Who remembers this line from the movie Zoolander, “the files are IN the computer?!” Twenty years later, we all painfully know the file structure of our computer. But as we continue to digitize clinical trials, has access to new streams of valuable data become easier because “the data is IN the cloud”?
Connected sensor technologies and tools, from wearables, smart speakers and bed mats to ingestible and implantable sensor capsules, are capturing unprecedented data streams with the potential to transform our clinical decision-making. Since the launch of DiMe in October 2019, there has been a 929% increase in number of digital terminals used by the life science industry for the safety and efficiency of new drug development. Yet we are still unprepared to access reliable and trustworthy sensor data at scale.
Today, it is incredibly difficult to use this data effectively and efficiently, while adhering to privacy and security standards. To accommodate privacy and security, sensor technology data is collected, processed, stored and transmitted through systems and infrastructure developed and owned by individual companies. Data users must integrate in a custom way to meet the specifications of each data partner. It takes time – building 1:1 “point” solutions is neither efficient nor scalable.
This requires specialized knowledge: data scientists and software engineers are needed in real time to create, troubleshoot and repair each individual connection. Not only is this cost prohibitive, but if we consider all the many streams of sensor data that could help us make better decisions, faster, and then multiply that by the number of decisions, it’s clear that this approach doesn’t will not be scalable.
Data integration refers to the technical and business processes used to aggregate and combine data from multiple sources to provide a unified, single, and usable view of data. Efficient sensor data integration has the power to realize the promise of these new streams of data to make better, faster decisions across the healthcare continuum and advance new drugs, devices and therapies that will improve patient outcomes.
But the current influx of data from sensor technologies exceeds industry’s ability to effectively collect, store, analyze, protect and use this data for research and patient care. The promise of collected data will not be realized until we have systems and infrastructure in place to manage the wave of new data efficiently, effectively and affordably.
DiMe is ready to help get data out of the cloud and into the hands of those who need it to improve the lives of patients. In our role as a society bringing together experts from all disciplines comprising the diverse field of digital medicine, and in partnership with we have developed a set of free resources to help organizations define, select and implement the integration sensor data and algorithms for appropriate platforms so they can manage research data and guide clinical care.
We created four free complete toolkits for data producers, processorsand consumers to effectively and efficiently utilize this new wave of data from wearable devices and large-scale digital sensing products, while maintaining data privacy and security.
Data architecture translates business needs into data and system requirements and seeks to manage data and its flow through the business. Data architecture describes the structure of an organization’s data assets. It includes the models, policies, rules, and standards that govern the collection, storage, organization, integration, and use of data within an ecosystem. An optimized data architecture is essential to realize the value of sensor-generated data for clinical decision-making. This toolkit includes:
- Logical Data Architecture – Provide a high-level model for sensor data in the health data ecosystem, independent of platform, operating system, file structure, or database technology. It defines the entire data landscape needed to use sensor-generated data to guide clinical decision-making.
- Master Data Architecture — A library of sample data architectures that 1) implement the logical data architecture, and 2) have been successfully implemented to support successful data integrations. Find what works for you and your team.
- Sensor Data Flow Design Tool – Using this tool, you can map the sensor data stream from any connected sensor technology you choose to a final dataset for analysis and querying, whether either in the context of care delivery or research. After answering a few questions to define the right steps, you have the opportunity to use an easy-to-use design tool to create and annotate your custom sensor data stream for documentation and collaboration with your team and partners. .
Every data ecosystem needs a common language and a shared approach to distribute, store and interpret information. It is critical to meet sensor data standards that protect patients and enable data exchange. By adopting standards, we ensure that we use accessible, relevant and trustworthy (ART) sensor data.
Start differentiating your products and the data they generate by building to the right standards. Distinguish between products and solutions to identify those that meet your needs. We have designed our interactive standards landscape to guide you as you consider how your data may be collected, stored, adapted, safeguarded or retrieved for clinical decision making in healthcare and research. This toolkit includes:
- Interactive standards landscape – All standards related to sensor data integrations in an interactive landscape tool.
- Library of standards – Standards database. Note: DiMe is committed to keeping this information current as new standards are developed.
Are you looking to start or advance your sensor data integration journey? Six criteria are essential to your success: collection, transmission, processing, security, confidentiality and quality of data. Together, these key areas provide the building blocks for a successful sensor data integration strategy so you can make better, faster decisions in healthcare and research. This toolkit includes:
- Accessible, Relevant and Trustworthy (ART) Criteria – Six essential criteria for delivering ART sensor data to your downstream partners for clinical decision making.
- Considerations and Best Practices – Short “cheat sheets” of key considerations and best practices aligned with each criteria.
- ART criteria prioritization tool – A tool to optimize sensor data integration approaches to meet the needs of organizations and individuals using your products and the data they generate.
- Case studies – Concrete examples so you can see the ART criteria in action.
Are you looking to start your sensor data integration journey? The Organizational Readiness Toolkit can help you identify areas in your team or business that could be developed for success with a sensor data strategy. It can also help you assess your organization’s level of progress toward the goal of making sensor-generated data accessible, relevant, and trustworthy to enable better clinical decisions, faster. This toolkit includes:
- Capability Maturity Model – A model to guide your understanding of your partners in the downstream market.
- Capability Maturity Calculator – An assessment tool to assess your partners’ readiness and guide you to the right resources to meet them where they are for shared success.
As we increasingly rely on sensor-generated data to power decentralized clinical trials, we must prioritize building to scale. Our potential to dramatically improve patient outcomes and advance the safe, effective, equitable, and ethical use of digital medicine to improve human health is within reach. And when we unite around standards and align on sensor data integration practices and infrastructure, our ability to make better and faster decisions will transform the healthcare system.
About the Author
Jennifer C. Goldsack founded and serves as CEO of the Digital Medicine Society (DiMe), a 501(c)(3) nonprofit organization dedicated to advancing digital medicine to optimize human health. Previously, Jennifer spent several years at the Clinical Trials Transformation Initiative (CTTI), a public-private partnership co-founded by Duke University and the FDA, and . working in research at the University of Pennsylvania Hospital, first in outcome research in the Department of Surgery and later in the Department of Medicine. Most recently, she helped launch the Value Institute, a hands-on research and innovation center embedded in a major academic medical center in Delaware. Jennifer holds an MS in Chemistry from the University of Oxford, England, an MS in History and Sociology of Medicine from the University of Pennsylvania, and an MBA from George Washington University.
For more on DiMe, and more, check out the latest episode of our Health Innovators video series here.