Clinical Decision Support Track

---

The goal of clinical decision support (CDS) is to ensure that optimal, usable, and effective clinical knowledge is widely available to health care providers, patients, and individuals where and when they need it to make high quality, cost-effective health care decisions. The study of CDS encompasses a variety of tools and interventions such as computerized alerts and reminders, clinical guidelines, order sets, patient data reports and quality dashboards, documentation templates, diagnostic support, and clinical workfl ow tools. Achieving desirable levels of patient safety, care quality, patient centeredness, and cost-effectiveness requires that the health system optimize its performance through consistent, systematic, and comprehensive application of available health-related knowledge – that is, through appropriate use of CDS.

Learning Objectives

• To learn a framework for developing, deploying and assessing clinical decision support.
  
• To acquire techniques for implementing specific clinical decision support interventions.
  
• To appreciate how clinical decision support may be deployed to enhance patient safety and disease management.
  
• To review and gain an understanding of key lessons learned by clinical decision support implementers.
  

S03 Invited Panel
Tuesday, May 22, 2007, 10:30 am - 12:00 pm

Improving Outcomes with Clinical Decision Support

Clinical decision support (CDS) is increasingly recognized as a vital component of the electronic health record. It allows clinicians to take full advantage of knowledge and data across the spectrum of care in order to provide optimal, evidence-based care. This panel will address the continuum of decision support through two major foci: description of a six-step process for developing CDS based on an organization's goals; and case studies that help illustrate the process for developing, implementing and monitoring CDS and how this process can change clinical care.

- Robert Jenders, Cedars-Sinai Medical Center and University of California, Los Angeles
- Dean Sittig, Kaiser Permanente and Oregon Health & Science University
- Bimal Desai, University of Pennsylvania
- William Galanter, University of Illlinois Chicago
- Mark Graber, Department of Veteran Affairs

S07 Presentations--Developing Clinical Decision Support
Tuesday, May 22, 2007, 1:30 pm - 3:00 pm

Using Collaboration Technology to Achieve Order Set and Care Standardization: Contemporary Approaches, Trends, and Lessons Learned

This presentation provides a practical overview of collaboration technologies that can be used by health systems to standardize care, improve committee efficiency, reduce committee work, and achieve clinical consensus on the path toward CPOE deployment.

- Timothy McNamara, HealthGate Data Corp.

Knowledge Elicitation Methods and Development of an Evidence-based Knowledge Base for Clinical Decision Support

The objective of this mixed-methods study (RO1-LM-06967) was to develop an evidence-based knowledge base for the clinical problem of determining feeding readiness in premature infants. While the clinical problem is narrow, with well-defined boundaries, the purpose of this presentation is to demonstrate the richness of the methods for extracting knowledge from experts and from the literature. In addition, the author will address some of the challenges facing informaticians when attempting to construct decision support systems that focus on decisions about complex, costly problems faced by clinicians in delivering care to vulnerable populations.

- Teresa Panniers, George Mason University

IT Support for Clinical Workflow: Definitions, Requirements and an Example of a Health Care Workflow Management System

Many health care organizations recognize the value of information systems for providing data and information critical to the delivery of health care, and most use a variety of clinical and administrative systems to support patient care. However the use of information systems to support the processes related to care delivery, or workflow, is a relatively recent development. A unique process for the collection of baseline and post-implementation metrics will be described along with plans for the development of additional automated workflows.

- Cynthia Mascara, Siemens Medical Solutions

Assessing the Utility of a Cognitive Approach to Detect Adverse Drug Events (ADEs) in Clinical Notes

Manual chart review remains the gold standard for detection of ADEs but is expensive and resource intensive. Automating the chart-review process might improve efficiency. The presenter will discuss a cognitive approach to investigate textual signals used by chart-reviewers for ADE detection, with a subsequent aim of automating this reasoning process.

- Shobha Phansalkar, University of Utah

S12 Presentations--Assessing Clinical Systems
Wednesday, May 23, 2007, 8:30 am - 10:00 am

Lessons in Using Clinical Decision Support for Chronic Disease Management

Chronic diseases are prevalent, costly, and preventable. Health information technologies, such as clinical decision support systems (CDSS), can be implemented to support chronic disease care through delivery of evidence-based best practices and education to providers and patients. The presentation will outline grants in the Agency for Healthcare Research and Quality (AHRQ) health IT portfolio that demonstrate trends in application of CDSS to chronic disease care. Information is based on efforts by the AHRQ National Resource Center for Health IT to collect, discuss, and extract lessons from the relevant AHRQ projects. The practical lessons drawn from the portfolio will help the nation move forward with greater adoption and use of clinical decision support systems to improve the quality, safety, and efficiency of health care for all Americans.

- Brian Dixon, Regenstrief Institute

Assessing the Quality of Individualized Risk Estimates in Clinical Decision Support Systems

Predictive models are being developed in virtually every medical specialty for the purposes of diagnosis or prognosis. While some of these models are not widely published, others are endorsed by the NIH and other reputable institutions. Clinical predictive model tools are being used by lay persons as well as health care professionals. The goal of these tools is to provide an individualized prediction of future health care related events such as prognosis in breast cancer given specific information about the individual. These tools are usually based on models synthesized from data with a fine granularity of information.

- Lucila Ohno-Machado, Harvard Medical School

Representation of Smoking-related Concepts in an Electronic Health Record

Data-driven models, created using the process of knowledge discovery in databases (KDD), can be used to predict characteristics or outcomes of patients, and thus aid decision-making. These models possess the capacity to augment evidence-based decision making related to smoking cessation on the basis of individual patient characteristics. A first step in creating such clinical knowledge models, this study will (1) assess the feasibility of a KDD project related to smoking cessation, and (2) map strengths and gaps of representation of smoking cessation related concepts in a commonly used electronic health record system (EHR), EpiCare EMR. A sample of patient records will be assayed then mapped to smoking cessation guidelines. The results of this exploratory analysis will be summarized in a master table and used to guide planning of a KDD project. Additionally, a map of data elements linked to smoking cessation concepts will be created. This map will describe strengths and gaps in the representation of smoking cessation related concepts in an EHR, important for both secondary analysis and KDD projects related to smoking cessation.

