“What you measure will improve”- Aron Anderson
A Historical Perspective:
Big data analytics was used as early as in 1600s for predicting epidemics of bubonic plague in Europe. Though records of decision-making from health care data are available from Babylonian times, demographer John Graunt pioneered the principles of statistical analytics in health care which were used for predicting mortality at a population level in London. When mathematician Clive Humby declared “data is the new oil”, the analytics industry was just scratching the surface of what would be a “big data analytics” revolution.
While clinical trials are the cornerstone of research, the huge investments and long turn-around time leaves room for innovation in conducting clinical studies efficiently. Only 12% of drug candidates entering clinical trials are approved for use in therapy. With an average time of ten years for a new drug discovery and an investment of approximately USD 2.6 billion per new molecule brought to the market, the industry is exploring new avenues to cut down time and costs. A blockbuster drug takes approximately 12 years and $4 billion-$11 billion of investment.
Quantum of Data:
“Data that sit unused are no different from data that were never collected in the first place.” – Doug Fisher
Colossal amounts of data are continuously added to the health care space. As of 2019, ClinicalTrials.gov lists 325,860 studies with locations in 209 countries. In addition, the average hospital generates approximately 665 terabytes of data annually, 20% of which is unstructured forms of images, video, and doctor’s notes. Also, add to this the vast amounts of data generated from Internet of Things in health care.
Big Data Utilization:
“It’s time to move from reactive sick-care to proactive healthcare by default” – Koen Kas
The health care industry has been experimenting with the idea of combining the data generated from different sources to achieve efficiency in drug development across the drug lifecycle. With the vast amounts of data being generated from clinical trials and from real world data sets, comes the promise of shorter drug discovery times, targeted personalized medicine and low cost of drug development. The industry has evolved to accept big data from electronic medical records, patient registries, pharmacies, and claims data sets for prevention, prediction, diagnosis, planning, and, management of health conditions. It is not only used for tracking and estimating safety, efficacy and effectiveness of interventions, but also for improving efficiencies within health systems. Implementation of big data analytics for better diagnosis and disease predictions has the potential to save more than 25% in annual costs by decreasing the hospital readmission rate.
Technical Challenges:
“When we have all data online it will be great for humanity. It is a prerequisite to solving many problems that humankind faces.” – Robert Cailliau
The myriad challenges facing the storage, procurement, privacy, processing, homogenization, analysis, interpretation, and communication of this data are a constant work in progress. Methods for big data management and analysis are being continuously developed especially for real-time data streaming, capture, aggregation, analytics (using machine learning and predictive), and visualization solutions to integrate a better utilization of real-world data (RWD) in health care decision-making. On-site server networks are challenging to maintain and scale. With decreasing costs and increasing reliability, cloud-based storage using IT infrastructure is fast becoming a preferred option by multiple health care organizations. This also enables creation of massive analysable data lakes. Interoperability of shared data is a concern too. Solutions like Fast Healthcare Interoperability Resource (FHIR), public APIs, CommonWell, and Carequality are making data sharing easy and secure.
Compliance Guidelines:
“Having access to information when it’s needed is critical” – Hal Wolf
Technical challenges aside, data security of public health information is governed by rules, termed as HIPAA Security Rules, to guide organizations with storing, transmission, authentication protocols, and controls over access, integrity, and auditing. The “Breakthrough therapy” designation is a United States Food and Drug Administration (US FDA) designation that expedites drug development approved in 2012 to ensure the most efficient possible path to approval is available for manufacturers of drugs. The FDA released draft guidance for communication of healthcare information regarding drugs for reimbursements in the FDAMA Section 114 and Preapproval Information Exchange (PIE) in 2016 for enabling early conversations with payors. In 2018, the FDA issued guidelines regarding analytics and use of real-world evidence (RWE) for drug approval. The Priority Medicines scheme (PRIME) launched in 2016 builds on the existing regulatory framework and tools already available within the European Medicines Agency (EMA) framework for scientific advice and accelerated assessment of drugs. The EMA and European Network for Health Technology Assessment (EUnetHTA) are piloting projects (from 2017-2020) to explore areas of collaboration for identifying synergies for evaluation of interventions and types of accepted evidence for approval and reimbursement.
Proof of Success:
“The goal is to turn data into information and information into insight.” – Carly Fiorina
An impact analysis of breakthrough therapy designation initiative of 2012 found that the FDA approved more than 150 breakthrough therapies and granted more than 332 breakthrough therapy designations since its launch; the initiative also accelerated premarket drug development by 2.2 years. The increasing number of expedited FDA approvals for cancer drugs based on surrogate end points is encouraging many organizations to search for new ways to uncover efficacy and safety data to justify the costs associated with costly treatments. With the growth of data innovation and collaborations, organizations like Friends of Cancer Research (FOCR) are instrumental in driving change and have initiated processes to expedite drug development. FOCR partnered with several data partners, including the American Society of Clinical Oncology (ASCO)’s CancerLinQ, Cota, and Flatiron Health, to conduct pilot studies for creating better defined patient cohorts for improving outcomes and driving efficiencies.
What lies ahead:
“We need to redefine the rules for data ownership and draft something like a digital constitution for citizens across all industries” – Jaana Sinipuro
While we create frameworks and guidelines to drive the use of data and artificial intelligence solutions, data transparency, transparent algorithms, robust data protection rules, and international co-operation across countries and data providers will play an important role in defining the success of big data analytics in health care. Non-profit organizations like Bioethics International track biopharma companies for accountability for trial registration, reporting negative and positive results, publication, and data-sharing. Data transparency and democratization is important to expedite research and customize delivery of efficient seamless health care to all.
Data assimilated from clinical trials and RWE is the future of health care.