How to Write a Critical Analysis in Nursing

1. What a critical analysis is

A critical analysis in nursing is a structured evaluation of evidence, a practice, a policy, or a case. You judge its quality, relevance, and limitations, and you justify the judgment with evidence.

It is not:

• A summary of what a paper says
• A list of everything you read
• A description of a procedure or condition
• An opinion with citations attached

If your draft is more than 25% description, it is a summary, and you will lose marks for it. Markers look for judgment backed by reasoning.

2. The most common reasons these essays fail

1. Descriptive writing. "Smith (2021) conducted a study with 200 patients and found X." That is reporting. Critical writing says whether X can be trusted and why.
2. Citation stacking. Five references behind one claim, with no discussion of how they differ in quality or findings.
3. Vague criticism. "The sample size was small." Small compared to what? What does it do to the findings? Be specific: "With 24 participants, the study was underpowered to detect a difference in pressure injury incidence, so the non-significant result cannot be read as evidence of no effect."
4. No position. Fence-sitting with "there are advantages and disadvantages" is not analysis. You must reach a conclusion.
5. Weak sources. Relying on websites, textbooks only, or outdated studies when current primary research and guidelines exist.
6. Ignoring the question. Read the assignment brief again. Most lost marks come from answering a slightly different question than the one set.
7. Ignoring patient context. In nursing, evidence has to be tied to practice. A strong trial in a population unlike your patients is of limited use.

3. Step-by-step process

Step 1: Decode the task (15–30 minutes)

Before reading anything, extract three things from the brief:

• The command word. "Critically analyse," "evaluate," "discuss," and "appraise" all demand judgment. "Describe" and "outline" do not.
• The subject. A research article, a clinical practice, a policy, a case study, a nursing theory, or a combination.
• The marking criteria. Check the weighting. If 40% of marks go to "application to practice," that section needs 40% of your words, not a closing paragraph.

If the brief doesn't specify a framework, ask your tutor or check the module guide. Do not guess.

Step 2: Choose your sources properly

Use this hierarchy as a guide, not a rule:

1. Systematic reviews and meta-analyses
2. Randomised controlled trials
3. Cohort and case-control studies
4. Cross-sectional and descriptive studies
5. Qualitative studies (these rank differently, because the hierarchy is built for effectiveness questions, not experience questions)
6. Expert opinion and guidelines (check what evidence the guideline itself is based on)

Practical rules:

• Prioritise sources from the last 5–10 years unless you are discussing a landmark or foundational study. Say so explicitly if you cite an older one.
• Search CINAHL, PubMed/MEDLINE, Cochrane Library, and Joanna Briggs Institute databases. Google Scholar is acceptable for tracking citations, not as your only database.
• Use professional body standards where relevant (NMC Code in the UK, ANA Code of Ethics in the US, or your national equivalent, such as the Indian Nursing Council standards).
• If the question concerns a clinical practice, include at least one current guideline (NICE, WHO, AHA, etc.).

Step 3: Appraise using a framework, not instinct

Pick one validated appraisal tool and apply it consistently. Naming the tool in your essay shows methodological rigour.

If your institution mandates a specific tool, use that one.

For each source, work through these questions and take notes in your own words:

• Aim/question: Is it clearly stated? Does it match your topic?
• Design: Is the design appropriate for the question? (A cross-sectional survey cannot show that an intervention caused an outcome.)
• Sampling: How were participants chosen? How many? Is the sample representative? Any selection bias?
• Data collection: Were measures valid and reliable? For qualitative work, were interviews or observations suited to the question?
• Analysis: Appropriate statistics, or a transparent qualitative method (thematic analysis, grounded theory)? Were confounders controlled?
• Bias: Selection, performance, detection, attrition, reporting. Blinding? Drop-out rate?
• Results: Effect sizes and confidence intervals, not just p-values. A p-value of 0.04 with a wide confidence interval is weak evidence.
• Ethics: Approval, consent, conflicts of interest, funding source.
• Applicability: Do the setting and population match the one you are writing about?

Build a one-page appraisal grid for each source. This saves hours at the writing stage.

Step 4: Form your argument before you write

Write your thesis in one or two sentences. It must be a judgment, not a topic.

