The Study stage of the PDSA cycle is where you look for trends. This is the third step in the Plan-Do-Study-Act method, and its entire purpose is to analyze the data collected during the Do stage, compare results to your original predictions, and determine whether a meaningful pattern has emerged.
What Happens in the Study Stage
The Study stage has three core tasks: complete your data analysis, compare results to what you predicted would happen, and summarize what you learned. Before the cycle even begins, your team should have agreed on measurable outcomes. The Study stage is where you hold your results up against those benchmarks.
This goes beyond a quick glance at numbers. The Institute for Healthcare Improvement describes this step as analyzing data, studying results, and reflecting on lessons learned. You’re not just checking whether something improved. You’re trying to understand why it did or didn’t, what surprised you, and whether the change is sustainable over time. Teams that skip the reflection piece, or collect the wrong type of data, often draw incorrect conclusions and make poor decisions in the next stage.
One common pitfall is collecting only quantitative data without any qualitative feedback. Numbers can tell you that a change had an impact, but without understanding staff attitudes, workflow issues, or patient experiences, you won’t know the reasons behind the results. Another mistake is failing to notice unintended consequences, which can lead the team to adopt a change that causes problems elsewhere.
How Trends Are Detected
The standard tool for spotting trends during the Study stage is a run chart: a simple line graph plotting your measure over time, with a horizontal line at the median. Half the data points sit above the median, half below. The purpose is to detect non-random patterns, meaning signals of genuine improvement or degradation rather than normal fluctuation.
Three specific rules help you read a run chart:
- Shift rule: A shift is an unusually long run of consecutive data points all sitting on the same side of the median. This suggests something fundamental has changed in the process.
- Crossings rule: You count how many times the line crosses the median. Too few crossings signals non-random variation.
- Trend rule: A trend is an unusually long run of consecutive data points all moving in the same direction, either up or down.
These rules have been studied since the 1940s and remain the most practical way to distinguish real change from noise. Quality improvement experts recommend studying your data with run charts before moving to more complex statistical tools like control charts, which are better suited for fine-tuning processes that are already performing at acceptable levels.
How Study Differs From Check in PDCA
You may have seen the older version of this method called PDCA, where the third step is “Check” instead of “Study.” The distinction matters. The Check stage in PDCA focuses on measuring whether improvement occurred and then moving on. The Study stage in PDSA puts a spotlight on deeper analysis, asking not just “did it work?” but “what did we learn and why?”
This deeper analytical focus is one reason PDSA became the preferred model in healthcare, where small changes can have complex and unpredictable effects. The PDSA model emphasizes small, continuous improvements tested over repeated cycles, and the Study stage is the engine that drives learning between those cycles.
What Happens After You Find a Trend
The trends and patterns you identify in the Study stage directly shape your decision in the Act stage. Based on what the data shows, you choose one of three paths: adopt the change because it worked, adapt the change by modifying it and running another cycle, or abandon the change entirely and try something different.
This is where the cyclical nature of PDSA becomes important. A single cycle rarely gives you a definitive answer. If you encounter a barrier during the Study stage, you might retract the intervention and design a new approach for the next cycle. If the data shows a positive trend but raises new questions, you might expand the test to a larger group. Each cycle builds on the knowledge from the one before it, and the quality of that knowledge depends entirely on how thoroughly you analyzed the data during Study.
Monthly reporting using run charts or statistical process control charts helps teams track whether improvements are sustained across multiple cycles, not just in a single test. A change that looks promising in one small trial might not hold up over time, which is exactly why the Study stage exists: to give you the evidence you need before committing to a permanent change.