How to Write a Systematic Review in Nursing

In the hierarchy of evidence, the Systematic Review sits at the apex (Level I). Unlike a standard literature review that summarizes a topic broadly, a systematic review employs a rigorous, reproducible scientific protocol to answer a specific clinical question. It synthesizes all available high-quality evidence to minimize bias and inform practice guidelines. For DNP students and nurse researchers, mastering this methodology is the ultimate test of Evidence-Based Practice (EBP) competency. This guide details the step-by-step process of conducting a rigorous systematic review, from protocol development to data synthesis.

Systematic Review vs. Literature Review

Understanding the distinction between these two review types is critical for academic success. A Literature Review offers a narrative summary of current knowledge on a broad topic. It is often subjective, lacks a defined search protocol, and is prone to selection bias.

In contrast, a Systematic Review is a research study in itself. It uses specific inclusion and exclusion criteria to identify, appraise, and synthesize all high-quality evidence on a narrow clinical question. Its primary goal is to eliminate bias and provide a definitive answer regarding the effectiveness of an intervention. The Cochrane Library sets the global standard for this methodology, emphasizing reproducibility and transparency.

Step 1: The Protocol and Question

Every review begins with a clearly defined question using the PICOT format (Population, Intervention, Comparison, Outcome, Time). This framework ensures the search strategy is focused and retrievable.

• Protocol Registration: To prevent duplication of effort and ensure transparency, researchers should register their review protocol with PROSPERO (International Prospective Register of Systematic Reviews) before starting. This locks the methodology, preventing “outcome switching” later in the process.
• Inclusion/Exclusion Criteria: Researchers must define exactly what studies will be accepted. Criteria typically include study design (e.g., RCTs only), language (e.g., English), publication date (e.g., last 5-10 years), and participant demographics (e.g., adults >65).

For help refining your question, see our Guide to PICOT Questions.

Step 2: Comprehensive Literature Search

A systematic search must be exhaustive to avoid bias. You cannot simply select the first few articles that appear; you must attempt to find every relevant study ever published on the topic.

• Databases: Search at least three major databases (e.g., CINAHL, PubMed/MEDLINE, Embase, Cochrane Central) to ensure coverage of nursing, medical, and pharmacological literature.
• Search Strategy: Use Boolean operators (AND, OR, NOT) to combine keywords. Incorporate MeSH terms (Medical Subject Headings) to capture indexed articles that might use different terminology for the same concept (e.g., “cancer” vs. “neoplasm”).
• Grey Literature: Search for unpublished studies (dissertations, conference proceedings, clinical trial registries) to minimize publication bias (the tendency for journals to publish only positive results).

Step 3: Screening and PRISMA

Document the search results using the PRISMA Flow Diagram (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). This ensures the process is transparent and reproducible.

1. Identification: Record the total number of records found from all databases.
2. Screening: Remove duplicates. Screen titles and abstracts for relevance to the PICOT question. Discard obviously irrelevant studies.
3. Eligibility: Retrieve and read the full texts of remaining articles. Apply inclusion/exclusion criteria rigorously. Document specific reasons for excluding any full-text article (e.g., wrong setting, wrong outcome measure).
4. Included: The final number of studies eligible for synthesis.

For assignments requiring this level of detail, our EBP Paper Services can assist with PRISMA documentation.

Step 4: Critical Appraisal (Quality Assessment)

Evaluating the quality of included studies is mandatory to determine the Risk of Bias. Just because a study is published does not mean it is high quality. Use standardized tools specific to the study design:

• CASP Checklists: User-friendly tools for appraising Qualitative, Cohort, or RCT studies.
• Cochrane Risk of Bias Tool (RoB 2): The gold standard for assessing Randomized Controlled Trials. Evaluate domains such as randomization process, deviations from intended interventions, missing outcome data, and selection of the reported result.
• JBI Checklist: Excellent for various study designs, including cross-sectional and quasi-experimental.

Studies deemed “High Risk of Bias” should be noted in the synthesis or excluded, as they weaken the validity of the review’s conclusions.

Step 5: Data Synthesis

Synthesis involves combining the findings to answer the research question.

• Narrative Synthesis: Describe the findings in text, grouping studies by themes, interventions, or outcomes. Use a Synthesis Matrix to organize findings visually. This is common when studies are too heterogeneous (different) to pool statistically.
• Meta-Analysis (Quantitative): If the data is homogeneous (similar enough in design, population, and outcome measure), statistical software is used to pool the results. This generates a Forest Plot and calculates an overall effect size, providing the highest level of evidence.

Reporting Findings

Structure the final paper according to the PRISMA checklist items.

• Introduction: State the rationale and specific objectives.
• Methods: Detail the protocol, search strategy, and selection criteria so others can replicate the study.
• Results: Present study characteristics, risk of bias assessments, and results of individual studies.
• Discussion: Summarize the evidence, discuss limitations (of the included studies and the review itself), and provide conclusions for practice and future research.

FAQs on Systematic Reviews

Conclusion

A systematic review is the pinnacle of evidence synthesis. By rigorously locating, appraising, and summarizing research, nurses generate the Level I evidence necessary to transform clinical practice, influence policy, and improve patient outcomes globally.