• Weak: "This essay will critically analyse a study on hand hygiene compliance."
• Strong: "Although the study demonstrates a short-term rise in hand hygiene compliance, its single-site design, reliance on direct observation, and absence of follow-up limit its usefulness for supporting a ward-wide policy change."

Every paragraph afterward either supports, qualifies, or tests this claim.

Step 5: Plan the structure

For a 2,000–3,000 word analysis, use this proportion as a baseline (adjust to the marking criteria):

Organise the body by theme or criterion, not by source. A source-by-source structure ("Study A did this, Study B did that") reads as a series of summaries. A thematic structure ("Sampling and generalisability," "Measurement validity," "Strength of the effect") forces you to compare sources against each other. This is where most of the critical thinking marks are.

Step 6: Write the introduction

Include, in this order:

1. Context: One or two sentences on why the issue matters, using a statistic or a clinical fact. Example: "Hospital-acquired pressure injuries affect roughly 1 in 10 hospitalised patients and increase length of stay and cost."
2. Focus: Exactly what you are analysing.
3. Scope and approach: The framework you use and any limits.
4. Thesis: Your judgment.
5. Roadmap: One sentence on the order of the essay.

Do not open with "Since the beginning of time" or a dictionary definition. Do not write the introduction first if you struggle; draft it last, when you know what you actually argued.

Step 7: Write analytical paragraphs

Use PEEL (Point, Evidence, Explain, Link) and make sure the "Explain" part is the largest.

Weak paragraph (descriptive):

Jones (2020) carried out a randomised trial of 120 patients comparing two wound dressings. The results showed that the new dressing healed wounds faster. The study was published in a nursing journal.

Strong paragraph (critical):

The trial by Jones (2020) provides moderate support for the new dressing, but three design features weaken confidence in its findings. First, allocation was randomised but not concealed, which creates a risk of selection bias, since nurses who knew the next allocation could influence which patients were enrolled (Higgins et al., 2019). Second, wound assessors were not blinded, and healing time was partly judged by visual inspection, a subjective measure vulnerable to detection bias. Third, 18% of participants withdrew, with more drop-outs in the control group, which could inflate the apparent benefit. These limitations do not invalidate the result, but they mean the reported 3-day reduction in healing time should be treated as an upper estimate rather than a reliable figure. This is consistent with the smaller effect found in the larger systematic review by Patel et al. (2022).

What the strong version does:

• States a judgment in the first sentence
• Names specific flaws and explains the mechanism of harm to validity
• Weighs the seriousness ("do not invalidate")
• Connects to another source for comparison
• Ends by drawing a conclusion

Rules for every paragraph:

• One main point per paragraph
• Evidence in a paragraph must be interpreted, never just placed
• Every criticism needs a reason and an effect: what is the flaw, why does it matter, how does it change the conclusion?
• Balance: if you criticise, also acknowledge real strengths, then judge the net effect

Step 8: Compare sources against each other

Comparison is where marks are won. Techniques:

• Agreement: "Three studies report consistent reductions in falls (A, B, C), though only B used a validated falls measure."
• Conflict: "Findings diverge. Study A reports no effect, but it enrolled patients with a mean age of 52, whereas Study B, which found benefit, enrolled patients over 75. Age may moderate the effect."
• Quality weighting: "Because Study C is a Cochrane review pooling 14 trials, its estimate carries more weight than the single-centre findings of Study D."
• Gaps: "No included study examined outcomes beyond 6 months, so long-term sustainability is unknown."

Step 9: Apply the analysis to nursing practice

Many students write a strong critique and then bolt on a generic "implications" paragraph. Instead:

• Name the patient group and setting (e.g., older adults on a medical ward in a district hospital).
• State what the nurse should do differently, concretely: change an assessment frequency, escalate earlier, adopt a tool, modify education.
• Address barriers: staffing ratios, cost, training, local policy, patient preferences, cultural context.
• Link to professional standards and ethical principles (autonomy, beneficence, non-maleficence, justice) where they genuinely apply, not as decoration.
• Say whether the evidence is strong enough to act on. "Adopt now," "pilot with audit," and "do not adopt yet" are all legitimate conclusions.

