Education and Research
Domain 6 - Education and Research
This domain tests how you build real competencies in staff and how you read and conduct research well enough to base practice on it.
Core rules
- Education must change behavior, not just knowledge. Write measurable objectives; use teach-back, just-in-time coaching, and simulation; verify competency by direct observation and documented return demonstration.
- Document training: date, content, method, trainer, and evidence of competency. Audit compliance rather than assume it.
- Evaluate at levels: reaction, learning, behavior, results. A post-test score reaches only level 2.
- Study designs: descriptive (case series), cross-sectional, case-control (rare outcomes - odds ratio), cohort (exposed/unexposed - relative risk), randomized controlled trial (strongest for causation). Outbreak investigations commonly use case-control when the outcome is rare.
- Test statistics: sensitivity and specificity describe the test; PPV and NPV depend on prevalence. Low-prevalence settings generate many false positives.
- Human-subjects research requires IRB approval and consent; quality improvement may need ethics review depending on intent and publication.
Worked example
A 90%-sensitive, 95%-specific MRSA admission screen is used where 2% of admissions are colonized. Of 10,000 admissions: 200 colonized → 180 true positives; 9,800 not colonized × 5% → 490 false positives. PPV = 180 ÷ 670 ≈ 27%. Three of four positives are false, so confirm by culture before cohorting decisions and explain this to clinicians who assume a positive equals colonization.
Traps
Equating a knowledge quiz with competency; reporting sensitivity as if it were PPV; generalizing single-hospital results to all settings; ignoring confounding; launching a study without IRB review; treating statistical significance as clinical importance.