New mHealth Action Plan Released

The Duke Margolis Center for Health Policy (DMCHP) recently collaborated with the FDA to release an mHealth action plan entitled

. This action plan includes defining the different mHealth data types, understanding needs and incentives relating to patients, researchers, sponsors, and mHealth companies, and elaborating on challenges, such as fit-for-purpose determination, and offers valuable recommendations on approaching mHealth in clinical trials.

Among a group of experts in mHealth including personnel from the FDA, John Reites, Chief Product Officer at THREAD (www.THREADresearch.com), was a contributor to the action plan and elaborated on how this action plan makes a difference in approaching mHealth in clinical research.

What drove DMCHP to draft the paper on mHealth and what is the purpose of this action plan? Who is the target audience for this release?

The FDA has been especially focused and active in encouraging and supporting digital health initiatives with

recent releases of the pre-certification program, Nest and other programs. Clinical research studies are also now experiencing rapid growth in the use of mobile health. This presented an opportunity for the Robert J. Margolis, MD, Center for Health Policy at Duke University to bring together a group of experts from across the mHealth and healthcare ecosystem to identify the steps needed to advance the use of patient-facing mHealth data to support the evaluation of medical products. DMCHP led the development of this action plan to provide a clear structure as to how mobile health and medical devices can be categorized and defined in the health and research community.

The intent of the action plan is to provide definitions needed in mHealth, provide recommendations, and give real world examples of how companies are using mHealth to advance research and real world evidence collection across the industry. The action plan specifically targeted audiences from companies of all sizes and focus areas such as software, medical devices, consumer brands, academia, pharma, etc. The committee that developed the action plan was equally diverse with members from FDA, Duke University, Kaiser Permanente, Verily, Edwards Lifesciences, Sage Bionetworks, THREAD, PatientsLikeMe, Mt. Sinai, and University of Massachusetts all experienced conducting mobile health research for many years. 

Can you describe the different data types that can be collected through mHealth? Which of these data types is most applicable in clinical trials?

The Action Plan lays out four (4) data types: Patient/Consumer Reported Data, Task-Based Measures, Active Sensor Data, and Passive Sensor Data.  

Patient/Consumer Reported Data: This type of data is reported manually by the patient themselves or by their caregiver if the patient is unable to enter the data. ePROs, non-validated surveys and diaries are examples. This type of data is being collected from participants in research studies at a large scale now and is becoming a standard for the industry in trials of various phases.

Task-Based Measures: These data are objective measurements of a person's mental and/or physical abilities to perform a test or a series of tasks. To elaborate, examples of task-based measures are Active Tasks in Apple’s ResearchKit or what THREAD refers to as Electronic Device Reported Outcomes (eDROs). eDROs are used when a patient is given an activity to complete, like walk, move or use a device. While the participant completes the task via the step-by-step process, smartphone sensors, surveys, contextual data and/or external sensors from medical devices capture a number of data points around that activity.

Active Sensor Data:  This is a measurement of the person's daily activities, mental state or physiological status 

that requires an activation step. Examples include a participant being prompted to and then stepping on a connected scale or providing a glucose self-measurement.

Passive Sensor Data: The action plan defines this type of data as a measurement of a person's daily activities, mental state or physiological status that does not interrupt a patient's normal activity (i.e. measures what a person does in their daily life). For example, a patient is wearing an activity monitor that is tracking sleep. The patient does not have to do any tasks, they do not have to provide data, they do not have to wake up and tell the device to do something. They just sleep and the device collects the data. This represents a invisible, continuous way of collecting and integrating information.

What requests have researchers brought up to advance mHealth in clinical trials?

The group DMCHP assembled was very experienced, practical, and highly engaged on the topic. A majority of the members are conducting mHealth enabled research everyday in their work. Because of this experience, the group provided a large amount of short-term and long-term recommendations to advance mHealth, most included in the action plan. A couple of these recommendations, however, stood out for me personally. One request was for the research community to publish and/or make public more of the failures in mHealth, so that that community at large can learn from other’s lessons and not repeat those “successful mistakes” in their implementation of mHealth. Another request was for the research community to find more pre-competitive areas to partner in standardizing more aspects of mHealth. Lastly, the idea of “fit-for-purpose” use of mHealth balanced with standardization of best practices is really important. Many researchers are challenged with finding that balance as they learn which solutions work best for specific populations.

Do you believe biopharmaceutical enterprises and other sponsors will be using data from mHealth devices to support primary and secondary clinical trial endpoints in the future?

I do believe this, but recognize it is a matter of “when” as some data types are further along than others in becoming validated standards of use in clinical research. What has changed in recent years is an accelerated movement in using mHealth and its integrations to support secondary and exploratory endpoints. THREAD is also seeing a move towards our customers using more regulated, approved medical devices for providing the specific endpoints of interest in combination with the other data types now being collected.

Can you expand on Johns Hopkins Epilepsy EpiWatch iPhone and Apple Watch study that THREAD supported?

Johns Hopkins leads the EpiWatch study and THREAD was excited to bring our expertise to support its

development as was the first Apple Watch app built for research. The study uses mHealth (the patient-facing app and Apple Watch app) to engage participants and capture important seizure data throughout the participants normal life activities. As defined on Johns Hopkins’ website, “Johns Hopkins EpiWatch™ is an app for Apple Watch™ and research study. EpiWatch helps you manage your epilepsy by tracking your seizures and possible triggers, medications, and side effects. You can view this information at any time, and a dashboard lets you share a summary of the data with your doctor or caregiver if you want. With EpiWatch, you can also send a message to family members or caregivers to let them know when you are tracking a seizure.”

The researchers at Johns Hopkins have been open about the studies’ results to date and have made this information available via conferences, publications, etc. This is a great example of how mHealth research is engaging people in their daily life, providing them with value, and contributing to critical research.

This action plan represents exciting progress for the standardization of mobile health in clinical research. We at THREAD are excited to be a part of this movement to change how we conduct trials with remote patient research approaches as we strive to make these techniques repeatable, cost effective, and more intuitive for all the research stakeholders.

Learn more about the project, watch the release meeting and download the action plan here: https://healthpolicy.duke.edu/events/public-event-mobilizing-mhealth-innovation-real-world-evidence-generation