Textbook of Usability

Also known as: CDS, clinical decision support system, CDSS

Clinical decision support (CDS) systems are designed to assist clinical decision-making by providing timely, patient-specific information and recommendations. They range from simple alerts (drug-allergy interaction warnings) to complex diagnostic algorithms and machine learning models.

Well-designed CDS can reduce errors through:

• Alerts for drug interactions, contraindications, and allergies
• Order sets that bundle related actions for common conditions
• Dose calculators that compute correct doses based on patient weight and renal function
• Reminders for preventive care and follow-up
• Diagnostic suggestions based on presenting symptoms

The dominant usability problem in CDS is alert fatigue (see separate entry). Effective CDS uses a tiered approach:

1. Passive information — relevant data displayed in context, no interruption
2. Non-interruptive alerts — icons or colour changes that signal relevant information
3. Interruptive alerts — dialogs requiring acknowledgment, reserved for potentially dangerous situations
4. Hard stops — the system prevents the action entirely (reserved for absolute contraindications)

Hard stops are the most restrictive form: they prevent errors but also prevent clinical judgement. Deciding which alerts should be hard stops is a high-stakes policy decision that involves clinicians, pharmacists, and informaticists.

CDS is most effective when it delivers the right information to the right person at the right time in the right format through the right channel — the "Five Rights" of clinical decision support.