16.3 Evidence-Based Practice & Research

Domain III expects you to distinguish EBP from research and quality improvement, build a searchable clinical question, rank evidence, and know the bedside nurse's ethical duties when research happens on your unit.

What Evidence-Based Practice Is

Evidence-based practice (EBP) is clinical decision-making that integrates three elements: (1) the best available research evidence, (2) the clinician's expertise, and (3) the patient's values and preferences. All three matter — the strongest study in the world does not override a patient's informed preference, and habit ("we've always done it this way") is not evidence. EBP is not conducting research: it is finding and using existing evidence to solve a clinical problem.

The Five Steps of the EBP Process

1. Ask — convert a clinical uncertainty into a searchable, answerable question (PICO format).
2. Acquire — search the literature systematically (CINAHL, PubMed/MEDLINE, Cochrane Library), starting with pre-appraised sources such as systematic reviews and clinical practice guidelines.
3. Appraise — critically evaluate the evidence for validity, reliability, and applicability to your population.
4. Apply — integrate the evidence with clinical expertise and patient preferences; this step includes practice changes, pilot protocols, and policy revision.
5. Assess — evaluate the outcomes of the change (audit, indicators, patient outcomes) and refine.

Building a PICO Question

PICO structures the foreground clinical question:

• P — Population/Patient/Problem: who or what is the question about?
• I — Intervention: the treatment, test, or exposure of interest.
• C — Comparison: the alternative (usual care, placebo, another intervention).
• O — Outcome: the measurable result you care about.
• T — Time (optional, making it PICOT): the time frame.

Example: "In hospitalized adults with a central line (P), does daily chlorhexidine bathing (I), compared with standard soap-and-water bathing (C), reduce central line-associated bloodstream infections (O) during the ICU stay (T)?" A well-built PICO question dictates your search terms and your inclusion criteria — weak questions produce unusable searches.

Hierarchy of Evidence

When asked which source provides the strongest evidence for a practice question, choose the systematic review/meta-analysis or guideline over any single study.

Research Designs: The Basics

• Randomized controlled trial — participants are randomly assigned to intervention or control; the gold standard for testing cause and effect. Look for randomization, blinding, and a control group.
• Cohort study — follows exposed vs. unexposed groups forward (or backward) in time to see who develops the outcome; observational, so association — not proven causation.
• Case-control study — starts with the outcome (cases) and looks back at exposures compared with matched controls; efficient for rare outcomes.
• Qualitative research — explores meaning and lived experience (phenomenology, grounded theory, ethnography) using interviews and observation rather than numbers; answers "what is it like?" questions, such as the experience of living with an ostomy.
• Quantitative research — measures variables numerically and tests hypotheses statistically.

Critiquing Research: Validity, Reliability, Sample

• Validity — does the study measure what it claims to measure? Internal validity asks whether the intervention (not confounders) caused the outcome; external validity asks whether results generalize to your patients.
• Reliability — consistency: would the instrument produce the same result if repeated (test–retest) or used by different raters (inter-rater)?
• Sample — size and selection matter: small or convenience samples limit generalizability; random sampling reduces selection bias. A statistically significant result (p < 0.05) means the finding is unlikely due to chance, but ask whether it is clinically significant — a 0.2 mm Hg blood pressure difference can be statistically significant and clinically meaningless.

Implementing Evidence into Practice

Evidence only matters if it reaches the bedside. Implementation steps mirror the EBP process: form a team, assess organizational readiness, pilot the change on one unit, educate staff with supporting materials (protocols, algorithms, skills checklists, audit tools), measure outcomes, and then spread. Common barriers — time, access to databases, appraisal skills, and resistance to change — are addressed with journal clubs, EBP mentors/champions, librarian partnerships, and leadership support.

Nursing-Sensitive Indicators as Evidence

Nursing-sensitive quality indicators are outcomes directly influenced by nursing care — falls, pressure injuries, CAUTI, CLABSI, restraint use, and patient satisfaction — collected through programs such as the National Database of Nursing Quality Indicators (NDNQI). They serve two functions: they are evidence about your own unit's practice, and they let you benchmark against peer facilities to identify where an EBP question is worth asking (see Section 16.4).

The Staff Nurse's Role in Research

Human subjects protection is the core tested concept. Historical abuses (Tuskegee, Willowbrook) produced the Belmont Report principles — respect for persons, beneficence, justice — and today's safeguards:

• Institutional Review Board (IRB) — every study involving human subjects must have IRB approval before enrollment begins; the IRB reviews risk–benefit balance and consent procedures.
• Informed consent in research — must disclose the study's purpose, procedures, risks, benefits, the right to withdraw at any time without affecting care, and alternatives. Research consent is the investigator's responsibility, but the staff nurse protects subjects by verifying the patient still consents and is not being coerced.
• Voluntary participation — no penalty for declining or withdrawing; the nurse continues to provide the same standard of care.
• Vulnerable populations (children, prisoners, pregnant women, cognitively impaired adults) receive extra protections.
• The staff nurse may also identify researchable clinical problems, collect data per protocol, and serve on evidence review committees — but never enrolls or consents patients outside the approved protocol.