This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on http://www.researchprotocols.org, as well as this copyright and license information must be included.
Mobile and wearable technology have been shown to be effective in improving diabetes self-management; however, integrating data from these technologies into clinical diabetes care to facilitate behavioral goal monitoring has not been explored.
The objective of this paper is to report on a study protocol for a pragmatic multi-site trial along with the intervention components, including the detailed connected health interface. This interface was developed to integrate patient self-monitoring data collected from a wearable fitness tracker and its companion smartphone app to an electronic health record system for diabetes self-management education and support (DSMES) to facilitate behavioral goal monitoring.
A 3-month multi-site pragmatic clinical trial was conducted with eligible patients with diabetes mellitus from DSMES programs. The Chronicle Diabetes system is currently freely available to diabetes educators through American Diabetes Association–recognized DSMES programs to set patient nutrition and physical activity goals. To integrate the goal-setting and self-monitoring intervention into the DSMES process, a connected interface in the Chronicle Diabetes system was developed. With the connected interface, patient self-monitoring information collected from smartphones and wearable fitness trackers can facilitate educators’ monitoring of patients’ adherence to their goals. Feasibility outcomes of the 3-month trial included hemoglobin A1c levels, weight, and the usability of the connected system.
An interface designed to connect data from a wearable fitness tracker with a companion smartphone app for nutrition and physical activity self-monitoring into a diabetes education electronic health record system was successfully developed to enable diabetes educators to facilitate goal setting and monitoring. A total of 60 eligible patients with type 2 diabetes mellitus were randomized into either group 1) standard diabetes education or 2) standard education enhanced with the connected system. Data collection for the 3-month pragmatic trial is completed. Data analysis is in progress.
If results of the pragmatic multi-site clinical trial show preliminary efficacy and usability of the connected system, a large-scale implementation trial will be conducted.
ClinicalTrials.gov NCT02664233; https://clinicaltrials.gov/ct2/show/NCT02664233 (Archived by WebCite at http://www.webcitation.org/6yDEwXHo5)
Obesity and type 2 diabetes mellitus (T2DM) are serious chronic illnesses in the US. Compared to a standard diabetes education, behavioral lifestyle interventions were found to be more effective in weight loss and diabetes control among overweight or obese patients with T2DM [
Research supports diabetes education as a cost-effective way to coordinate diabetes care. Diabetes self-management education and support (DSMES) programs located throughout the US are integrated within the existing health care system. Thus, DSMES programs present an ideal setting for testing the implementation of an evidence-based self-monitoring intervention. Chronicle Diabetes is a Web-based electronic health record (EHR) system designed to facilitate behavioral goal monitoring for the American Diabetes Association (ADA)–recognized DSMES Programs [
We used the Chronicle Diabetes system currently freely available to diabetes educators through ADA–recognized DSMES programs to set patients’ diet and physical activity goals and to integrate patients’ self-monitoring information collected from smartphones and wearable fitness trackers to improve educators’ ability to monitor patients' adherence to goals. Moreover, using a central location for tracking patients’ behaviors and progress enables long-term self-management support for sustained behavior change. A system with good usability should foremost have a functionality design that matches the work domain and a user interface that supports efficient task performance by the users. We used a usability framework developed to ensure high usability of connected health systems to guide the development of our proposed connected system [
Our study developed such an interface and tested its usability, acceptability, and feasibility in a multi-site randomized clinical trial. The objective of this paper is to report on a study protocol for a pragmatic multi-site trial and the intervention components, including the detailed connected health interfaces.
Eligibility was assessed prior to participant enrollment. To participate in the study, individuals had to have been diagnosed with T2DM, be 18 years or older, own a smartphone compatible with the Jawbone UP24 fitness tracker, and be overweight or obese as classified by their body mass index >25kg/m2. A list available on the Jawbone website was used to determine whether participants’ smartphones were compatible with the Jawbone fitness tracker. Research assistants also searched the smartphones’ respective app stores to determine whether the companion app to the Jawbone UP24,
We enrolled 60 patients for the study: 30 in Houston and 30 in Pittsburgh. In a national study testing the behavioral lifestyle intervention in a nontranslational setting [
Individuals were recruited from ADA–recognized DSMES programs in Houston, Texas and Pittsburgh, Pennsylvania. Patients were asked if they had a smartphone and whether they would be interested in participating in a research study in which they would monitor their diet and physical activity using a fitness tracker and smartphone app. Those who expressed interest were provided with more detailed information about the study and screened for eligibility. Informed consent was provided by all eligible patients. This study was approved by the Institutional Review Boards of The University of Texas Health Science Center at Houston and the University of Pittsburgh.
