How patients can benefit from big data

"By changing the way we use data in health care, making that data actionable, we can save lives, save money and improve the health of our populations." - Denise Hatzidakis, Chief Technology Officer, Premier Healthcare Alliance

Connected citizens consume or barely stay afloat in a sea of digital information. Healthcare stakeholders (IBM, Microsoft, Verizon, Flatiron Health, WebMD, etc.) guard or aggregate the zettabytes of data (‘big data’) that define our electronic existence. Analyzing and storing meaningful digital health information (think volume [Google-size], velocity [e.g. Twitter] and variety [text, images, links, videos, etc.]) can optimize the patient health.

Healthcare stakeholders are identifying how patients, providers and payers can make better use of the available information. For instance, Tableau Software Inc. has provided 7 general tips geared towards businesses on how to organize the variety, volume and velocity of information collectively known as big data in order to empower their organizations, improve operational efficiency, and enhance revenues. The tips are:

1.    Simplify (adopt a strategic approach)

2.    Coexist (integrate database platforms into your organization’s data architecture)

3.    Visualize (explore data visually with the appropriate software in order to detect patterns)

4.     Empower (creativity of all employees can be enhanced through rapid collection of actionable data from disparate sources – from tackling new business problems to developing entirely new products and services)

5.    Integrate data from disparate sources without a large upfront cost e.g. social media, technology, email databases, fundraising databases and consumer market data  

6.    Govern i.e. organizational master data management systems and global standards for the protection of privacy will have to be put into place (an international Health Insurance Portability and Accountability Act [HIPAA] for the digital age that minimizes privacy concerns)

7.    Evangelize across your organization in order to increase the adoption of analytics and collaboration

Ever-increasing amounts of data will be collected thanks to health information exchanges created by the HITECH and Affordable Care Acts.1 The data pools typically accessible to patients and other healthcare stakeholders are: clinical data (electronic medical records, medical images), patient behavior and sentiment data (patient behaviors and preferences, retail purchase history and exercise data), healthcare cost estimates and pharmaceutical research and development data (clinical trials, high-throughput screening libraries). Evangelists predict that a meaningful analysis of these data pools could lead to an in-depth understanding of population health and the effects of healthcare interventions in the real world.

What value can the modern patient derive from the health ‘datanami’?

Patients rely on experts and regulatory agencies to translate the evidence supporting the authorization and safe use of medications and devices. Broad categories or STEPS (([S], satisfaction; [T], treatment/clinical; [E], electronic information/data; [P], prevention and Patient Education; [S], savings) have been developed by The Healthcare Information and Management Systems Society (HIMMS) to address the following questions:

1.    How does health information technology (HIT) improve patient care for a physician serving patients in a medical practice?

2.    How does HIT benefit the patient experience?

3.    How can hospitals, medical practices, clinics and other points of care save money while providing quality and safe patient care using health IT?

Sixty one percent of survey respondents (physicians, pharmacists, and nurses) in the 2013 iHIT study agreed that HIT tools helped physicians to be more problem-focused in their communications and 83% agreed that these tools and applications supported clinical processes (versus 2006). Electronic portals designed by payers, such as Express-Path, can also reduce administrative costs and shorten the path to patient treatment. However there is still a huge gap between the ability to amass data from millions of patients and harnessing this information to tailor treatments to individuals. Time and resources may also limit physicians and allied healthcare providers’ abilities to fully integrate and explain medical information to patients. Therefore non-clinical providers such as patient navigators could fulfill a valuable role in assisting patient understanding of all healthcare options.

In addition, patients requiring one or many different medications/devices need to pay special attention to the benefit/risk profiles of these interventions. Big data analytics may aid physicians and patients in presenting a more accurate picture of drug efficacy and safety in the real world compared with results obtained from often-healthier patients participating in rigorously controlled clinical trials. Large-scale pharmacovigilance studies incorporating big data may also address concerns about bias associated with voluntary adverse event reporting to the Food and Drug Administration’s (FDA) Sentinel system. Blending this data with information from prescriptions and electronic medical records could aid in predicting and personalizing the healthcare needs of patients.

Why should the volume of information matter to patients?

Patients and their care teams need access to all the medical evidence in order to make informed decisions about disease management. As big data analysts sort through information focused on disease etiology and cost (e.g., HealthData.gov.),2 it is important to consider the quantity and quality of medical information accessible to the public. Medical information about approved interventions are typically garnered during a costly clinical trial process, which may result in only one out of 10,000 potential new medicines being approved by the FDA.3 While results from 75 trials are published daily in journals,4 a significant proportion of trial results remain unpublished or are reported in a selective manner in papers. Knowledge may be sequestered for various reasons. However, timely access to comprehensive clinical trial results will provide physicians with a totality of objective information in order to enhance patient care. Broader access to this information may also enable trial investigators to target populations most likely to benefit from interventions or shorten the duration of expensive trials. New programs adopted by for-profit industry leaders such as GlaxoSmithKline and Medtronic, guidance (joint principles adopted by PhRMA-European Federation of Pharmaceutical Industries) and requirements by the National, Heart, Lung and Blood Institute and available clinical trial registries are positive steps towards responsible data sharing, the step preceding master data management and publication. The Infosphere platform developed by IBM provides the foundation for data integration, data warehousing, master data management, big data and information governance.

Which e-resources could educate and enhance clinical decision support systems?

Imagine that the controversy surrounding data sharing can be resolved. How could large volumes of data rapidly enter the public domain or be digitized from existing print sources? Ongoing research across different disciplines is reshaping the processing of big data to make it more accessible to patients and physicians. Clinical learning, a unique interactive e-resource, has been developed by Elsevier to teach and enhance the clinical and practical skills of undergraduate medical students. Additional digital resources that could aid researchers in searching for information, analyzing data, enhancing writing, archiving manuscripts, collaborating and integrating changes to documents and presenting their results at suitable congresses or in peer-reviewed journals are presented in the Table. Together with responsible big data analytics, these “writing tools of the trade” may enhance clinical decision support and recommendation systems and enable more efficient post-market surveillance of medical interventions.

References

1.         Abbott R. Big data and pharmacovigilance: using health information exchanges to revolutionize drug safety. Iowa Law Review [e-pub]. 2013:87.

2.         Groves P, Kayyali B, Knott D, S VK. The 'big data" revolution in healthcare. Accelerating value and innovation.: McKinsey & Company (Center for US Health System Reform Business Technology Office),;2013.

3.         Fischer S. What we learn from clinical trials. 2013; http://www.phrma.org/what-we-learn-from-clinical-trials. Accessed August, 2013.

4.         Ross JS, Krumholz HM. Ushering in a new era of open science through data sharing: the wall must come down. JAMA. 2013;309(13):1355-1356.