Clinical Decision Support Tools

Abstract

It is difficult to imagine that practice guidelines, randomized controlled trials, evidence-based medicine, and similar tools of current practice did not exist a generation ago. We now have a rich repository of guidelines, and with the advent of electronic health records, they can be translated into patient care through the use of clinical decision support tools. The AGA understood the importance of its leadership in this area and organized development of such tools within its “Roadmap to the Future of GI Practice.” In this month’s Practice Management section, Dr Lawrence R. Kosinski, an expert in the field of clinical decision support, shares some of his insights into this segment of The Roadmap.John I. Allen, MD, MBA, AGAF Special Section Editor First we build the tools, then they build us. Marshall McLuhan1 The field of information technology has become embedded within the practice of medicine. As a result, several aspects of our profession that traditionally have been the sole province of human intelligence are now conducted by computers with greater precision and at a vastly greater speed. Every time you send an e-mail or connect a cell phone call, intelligent algorithms optimally route the information. Intelligent algorithms automatically detect credit card fraud, fly and land airplanes, guide intelligent weapons systems, assemble products in robotic factories, and play games such as chess and routinely defeat human beings.1Kurzweil R. How to create a mind. Penguin Books, Ltd, London2012Google Scholar We are building tremendous tools and they are now increasing what we can accomplish. The increase in the use of health care information technology also has come at a very critical point for our profession as discussed in the “Practice Management: The Road Ahead” section earlier this year by Weinstein.2Weinstein M. How a practice chooses an electronic health record.Clin Gastroenterol Hepatol. 2012; 10: 1187-1189Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar The cost of health care continues to increase at a rate that is no longer sustainable (soon to reach 18% of the gross domestic product). Purchasers and payers of health care are no longer willing to bear the financial risk alone and are altering our reimbursement by transferring some risk to providers. We now will be responsible for both provision of care and some of its expense. To decrease our cost and successfully manage the financial risk of health care, gastroenterologists must do the following: (1) ensure that each health care professional is working up to the level of his/her licensure; (2) direct provision of care to the lowest level of capable professional; (3) define and limit the scope of lower-level professionals so they work at maximum efficiency; (4) establish seamless communication within a health care system; and (5) report pertinent outcomes data to external repositories and registries.3Clayton C.M. The innovator's prescription. McGraw-Hill, New York2009Google Scholar Electronic health records (EHRs) are essential to accomplishing these objectives, but not without substantial changes. The EHR of the future will not be the proprietary and isolated version we use today, but rather one whose user interface is tailored to usual workflow and actual clinical experience and whose database is tied to a health information exchange that is integrated with multiple outcomes registries. In addition, Korman4Korman L.Y. Standardization of endoscopy reporting: deja vu all over again?.Clin Gastroenterol Hepatol. 2012; 10: 956-959Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar pointed out the need for standard language in developing gastroenterology-related EHRs. The user interface will derive much of its strength and power from judicious use of clinical decision support (CDS) tools: information technology artifacts that create a process for enhancing health-related decisions and actions with pertinent, organized, clinical knowledge and patient information, to improve health and health care delivery. Most of us are accustomed to decision support tools such as global positioning system devices that tell us where we are or Internet search engines that help us make choices. Health care–related CDS tools are similar types of devices designed to help us make better medical decisions. We see them regularly as drug–drug interactions, drug–allergy interactions, and dose range checking. Those who have implemented EHRs are using them with evidence-based order sets and standardized system-wide clinical policies.5Osheroff J.A. Teich J.M. Levick D. et al.Improving outcomes with CDS: an implementer's guide. HIMSS, Chicago, IL2011Google Scholar The Centers for Medicare and Medicaid Services describe CDS tools as “health information technology functionality that builds upon the foundation of an EHR to provide persons involved in care processes with general and person-specific information, intelligently filtered and organized, at appropriate times, to enhance health and health care.”6Osheroff J.A. Teich J.M. Middleton B.F. et al.A roadmap for national action on clinical decision support.J Am Med Inform Assoc. 2007; 14: 141-145Crossref PubMed Scopus (391) Google Scholar The 2012 Institute of Medicine report, “Best Care at Lower Cost: The Path to Continuously Learning Health Care in America,” makes recommendations for CDS tools by stating, “Accelerate integration of the best clinical knowledge into care decisions. Decision support tools and knowledge management systems should be routine features of health care delivery to ensure that decisions made by clinicians and patients are informed by current best evidence.”7Best care at lower cost: the path to continuously learning health care in Americahttp://www.nap.edu/catalog.php?record_id=13444Google Scholar These Institute of Medicine recommendations reflect the value of CDS tools because they improve quality, maintain safety, and decrease cost. Outcomes also are improved by CDS because they decrease errors of omission and commission; reduce unnecessary, ineffective, or harmful care; and promote adherence to evidence-based care.8Wolfstadt J.I. Gurwitz J.H. Field T.S. et al.The effect of computerized physician order entry with clinical decision support on the rates of adverse drug events: a systematic review.J Gen Intern Med. 2008; 23: 451-458Crossref PubMed Scopus (235) Google Scholar The final rule on meaningful use stage 29Department of Health and Human Services, Centers for Medicare and Medicaid Services. 