Every nursing capstone requires a literature review built on peer-reviewed evidence. Finding the articles is only the first step. Your committee expects you to appraise each source critically — evaluating its design, methodology, validity, and relevance to your PICOT question — and then synthesize the evidence into a coherent argument. This guide walks you through exactly how to do that, from selecting the right CASP checklist to writing the synthesis paragraph that turns five individual studies into one cohesive body of evidence.
What "peer-reviewed" means — and how to verify it
A peer-reviewed (also called refereed or scholarly) article has been evaluated by independent experts in the field before publication, in addition to editorial review. This process filters out work with fatal methodological flaws, unsupported conclusions, or insufficient rigor. For nursing capstones, all primary literature cited in the evidence review must be peer-reviewed.
How to verify peer-review status:
- In CINAHL or PubMed, apply the "Peer Reviewed" or "Journal Article" filter before running your search.
- In EBSCO databases, check the box "Peer Reviewed" under Limit Results.
- If unsure, look up the journal in Ulrichsweb (ulrichsweb.serialssolutions.com) — peer-reviewed journals are marked with a referee symbol.
- Gray literature (government reports, clinical guidelines from organizations like AHRQ, CDC, and the Joint Commission) is not peer-reviewed in the traditional sense but is acceptable as supplemental evidence in nursing capstones. Cite it as gray literature, not as a research study.
Recency requirement
Most nursing programs require literature published within the past 5 years (some extend to 7–10 years for landmark or foundational studies). Review your program's evidence currency requirement before selecting sources. Articles published before 2019 should generally be excluded unless they are seminal studies (e.g., the original STRATIFY tool validation study, the founding publication of a theoretical framework) that are foundational to your topic and still cited in current literature.
Melnyk's levels of evidence hierarchy
The most widely used evidence hierarchy in nursing is the Melnyk & Fineout-Overholt (2019) seven-level scale. Higher levels (I–II) provide stronger evidence for clinical decision-making; lower levels (VI–VII) provide contextual, descriptive, or expert-opinion evidence. Your capstone should primarily cite Level I–IV evidence, with Levels V–VII used sparingly for context or when higher-level evidence does not exist for your specific intervention.
The CASP checklist — how to use it
The Critical Appraisal Skills Programme (CASP) provides separate checklists for different study designs. Select the checklist that matches the design of the article you are appraising. The most commonly used checklists for nursing capstones are:
| Study design | CASP checklist to use | Most common in nursing capstones? |
|---|---|---|
| Randomized controlled trial | CASP RCT checklist (11 questions) | Yes — for Level II evidence |
| Cohort study | CASP Cohort checklist (12 questions) | Yes — for Level IV evidence |
| Case-control study | CASP Case-Control checklist (11 questions) | Sometimes |
| Systematic review | CASP Systematic Review checklist (10 questions) | Yes — for Level I evidence |
| Qualitative study | CASP Qualitative checklist (10 questions) | Yes — for Level VI evidence |
| Diagnostic / prognostic | CASP Diagnostic / Prognostic checklist | Rarely |
CASP checklists are available free at casp-uk.net. You do not need to reproduce the entire checklist in your capstone — instead, you report your appraisal conclusions in the evidence table and the narrative synthesis.
- Did the study address a clearly focused issue?
- Was the assignment of participants to interventions randomized?
- Were all participants who entered the trial accounted for at its conclusion?
- Were participants, staff, and study personnel blinded to participants' group assignment?
- Were the groups similar at the start of the trial?
- Aside from the experimental intervention, were the groups treated equally?
- How large was the treatment effect? (Report effect size, CI, p-value)
- How precise was the estimate of the treatment effect?
- Can the results be applied to the local population/context?
- Were all important outcomes considered?
- Are the benefits worth the harms and costs?
What "critically appraising" actually means in practice
Many students confuse critical appraisal with summarizing an article. Summarizing describes what the study found. Critical appraisal evaluates how trustworthy the findings are and whether they apply to your clinical question. The distinction is crucial: your committee can read the abstract themselves. What they want to see is your judgment about the evidence.
Summary vs. critical appraisal — the difference in practice
Summary (weak — do not write this):
"Jones et al. (2022) conducted a study of 200 hospitalized patients in which they tested a nurse-led fall prevention protocol. The study found a 28% reduction in fall rates over 12 weeks."
Critical appraisal (strong — write this instead):
"Jones et al. (2022) conducted a quasi-experimental pre-post study (Level III evidence) of 200 medical-surgical patients across two inpatient units. The nurse-led fall prevention protocol produced a statistically significant 28% reduction in fall rates over 12 weeks (p = .03, 95% CI: 12–44%). Methodological strengths include a large sample size, fidelity monitoring via weekly audits, and consistent staff training. Limitations include the absence of a concurrent control group, potential contamination between units, and lack of long-term follow-up beyond 12 weeks. Despite these limitations, the moderate effect size and feasibility in a comparable acute care setting support its applicability to the proposed implementation context."