Eligible patients were randomly assigned to the intervention program or a standard diabetes education program in a 1:1 allocation ratio. The study statistician created a randomization sheet to randomly assign the patients at the time of enrollment after written informed consent was obtained.
The recruiting sites all offer ADA–recognized diabetes education programs. During the study, patients in the standard diabetes education group saw their diabetes educators at baseline and for the follow-up data collection visit at 3 months. The patients’ interaction with their diabetes educators included setting and modifying patients’ goals related to nutrition, physical activity, risk prevention, self-monitoring of blood glucose, and medication based on their self-report of their progress. Additional visits could be scheduled as usual care based on patients’ conditions. These visits were recorded as confounding factors that would indicate any treatment or patient condition changes during the study period.
Participants in the connected group received standard diabetes education as described above. In addition, participants randomly assigned to this group were exposed to the following intervention components and procedures:
The Jawbone UP24 smartphone app,
According to the national standards for DSMES, diabetes educators are expected to establish and track patients’ behavioral goals [
A partial snapshot of the Chronicle Diabetes system with goal setting function.
While educators found Chronicle Diabetes useful for setting and tracking behavioral goals, one of the major barriers they identified was the lack of a feature to attach a food diary for educators to evaluate patients’ adherence to the prescribed goals. Thus, in this study, we developed a connected interface in the Chronicle Diabetes System to connect data from the Jawbone UP24 fitness tracker and its companion smartphone app for patient diet and physical activity data. Within each patient’s documentation record in Chronicle Diabetes, two new sections (tabs) of the record were added: “Self Monitoring–Nutrition” and “Self Monitoring–Activity.” These two tabs were used by educators to navigate to two pages showing self-monitored diet and activity information. When educators click the “Self Monitoring–Nutrition” tab in the navigation panel for a patient for the first time, a link to connect to the patient’s account will appear. In the beginning of the study, educators needed to click a link in this page to connect the patient’s Jawbone account to their account in the Chronicle Diabetes system. The link took the educator into a page to select what they wanted to connect to from a list of the devices or apps that were supported by Validic, an intermediary platform that provides connection to data from a wide variety of devices and apps. In our study, educators were instructed to select Jawbone and enter their account information to complete the connection process. The link can also be sent to participants via a previously generated email template. The educator can encourage the patient to complete the connection process during the first visit after enrolling in the study. Once the connection process is complete, a monthly calendar view will appear in the “Self Monitoring–Nutrition” page with cells for calories, carbs, saturated and unsaturated fat, fiber, and protein for each day (See
The “Self Monitoring–Activity” page has the same monthly calendar view as the “Self Monitoring–Nutrition” page once connected: calories burned, steps, and planned exercise duration for each day (
Partial Screenshot of Monthly Calendar View of Diet Monitoring in Chronicle Diabetes.
Partial Screenshot of Monthly Calendar View of Activity Monitoring in Chronicle Diabetes.
We used Validic as an intermediary platform to connect the Jawbone UP24 diet and physical activity data to the Chronicle Diabetes system. We adopted this approach rather than directly connecting mobile data and Chronicle Diabetes owing to 1) the flexibility of connecting to additional fitness tracking systems in the future, 2) robust screening of data and high-level security through Validic, and 3) Validic’s experience working with large EHR systems. We believe the incorporation of Validic not only enables a seamless transition for national dissemination of our developed connected interface but also enables its future integration with EHRs seamlessly to make connected technology available to any health care system using an EHR system.
The security features are based on the premise that any data at rest or in motion must be encrypted and unreadable to outsiders. User profile, clinical data, and progress stored in the local SQL lite database on the smartphones are encrypted using Advanced Encryption Standard/Rivest-Shamir-Aldeman algorithms. When the data are sent to the server, the data are encrypted using the session ID received from the server in its initial authorization token. This encrypted data is sent over HTTPS, thus allowing 2 levels of security. When the data reach the server, they are decrypted using the session identification and stored in the database by encryption again using a server-specific encryption algorithm. Thus, the basic premise is satisfied and the data are completely secure when collected through smartphones, in transit, and when accessed from the Chronicle Diabetes system. All data from the Chronicle Diabetes system are stored securely in an HIPAA–compliant manner and restricted based on access privileges. An HIPAA Business Associates Agreement is signed with each facility using this system. All data access is recorded and a full audit trail can be produced.
