How Many PDSA Cycles Does a Quality Improvement Project Need?

Why This Topic Matters

Underestimating or overestimating the number of PDSA cycles can significantly impact a QI project's success. Too few cycles might lead to premature implementation of untested changes, potentially causing unintended consequences or failing to achieve desired outcomes. Conversely, conducting an excessive number of cycles without clear purpose can consume valuable resources, delay widespread implementation, and lead to team fatigue.

For junior doctors, trainees, and QI leads, understanding the strategic application of PDSA cycles is crucial for designing robust QI projects that deliver sustainable improvements. It enables teams to learn rapidly, adapt changes, and build confidence in their interventions before scaling up.

Practical Explanation: The Iterative Nature of PDSA

The PDSA cycle is a scientific method for testing a change. It's a structured approach to learning and improvement, guiding teams through:

• Plan: Define the objective, predict the outcomes, and plan the change and data collection.
• Do: Carry out the plan, collect data, and observe.
• Study: Analyse the data, compare with predictions, and summarise what was learned.
• Act: Decide to adopt, adapt, or abandon the change, and plan the next cycle.

The key principle is iteration – learning from each cycle to refine the change and inform the next. This iterative process allows for incremental adjustments and deeper understanding, rather than attempting a 'big bang' change from the outset.

The Importance of Starting Small

Effective QI teaches us to ‘think big, start small, and scale fast’. The initial PDSA cycles should focus on testing changes on a very small scale – perhaps with one patient, one shift, or one specific scenario. This minimises risk, allows for quick learning, and makes it easier to troubleshoot problems before wider implementation. As confidence grows and the change is refined, the scale of testing can gradually increase.

Common Pitfalls

Several common issues can hinder effective PDSA cycling:

• 'One and Done' Mentality: Treating the first test as the definitive solution without further refinement. This overlooks the iterative nature of PDSA and risks embedding an inefficient or flawed process.
• Lack of Clear Prediction: Failing to articulate a clear prediction in the 'Plan' phase. Without a prediction, it's difficult to 'Study' what was learned effectively.
• Poor Data Collection: Not planning or executing data collection rigorously in the 'Do' phase, making the 'Study' phase anecdotal rather than evidence-based.
• Skipping the 'Study' Phase: Moving directly from 'Do' to 'Act' without analysing the results and understanding why the change worked or didn't work.
• Scaling Too Quickly: Implementing a change across a whole service after only one or two small-scale tests, often leading to unforeseen problems and resistance.
• Testing Too Many Variables at Once: Making multiple changes simultaneously within a single cycle, making it impossible to determine which specific change led to the observed outcome.
• Abandoning Too Early: Giving up on a change after a single unsuccessful cycle, rather than adapting and re-testing.

A Practical Framework: Determining the 'Right' Number of Cycles

Instead of a fixed number, consider the following factors and questions to guide your PDSA strategy:

1. Project Complexity and Scope

• Simple Change: A straightforward process adjustment (e.g., changing the location of a form) might require fewer cycles (2-4) to test and embed.
• Complex Change: A multi-component intervention affecting various teams or patient pathways (e.g., redesigning a referral pathway) will likely require many more cycles (5-10+) to test each component, understand interdependencies, and integrate the overall change.

2. Risk and Impact of Failure

• Low Risk: If a failed test has minimal impact, you can be more agile and potentially iterate faster with fewer cycles between refinements.
• High Risk: If a failed test could significantly impact patient safety, staff workload, or finances, more rigorous, small-scale testing and multiple cycles are essential to mitigate risks.

3. Rate of Learning and Data Feedback

• Quick Feedback Loop: If you can gather data and learn quickly from each test (e.g., within an hour or a single shift), you can perform more cycles in a shorter timeframe.
• Slow Feedback Loop: If data takes longer to collect or process (e.g., monthly audit data), each cycle will naturally take longer, potentially limiting the total number of cycles practical within a project timeframe. This necessitates careful planning for robust data collection.

4. Degree of Confidence and Understanding

• Early Cycles (Learning & Understanding): Focus on understanding the current process, identifying root causes, and testing initial hypotheses. These might be very small, rapid cycles to learn quickly.
• Mid Cycles (Refinement & Optimisation): Once the basic change concept works, subsequent cycles focus on refining the process, addressing unforeseen issues, and optimising performance.
• Later Cycles (Confirmation & Implementation): The final cycles confirm that the change reliably produces the desired outcome in varying conditions and prepares for wider implementation and standardisation.

