WASHINGTON WATCHLINE. The Big Data Revolution in Intensive Care

Transcription

1 WASHINGTON WATCHLINE PHYSICIAN ADVOCACY FOR EXCELLENCE IN THE DELIVERY OF PULMONARY, CRITICAL CARE AND SLEEP MEDICINE June VOLUME 27 No. 6 The Big Data Revolution in Intensive Care In many areas of science, government and business, analyzing very large amounts of information, Big Data, has become a major driver of innovation and success. A talk at the 2017 NAMDRC Annual Meeting has raised our awareness of impending changes in patient evaluation and intervention strategy in the intensive care unit. The talk, Big Data Approaches for Critical Care Medicine, was delivered by John R. Hotchkiss, MD, Associate Professor of Medicine at the University of Pittsburgh School of Medicine. This talk highlighted the dynamics that are going to make significant changes in the practice of critical care medicine in the near future. This Big Data Revolution is a result of cooperative efforts between leading academic centers and the information technology industry. For the most part, treatment decisions in intensive care units have been based on a physician s judgement of the likelihood of a patient s physiologic deterioration or improvement. Traditionally the computing of risk of deterioration has occurred in the physician s mental processes often assisted by conferences with colleagues. The amount of patient data available in intensive care units is expanding, placing a substantial burden on the acting clinician to make quick and effective patient care decisions while making sense of mountains of information. In critical care settings clinicians need to make timely care decisions, and navigating the vast amounts of clinical data available at the patient bedside can be difficult. In 2013, in an AJRCCM Editorial, Dr. Bart Celli observed: The WASHINGTON WATCHLINE is published monthly and provides timely information to NAMDRC members on pending legislative and regulatory issues that impact directly on the practice of pulmonary medicine. NAMDRC s primary mission is to improve access to quality care for patients with respiratory disease by removing regulatory and legislative barriers to appropriate treatment. INSIDE THIS ISSUE: About NAMDRC Membership Benefits NAMDRC Application...6 NAMDRC Leadership...3 Product and Technology News...4 NAMDRC 41th Annual Meeting and Educational Conference will be held: March 22 24, 2018 Omni La Costa Resort Carlsbad, CA NAMDRC 8618 Westwood Center Drive, Suite 210 Vienna, VA Phone: Fax: ExecOffice@namdrc.org In the absence of a practical way to systematically capture, analyze, and integrate the information contained in the massive amount of data generated during patient care, medicine has remained a highly empirical process in which the disconnected application of individual experiences and subjective preferences NAMDRC will directly affect your practice more than any other organization to which you belong.

2 June 2017 VOLUME 27 NO 6 Page 2 continues to thwart continuous improvement and consistent delivery of best practices to all patients. The intensive care unit (ICU) presents an especially compelling case for clinical data analysis. The value of many treatments and interventions in the ICU is unproven, and highquality data supporting or discouraging specific practices are embarrassingly sparse. Guidelines developed to standardize practice are dependent on an evidence base that is surprisingly thin considering the copious data generated in the ICU. In 2015 Dr. Celli wrote: As target populations subdivide along combinations of comorbid conditions and countless genetic polymorphisms, as diagnostic and monitoring devices, including wearable sensors, become more ubiquitous and as therapeutic options expand beyond the evaluation of individual interventions including drugs and procedures, it is clear that the traditional approach to knowledge discovery cannot scale to match the exponential growth of medical complexity. In the last several years there has been a proliferation of increasingly accurate and sophisticated monitoring devices able to provide multiple physiologic parameters. Concomitantly, the technology to collect and store this data has improved and mathematical algorithms have been developed that can superimpose the physiologic data on a patient s electronic medical record. Risk analytics algorithms combine and continuously convert multiple streams of raw physiological data into near real-time clinical information enabling the presentation of patient data in terms of estimated risks to the patient. Cognitive analytics, using evidence-based guidelines, are being studied to help clinicians make more informed decisions when assessing interventions and developing patient care plans. These technologies are ushering in a revolution in ICU practices both to better understand an in-patient s trajectory and optimize therapeutic interventions. Public/Private Partnerships The Massachusetts Institute of Technology (MIT) in partnership with Beth Israel Deaconess Medical Center and Philips Healthcare has developed and maintains the Multiparameter Intelligent Monitoring in Intensive Care database. This database is comprised of highly granular structured data including minute-by-minute changes in physiologic signals as well as time-stamped treatments with dosages enabling modeling of the individual dynamic response to a physiologic insult or clinical intervention. MIT researchers are exploring a cloud-based clinical decision support solution with a novel user interface for the ICU. Boston Children's Hospital initiated a formal partnership with a Boston-based health care technology company, Etiometry, in 2012 to build software to provide ICUs with actionable information through data consolidation, enhanced visualization and predictive analytics. Volumes of data from each ICU patient are captured, anonymized, and archived as big data resources. Etiometry has now recorded over 10 million hours of patient monitoring data accessible to approved research projects. The Mayo Clinic, Philips Healthcare, Ambient Clinical Analytics and The United States Critical Illness and Injury Trials Group received an award from the Center for Medicare and Medicaid Innovation (CMMI) to research the impact of using a cloud-based clinical decision support solution with a user interface for the ICU. This group developed the IntelliSpace Console under the product name AWARE (Ambient Warning and Response Evaluation). The Mayo Clinic has also studied an analytics platform developed by Israeli company, Intensix, for detecting and predicting sepsis in the ICU.