Feasibility of the study was evaluated through participant attrition rates and qualitative and quantitative assessments of the usability of the connected system. Usability of the connected interface technology was measured using the System Usability Scale [
Preliminary efficacy was measured by changes in patients’ HbA1c levels and weight from baseline to 3 months. HbA1c levels were measured using the
At baseline, a sociodemographic and medical history questionnaire is used to collect study participants’ socio-demographic information, diabetes treatment plan, and other medical history. At 3 months, we asked patients how many medical, emergency, and diabetes education visits were made in between the 2 study visits and recorded this information as potential confounding factors of the study outcome.
SAS (version 9.2, SAS Institute, Inc, Cary, NC) was used for data screening and analyses. All the statistical tests were performed at 5% level-of-significance. The outcome variables HbA1c (%) level and weight change were used to measure the preliminary efficacy of this study. Qualitative thematic analysis was used to analyze interview data.
We recruited 30 patients from Houston, Texas and 30 patients from Pittsburgh, Pennsylvania, through various diabetes education programs recognized by the ADA. At each site, 30 enrolled patients were randomly assigned to the intervention group or standard diabetes education group. Data collection is completed. Data analysis is in progress. The study results will be reported in mid-2018.
This study leveraged existing DSMES programs, resources, and diabetes educators to deliver the technology-based program. Although mobile health interventions were developed to improve patient self-management in various research efforts [
Using smartphones and connected wearable fitness trackers not only reduces the burden of patient self-monitoring but also enables the connection of daily patient information to the Chronicle Diabetes clinical information system, where educators can track patients’ behaviors between visits in a graphical format and prepare for the next intervention session. This connection could also serve as an interactive platform to deliver intervention and provide feedback from diabetes educators in real time in the future. Connection to a national recognition data base through Chronicle Diabetes also offers the potential to collect behavioral and clinical information on unique populations and practices nationwide. Connecting patients’ self-monitoring information collected through smartphones to Chronicle Diabetes can facilitate education processes by allowing educators to more efficiently coordinate care plans with patients and ensure delivery of effective and tailored interventions. Moreover, using a central location for tracking patients’ behaviors and progress enables long-term self-management support for sustained behavior change. A system with good usability should foremost have good design of functionality that matches the work domain and a user interface that supports efficient task performance by the users [
This study could be easily and widely disseminated in future studies and practice. There are increasing numbers of smartphone users in the United States, including minority populations. As ADA–recognized education programs with access to the Chronicle Diabetes system are located throughout the US, this study could be widely disseminated. We anticipated that diabetes educators would use this connected tool to engage diabetes patients in lifestyle changes in between diabetes education visits and facilitate conversation on meeting or changing behavioral goals at follow-up diabetes education visits, rather than use data from the connected tool as a stand-alone piece of information to make treatment changes. The smartphone and wearable tracker usage in this study is only a tool to assist with diabetes educators, not to replace the role of a diabetes educator. The wearable fitness tracker and its companion smartphone app are not approved by the US Food and Drug Administration and their accuracy on measuring physical activity levels and dietary information is not guaranteed. Thus, the data connected from these devices and smartphone apps to Chronicle Diabetes system should be used with caution.
In summary, if proven effective, the study will not only advance nursing and behavioral science by leveraging existing resources to disseminate an evidence-based behavior intervention via emerging technology, but it will also provide theoretical and methodological guidance for other researchers conducting usability evaluations connecting mobile device information collection with EHR systems for managing diabetes and other chronic conditions.
American Diabetes Association
diabetes self-management eucation and support
electronic health record
Health Insurance Portability and Accountability Act
type 2 diabetes mellitus
The study is sponsored by the Robert Wood Johnson Foundation Nurse Faculty Scholars Program (Mentors: Jiajie Zhang, Nancy Bergstrom, Janet Meininger, Suzanne Bakken). The fitness trackers were primarily sponsored by the Jawbone Company. We thank colleagues at Flipside Media and Validic for technological assistance. We appreciate the tremendous dedication of the diabetes educators at our recruitment sites at University of Pittsburgh Medical Center (UPMC), and all the research assistants and students who helped with study recruitment and data collection. We thank all study participants who graciously gave their time to participate in the study and complete study-related assessments. JW’s contribution to this manuscript is also sponsored by the Josiah Macy Jr. Foundation Macy Faculty Scholars Program and John P. McGovern Distinguished Professorship at UTHealth Cizik School of Nursing. The authors thank Markeda Wade, ELS, at UTHealth Cizik School of Nursing for editorial review of the manuscript.
BU is the President of Flipside Media. Flipside Media was the contractor that originally developed Chronicle Diabetes for the University of Pittsburgh Medical Center. Flipside Media is currently the contractor that maintains Chronicle Diabetes for the American Diabetes Association. Neither BU nor Flipside Media have any financial interest in Chronicle Diabetes.