5. Sustainability and Spread

Beyond simply proving a change works, PDSA cycles are needed to embed it sustainably. This includes:

• Testing in Different Contexts: Does the change work on different shifts, with different staff members, or in different sub-specialties?
• User Acceptability: Is the change practical and acceptable to those who need to enact it?
• Monitoring Plan: How will the sustained change be monitored? The 'Act' phase of later cycles often involves planning for control and maintenance.

Guiding Questions for Each Cycle:

Before embarking on a new PDSA cycle, ask:

• What specific question are we trying to answer with this cycle?
• What precisely are we testing this time?
• What do we predict will happen?
• How will we know if our prediction was correct (what data will we collect)?
• What is the smallest scale we can test this on?

Example in Clinical Practice: Improving Discharge Summaries

Project Aim: Reduce the time from patient discharge to the completion of a GP-ready discharge summary for patients on Ward X to under 24 hours.

Initial Observation: Average time is 48 hours, often due to missing information or delayed dictation/typing.

PDSA Cycle 1: Testing a new prompt

• Plan: Introduce a sticky note prompt on patient whiteboards for junior doctors to complete discharge summaries immediately post-discharge. Predict: A slight improvement, but not consistently under 24 hours. Test on 5 discharges.
• Do: Apply prompt for 5 patients. Collect time-to-completion data.
• Study: 2/5 summaries completed within 24 hours. Doctors noted forgetting the prompt or being too busy.
• Act: The prompt alone isn't enough. Need to integrate it more effectively into workflow or add a reminder. Conclusion: Adapt.

PDSA Cycle 2: Integrating prompt with handover

• Plan: Incorporate discharge summary completion as a mandatory item on the handover checklist for the outgoing junior doctor, prompting them to check their patients' status. Predict: Better compliance. Test on 10 discharges.
• Do: Implement checklist change for 10 patients. Collect data.
• Study: 6/10 summaries completed within 24 hours. Better, but still not consistent. Some doctors still delayed due to perceived time pressure.
• Act: The prompt needs to be combined with protected time or prioritisation. Conclusion: Adapt.

PDSA Cycle 3: Designated 'Discharge Slot' & Admin Support

• Plan: Implement a 15-minute 'discharge summary slot' within the junior doctor's afternoon on-ward time, supported by a ward administrator to chase initial dictation/electronic entry for patients discharged that morning. Predict: Significant improvement, consistently meeting the target. Test on all discharges for one week.
• Do: Implement slot and admin support for a week. Collect data on all discharges.
• Study: 85% of summaries completed within 24 hours. Feedback positive from junior doctors (protected time) and admin (clear task). Some remaining issues with complex summaries requiring senior input.
• Act: Adopt this core change. Plan a further cycle to address complex summaries (e.g., senior doctor review slot). Begin to roll out to a second ward for a broader test. Conclusion: Adopt & Plan for Spread.

This example shows that multiple cycles were needed to refine the intervention from a simple prompt to a multi-faceted approach involving workflow changes, protected time, and administrative support. Each cycle built on the learning from the last.

How Lazomis Can Help

Lazomis provides structured tools to help NHS teams manage their QI projects and PDSA cycles effectively:

• QI Project Setup: Define your project aim, measures, and initial change ideas, setting the stage for robust PDSA planning.
• PDSA Cycle Tracking: Document each cycle's plan, predictions, data, learnings, and next steps in a standardised format. This ensures clear record-keeping and facilitates team collaboration.
• Data Dashboards: Visualise your run charts and control charts, allowing for quick analysis of 'Study' phase data and identification of patterns or shifts indicating improvement or decline.
• Knowledge Library: Access resources like this one to deepen your understanding of QI methodology and best practices.

By centralising your QI efforts within Lazomis, teams can ensure consistent application of PDSA methodology, track progress transparently, and make data-informed decisions about when to adopt, adapt, or abandon a change.

Key Takeaways

• There is no magic number of PDSA cycles; it's a dynamic process driven by learning.
• Start small, learn fast, and iterate based on your findings, gradually increasing the scale of your tests.
• Rigorous data collection and analysis in the 'Study' phase are critical for informing subsequent 'Act' phases.
• Avoid common pitfalls like skipping phases or scaling too quickly without sufficient testing.
• The number of cycles depends on project complexity, risk, feedback speed, and confidence in the change.
• Use multiple cycles to refine changes, address different contexts, and ensure sustainable implementation.

This resource supports, but does not replace, clinical judgement. Local policy, formulary and specialist advice should be followed.