Oracle Pursues mHealth in Clinical Trials

With changing regulations, and as numerous clinical trial mHealth technology enterprises emerge, the field of mHealth in clinical trials is becoming an attractive venture for large technology organizations, such as Oracle. Oracle Health Sciences recently launched the mHealth Connector Cloud Service, which is a platform that focuses on integrating a variety of mHealth data sources, and securely storing the data in the cloud. In this interview, Jim Streeter, Global Vice President of Life Sciences Product Strategy at Oracle, will elaborate further on Oracle’s strategy with mHealth in clinical trials.

What challenges exist with the way data is structured from mHealth devices?

Jim Streeter

The challenges involve the structure of the data, and frequency of data collection. There are two types of data that come in: you have raw data and the derived data that the device is sending (depending on the device). To elaborate, you could be checking blood pressure twice a day, you could be collecting data from an activity tracker, or an ECG that is collecting data at a high frequency throughout the day. Another issue involves data loss and warehousing; we often find that researchers want access to raw data, but, can’t access it because of limitations pertaining to mHealth device connectivity and data warehousing, so they end up with aggregated datasets providing limited or imprecise information.

At Oracle, we have implemented an mHealth adapter, which allows researchers to send the data to our Data Management Workbench once it comes in. This IoT adapter has an industrial-strength infrastructure (as it is being used in other industries), and it is designed to handle thousands of devices that can be simultaneously connected and concurrently collecting data. We unveiled a new product called mHealth Connector Cloud Service, which is designed for sending structured data directly from mHealth devices into Oracle Health Sciences InForm integrated clinical data capture and management cloud platform. This product is made for structured data types, however, if you are collecting raw data, you can aggregate and average that within a certain amount of time and have the data re-encoded and entered into InForm EDC based on specified parameters. Moreover, we have tools around data warehousing and connectivity to prepare raw data for additional analysis. Researchers default to a limited amount of data going into current regulatory systems (i.e., averages of datasets), whereas others want to analyze raw data for additional signals. This system is capable of facilitating the connectivity of thousands of mHealth devices, and storing large amounts of raw data.

How is AI being used with mHealth?

A lot of data scientists are looking at how to piece mHealth data, especially when interacting with raw data, and this is where I see AI coming into play at the moment; it is currently being used to structure data more efficiently. I envision AI as a tool that will look at how mHealth data interacts with clinical data in existing and previous studies. Although AI is in its infancy in clinical trials, from an analytics point of view, the time to bring all that data together is shortened from days and months to a few minutes. For example, we have brought in data from mHealth devices and have joined it with CRF and ePRO data very rapidly.

How can the clinical trials industry ensure that their digital study data is secure?

Data encryption is key. We tend to look at the transfer of data from the mHealth device up to the cloud; some mHealth devices will allow sponsors to access the data via their cloud service, but the data needs to be encrypted throughout the journey from the source to the backend target. At Oracle, whether we take the data from the cloud or mHealth device, we want it to be encrypted and ensure patient information is deidentified. We built our mHealth system with data security in mind, so once data lands in the Oracle Cloud, it is secured and encrypted.

How will future clinical trials look like?

In the future, I envision the use of real world data and see a lot more use of mHealth in clinical trials especially with collecting primary endpoints in regulatory submissions. We've started to see the use of mHealth to promote patient centricity, as we are trying to understand the day to day activities of patients. Anyone using an mHealth device needs to understand new endpoints, and safety data. A question we need to ask is whether mHealth will reduce clinic visits, or whether those visits are needed at all. Patient centricity coming together with raw data and mHealth devices lead to a much different way of running clinical trials and how much data we collect. I believe we will see event-based clinical trials, where we are not only going to run a few ECGs, but, it will be continuously monitored through mHealth devices. Similarly, we may currently only take vitals once during a study visit, but soon we will be collecting that data continuously. The patients will be wearing devices that will allow the data to come together or it will be collected while they are in the hospital or at the doctor’s office. We will approach clinical trials and obtain data in a much different way and so, not only analysis methods will have to change, but, also the regulatory environment. 

Moe Alsumidaie, MBA, MSF is Chief Data Scientist at Annex Clinical, and Editorial Advisory Board member for and regular contributor to Applied Clinical Trials.