- Mollie Poynton, University of Utah

Improving Clinical Decision Support: the VHA Experience with Terminology Standardization

Sharing of computable data and use of automated functionality are greatly enhanced when terminology is standardized within and across facilities. This presentation will describe the Veterans Health Administration's success with data standardization in two critical areas across 1400 facilities. The session will focus on methods used in the standardization of allergies and clinical document note titles. More than 700,000 historical free text allergies were standardized and are now functional in automated decision support and 156,000 obscure and cryptic document note titles are now represented by clear, concise standard titles. The resulting impact on patient safety has been applauded by clinicians and informaticians alike.

- Catherine Chandler Hoang, US Veteran's Administration

S17 Panel
Wednesday, May 23, 2007, 10:30 am - 12:00 pm

Integrating On-line Patient Safety Reporting into Hospital Leadership, Nursing Practice, PI, and Risk Management

The strategies for incorporating patient safety principles into an integrated program will be explored by the panel. The latest phase of their patient safety journey has led them to empower their staff and physicians to report on-line. The panel presentations will feature administration, nursing, risk and performance improvement leadership who are intimately involved with the entire patient safety program. They will each describe from their role, the unique challenges and successes they have seen over the last 10 years. Included in their presentations will be the structure, process, and outcomes achieved with this model. They will discuss the advantages and opportunities this endeavor has created in each of their focus areas.

- Donna Scott, SafeCare Systems, Division of Apptis, Inc.
- Calvin Glidewell, University Community Hospital
- Debora Martoccio, University Community Hospital
- Brad Bjornstad, University Community Hospital

S22 Panel
Wednesday, May 23, 2007, 1:30 pm - 3:00 pm

Decision Support for Clinical Care Teams: System Strategies and Challenges

The timing of decision support is not just at the point of care, but at the point of need. The knowledge bases underlying these decision support systems must be both up to date, and provide accurate and relevant information that recognizes the roles, responsibilities, and capabilities of the disparate set of stakeholders. This panel will discuss the ways electronic health records can be used to both implement and evaluate clinical decision support systems in the management of chronic disease and help health care team members work better with each other.

- Mark Weiner, University of Pennsylvania School of Medicine
- Khalid Moidu, Purdue University
- Stephen Morgan, Partners Health Care
- David Dorr, Oregon Health & Science University

S27 Panel
Thursday, May 24, 2007, 8:30 am - 10:00 am

Electronic Health Care--Addressing the Information Security and Privacy Issues

Electronic health systems are becoming a prevalent entity in the American landscape. The deployment of the Vista system by the Veterans Adminstration offers a proof point on the viability and usefulness of these systems, hints to the evolution of these systems, and highlights the concerns that must be addressed for acceptance and use by practitioners and patients. This panel will discuss the issues; going from policy to technology enablers.

- Tyrone Grandison, IBM Almaden Research Center
- John Mike Davis, Veterans Health Administration
- Ted Cooper, Stanford University
- Bernd Blobel, University of Regensburg
- Bill Cody, Almaden Research Center

Clincial Decision Support Posters

Poster Session One
Tuesday, May 22, 2007, 3:30 pm - 5:00 pm

Use of Clinical Decision Support Software to Increase Documented Follow-up for Outpatients who Screen Positive for Alcohol Misuse, Carol Achtmeyer, Puget Sound VA Medical Center

Clinical Decision Support Tools: Focus on Dietary Supplement Databases, Kevin Clauson, Nova Southeastern University

Using Wireless Handheld Computers with Clinical Decision Support to Improve Procedure Documentation and Enhance Patient Safety, Nhan Do, Tricare Management Activity

Assessment of SNOMED CT® Implementation in EMR/EHR Applications, Kathy Giannangelo, AHIMA

One Hospital's Experience with Failure to Rescue, Juliana Hart, Cardinal Health-MediQual

MRSA Infection Control with Workflow Technology, Laura Koetter, Siemens Medical Solutions

Veterans Healthcare Administration Decision Support System (DSS); Transforming Data into Clinical Management and Outcomes, Nancy Lapointe, VHA

Combining the Power of Graphical User Interfaces with Command-based Input: A Medical Literature Search Tool with No Buttons, Stephen Piccolo, University of Utah

How Do Computerized Patient Record Systems Affect the Construction of Medical Case Knowledge?, Philip Scott, Portsmouth Hospitals NHS Trust

Poster Session Two
Wednesday, May 23, 2007, 3:30 pm - 5:00 pm

Workflow / BPM Standards, Jonathan Emanuele, Siemens Medical Solutions

Healthcare Maintenance in Ambulatory EMR, Raj Gopalan, University of North Carolina Healthcare Systems

Running Decision Support Logic Retrospectively to Determine Guideline Adherence: a Case Study With Diabetes, Vojtech Huser, University of Utah

The Scheduling Support System Used in Clinical Work Allocation, Kuan-Liang Kuo, Taipei City Hospital

Decision Support Tool for Emergency Medical Dispatch of Trauma Cases, Muzna Mirza, University of Alabama at Birmingham, AL

Designing an EHR framework to promote and support high quality clinical documentation, Ruthann Russo, Navigant Consulting

An Evidence-based System for Evaluating QTC-related Drug-drug Interaction Alerts, Eileen Yoshida, Partners Healthcare