Clinical Decision Support System

Problems with INDICATION

• Unnecessary drug therapy

• Needs additional drug therapy

Problems with EFFECTIVENESS

• Ineffective drug

• Dosage too low

Problems with SAFETY

• Adverse drug reaction

• Dosage too high

Problems with ADHERENCE

• Non-adherence

Clinical Decision Support Systems

Health information technology system that assists clinical decision-making tasks when it comes to their patients

Passive CDS

Directs user to the most appropriate practices unobtrusively

Examples of Passive CDS

• Order sets

• Tallman lettering

• Limited selections on drop-down lists

Active, Non-interruptive CDS

New patient information is posted to work queues, lists or forms for resolution at a time conventient to the clinician

Active, Interruptive CDS

Just-in-time alerts are presented directly to the user who is required to take some action to respond to the alert

Free-form, rule-based alert

flexibility to develop the characteristics of the message

Database Driven Alerts

Utilize a large database containing drug interaction, dose range and condition, and allergy checking

Clinical Order Checking checks for the following:

• Allergy

• Dr-Dr Interaction

• F-Dr interaction

• Duplicate therapy

• Dose range checking

Challenges with allergy checking

• Allergies must be coded

• System no distinguishment between true allergies, intolerances, or side effects

Challenges with Drug Interaction checking

• Alert levels are very conservatively assigned by the vendor

• Research and ongoing maintenance of the interactions are a lot of work

Challenges in Duplicate Therapy Checking

• Alert specificity based on strength, salt, dose form, and route of administration

• Nuisance warnings

• Specificity of therapeutic class designations

Order Sets

group of pre-established orders related to a symptom or diagnosis based on evidence or established best practices

Evidence-based Medicine

conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients

Best-Practice determined by service line

collective techniques and methodologies that have been proven through repeated experience to achieve optimal outcomes

Challenges of Order Sets

• Lack of clinically useful articles

• Different opinions within and between service lines

• Difficulty in measuring outcomes

Cognitive Computing

Simulation of human thought processes in a computerized model. Self-learning systems that use Data mining, pattern recognition, natural language processing, probability analysis

Categories of Rules

• Suggest screenings

• Suggest treatment plans

• Warn of abnormal results

Purpose of CDSS

• Assist in problem-solving with semi-structural problems

• Improve effectiveness of the decision-making process

• Support clinical coding and documentation

• Authorize procedures and referrals

• Manage clinical complexity and details

• Monitor medication orders

• Avoid duplicate or unnecessary tests

Data aggregators

grouping of relevant information that are synthesized that will help you arrive at a decision

Heuristic models

how care is given today, constructed by clinical societies, with scoring mechanisms

Alert fatigue

excessive alerting can desensitize clinicians to the clinical significance of the alerts, potentially leading to disregard and overrides of important messages