A nursing quality improvement project is strongest when it shows a transparent path from a specific care problem to measured learning. The task is not simply to describe an intervention. You need to understand the local system, define the improvement aim, test change safely, measure what happens and explain what the findings do and do not support.
This guide focuses on nursing and nurse-led improvement projects in hospitals, community services, mental health settings, care homes and other health services. Local governance requirements always take priority over a generic template.
What a strong nursing QI project needs
- A specific local problem supported by evidence.
- A measurable aim with a population, setting, target and time frame.
- Process analysis before selecting the intervention.
- Outcome, process and balancing measures with clear definitions.
- Small, genuine tests of change where PDSA is used.
- Results presented over time rather than as one final percentage.
- Proportionate claims about causation and transferability.
- Accurate governance, confidentiality and data-handling statements.
- Reporting that reflects SQUIRE 2.0 where appropriate.
Quality improvement, audit and research are not interchangeable
Audit compares current practice with an explicit standard. Research is designed to generate knowledge through a defined research method. Quality improvement usually tests how a change works within a local service or system.
Real projects can contain overlapping features, so the project label should not replace a formal governance decision. Before collecting data, confirm the correct organisational route for approval, information governance, confidentiality, consent and data protection.
1. Define a focused nursing problem
Begin with an observable care or service problem rather than a broad topic. “Patient safety” is too wide. “Incomplete allergy-status documentation during adult admission to Ward A” identifies a process, population and setting that can be investigated.
A useful problem statement answers:
- What is happening now?
- Who is affected?
- Where and when does the problem occur?
- Why does it matter to patients, staff or the service?
Use local evidence where possible, such as audit findings, incident reports, waiting-time data, complaints, patient feedback or staff observations. Explain the limitations of that evidence rather than treating a small local sample as though it represents the whole organisation.
2. Explain the evidence and local significance
Connect the local problem with relevant evidence, professional standards, patient-safety guidance and service expectations. Then return to the local setting and explain why improvement is needed there.
Do not turn this section into a general literature review. Select evidence that helps explain the problem, expected good practice, previous improvement approaches and contextual factors likely to affect implementation.
Use the nursing literature review support route where evidence synthesis is the main difficulty.
3. Map the current process
Before choosing a solution, show how work currently happens. A simple process map can identify handovers, delays, duplicated work, missing information and decision points.
Useful diagnostic tools can include fishbone diagrams, the five whys, Pareto charts and stakeholder mapping. These tools organise possible explanations; they do not prove causation. Check suspected causes against records, observation and frontline experience.
4. Write a measurable aim
A strong aim states what will improve, by how much, for whom, where and by when. Institute for Healthcare Improvement guidance similarly emphasises defining the population, location, level of improvement and time frame (IHI, n.d.-e).
Increase the proportion of adult admissions to Ward A with complete allergy-status documentation from 68% to at least 90% by 30 November, while monitoring admission time and documentation errors.
The baseline and target should be justified rather than selected because they sound impressive.
5. Build the improvement team
Include people who understand the parts of the process being changed. Depending on the project, that may include nurses, pharmacists, administrative staff, allied professionals, managers, information staff and service users.
Participation should be meaningful. Asking a patient representative to approve a finished intervention is not the same as involving service users in defining the problem or designing the change.
6. Choose a change idea that addresses the problem
The intervention should respond to a plausible cause identified during process analysis. Examples may include a structured prompt, revised checklist, improved handover sequence, redesigned workflow or easier access to guidance.
Training should not be the automatic solution to every nursing problem. If the main barrier is workload, inaccessible equipment, confusing documentation or a poorly designed workflow, education alone may have little effect.
7. Define outcome, process and balancing measures
| Measure | Question | Example |
|---|---|---|
| Outcome | Did the overall result improve? | Percentage of admissions with complete allergy status |
| Process | Was the intended action completed? | Percentage of admission forms checked before handover |
| Balancing | Did the change create another problem? | Median admission-documentation time |
For every measure, specify the numerator, denominator, inclusion criteria, exclusions, data source, collection frequency and person responsible. “Documentation compliance” is not useful until the required fields and eligible cases are defined.
8. Establish a defensible baseline
Collect baseline data before testing the change and use the same operational definitions throughout the project. If the measurement method changes after the intervention begins, apparent improvement may reflect the data-collection method rather than the care process.
Report actual counts as well as percentages where denominators change.
9. Use PDSA cycles as genuine tests of change
The Model for Improvement asks what the team wants to accomplish, how it will know whether a change is an improvement, and what change can be tested. PDSA cycles provide one route for testing and learning (IHI, n.d.-a).
Plan
State the change, prediction, test population, responsibilities, data and time frame.
Do
Carry out the test and record deviations, practical difficulties and unexpected observations.
Study
Compare the result with the prediction and examine what the data and feedback suggest.
Act
Decide whether to adopt, adapt or abandon the change, then use that learning to design the next test.
Repeating the same intervention four times without meaningful adaptation is not evidence of four informative PDSA cycles.
10. Analyse change over time
Use displays such as run charts where appropriate so readers can see baseline variation, intervention timing and subsequent performance.
Improvement after an intervention does not automatically prove that the intervention caused it. Staffing, workload, case mix, seasonal demand and concurrent policy changes may also affect the outcome.
