Clinical Analytics and Health Information Exchange: Two Technologies Come Together at the Moment of Need Author: Rodney Hamilton, MD Chief Medical Information Officer, Informatics Corporation of America
Clinical Analytics and Health Information Exchange: Two Technologies Come Together at the Moment of Need Rodney Hamilton, MD, Chief Medical Information Officer, ICA It s an old story: healthcare is awash in data but starved for information. The advent of electronic and digital processing has only accentuated the problem by enabling and encouraging the creation and storage of more data. However, storing, transferring and, above all, making that data usable remains the key nagging issue as ever more data is created and stockpiled. And while even today, many, if not most, physician offices contain vast shelves of handwritten clinical notes on their patients, the relentless advance of electronic clinical information will eventually make those hand-written mountains look like mole-hills by comparison. And what will we have gained? Because of the sporadic and ad-hoc nature of how IT slowly worked its way into healthcare over several decades, clinical data has been input, displayed and analyzed by countless clinical systems that have no, or very little, alignment with one another. Now, as the exchange of clinical information is increasingly seen as the critical linchpin to improving outcomes and lowering costs, we find many, if not most, clinical datasets to be a mess and very difficult to work with. 1 In fact, with the rapid proliferation of healthcare information technology as a result of recent government programs such as HITECH and the Accountable Care Act, healthcare data is making quantum leaps in volume, velocity, variety and complexity. Consequently, healthcare is looking for tools to help get its arms around this volume of data to reorganize, re-interpret and re-align it for use by multiple stakeholders across multiple institutions and regions. Clinical analytics and business intelligence tools have emerged as a top priority for hospital IT leaders who are moving towards accountable care adoption, according to a Black Book Rankings survey. The survey finds that more than 1,340 hospital IT leaders nationwide indicated clinical analytics to be their highest priority system for acquisition over the next year. 2 Of course the great booster rocket in all of this has been Healthcare Reform. It is entirely likely that healthcare would have continued muddling along in its fragmented fashion, building unaligned systems that create, store and deposit unaligned data of limited use. But the incentives built into the physician use of EMRs, and the broader use of clinical information exchange mechanisms, such 1 John Moore, Founder and Managing Partner, Chilmark Research, Digging into Analytics, July 16, 2012 2 Erin McCann, Associate Editor, Healthcare IT News, Clinical analytics next big thing for health IT, August 01, 2012 1 1
as health information exchange (HIE), have forced the industry to see the light and events are now moving rapidly and relentlessly. In fact, the recently released Meaningful Use Stage 2 rules require, and ONC has announced that it will enforce, cross-vendor data exchanges in an attempt to eliminate silos or walled gardens 3, making the exchange of clinical information not only nice but imperative. Breaking down these silos and garnering collaboration within the vendor community will be an essential step in the process of true transformation of the healthcare delivery system into a system of delivering health. The Current Buzz around Analytics As John Moore of Chilmark Research reported from HIMSS12, the conference buzz was analytics. [It] seemed as if every vendor in attendance was promoting some sort of analytics capability. It was a tad overwhelming and if I was having a problem trying to separate out all these countless vendor offerings (this is exactly what I am trained to do as an industry analyst) I couldn t imagine what it must have been like for those representatives of healthcare IT departments. 4 As it turns out, clinical analytics, and its broader umbrella, business intelligence, is more than a buzz. More than 80 percent of healthcare organizations are investing in business intelligence tools for clinical performance improvement, according to a new report from the Healthcare Financial Management Association (HFMA). 5 About a quarter are spending money for data warehouses, and a slightly smaller portion is also buying new costing systems. Despite the low percentage of institutions that have the latter capability, the HFMA points out that they will need it to succeed in a value-based payment environment. Very small percentages of respondents today have significant ability to attribute per patient costs across the care continuum and few organizations are able to quantify the financial impact of quality improvements, the report noted. This skill will be important as organizations determine how to reduce their cost structure over time to remain market-competitive. 6 Making these challenges more acute is the rise of forward thinking accountable care models such as patient-centered medical homes and ACOs. These models of care and reimbursement require the sharing of comprehensive patient data among care teams, health plans, and government entities. The operating environment may consist of various EHR systems, a regional HIE, multiple hospital information systems and departmental applications all deployed within the 3 Farzad Mostashari, M.D., National Coordinator for Health IT, FierceHealthIT, Mostashari: ONC will enforce cross-vendor data exchanges, Julie Bird, August 29, 2012 4 John Moore, Founder and Managing Partner, Chilmark Research, Digging into Analytics, July 16, 2012 5 Healthcare Financial Management Association, Defining and Delivering Value, June 2012 6 Healthcare Financial Management Association, Defining and Delivering Value, June 2012 2
