banner_data numbers

Advances in electronic health records, personal fitness tracking devices, and the ‘Internet of things’ have made it possible for biopharma researchers to gather exponentially more data about patients to inform their trials. Now the big question is -- what do we do with all the data?

There is tremendous value in clinical data, but only if we as an industry develop standards and protocols for capturing, mining, and analyzing that data so we can foster more informed decisions. But when you consider the vast and growing collection of data coming from an increasing number of sources, gaining that clarity and control becomes a deeply complex matter.

Consider the recent adoption of continuous glucose monitoring (CGM) devices for diabetes patients. In past trials, a patient might see their physician once a month to have their HbA1c level tested, creating a single piece of data to judge their disease status. But with a CGM, that same patient’s glucose is tested every five minutes round-the-clock, and that data is instantly sent to a trial database, creating 288 unique pieces of data every single day for duration of the trial. Multiply that by hundreds of patients over thousands of days, and you can begin to envision the deluge of information that researchers now have access to. And this is just one device used by one category of patients. The internet of things is enabling countless devices to gather more health specific data -- from Fitbits and Jawbones, to ‘smart contact lenses’ and heart monitoring patches – which means the surfeit of data will only continue to expand. Until we figure out how to manage all of this data and use it in meaningful ways, it won’t generate the value that’s intended.

Therein lies the challenge. While industry experts love to talk about the value of Big Data, analytics and the cloud, these are only the first part, and arguably easier piece of the equation.

Connectivity is key

To make the most of big data in healthcare, we first need to break down the physical and cultural silos that prevent us from taking advantage of the data we collect. Then we need to embrace collaboration, among biopharmaceutical industry players, and with academia, government and the tech sector, to define tools and standards to make it possible for us to confidently use this data to drive decisions.

Many innovative projects are already underway. The 2009 Health Information Technology for Economic and Clinical Health (HITECH) Act, for example, drove rapid adoption of health information technology and standards for data sharing via electronic medical record systems and Health Information Exchanges (HIEs). Partnerships between pharma and tech giants, like Google and Apple, have led to rapid growth in the development of mHealth apps and platforms, including Apple’s HealthKit, and the Google Fit platforms where users can gain access to summarized views of all their relevant health and fitness data. We have also seen several cross- industry collaborations between pharma companies to improve the link between technology and trial data, including the 2014 launch of Project Data Sphere, which is enabling the sharing of clinical trial data to speed research in critical disease areas.

But we’ve only just begun. As an industry we have a wealth of data that has the potential to change the way we run trials and make research decisions, but to get there we need to get smarter about developing sophisticated algorithms to answer specific data questions, and take advantage of machine learning technology to hone our search criteria and develop more predictive analytics. We also need to support efforts to develop better tools to mine unstructured data, like physician notes and real-world information reported by patients themselves. Such tools can enable researchers to identify trends among patient populations sooner, and support precision medicine goals, like offering customized dosing based on patient genomics.

Focus on the future

To achieve this future state where big data analytics are an integral part of the trial environment, we need to focus on four key areas:

  1. Using data to gain real-time insights and support decision making throughout the clinical trial lifecycle. That includes implementing risk-based monitoring of participating patients, and using analytics to support decision making. When looking at ‘big data’ and new smart networking capabilities,  we will see that the decisions and mining of information will not only happen in the back office within data “lakes” and data warehouses. The future is about ‘edge intelligence’ with smarter remote devices. It’s also possible to analyze data while it is being streamed within the network. Lastly, near real-time federation across multiple data sources will become available.
      
  2. Focusing on patient centricity and service personalization. That means looking at the data you collect and the way you collect it from the patient’s perspective and adapting your process accordingly. Patients today are more informed and connected, and they have expectations for how their data will be collected and used. They want to participate in the healthcare discussion and maintain control of their results, which they will share within their social networks. Clinicians need to factor these preferences in when planning the trial and communicating with patients. In addition with genomics each patient knows that they are unique and will demand more personalized care and medication in the future.
      
  3. Using event and rule driven orchestration of business processes. Events that impact a patient’s health, ranging from taking a run to missing a doctor’s appointment, happen every day. Researchers cannot rely on patients to report every event that occurs, so we need to build tools that codify these experiences into rules-based engines to automatically process this data so we can better monitor and treat the patients in our care. The alerts from the field may come by way of existing big data, or from new workflow and event management tools. These tools can respond in predictable ways through therapeutic expert-managed rules. They also record decisions made for audit purposes. I believe that more IT solutions will be workflow and event-based in the coming years.
      
  4. Pursuing collaborative innovation and focused ecosystem business models. The future will be built on dynamic collaborations through which industry leaders come together to solve specific problems, establish best practices, and move on to new opportunities. It is only through such agile partnerships that we can efficiently overcome the big data hurdles we face, and prepare for a more connected, data driven future.

If we continue to invest in data analytics tools and standards, and align our efforts with these trends, we can enable the industry to make the most of the data it already owns, and to better meet the needs of patients around the world.