# Quality Improvement Methodology: PDSA, Lean, and Six Sigma

## Introduction

**Quality improvement (QI)** is a systematic, data-driven approach to improving healthcare processes and outcomes. QI is a core competency for preventive medicine residents under ACGME requirements for **practice-based learning and systems-based practice**. The Institute of Medicine's **six domains of healthcare quality**: Safe, Effective, Patient-Centered, Timely, Efficient, and Equitable (STEEEP) Three dominant QI methodologies are used in healthcare: the **Model for Improvement (PDSA)**, **Lean**, and **Six Sigma**, each with distinct origins, philosophies, and tools.

## The Model for Improvement and PDSA Cycles

### The Model for Improvement

Developed by **Associates in Process Improvement** and promoted by the Institute for Healthcare Improvement (IHI) Built around three fundamental questions: **What are we trying to accomplish?** (Aim statement) **How will we know that a change is an improvement?** (Measures) **What change can we make that will result in improvement?** (Change concepts) Changes are tested through iterative **Plan-Do-Study-Act (PDSA) cycles**.

### PDSA Cycle Components

**Plan**: State the objective, make predictions, develop a plan for the test including what data to collect, who will do what, and when. **Do**: Carry out the test on a small scale, document observations and problems encountered, begin data analysis. **Study**: Complete data analysis, compare results to predictions, summarize what was learned. **Act**: Determine what modifications should be made; adapt, adopt, or abandon the change; plan the next cycle. PDSA cycles are designed to be **rapid, small-scale tests** that build knowledge iteratively; start with one patient, one provider, one day.

### Writing Effective Aim Statements

Aim statements should be **SMART**: Specific, Measurable, Achievable, Relevant, and Time-bound. Example: "Reduce the average time from patient arrival to influenza vaccination from 45 minutes to 20 minutes in the adult medicine clinic by June 30, 2026". Include the **population, timeframe, and measurable target**.

### QI Measurement

**Outcome measures**: What are we trying to improve? (e.g., vaccination rate, A1c levels, readmission rate) **Process measures**: Are we doing the right things? (e.g., percentage of patients screened, percentage of orders placed) **Balancing measures**: Are we causing unintended consequences? (e.g., increased wait times, staff burnout, patient dissatisfaction) Data should be displayed on **run charts or control charts** to distinguish true improvement from random variation.

<image>Diagram of the PDSA cycle showing four quadrants (Plan, Do, Study, Act) arranged in a clockwise circle, with an upward-sloping ramp beneath showing iterative cycles building upon each other over time, each cycle smaller and faster as the change is refined, leading toward the aim at the top of the ramp</image>

## Lean Methodology

### Origins and Philosophy

Lean originated from the **Toyota Production System (TPS)** developed by Taiichi Ohno and Shigeo Shingo in post-WWII Japan. Core philosophy: Maximize **value** (defined by the customer/patient) while eliminating **waste** (muda) Lean was adapted for healthcare by organizations including **Virginia Mason Medical Center** and **ThedaCare**.

### The Eight Wastes in Healthcare (DOWNTIME)

**D**efects: Errors requiring rework (medication errors, mislabeled specimens) **O**verproduction: Doing more than needed (unnecessary tests, redundant documentation) **W**aiting: Idle time for patients, staff, or information. **N**on-utilized talent: Underusing staff skills and knowledge. **T**ransportation: Unnecessary movement of materials, specimens, or equipment. **I**nventory: Excess supplies, medications, or equipment beyond what is needed. **M**otion: Unnecessary movement of people (walking to distant supplies, searching for equipment) **E**xtra processing: Steps that add no value to the patient (redundant approvals, unnecessary forms)

### Key Lean Tools

**Value stream mapping (VSM)**: Visual representation of all steps in a process from beginning to end, distinguishing value-added from non-value-added steps. **5S workplace organization**: Sort, Set in order, Shine, Standardize, Sustain; creates organized, efficient workspaces. **Standard work**: Documented best practice for performing a task consistently; the foundation for improvement. **Gemba walks**: Leaders go to the actual workplace ("gemba") to observe processes, engage with frontline staff, and identify improvement opportunities. **Kaizen events**: Focused, short-duration (3-5 day) improvement workshops targeting specific processes. **A3 problem-solving**: Structured problem-solving approach documented on a single A3-sized paper; includes background, current condition, root cause analysis, target condition, countermeasures, and follow-up plan.

