collecting any data created by or for any covered entity. HIPAA offers no protection to data voluntarily disclosed by a consumer to entities outside the health system. Additionally, self-help tools such as GluCoach does not involve a medical profession or insurer hence we are outside the scope of HIPAA.
GluCoach has a very stringent privacy policy which is very detailed and is specified above on pages 18 and 19. We may ask the user to provide some personally identifiable information such as; name, date of birth and emails. This information will be retained by GluCoach and will not be sold or supplied to any other party. Our privacy terms and conditions will be periodically updated and users will be notified each time this is done.
The team, in concepting the GluCoach Assist platform is aware of the possibilities of constraints with the use of any technological tool, especially for an elderly population. We expect that members of this subpopulation may experience some challenges due to the potential natural declines in cognitive function related to normal aging process along with thoughts or attitude related to acceptance and lack of education towards some CHI tools. It was therefore important that the possible constraints were identified in the UCD process and the solutions to these constraints were considered in the overall design of the system
There is a Digital Health Innovation Action Plan generated by the FDA to safeguard high-quality, safe and effective digital health products to the consumer. According to the FDA “This plan lays out the Center for Devices and Radiological Health’s (CDRH) vision for fostering digital health innovation while continuing to protect and promote the public health.”
To ensure we are in compliance with FDA policies and standards we carefully examined and followed the guidance proved by the FDA. The21stCentury Cures Act (12/13/2016) amended the definition of “device” in the Food, Drug and Cosmetic Act to exclude certain software functions. GluCoach is a Non-device function so it is currently not regulated by the FDA. Therefore GluCoach is not subjected to FDA rules and regulation.
The team prudently studied the Resource Guide for Implementing the Health Insurance Portability and Accountability Act (HIPAA) Security Rule document (2008) presented by the U.S. Department of Commerce. Particularly in section 2.1. and 2.1.1. on page 6 of the document. GluCoach does not fall under the covered entities handling EPHI and as such would not be subjected to HIPAA’s security rules. GluCoach will not be collecting any data created by or for any covered entity. HIPAA offers no protection to data voluntarily disclosed by a consumer to entities outside the health system. Additionally, self-help tools such as GluCoach does not involve a medical profession or insurer hence we are outside the scope of HIPAA.
GluCoach has a very stringent privacy policy which is very detailed and is specified above on pages 18 and 19. We may ask the user to provide some personally identifiable information such as; name, date of birth and emails. This information will be retained by GluCoach and will not be sold or supplied to any other party. Our privacy terms and conditions will be periodically updated and users will be notified each time this is done.
The team, in concepting the GluCoach Assist platform is aware of the possibilities of constraints with the use of any technological tool, especially for an elderly population. We expect that members of this subpopulation may experience some challenges due to the potential natural declines in cognitive function related to normal aging process along with thoughts or attitude related to acceptance and lack of education towards some CHI tools. It was therefore important that the possible constraints were identified in the UCD process and the solutions to these constraints were considered in the overall design of the system
Adoption and Awareness
According to Calvin K.L, there are several factors involved in promoting and marketing consumer health informatics technology to patients and healthcare consumers. These include patient, human-technology interaction, organizational, environmental, social, and task factors and they determine how our sub-population of older adults can accept and adapt to our new technology in order to manage their diabetic condition. Studies such as Calvin, 2009 have shown that acceptance of new technology defined in four ways including satisfaction with the technology, use or adaption of the technology, efficient or effective use of technology, and intention and willingness to use the technology (Calvin K.L. Or, 2009). As an example, promoting awareness about a new glucose monitoring device called GluCoach Assist to our sub-population (older adults ages from 55–75) with Type 1 diabetes can be challenging, but we must find the way to do so in order to provide an effective solution. Social factors also play a crucial role when promoting the prototype GluCoach Assist glucose device in diabetes management for the targeted population. For our sub-population, the GulCoach Assist platform can be of benefit as a direct result of its usefulness, usability, and rapid performance. GluCoach Assist technology is the solution to the needs of this population and provides for the opportunity to empower the individuals within this population through the self-management of their disease. Also, organizational factors including support, justice, satisfaction with training about the new device, and technical support can be a great source when promoting the new device, and at the same time, it would improve acceptance.
Environmental concerns are also a factor when promoting new products. These considerations occur when a new product is designed to use in certain conditions or areas including in the home environment or elsewhere. Calvin describes the environmental factors that play an essential role because they will facilitate an individual’s ability to use technology effectively and efficiently, resulting in the acceptance of the product. The product design must meet all the environmental and human factors needs of the patient so that they can access the device anywhere. Social factors have a prominent predictive in technology acceptance when offering this particular subpopulation, a new product such as the GluCoach Assist platform and associated services (Calvin K.L. Or, 2009).
