Overcoming data paralysis to build next-level data-centric orgs and realize better patient care
Many life sciences companies have found that moving from data acquisition to imbuing data and analytics into decision-making processes throughout their operations is a steep climb.
DRG experts discuss common challenges in the industry — and share a roadmap for how to surmount them.
Obstacles on the road to data transformation
When Novartis’ Vas Narasimhan took over as CEO in 2017, his first order of business was to reposition the company as a “medicines and data science company” by “pivoting to become a data centric, digitally enabled organization” that would use these technologies to realize a “productivity revolution” in drug development.
“As an industry we have vast datasets built from having conducted countless studies in thousands of diseases and in some cases, like ours, going back decades,” said Narasimhan, in a LinkedIn blog post. “At Novartis, this is our goldmine: our wealth of clean, curated longitudinal and interventional data that if we take full advantage of, has the potential to completely transform the way we develop medicines.”
In placing data science at the center of its long-term strategic vision, Novartis has staked a bold claim on leadership in data innovation. However, even for an innovation leader, realizing data transformation in the life sciences requires summiting a learning curve whose steepness rivals the mountains that ring Novartis’s home offices. “As we’ve gotten quite scaled and working on digital health and data science, we’ve learned there’s a lot of talk and very little in terms of actual delivery of impact,” said Narasimhan in a recent podcast.
Moving from data hype to data delivery
What is the holdup? Life sciences companies are struggling with mastering and integrating vast and disparate data sets as they go about building their capabilities for data-driven development, commercialization and customer engagement.
Beyond acquiring data, many organizations wrestle with prioritization and understanding which inputs are needed as the volume of available health and medical data has exploded and the number of data sources grown exponentially.
Often, companies have amassed large data sets without first ascertaining what business questions they want to answer—and whether or not the data they have acquired includes the requisite fields to do so.
Identifying the data sets you need and getting them to talk to one another are heavy lifts. Leveraging that data into insights that speed new therapies to market and improve patient care is yet an order of magnitude more difficult, but promises to change the game—and just in time, as mounting costs and correspondingly high prices are rendering traditional modalities of R&D unsustainable.
Companies are investing in building data expertise for the long haul
Novartis is hardly alone among life sciences companies in having to level-set expectations after investing heavily in data management expertise. In February, 2018, Roche plunked down $1.9 billion to buy up the remaining shares of Flatiron Health, the oncology-focused EHR platform. Then-Roche CEO Daniel O’Day called it “an important step in our personalized healthcare strategy for Roche, as we believe that regulatory-grade real-world evidence is a key ingredient to accelerate the development of, and access to, new cancer treatments.”
In placing Roche’s initial bet on Flatiron Health, however, O’Day was careful to emphasize its long tail nature. “This is a long-term strategic investment, and the strategic collaboration benefits us, he said. “The value in the investment is not to generate a financial return.”
Some smaller biopharma companies have taken a more incremental approach to data applications. For example, UCB is developing a suite of predictive algorithms for epilepsy dubbed Eliprio, using de-identified patient information to predict drug-resistant epilepsy, among other use cases. Meanwhile, Takeda has used claims data to better understand factors in drug switches for severe depression patients.
And medical device companies are looking for ways to realize the potential value of the rich data streams their products generate. Boston Scientific has partnered with Accenture to give providers population-level insights into cardiac patient outcomes and costs. Medtronic is working on implantable remote monitoring systems to help heart disease patients and providers better manage outpatient care, while partnering with IBM Watson and Sugar.IQ to head off episodes of high or low blood sugar and provide personalized sugar-level adjustments.
Getting to data maturity is a steep climb for life sciences orgs
For all its potential to speed up clinical trials, weed out the duds before Phase III, deliver much more targeted therapeutics and improve patient outcomes at scale, data science is hard, and these technologies are still in their infancy.
As excitement about the health applications of data technology gives way to a more sober and realistic appraisal of the challenges involved, a weary cynicism has begun to set in.
However, as innovators know, getting it right will be essential to delivering and commercializing the costly, highly specialized medicines of tomorrow in an increasingly value-driven healthcare environment.
Doing so will require an understanding of the key challenges and success factors for integrating data-driven decision support into life sciences businesses.
Download the full white paper to learn:
- The three stages of life sciences data maturity
- Framework for Success: Common characteristics that define data-mature organizations
- Accessing the right data
- Mastering and integrating
- Analysis to insight
- Making it work within the org