Notice that the critical appraisal paragraph does four things the summary paragraph does not: (1) identifies the evidence level, (2) reports the statistical result with precision (effect size, CI, p-value), (3) names specific strengths and limitations, and (4) makes an applicability judgment for your specific context.
Key appraisal questions for any design
Regardless of which CASP checklist you use, the following four questions should guide your appraisal of every article:
| Question | What to look for |
|---|---|
| 1. Is the design appropriate to the research question? | An RCT is appropriate for testing intervention effectiveness. A qualitative study is appropriate for exploring patient experience. A cross-sectional survey is not appropriate for establishing causation. If the design doesn't fit the question, the findings are unreliable regardless of how large the sample is. |
| 2. Are the sample and setting comparable to yours? | A study of ICU patients may not generalize to medical-surgical. Pediatric findings may not transfer to adults. Population, acuity level, unit type, and country/health system affect applicability. |
| 3. Are the results statistically and clinically significant? | A p-value <.05 means statistically significant. But clinical significance is about effect size: a 2% fall rate reduction is statistically significant with 10,000 patients but clinically meaningless. Look for both. |
| 4. What are the key limitations, and do they invalidate the findings? | All studies have limitations. The question is whether the limitations are fatal (e.g., no control group in an efficacy trial, unblinded outcome assessment in an RCT) or manageable (e.g., single-site study, 3-month follow-up). Name the limitations and explain why you still included the article. |
From appraisal to synthesis — the most important step
A literature review is not a chain of individual article summaries. It is a synthesis — a single narrative that draws on multiple sources to answer your PICOT question collectively. The synthesis identifies patterns, agreements, contradictions, and gaps across the body of evidence.
How to write a synthesis paragraph (not a summary chain):
- Group your articles by theme or finding, not by study.
- Start with the overall finding across the body of evidence: "Multiple studies support the effectiveness of nurse-led fall prevention protocols in reducing fall rates among hospitalized adults..."
- Cite supporting evidence parenthetically, combining multiple sources in a single sentence: "...with reported reductions ranging from 18% to 34% (Jones et al., 2022; Patel & Singh, 2021; Watkins et al., 2020)."
- Identify where studies agree: "All four RCTs included in this review demonstrated significant fall rate reductions, regardless of the specific risk assessment tool used."
- Identify where studies diverge or conflict: "However, two cohort studies found no significant difference in fall rates when the protocol was implemented without concurrent nursing education (Li et al., 2021; Moore, 2023), suggesting that training is a critical implementation component."
- Note the overall quality of the evidence and any significant gaps: "The evidence base is predominantly drawn from acute medical-surgical settings; evidence from rehabilitation and long-term care environments is limited, representing a gap in the literature relevant to this capstone's proposed implementation setting."
The most common literature review mistake: the annotated bibliography format
The most common error in nursing capstone literature reviews is writing article-by-article summaries rather than a synthesis — effectively producing an annotated bibliography disguised as a literature review. If every paragraph in your literature review begins with an author name ("Smith et al. found..."; "Jones et al. reported..."; "Patel & Singh demonstrated..."), you are summarizing, not synthesizing. Reorganize around themes, findings, and patterns. The articles should be evidence for the argument you are making, not the subjects of the argument themselves.
Evidence table — how to build it
Most programs require an evidence table in the appendix that documents your appraisal of each included article. Standard evidence table columns are:
| Column | What to include |
|---|---|
| Author(s) / Year | APA 7th in-text citation format (Author, Year) |
| Purpose / Research question | 1 sentence |
| Design | RCT, quasi-experimental, cohort, qualitative, etc. |
| Level of evidence | Melnyk Level I–VII |
| Sample / Setting | N=, population, unit/setting, country |
| Key findings | Primary outcome results with statistics (p-value, effect size, CI) |
| Limitations | 2–3 most important methodological limitations |
| Applicability | 1–2 sentences: how this supports your PICOT / intervention |
Need your literature review written and synthesized?
Our nurse-expert writers build evidence tables, critically appraise each source, and write a full APA 7th synthesis — not a summary chain.
Start your capstone Literature review guideRelated guides
Frequently asked questions
No — but all sources cited in the evidence review section must be peer-reviewed research. Clinical practice guidelines from the AHRQ, CDC, or professional organizations (e.g., American Nurses Association, NDNQI) are acceptable as supplemental evidence in the background or implications sections, cited as gray literature. Textbooks can support theoretical framework descriptions. The evidence table, however, should contain only peer-reviewed studies. If your program requires an evidence-based practice review, the inclusion criteria for that review should specify peer-reviewed sources from the past 5–7 years. If you are unsure which sections require peer-reviewed sources, ask your faculty advisor to clarify the requirement.
This varies by program and degree level. BSN capstones typically require 8–15 peer-reviewed sources. MSN capstones typically require 15–25. DNP projects commonly require 20–40+ sources, with more rigorous inclusion/exclusion criteria and a full PRISMA flow diagram. Check your program rubric for the specific number required, and note whether that count applies to the evidence review only or to the entire paper's reference list. Many students confuse the two — you may have 30 total references but only 12 in the evidence table; both may need to meet different thresholds.