same integrated delivery network (IDN). However, responsibility for providing data and making clinical decisions is shared among acute care, post-acute care, long-term care, primary care, and specialty providers. Yet these teams may not have the data-sharing and communication tools they need. 7 A recent randomized controlled trial sponsored by Brigham and Women s Hospital in Boston and published in JAMIA confirms that while clinical decision support lists are critical for patient care, clinical decision support, population reporting, quality improvement, research and problem lists are often incomplete or out of date. 8 This incompleteness results in less than effective diagnosis and treatment of patients that can impair outcomes and increase redundancies and costs. The Role of HIE in Clinical Analytics In his rumination about HIMSS12, John Moore also likened the current buzz in clinical analytics to a similar buzz a few years earlier about HIE, when that market was first emerging and bursting at the seams with excitement and promise. Turns out the two may go hand-in-hand and in fact both bolster the effects of the other. As stated, the problem now isn t a dearth of clinical data. The problem is rapidly becoming too much data housed in various and disparate storage systems in different formats that are difficult to access and often impossible to align in a usable way for physicians treating patients. Hence the phrase: messy data. Many of the elements of HIE have been built to address this quandary. It s one thing to pull and aggregate usable structured data from rational systems. It s another to access combinations of non-structured and discrete data from non-friendly systems with conflicting information and be able to interpret that data in real-time while treating a patient individually, or at a population level, to make broader management decisions. HIEs now have the ability, through laying down a foundation of interoperability, of accessing, aggregating and exchanging clinical data via the standards-based exchange of continuity of care documents (CCDs) between electronic health records. Through technology that utilizes integration standards, such as Direct, IHE, HL7, and others, clinicians are able to access exchanged data directly through their EHRs. By aggregating and normalizing both the HL7 and CCD content made available to the HIE, the content can be organized and stored in a clinical data repository (CDR). The CDR can then be accessed by other sources such as clinical analytics applications, care management tools, personal health records, alerting and notification applications, etc. These capabilities and tools form the basis for the access, synthesis and use of structured and unstructured data that will take the exchange of clinical information to the next level. 7 InterSystems Corporation, Active Analytics: Driving Informed Decisions Leading to Better Clinical and Financial Outcomes, 2012 8 Adam Wright, Justine Pang, Joshua C Feblowitz, Francine L Maloney, Allison R Wilcox, Karen Sax McLoughlin, Harley Ramelson, Louise Schneider,David W Bates, Improving completeness of electronic problem lists through clinical decision support: a randomized, controlled trial, JAMIA, July 27, 2012 3 3
The Lines Begin to Blur As such, clinical vendors, even niche vendors, while continuing to expand their HIE and analytics offerings to meet these emerging demands, are finding greater emphasis being placed on population health capabilities and tools that help non-affiliated organizations exchange and analyze both discrete and unstructured information: the line between HIE and clinical analytics is rapidly blurring. These tools must work together to identify, source, translate and analyze the available data. Hospitals will increasingly require both sets of tools as the exchange of clinical information becomes more ubiquitous, and the demand for accountability becomes more dominant. Given the demand for greater accountability, delivery organizations will need to use information and data from an increasingly broad community of care partners to help reduce costs and manage quality and outcomes. Hospitals and health systems will need to exchange data with non-affiliated organizations and actually use that data to influence care decisions. Providers will need to be alerted to gaps in care such as overdue tests and procedures, and it will be just as important to understand what services have already been provided so as to avoid duplication. Simply aggregating patient data from disparate systems across multiple settings of care will not be sufficient; organizations will also need robust analytics tools that can spot trends across a defined population, identify high risk patients, and gauge performance on required metrics. 9 The common ground of HIE and clinical analytics are data aggregation and storage, normalization, and workflow support and integration, all from a universe of capabilities that drive connectivity, messaging, population health management, predictive analytics and decision support (See graphic). The blending of HIE and analytics is the inevitable next step on the path toward the seamless exchange of clinical information drawn from often rival organizations using different systems that may house unaligned, incorrect and generally messy data. The power of these technologies together provides a solution that can extract, translate and synthesize that data to turn it into usable clinical information giving physicians and providers the information they need when they need it to improve patient outcomes and reduce costs. Converging HIE & Analytics Capabilities Data exchange & connectivity Patient indexing Provider messaging & communication Patient messaging communication & self-service Data aggregation & storage Data normalization Workflow support & integration Population health management / registries Predictive analytics & risk management Utilization & performance management Point-of-care decision support 10 9 Jason D. Fortin, The Imperative for Robust HIE and Analytics Tools, Healthcare Informatics, September 5, 2012 10 Jason D. Fortin, The Imperative for Robust HIE and Analytics Tools, Healthcare Informatics, September 5, 2012 4
About ICA Originating from clinical informatics solutions developed within Vanderbilt Medical Center, ICA s state-of-the-art CareAlign interoperability and informatics platform addresses and solves data and communication challenges for many healthcare entities, including hospitals, IPAs, IDNs, HIEs, payers and others. CareAlign delivers a flexible architecture to connect, collect, consume and intelligently distribute data through Direct, IHE, HL7, and custom methods for use in EHRs, third party applications, and ICA s applications. CareAlign unites a wide range of information and supports analytic needs associated with population health management, transitions of care communication, re-admissions reduction, meaningful use requirements and PCMH/ACO operations. Visit www.icainformatics.com, and follow us on Twitter, ICA HITme Blog, Facebook, LinkedIn and YouTube. 1801 West End Avenue, Suite 1000 Nashville, TN 37203 615-866-1500 www.icainformatics.com 2012 ICA. All rights reserved. No reproduction or reuse without permission. WP-12/03/12