42 CFR. Parts 412, 413, and 495.Google Scholar also has increased the requirements for CDS tools. In stage 1, eligible professionals needed only to implement one CDS rule relevant to specialty or high clinical priority. In stage 2, eligible professionals must use CDS to improve performance on 5 high-priority health conditions, support querying of immunization registries, and identify reportable conditions. Clearly, we all must embrace CDS tools. Key elements of a successful CDS implementation within your practice should include the following:•Support for the program should come from all levels of the organization•Key stakeholders must be involved in all aspects of design and implementation•A clinical champion should lead the effort•A multidisciplinary CDS committee should manage the entire process•CDS goals should align with organizational strategic goals•Ongoing monitoring and fair process communication with affected clinicians will increase the chances of successful implementation. The ultimate goal of a CDS program is to follow Osheroff's10Osheroff J.A. Improving medication use and outcomes with clinical decision support: a step by step guide. The Healthcare Information Management and Systems Society, Chicago, IL2009Google Scholar CDS Five Rights, which seek to provide the right information, to the right person, in the right format, through the right channel, at the right point in clinical workflow to improve health and health care decisions and outcomes. The CDS Five Rights approach is also a framework for setting up and optimizing CDS interventions to address priority objectives. It will be challenging for you to implement an effective CDS program in your practice, but the enhanced quality and efficiency that you will realize can make this effort worthwhile. Commercially available tools embedded in EHRs have generally unreliable alerts and insufficient application of usability standards. These inherent problems limit our ability to create CDS tools that are effective and efficient. Small practices do not have the required personnel or resources to create templates and deploy them in quantities sufficient to impact overall care. Ideally, CDS tools would be standardized and we could pull from a repository of validated and clinically useful examples. Unfortunately, there is currently a lack of standardization and sharing between and among health care organizations with respect to CDS tools. Each of our EHRs is proprietary with a unique database that would require customized programming to implement a CDS tool. Clearly, we need a better solution. The American Gastroenterological Association (AGA) is proactively trying to develop CDS tools and definitions for 3 clinical service lines: colon cancer prevention, inflammatory bowel disease, and management of chronic hepatitis C. The AGA is using or creating updated guidelines, deriving performance measures, creating standard order sets, and producing clinical management algorithms with grades of evidence for each decision point.11Braden G. Allen J. Organizing your clinical service line: colon cancer prevention.Clin Gastroenterol Hepatol. 2013; 11: 2-5Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar The AGA has developed a series of expert panels to aid in this effort and has engaged major EHR companies to develop test programs to validate processes and measures outcomes. The following 3 well-funded strategic national projects are determined to provide us with solutions to our CDS implementation issues. The National Quality Forum convened a CDS Expert Panel to develop a classification and categorization model of the CDS information necessary for quality improvement. It built on the work of the National Quality Forum Health Information Technology Expert Panel (HITEP) and is supported by the Agency for Healthcare Research and Quality. HITEP I released a model to facilitate the development, use, and reporting of quality measures from EHR systems. HITEP II developed the quality data set, which was designed to centralize and maintain a repository of quality data requirements and definitions.12National Quality Forum, Driving Quality and Performance MeasurementA foundation for clinical decision support: a consensus report. National Quality Forum, Washington, DC2010Google Scholar A second endeavor is the CDS Consortium, another Agency for Healthcare Research and Quality–funded research collaborative involving multiple Academic Medical Centers and EHR vendors based at Partner's Healthcare in Boston. Their clinical partners include Partners Healthcare, the Regenstrief Institute, Veterans Health Administration, University of Texas Health Science Center at Houston, Oregon Health Sciences University, Kaiser Permanente, Geisinger Health System, Mayo Clinic, and others.13CDSCOverview.http://www.partners.org/cird/cdsc/overview.aspGoogle Scholar The vision of the CDS Consortium is to create a system that permits sharing of CDS elements across organizations, provides rules and other CDS elements in file formats that can be directly instantiated as executable decisions supported locally and within any vendor's EHR, and provides CDS in the Cloud. The Office of the National Coordinator for Health Information has initiated a bold project that will identify, define, and harmonize standards to facilitate emergence of systems and services to share CDS tools by standardizing structured medical knowledge in a shareable and executable format, and defining how a system can interact with and use a Web service to provide helpful, actionable clinical guidance. The purpose of the initiative is to facilitate integration of CDS interventions into an EHR in a standard manner and to create standards to link recommendations to implementable actions.14Health edecisionshttp://wiki.siframework.org/Health+eDecisions+HomepageGoogle Scholar CDS tools will mold and shape the practice of medicine in the future. We must succeed in standardizing the process and making the information uniformly available (Figure 1). For CDS tools to have an impact they will require creation of cloud-based tools that are no longer proprietary to specific EHRs. Providers cannot do this in isolation. We will need to bring patients into the process through the development of “hovering tools” designed to provide us with information about them on a real-time basis.15Asch D.A. Muller R.W. Volpp K.G. et al.Automated hovering in health care--watching over the 5000 hours.N Engl J Med. 2012; 367: 1-3Crossref PubMed Scopus (126) Google Scholar Several companies have been focused on developing hovering tools that mirror experiences people see in computer-based games to promote participation. CDS tools have been used recently as part of a multipronged intervention to manage patients with chronic liver disease.16Wigg A.J. McCormick R. Wundke R. et al.Efficacy of a chronic disease management model for patients with chronic liver failure.Clin Gastroenterol Hepatol. 2013; 11: 850-858Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar The Illinois Gastroenterology Group recently developed a program we call Project Sonar. It is an initiative to improve communication with our patients with inflammatory bowel disease. We created a cloud-based repository of CDS tools that are accessible from our EHR. Patients are sent queries in real time that contain an abbreviated Crohn's Disease Activity Index to assess their symptom quotient. These data then are fed into our database where we are developing artificial intelligence to determine the course of action based on the CDAI score and its rate of change. The value of Project Sonar is that it is scalable to other practices, even using other EHRs. There is just not enough bandwidth in most practices to create them in-house. If we are to succeed and thrive in a future value-based reimbursement system, we need to efficiently deploy our practice assets to obtain maximum effect. We must increase our return on assets to succeed in an era of diminishing resources.