Commentary
‘Big Data’ is coming. Are you ready?
If there is one phrase you are likely to hear repeated this year, it will be Big Data. IBM has estimated that we create 2.5 quintillion bytes of data every day, which means that 90% of the world’s data have been created in the last 2 years alone. These data come from everything around us – weather sensors, Facebook postings, financial transactions, and, of course, electronic health records.
Big Data analytics can be useful across every human endeavor and health care is no exception. Today’s steady adoption of electronic health record (EHR) systems by the majority of medical practices in the United States is accelerating this trend. Information from these EHR systems allows for analysis of larger data sets than was previously possible as few as 3 years ago. Perhaps equally important, we can now look at a single large set of related data from a broad population of clinicians and compare those data to separate smaller sets from individual physicians or practices. This allows us to utilize those data for a range of purposes, including monitoring prescription trends, carrying out quality assurance activities, embarking on population management for both chronic disease management and preventive health, and measuring outcomes to assess treatment effectiveness.
Making learning more meaningful
The availability of these data creates an opportunity both to increase the effectiveness of continuing medical education (CME) and to measure that increase through its impact on both physician behavior and patient outcomes. Let’s take a look at three distinctly different ways that the integration of EHR data analytics and CME can deliver benefits to clinicians and their patients through improved measurement of health care educational outcomes.
Closing the gap between evidence and practice has always been one of the goals of CME, but the challenge is for clinicians to understand how to translate knowledge into clinical strategy and action in a timely manner. Many of us struggle to keep up with the rapidly expanding amount of medical knowledge (some estimates suggest the world’s knowledge doubles every 3 years). Chances are great that the latest evidence-based recommendations have changed since we last learned about them.
Dr. Jonathan Bertman
An earlier EHR Report ("Clinical decision support," February 2013, p. 51) pointed out some of the limitations of currently available clinical decision support systems. Clearly, there is room for improvement when it comes to reminding clinicians about the importance of evidence-based recommendations during therapeutic and diagnostic interventions.
CME providers are naturally well positioned to offer the latest medical information at the point of care. An intelligent EHR system that identifies correlative patterns revealed through data analytics can present evidence-based recommendations to the clinician in an active, contextual manner based on the needs of an individual patient.
Learning done your way
The ability to customize individual learning plans that address identifiable gaps in care is another important benefit we can derive from Big Data analytics in EHR systems. Physicians and other specialists have specific needs and limited time to meet those needs, making it imperative that CME programming is as targeted as possible. As such, varying patient needs should be reflected in the education of primary care physicians.
With analytics of EHR data, CME providers can start to tailor their programs to address the needs of a clinician’s specific population of patients and patterns of practice. They can apply best-practice metrics to performance to identify specific gaps in care and then develop specific learning interventions to close those gaps.
The ultimate measure of success – patient outcomes
The standard of EHR data analytics is the ability to measure the impact of CME offerings on patient health outcomes and build feedback mechanisms based on educational content development and research. Drawing a direct relationship from learning to physician behavior to patient outcome won’t always be possible, but the data in EHR systems hold the promise of making such correlations with accuracy and timeliness.
The feedback mechanism is perhaps equally important in the development of new learning activities and content that take into account real-world results. A simple 1 to 5 score on an evaluation form doesn’t tell a medical educator anything about what happens after a clinician leaves the lecture hall. With a better understanding of the real-world impact of learning, CME developers can continuously improve their content to better serve the needs of their audiences and patients.
Data privacy concerns
Big Data has the power to transform every facet of our lives. How we use the data from EHRs – for clinical or commercial purposes – is up to us as a community. One way to ensure protection of data privacy rights would be to apply the same principle of informed consent that we use today for decisions related to the course of action we take with a patient. This means clinicians would have the right to know if and how information about their practice and/or patients could be used by another company, as well as details about how these data are identified/de-identified. Clinicians also should have the right to never be automatically opted into a process that uses any practice or patient data. Likewise, the right to opt out should always be one click away.