There are some reasonable barriers to using Consumer Health Informatics technology among older adults. The basic functionality of information sharing must keep up with patient expectations. The usability of health technology information can also impact how much knowledge our sub-population gains when using GluCoach Assist. Further, issues are lack of information about the technology including apps and websites, lack of motivation, and the existence of computer problems. Wetter describes the importance of how website functionality can play an essential role when creating a PHR (Wetter, 2016). GluCoach Assist is carefully designed to fit all the requirements of diabetic people’s experience. Most devices offer only a glucose monitoring option, however, GluCoach Assist device is user-friendly and has several links that to our subpopulations lifestyle including glucose level monitoring, vital sign monitoring, medication management, and scheduling, a meal planning automated decision support, data and report sharing, and personalized diabetes management coaching that incorporates health goal settings. Implementing all these specifications would eliminate many of the barriers encountered by this specific sub-population and increases the acceptance of our technology within this group.
GluCoach Assist technology has been specifically created for our subpopulation, and it is the friendliest device that does not require much-advanced training. One of the most important embedded features is the voice assistant that can be essential for older adults with diabetes condition. This feature can control an entire application system including guides for device, glucose reading results, medication schedule, appointments and sending email messages just using the voice control
Environmental concerns are also a factor when promoting new products. These considerations occur when a new product is designed to use in certain conditions or areas including in the home environment or elsewhere. Calvin describes the environmental factors that play an essential role because they will facilitate an individual’s ability to use technology effectively and efficiently, resulting in the acceptance of the product. The product design must meet all the environmental and human factors needs of the patient so that they can access the device anywhere. Social factors have a prominent predictive in technology acceptance when offering this particular subpopulation, a new product such as the GluCoach Assist platform and associated services (Calvin K.L. Or, 2009).
There are some reasonable barriers to using Consumer Health Informatics technology among older adults. The basic functionality of information sharing must keep up with patient expectations. The usability of health technology information can also impact how much knowledge our sub-population gains when using GluCoach Assist. Further, issues are lack of information about the technology including apps and websites, lack of motivation, and the existence of computer problems. Wetter describes the importance of how website functionality can play an essential role when creating a PHR (Wetter, 2016). GluCoach Assist is carefully designed to fit all the requirements of diabetic people’s experience. Most devices offer only a glucose monitoring option, however, GluCoach Assist device is user-friendly and has several links that to our subpopulations lifestyle including glucose level monitoring, vital sign monitoring, medication management, and scheduling, a meal planning automated decision support, data and report sharing, and personalized diabetes management coaching that incorporates health goal settings. Implementing all these specifications would eliminate many of the barriers encountered by this specific sub-population and increases the acceptance of our technology within this group.
GluCoach Assist technology has been specifically created for our subpopulation, and it is the friendliest device that does not require much-advanced training. One of the most important embedded features is the voice assistant that can be essential for older adults with diabetes condition. This feature can control an entire application system including guides for device, glucose reading results, medication schedule, appointments and sending email messages just using the voice control
Evaluation
The successful implementation of this concept will be evidenced by satisfaction among diabetic individuals within the targeted population. This operation can be achieved through constant monitoring and evaluation of their bodies and the associated data provided by the system. From the literature presented in the design concept, it is evident that the system provides many of the answers to patients about their condition to avoid fear and uncertainty to the individuals. Also, it is vital in the management of one’s diet and reminding patients of their appointments with their physicians. This system also is compliant with many healthcare standards that ensure the health of patients is both safe and private.
Patients who have Type 2 Diabetes that have confidence that the GluCoach assist will work, will be interviewed and from there, we can find out the outcomes of this concept design. According to research, most of the practitioners who use the systems to monitor their patients took 6 to 12 months to achieve their targets (Hirsch, 2004). For this application, the first six months will be considered for the first evaluation. To achieve targets, the set strategies include examining the uptake of the application from the application store and examining experiences from other patients concerning these systems. These are necessary for the identification of the areas to improve on (Kvedar et al., 2014) Also another key effective strategy is ensuring that the system is cost-effective in that, it favors the patients in terms of cost.
Patients who have Type 2 Diabetes that have confidence that the GluCoach assist will work, will be interviewed and from there, we can find out the outcomes of this concept design. According to research, most of the practitioners who use the systems to monitor their patients took 6 to 12 months to achieve their targets (Hirsch, 2004). For this application, the first six months will be considered for the first evaluation. To achieve targets, the set strategies include examining the uptake of the application from the application store and examining experiences from other patients concerning these systems. These are necessary for the identification of the areas to improve on (Kvedar et al., 2014) Also another key effective strategy is ensuring that the system is cost-effective in that, it favors the patients in terms of cost.
Disclosure
Notice: upon completion of this academic project, the project team became aware of a 2017 Alexa Diabetes Challenge submission called My GluCoach, HCL America, Inc. No copyright or commercial infringement intended.