Research, Evidence-Based Practice, and Quality Improvement
Research, Evidence-Based Practice, and Quality Improvement asks you to appraise studies, apply evidence, and improve systems. You need design hierarchy, basic biostatistics, and QI methods.
Core rules
- Evidence hierarchy: systematic reviews/meta-analyses of RCTs, then RCTs, cohort, case-control, cross-sectional, case series, expert opinion.
- PICO frames a question: population, intervention, comparison, outcome.
- A p value is the probability of data this extreme if the null hypothesis is true-not the probability the hypothesis is true. Confidence intervals show precision.
- Sensitivity = true positives / all with disease; specificity = true negatives / all without. Positive predictive value depends on prevalence.
- EBP steps: ask, acquire, appraise, apply, assess. QI uses PDSA cycles, flowcharts, fishbone diagrams, and run charts; QI improves local practice, while research creates generalizable knowledge.
Worked example
A screening test has 90% sensitivity and 80% specificity in a population with 10% prevalence. Of 1,000 people, 100 have the condition: 90 are true positives and 10 false negatives. Of 900 without it, 720 are true negatives and 180 false positives. PPV = 90/(90+180) = 33%. If prevalence rose to 30%, PPV would rise sharply-so a good test can still have a low PPV in a low-prevalence group.
Traps
Reading a p value as the chance the intervention works; ignoring confidence intervals; confusing sensitivity with PPV; calling a QI project research without IRB review; and generalizing from a single small study.