Module Components

• Overview
• Assignments

Topics

• Decision Theories
• Clinical Decision Support Systems
• Clinical Decision Support Systems - Examples
• Clinical Decision Support Systems - Benefits and Challenges
• References

Nurs 460 - HEALTH INFORMATICS FOR REGISTERED NURSES

Module 6: Technology to Support Decision-Making

Clinical Decision Support Systems - Examples

Learning Activities

View the video below on CDSS. Although this is UK sponsored research it provides a good idea of how technology can be used by providers and patients to support decision making.

Clinical Decision Support for Evidence Based Care (11:07) at https://www.youtube.com/watch?v=zoxpuzH4B_0

Standard CDSS

Clinical Decision Support System programs offer alternatives for structured, semi structured, and unstructured processes. In addition, they offer four phases of decision-making: intelligence, design, choice, and implementation. Using logic, optimization methods, and arithmetic tools, CDSS provide mathematically structured alternatives. The advantages of these systems is that the alternatives are often exhaustive, relatively easy to understand, and quickly available. However, the disadvantages are that the alternatives may be unrealistic, causing practitioners to underuse systems. For more than 30 years, informaticians have been developing a functional EHR with active CDSS components. These components include rules, alerts, and reminders for safe care such as flagging abnormal lab values, duplicate orders, drug-drug interactions, allergy alerts, preventive care reminders, trending of patient data and other condition of patient specific information.

Expert Systems

Expertise is extensive, task-specific knowledge gained from education and experience. Expert systems attempt to encapsulate the knowledge of experts using non-mathematical operations, such as heuristics, to make knowledge widely available to others. An early expert systems, known as MYCIN was developed in the 1970's at Stanford University. MYCIN was designed to aid physicians in diagnosing meningitis and prescribing treatment. MYCIN’s record of correct diagnoses and prescribed treatments has equaled the performance of top human experts. Specifically, the system’s objective is to aid physicians during a critical 24-48 hour period after the detection of symptoms. Early diagnosis and treatment can save a patient from brain damage or even death. MYCIN’s success has been so great that many new systems use MYCIN’s programming as the basis for new rules-based CDSS.

Cognitive Computing and Artificial Intelligence

Using models, databases, and neural networks, artificial intelligence (AI) “thinks through” problems, and then gives alternatives that maximize possible outcomes. These systems, are very expensive and not widely used, but we are beginning to see more applications for this type of sophisticated computing.

Current research in cognitive computing and AI includes bioinformatics, assistance for the aged and disabled, biomedicine, and data mining. Programming in these types of systems is being designed to account for the complexities of healthcare so that they can integrate more fully into clinical settings. Further, providers will soon be able to communicate and interact with multiple CDSS to provide better care (Bichindaritz & Marling, 2006). One of the most impressive applications today using cognitive computing is Watson, developed by IBM. Watch the short videos below to learn more about Watson, how it works, and the potential for this type of application in healthcare.

CDSS EXAMPLES

• Click link to read about 10 Innovative Clinical Decision Support Programs
  
• Learn about the EBP InfoBot at https://lhncbc.nlm.nih.gov/project/infobot
  Then try out for yourself at https://ceb.nlm.nih.gov/ridem/HCCRuleSet.html
  
• IBM’s Breakthrough: Watson May Help Beat Cancer https://www.youtube.com/watch?v=HkEOJnn_zlg