3 June 2017 VOLUME 27 NO 6 Page Officers & Board of Directors OFFICERS Charles W. Atwood, MD President James P. Lamberti, MD President-Elect Maida V. Soghikian, MD Secretary/Treasurer Timothy A. Morris, MD Past President Board of Directors Robert J. Albin, MD Albee Budnitz, MD Kent L. Christopher, MD Gerard J. Criner, MD Peter C. Gay, MD Thomas B. Hazlehurst, MD Nicholas S. Hill, MD Theodore S. Ingrassia, III, MD Steve G. Peters, MD Valerie A. Schneider, MD PRESIDENT S COUNCIL George G. Burton, MD John Lore, MD Louis W. Burgher, MD, Ph.D. Alan L. Plummer, MD E. Neil Schachter, MD Frederick A. Oldenburg, Jr., MD Paul A. Selecky, MD Neil R. MacIntyre, MD Steven M. Zimmet, MD Joseph W. Sokolowski, MD Peter C. Gay, MD Steve G. Peters, MD Lynn T. Tanoue, MD Dennis E. Doherty, MD Executive Director Phillip Porte Associate Executive Director/ Legislative Affairs Sarah Walter Associate Executive Director Karen Lui, RN, MS Director Member Services Vickie A. Parshall At Emory University Hospital, an ICU research team is making use of a streaming analytics platform developed by IBM that works together with an Excel Medical Electronics bedside monitor data aggregation application to provide quick analysis of big data. Clinicians are able to identify changes in a patient's condition including anomalies through analysis of continual data that comes in through bedside monitors. The University of Texas Health Science Center at Houston, decided to explore methods to change the retrospective nature of analytics. Their research resulted in a technology platform that was taken private as Decisio Health. The platform has been adopted at the University of Cincinnati Medical Center neuroscience ICU. The Problem of Comprehensibility The potential power of these systems will raise the importance and value of comprehensibility i.e. the ability of the various stakeholders to understand relevant aspects of the modeling process. As researchers construct increasingly sophisticated algorithms, the processes that result in patient outcome predictions and therapeutic recommendations will be increasingly difficult for the clinician to comprehend. The accuracy of predicted clinical outcomes is dependent on the specific mathematics chosen to develop the algorithms as ever-growing datasets are used as input to sophisticated computational processes constructing complex models subjected to human interpretation. In his talk, Dr. Hotchkiss touched on several important issues. The accuracy of predicted clinical outcomes is dependent on the specific mathematics chosen to develop the algorithms. Beginning with an accurate grouping of patients, disease states and physiologic parameters are all crucial to developing useful clinical models. There are a variety of techniques to group clinical data points including, clustering approaches, artificial neural networks and natural language processing. Each of these have areas of strength and weakness. Using the wrong data grouping technique at the very beginning of applying data analytic processes can result in misleading clinical recommendations. Cyber Attacks on Big Data It is prudent to recognize that ransomware attacks increasingly target medical institutions. Unlike traditional cyberattacks designed to exfiltrate records, delete data or physically damage computing systems, ransomware attacks appear to be on the rise due to the ease in which such extortion translates directly to money in the pockets of cyber criminals. U.S. medical facilities have been subject to ransomware attacks, including hospitals in California, Indiana, Kentucky, Maryland and Texas. Developers of the software products that will affect clinical decisions in the intensive care unit must find ways to protect them.