Step 10: Write the conclusion

• Restate your judgment in different words
• Summarise the two or three reasons that carry it
• Give a specific recommendation for practice, and if relevant, for research
• No new evidence or new arguments
• Do not use "In conclusion, there are many pros and cons"

Example:

The evidence supports adopting the repositioning protocol on high-risk orthopaedic wards, provided it is audited for adherence, but it does not justify hospital-wide mandation. Methodological weaknesses in the primary trials and the absence of data from community settings limit generalisability. Future research should use blinded outcome assessment and follow patients beyond discharge.

4. Language: how to sound critical without sounding vague

Use evaluative verbs: argues, demonstrates, fails to account for, overlooks, supports, undermines, qualifies, contradicts, oversimplifies.

Use hedged but precise claims: "suggests," "indicates," "is consistent with" for moderate evidence; "demonstrates" only for strong evidence. Never write "proves."

Avoid:

• "Very good study" / "bad study." Say what makes it strong or weak.
• "Obviously," "clearly," "everyone knows"
• First-person overuse. Check your institution's rule. Reflective assignments allow "I"; formal critiques usually prefer "this essay argues."
• Long quotations. Paraphrase and cite. One short quote is acceptable only where wording matters.

Signpost the logic: however, in contrast, consequently, this is significant because, despite this, taken together.

5. Referencing and academic integrity

• Follow the required style exactly (Harvard, APA 7th, Vancouver). Check in-text and reference list formats for journals, guidelines, and web documents.
• Every factual claim needs a source. Every source in the list must appear in the text.
• Cite the primary source wherever possible. If you only read a summary, use "cited in" and be aware that markers treat heavy secondary citing as a weakness.
• Use a reference manager (Zotero is free, Mendeley and EndNote are alternatives) from day one. Fixing references manually at the end wastes hours and creates errors.
• Run your institution's similarity check if available. AI-generated text and close paraphrasing can both be flagged. Write the analysis yourself from your own appraisal notes.
• Patient confidentiality: If you discuss a clinical case, anonymise it completely. Use a pseudonym, omit dates, location, and identifying details, and include the confidentiality statement your institution requires (typically referencing the NMC Code or equivalent).

6. A worked mini-outline

Task: Critically analyse a randomised trial on early mobilisation after cardiac surgery.

1. Introduction: Importance of post-operative complications. Thesis: the trial shows promise but design limits restrict generalisation.
2. Overview of the study: Aim, sample, intervention, main outcomes (150–250 words, no more).
3. Analysis:
   • Design and risk of bias (randomisation, blinding, allocation concealment)
   • Sample and generalisability (single site, age range, exclusion of high-risk patients)
   • Outcome measures (validity of the mobility scale, follow-up length)
   • Statistical interpretation (effect size, confidence intervals, clinical vs. statistical significance)
   • Comparison with other evidence (two trials and one systematic review)
4. Application to practice: What a cardiac ward nurse can implement, barriers (staffing, pain control, protocols), ethical points (patient consent and autonomy in mobilisation).
5. Conclusion: Judgment plus recommendation: pilot with audit, not full adoption.

7. Time plan for a 2,500-word piece (assuming 2 weeks)

If you have less time, protect the appraisal and planning stages. Rushed writing from a good plan beats polished writing with no argument.

8. Revision checklist

Argument

• Is there a clear, arguable thesis?
• Does every paragraph support it?
• Is the conclusion a judgment, not a summary?

Criticality

• Is each criticism explained with a reason and an effect?
• Are strengths acknowledged, not just flaws?
• Are sources compared with each other?
• Is description under roughly 25%?

Evidence

• Are sources recent, relevant, and mostly primary research or guidelines?
• Is the appraisal tool named and applied consistently?
• Are claims worded in proportion to evidence strength?

Practice

• Is a specific patient group and setting identified?
• Are recommendations concrete and realistic?
• Are barriers and ethical issues addressed?

Format

• Word count within ±10%
• Referencing style correct, all citations matched
• Confidentiality respected
• No paragraph longer than about 200 words

9. Final advice

• Highest-return action: Appraise with a checklist, then write your argument from the notes. Students who skip appraisal produce summaries.
• Biggest risk: Describing instead of judging. After each paragraph, ask: "So what? Does this change how far I trust the evidence or how I would act?" If not, cut or rewrite it.
• Unrealistic idea to drop: Trying to cover many sources. Four to six well-appraised sources analysed in depth score higher than fifteen skimmed ones.
• Check before you start: Your marking rubric and referencing style. Both vary by institution, and I can't know yours.