Qualitative feedback can help explain the pattern, but quotations and staff or patient comments should be handled under the project’s governance and confidentiality arrangements.
11. Plan sustainability and spread
Testing shows whether a change is promising. Implementation requires a plan for routine ownership, staff induction, ongoing measurement, resources and action if performance declines.
Do not claim sustainability after a very short follow-up. State the monitoring period and make the conclusion proportionate to the available evidence.
Likewise, a successful local project should not be assumed to work identically in another ward or service. Spread usually requires renewed local testing.
12. Report the project transparently
SQUIRE 2.0 provides a reporting framework for systematic efforts to improve healthcare quality, safety and value (Ogrinc et al., 2016). Relevant elements include the local problem, rationale, intervention, context, methods, measures, results, interpretation and ethical considerations.
Use the reporting framework during project planning where possible so important details are not reconstructed retrospectively.
Recommended report structure
| Section | Main task |
|---|---|
| Introduction | Define the local problem, evidence and measurable aim |
| Methods | Describe context, intervention, measures, data collection, governance and testing |
| Results | Show baseline, tests of change, adaptations, outcome trends and balancing measures |
| Discussion | Interpret results, context, limitations and alternative explanations |
| Conclusion | State what was learned and what should happen next without overclaiming |
Worked nursing QI example
Imagine a community clinic where follow-up appointments are frequently arranged without recording the patient’s preferred contact method.
Problem: A review of 60 consecutive records finds that 55% contain a documented contact preference.
Aim: Increase documentation from 55% to 90% within ten weeks without increasing median booking time by more than one minute.
Change: Add a contact-preference prompt to the booking workflow.
Measures: Outcome = documented preference; process = checklist use; balancing = booking time.
Testing: Begin with one administrator and one clinic, revise the prompt after feedback, then test with more staff and different sessions.
Interpretation: Report the weekly pattern, adaptations and contextual changes. Do not claim that the same intervention will produce identical results in every service.
Common nursing QI mistakes
- Describing meetings or teaching as though activity itself proves improvement.
- Changing the aim silently during the project.
- Reporting only one final percentage.
- Ignoring balancing measures.
- Creating retrospective PDSA cycles that were not genuinely performed.
- Overclaiming causation from a simple before-and-after comparison.
- Ignoring equity or differential effects across relevant groups.
- Leaving governance and confidentiality considerations until the final report.
Final checklist
- The problem is specific and supported by local evidence.
- The aim defines the population, setting, target and deadline.
- The current process was investigated before selecting the intervention.
- Outcome, process and balancing measures have operational definitions.
- Baseline and intervention data were collected consistently.
- PDSA cycles show genuine predictions, learning and adaptation where used.
- Results are presented over time with denominators where relevant.
- Limitations, context, equity and unintended effects are discussed.
- Governance and data handling are reported accurately.
- The sustainability plan identifies ownership and continued measurement.
Frequently asked questions
Does every QI project need PDSA?
No. Use the improvement method required by the programme or organisation. If PDSA is used, report real iterative testing rather than adding retrospective cycles to fit a template.
How many PDSA cycles are required?
There is no universal number. The number should reflect risk, complexity and what the team learns.
Can patient feedback be included?
Yes, where it fits the project and governance route. Protect confidentiality and make involvement meaningful rather than tokenistic.
What is the difference between an outcome and a balancing measure?
The outcome shows whether the intended result improved. A balancing measure checks whether the change created an unintended problem elsewhere.
Can local QI results be generalised?
Usually only cautiously. Local QI can support service decisions and generate useful learning, but context and design limit universal causal claims.
Related Nursing Guides
- How to Diagnose and Fix DNP Project Manuscript Problems
- Fishbone Diagram Healthcare Example: A Nursing QI Guide
- How to Write a Nursing Root Cause Analysis
Conclusion
A defensible nursing quality improvement project shows how a real care problem was understood, how change was tested, what happened and what the team learned. Strong reporting makes the reasoning, measures, limitations and context visible rather than presenting improvement as a guaranteed outcome.
For project-specific review of an existing QI or EBP draft, use nursing evidence-based practice project support or the contact page.
References
- Institute for Healthcare Improvement. (n.d.-a). How to improve: Model for Improvement. https://www.ihi.org/library/model-for-improvement
- Institute for Healthcare Improvement. (n.d.-b). How to improve: Model for Improvement—Establishing measures. https://www.ihi.org/library/model-for-improvement/establishing-measures
- Institute for Healthcare Improvement. (n.d.-c). How to improve: Model for Improvement—Implementing changes. https://www.ihi.org/library/model-for-improvement/implementing-changes
- Institute for Healthcare Improvement. (n.d.-d). Plan-Do-Study-Act (PDSA) worksheet. https://www.ihi.org/library/tools/plan-do-study-act-pdsa-worksheet
- Institute for Healthcare Improvement. (n.d.-e). Setting aims. https://www.ihi.org/library/model-for-improvement/setting-aims
- Ogrinc, G., Davies, L., Goodman, D., Batalden, P., Davidoff, F., & Stevens, D. (2016). SQUIRE 2.0: Revised publication guidelines from a detailed consensus process. BMJ Quality & Safety, 25(12), 986–992. https://doi.org/10.1136/bmjqs-2015-004411