4 June 2017 VOLUME 27 NO 6 Page 4 PRODUCT AND TECHNOLOGY NEWS! NAMDRC is providing this space to our benefactors and patrons who provide us with information about new products and innovations related to pulmonary medicine. NAMDRC reserves the right to edit this copy as appropriate.

5 June 2017 VOLUME 27 NO 6 Page 5 NAMDRC Membership Benefits AT A GLANCE... Monthly publication of the Washington Watchline, providing timely information for practicing physicians; Publication of Current Controversies focusing on one specific Pulmonary/Critical Care Issue in each publication; Regulatory updates; Discounted Annual Meeting registration fees; The Executive Office Staff as a resource on a wide range of clinical and management issues; and The knowledge that NAMDRC is an advocate for you and your profession. One of NAMDRC s primary reasons for existence is to provide both clinicians and patients with the most up-to-date information regarding pulmonary medicine. Bookmark this page! The complexity of our nation s health care system in general, and Medicare in particular, create a true challenge for physicians and their office staffs. One of NAMDRC s key strengths is to offer assistance on a myriad of coding, coverage and payment issues. In fact, NAMDRC members indicate that their #1 reason for belonging to and continuing membership in the Association is its voice before regulatory agencies and legislators. That effective voice is translated into providing members with timely information, identifying important Federal Register announcements, pertinent statements and notices by the Centers for Medicare and Medicaid Services, the Durable Medical Equipment Regional Carriers, and local medical review policies. ABOUT NAMDRC: Established over three decades ago, the National Association for Medical Direction of Respiratory Care (NAMDRC) is a national organization of physicians whose mission is to educate its members and address regulatory, legislative and payment issues that relate to the delivery of healthcare to patients with respiratory disorders. NAMDRC members, all physicians, work in close to 2,000 hospitals nationwide, primarily in respiratory care departments and critical/intensive care units. They also have responsibilities for sleep labs, management of blood gas laboratories, pulmonary rehabilitation services, and other respiratory related services.

6 June 2017 VOLUME 27 NO 6 Page 6 NAMDRC MEMBERSHIP APPLICATION MEMBERSHIP DUES SCHEDULE (Dues for first year include $75.00 Initiation Fee) Individual and Small Group Dues. $ Includes groups of up to 6. Please include contact information for all members. TWO EASY WAYS TO BECOME A NAMDRC MEMBER 1. Go to and register for membership online. 2. Mail this application to: NAMDRC 8618 Westwood Center Drive, Suite 210 Vienna, VA Please print clearly or type: NAME (LAST) (FIRST ) (MIDDLE INITIAL) DEGREE ADDRESS CITY STATE ZIP CODE TELEPHONE FAX FACILITIES WITH WHICH YOU ARE AFFILIATED GROUP MEMBERSHIP DUES (For larger groups, please attach a list of names. If a group member wishes to receive mailings at an address other than that indicated above, please attach appropriate information.) Groups of $1, Groups of $1, Groups of $1, TOTAL PAYMENT DUE $ PAYMENT Enclosed is a check payable to NAMDRC (U.S. Dollars) Change my credit card for total payment due American Express VISA MASTER CARD CREDIT CARD NUMBER _ EXPIRATION DATE SECURITY CODE NAME AS IT APPEARS ON CREDIT CARD BILLING ADDRESS (IF DIFFERENT FROM REGISTRATION) _ SIGNATURE ln accordance with IRS Regulations, 95% of your 2O17 Annual Dues are tax deductible. NAMDRC s Federal TAX ID # is FOR MORE INFORMATION, CONTACT NAMDRC Phone: Fax: ExecOffice@namdrc.org Web Site: