QUERI – Quality Enhancement Research Initiative

Go to the HSRD website

Quality Improvement Methods

44. Six Sigma

a. Definition: Six Sigma is an improvement approach incorporating many tools with a particular focus on quality as well as statistical control over processes. Based on data, improvement particularly comes from reducing unwanted variability. It shares many attributes with the Lean Approach (see Section 24). Six Sigma is a registered service mark and trademark of Motorola Inc. Other early adopters of Six Sigma who achieved well-publicized success include General Electric. By the late 1990s, about two-thirds of Fortune 500 organizations had begun Six Sigma initiatives with the aim of reducing costs and improving quality.[2] Training and certification in Six Sigma is available from many organizations. The term "six sigma" comes from a statistical reference to variability. A standard deviation is used to compute variability and is often labeled with the Greek letter sigma (σ). A good design results in only 3.4 defects out of a million, which is 6 standard deviations or Six Sigma, although this is not always achieved.

Six Sigma focuses on understanding and controlling a process with the use of quantitative tools. It makes a nice fit with research projects because both focus on understanding processes and often use extensive data analysis. Six Sigma's ideas offer the researcher opportunities for quality improvement in both conducting research and producing improved results. For example, Six Sigma is used to speed up the completion of the development of a medical device and it is used to track changes in quality and to reduce laboratory errors.

Many of the methods covered in this handbook are commonly referenced as part of Six Sigma as well as Lean or the Toyota Production Process. As a result of the overlap of Six Sigma and Lean, some refer to the "Lean Six Sigma" (LSS) method. The following are tools which are particularly part of Six Sigma but less often part of Lean.

  • Control chart (see Section 9)
  • Statistical analysis such as design of experiments (see Section 11) and analysis of variance (ANOVA)
  • Modeling and optimization (see Section 27)
  • DMAIC (see Section 12)

b. Literature: There is an extensive literature on Six Sigma but much of it is not focused on healthcare or research.

  • Schweikhart, Sharon A., and Allard E. Dembe. "The applicability of Lean and Six Sigma techniques to clinical and translational research." Journal of investigative medicine: the official publication of the American Federation for Clinical Research 57.7 (2009): 748.
  • Gras, Jeremie M., and Marianne Philippe. "Application of the Six Sigma concept in clinical laboratories: a review." Clinical Chemical Laboratory Medicine 45.6 (2007): 789-796.
  • Pyzdek, Thomas, and Paul A. Keller. The six sigma handbook. Vol. 486. New York, NY: McGraw-Hill, 2003. Gygi, Craig, and Bruce Williams. Six sigma for dummies. John Wiley & Sons, 2012.
  • Lean Six Sigma for Hospitals: Simple Steps to Fast, Affordable, and Flawless Healthcare by Jay Arthur (2011)
  • Improving Healthcare Quality and Cost with Six Sigma, by Brett E. Trusko, Carolyn Pexton, Jim Harrington and Praveen K. Gupta (2007)
  • Koning, Henk, et al. "Lean six sigma in healthcare." Journal for Healthcare Quality 28.2 (2006): 4-11. Adams, Rella, et al. "Decreasing turnaround time between general surgery cases: a six sigma initiative." Journal of nursing administration 34.3 (2004): 140-148.

c. Example: Six Sigma improvement projects generally follow a cycle called DMAIC (Define, Measure, Analyze, Improve, and Control) (see Section 10). For example, a research group wanted to improve the outcomes from a particular surgical procedure. First they defined in detail the sequences of processes involved. This is the Define step in DMAIC whereby a team representing all disciplines mapped the workflow, gathered statistical data and determined the issues that should be addressed. The team developed a clear written definition of the clinical decision making, the equipment involved and the patient mix. The next step was to Measure and Analyze the data available. Data was needed on a sufficient number of surgeries and the various aspects of each case. Also relevant literature was an input to the team. Of particular interest was variability and trends defined in statistical terms so that inferences could be gained about the system they were studying. This provided an insight into the factors causing problems and what was related to patient outcomes. By gaining such an understanding, particularly quantifying the relative importance of each factor, the team could know where improvements were necessary and which changes were important. Then, improvements were developed. The measurement and analysis also provides data for determining the cost-effectiveness of the changes. The source of improvements can come from several of the methods as discussed in the handbook such as a Kaizen, brainstorming or simulation modeling. The analyses above provided what should be expected in quantitative terms. The final step in the Six Sigma DMAIC cycle is to standardize the improvements and make sure the changes remain. Thus control, the C in DMAIC, is needed to sustain the changes.

d. Steps: Six Sigma improvement projects generally follow the DMAIC cycle and assume that there is always room for additional improvement. Thus the DMAIC cycle should be repeated (Define, Measure, Analyze, Design, Verify - or DMADV). These steps are similar to the Lean approach's PDSA (Plan, Do, Study, Act) where the emphasis is more on the change step whereby Six Sigma places relatively more emphasis on the analyze aspect. However both these sequences of steps are useful and can be used together. When the objective is to develop something new a DMADV sequence is followed, which is sometimes referred to as Design Six Sigma.