Clinical Decision Support and Evidence-Based Care
Core concepts
Clinical decision support (CDS) spans alerts, order sets, clinical pathways, reference information, and predictive risk scores. Know the classic taxonomy: passive vs. active, interruptive vs. non-interruptive, and patient-specific vs. general knowledge. The Five Rights framework (right information, right person, right format, right channel, right time) is frequently tested. Understand CDS governance - who approves, monitors, and retires rules - and the role of the EHR's CDS Hooks or rules engine. Evidence-based care requires grading evidence (e.g., strength of recommendation vs. quality of evidence), guideline adaptation to local context, and measuring CDS effect on process and outcome.
Traps
Questions often conflate alert fatigue with low specificity rather than low sensitivity; ask which attribute causes over-riding. Another trap: assuming more alerts improve safety. Also watch for confusion between clinical guidelines (advisory) and hard stops (mandatory), and between a CDS intervention's process measure (order rate) and outcome measure (mortality).
Rule of thumb
Before deploying an alert, state the desired action, the evidence grade, the target population, and the metric you will use to evaluate it. If you cannot name the metric, do not deploy.
Worked example
A drug-drug interaction alert fires for 80% of orders and is overridden 95% of the time. The problem is specificity, not sensitivity. Tighten the rule to high-risk pairs, add a suggested alternative, and re-measure override rate and adverse events.