Powering Digital Health Through a Data Science Approach

Updated on September 25, 2021

By Sheila Rocchio

The pharmaceutical industry is undergoing significant transformation, as life sciences companies transition from treating disease to managing health. The process includes digital transformation across all areas of the life sciences value chain including major shifts in technology architectures, trends and stakeholders including clinical development. As development pipelines shift towards more targeted therapies where there is significant demand and unmet medical needs, the volume and variety of data sources are ever increasing, with over two-thirds of sponsors using or piloting at least four sources of primary data. Those new sources are largely digital, from mobile health and wearables to electronic health and medical records. Clinical trials are also increasingly relying on digital platforms to recruit patients and design trial protocols, with artificial intelligence being implemented throughout the process. In order to power AI-driven capabilities to speed clinical development, the life sciences industry needs to harness high-quality data to deliver results. This requires a new approach to clinical trials using data science and digital health best practices.

The Need for Data

“Alexa, tell me what medicines I should take today,” or “Alexa, what clinical trials am I eligible for,” will likely be options on “smart speakers” in a few years. Amazon has been able to track staggering volumes of consumer data to create new and valuable offerings targeted at consumers engaging with their products. Life sciences companies can follow suit by finding new, data-driven ways to deliver value to both physicians and patients looking to manage their own health. Every day, the human body generates the equivalent of two terabytes of data. This data includes information such as heart rate, sleeping patterns, stress levels and brain activity, but it can also include information on DNA and potential diseases. Our ability to modify genes and harness the power of existing cells to fight disease presents incredible opportunities to treat and potentially cure formerly devastating diseases. CAR T-cell therapy uses a patient’s own T-cells to recognize cancerous cells and tumours, and to attack and kill those harmful cells. Gene therapies, which replace unhealthy, mutated cells with healthy genes, are being used to fight viral infections and genetic disorders. These therapies are more complicated to develop and test. They also generate exponentially more data than former classes of drugs, including high volume “omics” data from numerous speciality labs. These data present an untapped resource that can offer value to numerous stakeholders that participate in the life sciences value chain. 

Unfortunately, the high volume of data from numerous sources has also contributed to delays in clinical trials. Since 2017, large pharmaceutical companies have experienced a 32 percent increase in the Last Patient Last Visit to Database Lock time cycle time metric. Although there are a variety of factors behind these delays, data chaos is one of the main causes. The evolution of clinical trials has been rapid, and many of the existing roles in clinical trials are unprepared or unable to handle the diversification of data. Sponsors are beginning to recognize the importance of data science in clinical trials, and nearly 75 percent of sponsors are establishing or expanding the role of data scientists, leveraging their experience to increase data-driven insights. 

The Data Science Approach 

According to the 2020 LinkedIn Emerging Jobs Report, data science roles are multiplying in every industry, with tremendous growth expected to continue to take place. The life sciences and pharmaceutical industries will not be an exception. Traditionally, clinical development has relied on data managers and clinicians to collect, clean and analyze results, but the increases in both submission path and exploratory data sources and types are causing delays. The amount of data collected and analyzed will soon exceed the capabilities of the traditional data management function and the old approach of cleaning every data point no longer applies. The application of data science principles to clinical development processes has the opportunity to positively impact numerous aspects of the clinical development process. 

Data science augments the role of technology to drive data and analytics capabilities in predicting outcomes, which can lead to faster and more accurate trial results. Within clinical trials, taking a data science approach can improve analysis, including a greater ability to identify patterns, test hypotheses or assess risk. It can also be used predictively where, in conjunction with AI and machine learning, outcomes and risks can be forecasted based on past experiences. Only 30 percent of life sciences organizations currently have predictive analytics in place, yet in clinical trials, predictive analytics can help to alert both data, operational and clinical teams to problems much earlier. Data science can be applied to protocol designs, site selection, data collection, data review, site monitoring and trial close out processes, resulting in trial timelines that are more predictable and yield higher quality results. 

Laying the Foundation for Data Science and AI

In order to address clinical trial inefficiencies, and to begin benefitting from more robust data science problem solving capabilities, life sciences organizations must first lay the foundation for their data. This means planning and implementing a holistic, data-driven strategy to clinical trials. AI-driven drug discovery must have clean, usable data to draw insights from, yet only one-third of sponsors currently have a formalized data strategy in place. With the majority of sponsors failing to utilize all of their data, they will be unable to take advantage of the major benefits AI can deliver in automating labor and time intensive tasks and delivering new insights. Creating and implementing a robust and formalized data pipeline strategy and centralizing all data sources into one hub for transformation and analytics is an important first step for any organization looking to advance their analytics competencies and to leverage the power of AI across the clinical development enterprise. 

The Future of Clinical Trials

To advance clinical development, the pharmaceutical industry must adopt a data science approach. While most sponsors understand the benefits of a formalized data strategy to harness new data sources and increase analytics capabilities, clinical trials have progressed to the point that data strategies are now a necessity. Data is growing rapidly and fueling greater innovation, new products and capabilities in life sciences. In order to move forward with this transformation, gaining control of data before it becomes data chaos and delivering new, engaging customer experiences are no longer lofty objectives: they are strategic imperatives for the pharmaceutical industry.

Sheila Rocchio is the Chief Marketing Officer of eClinical Solutions. She enjoys finding creative ways to tell customer stories and building products and services that help clinicians, data scientists and technologists do the challenging and important work of bringing life-saving new therapies to market.

14556571 1295515490473217 259386398988773604 o

The Editorial Team at Healthcare Business Today is made up of skilled healthcare writers and experts, led by our managing editor, Daniel Casciato, who has over 25 years of experience in healthcare writing. Since 1998, we have produced compelling and informative content for numerous publications, establishing ourselves as a trusted resource for health and wellness information. We offer readers access to fresh health, medicine, science, and technology developments and the latest in patient news, emphasizing how these developments affect our lives.