<image>Value stream map of a patient visit in a primary care clinic showing each step from arrival through check-in, vital signs, provider assessment, orders, checkout, and follow-up scheduling, with time in minutes noted for each step, value-added steps highlighted in green and non-value-added steps (waiting, redundant data entry, searching for supplies) highlighted in red, with total lead time and total value-added time calculated at the bottom</image>

| QI Methodology | Origin | Core Philosophy | Key Tools | Best For |
|---|---|---|---|---|
| Model for Improvement (PDSA) | Associates in Process Improvement / IHI | Iterative testing of small changes | PDSA cycles, run charts, aim statements | Rapid, small-scale improvement in clinical settings |
| Lean | Toyota Production System | Eliminate waste; maximize value | Value stream mapping, 5S, Gemba walks, Kaizen events | Improving flow, reducing wait times, efficiency |
| Six Sigma | Motorola / GE | Reduce variation and defects | DMAIC, statistical process control, root cause analysis | Reducing errors, standardizing processes |

| Lean Waste (DOWNTIME) | Definition | Healthcare Example |
|---|---|---|
| Defects | Errors requiring rework | Medication errors, mislabeled specimens |
| Overproduction | Doing more than needed | Unnecessary lab tests, redundant documentation |
| Waiting | Idle time | Patient waiting for results, delays between steps |
| Non-utilized talent | Underusing staff skills | RNs doing clerical tasks, pharmacists not consulted |
| Transportation | Unnecessary movement of materials | Specimens sent to distant lab |
| Inventory | Excess supplies | Expired medications, overstocked supply rooms |
| Motion | Unnecessary human movement | Walking to distant supply closet |
| Extra processing | Steps adding no value | Redundant approvals, unnecessary forms |

## Six Sigma

### Origins and Philosophy

Developed at **Motorola** in the 1980s and popularized by General Electric under Jack Welch. Goal: Reduce process **variation and defects** to 3.4 defects per million opportunities (6 sigma level) Six Sigma is more **statistically rigorous** than Lean, emphasizing measurement and data analysis. Healthcare applications include reducing medication errors, surgical site infections, and diagnostic delays.

### DMAIC Framework

**Define**: Identify the problem, scope, customer requirements, and project goals; develop a project charter. **Measure**: Collect baseline data, determine current process capability, validate measurement systems. **Analyze**: Identify root causes of defects and variation using statistical tools. **Improve**: Develop and implement solutions; pilot and validate improvements. **Control**: Sustain improvements through monitoring, control plans, and process standardization.

### Six Sigma Tools

**Process capability analysis**: Measuring how well a process meets specifications. **Fishbone (Ishikawa) diagram**: Systematic identification of potential root causes organized by categories (people, process, equipment, materials, environment, management) **Pareto charts**: Identifying the vital few causes that account for the majority of defects (80/20 rule) **Statistical process control (SPC) charts**: Monitoring process performance over time to detect special cause variation. **Failure mode and effects analysis (FMEA)**: Proactive risk assessment identifying potential failure modes, their severity, likelihood, and detectability. **Root cause analysis (RCA)**: Retrospective investigation of adverse events to identify systemic causes.

## Lean Six Sigma and Integration

**Lean Six Sigma** combines the waste elimination focus of Lean with the variation reduction emphasis of Six Sigma. Lean addresses **process flow and efficiency**; Six Sigma addresses **quality and defect reduction**. Many healthcare organizations use an integrated approach, selecting tools from both methodologies as appropriate. Certification levels follow the **belt system**: White Belt (awareness), Yellow Belt (team member), Green Belt (project leader), Black Belt (expert/mentor), Master Black Belt (organizational leader)

## Applying QI in Preventive Medicine

**Immunization rate improvement**: PDSA cycles to test reminder systems, standing orders, and workflow redesign. **Screening optimization**: Lean value stream mapping to reduce time from screening order to completed test. **Population health management**: Six Sigma analysis of variation in chronic disease outcomes across clinic sites. **Public health emergency preparedness**: FMEA to identify vulnerabilities in outbreak response plans. **Health equity**: QI methods applied to identify and reduce disparities in preventive service delivery.

<image>Comparison table of the three QI methodologies showing PDSA (origin: Deming/IHI, focus: testing changes, key tools: PDSA cycles and run charts, strengths: simple and iterative), Lean (origin: Toyota, focus: eliminating waste, key tools: value stream mapping and 5S, strengths: efficiency and flow), and Six Sigma (origin: Motorola/GE, focus: reducing variation, key tools: DMAIC and statistical analysis, strengths: data-driven rigor), with a central area showing their overlap and integration in healthcare QI</image>

## Key Clinical Pearls

Start small with PDSA cycles; testing a change with one patient on one day is more valuable than spending months planning a perfect intervention. Run charts are the most important QI tool for distinguishing true improvement from random variation; learn to identify shifts, trends, and runs. Lean thinking is valuable for any process redesign; mapping the current state before proposing changes prevents solving the wrong problem. QI is not research; it does not require IRB approval when the primary intent is to improve local processes rather than generate generalizable knowledge.

## References

1. Langley GJ, Moen RD, Nolan KM, Nolan TW, Norman CL, Provost LP. *The Improvement Guide: A Practical Approach to Enhancing Organizational Performance*. 2nd ed. San Francisco: Jossey-Bass; 2009.
2. Womack JP, Jones DT. *Lean Thinking: Banish Waste and Create Wealth in Your Corporation*. New York: Simon & Schuster; 2003.
3. Chassin MR. Improving the quality of health care: what's taking so long? *Health Aff (Millwood)*. 2013;32(10):1761-1765.
4. Perla RJ, Provost LP, Murray SK. The run chart: a simple analytical tool for learning from variation in healthcare processes. *BMJ Qual Saf*. 2011;20(1):46-51.