Tools, Technologies and Training for Healthcare Laboratories

Six Sigma Basics: Process improvement, goals and measurements

Just what is Six Sigma? What has it accomplished? Does it apply to healthcare laboratories? Do I really have to learn it? All these questions (and more) are answered in this lesson. This is a gentle introduction to Six Sigma -- in fact, this lesson provides a "statistic-free" description of Six Sigma.

Why do I need to learn Six Sigma?

Now that Six Sigma books are regularly making the business bestseller list, it's clear that the next wave of the quality movement is upon us. Total Quality Management and Reengineering have given way to Six Sigma. Soon you won't be able to attend a meeting without seeing the words "Six Sigma" somewhere in the presentations.

So you're probably wondering if this is just another fleeting business trend, the corporate equivalent of a fad diet. It's undeniable that a good amount of hype is involved with Six Sigma, as happens with every new management movement. Here at Westgard QC, we have taken a long look at Six Sigma and discovered that behind the hype there are some truly valuable techniques to be learned. We also feel like we've been advocating Six Sigma for some time. In a paper published in 1990 on "Precision requirements for cost-effective operation of analytical processes," it was recommended that laboratories should required 4 sigma performance for any new method and that even more demanding criteria - 5 sigma to 6 sigma - may be necessary for applications where personnel have little training in analytical techniques [1]. You might say we were "Six Sigma'ing" before Six Sigma came around.

In learning about Six Sigma, a good place to start is to take a look at what Six Sigma has achieved in other industries. You see, Six Sigma isn't a new technique - it's been around since the 1980's. But at that time, it was only being used in select industrial and manufacturing companies. Slowly it has spread to service industry and other sectors. This is a good thing, because it means that there's been enough time to "work out the kinks" with the Six Sigma approach. It also means that Six Sigma has a proven track record. We can look at what Six Sigma has accomplished in other companies and use that to judge whether or not we need to learn more about it.

Here's a very short list of companies that have embraced Six Sigma: Motorola, General Electric, Allied Signal/Honeywell, Black & Decker, Dupont, Dow Chemical, Federal Express, Johnson & Johnson, Kodak, Navistar, Polaroid, Sony, Toshiba, and Quest Diagnostics.

What has Six Sigma accomplished?

General Electric is most often used as the example for Six Sigma results, in part because they have been so zealous about using this technique. GE estimates that in 1998, they made $750 million dollars due to Six Sigma improvements, with another estimated $1.5 billion in 1999. Their operating margins continue to make records, and the leaders of GE attribute this extraordinary performance directly to Six Sigma [2].

Motorola, the company that laid the foundations of the Six Sigma concept, has spent over a decade using Six Sigma improvement techniques - and has enjoyed tremendous long term success: 500% growth in sales, estimated savings of $14 billion, nearly 20% profit growth for a decade. After two years of Six Sigma, Motorola also won the prestigious Malcolm Baldridge National Quality award [3].

These are just two of the success stories. There are more and will be more in the future.

What exactly is Six Sigma, any way?

A complete description of Six Sigma would take a book. In fact, there now are several complete books on the market. The best known is Mikel Harry and Richard Schroeder's "Six Sigma: The breakthrough management strategy revolutionizing the world's top corporations" [4]. A long title, but only 300 pages - maybe that's why this book was ranked in the top twenty business books for the year 2000.

A simple description - Six Sigma is about PROCESS IMPROVEMENT, with emphasis on the general goal of achieving six-sigma process performance and the use of quantitative techniques to measure the actual performance of a process.

A Process can be defined simply, too. It's a sequence of activities that take an input and produce an output. In business, a process is supposed to add value to the input before producing the output (In government, a process is supposed to add confusion). Every business can be broken down into smaller and smaller processes. At the core, a process is the fundamental unit of business that needs to be managed; it describes how the business produces value.

Six Sigma takes the approach that you should first analyze your processes to determine how well they work. The measure of process performance is the number of defects per million products or million opportunities. The goal of Six Sigma performance is to achieve a very low number of Defects Per Million, less than 1 DPM, which is achieved when 6 sigmas of process variation fit within the tolerance specification or quality requirement for the product. This is better than 99% performance - it's 99.9997% perfection.

Once the "sigma-capability" of the process is known, you can determine what level of improvement is necessary. [A little bit of history: here's where the TQM "PDCA or plan-do-check-act" methodology comes into play, though it's been re-cast in various ways, e.g., "MAIC for measure-analyze-improve-control".] Much of the methodology for process improvement is given a proprietary spin in the different texts and by the different training organizations, but it comes basically from Juran's earlier project methodology for process improvement [5].

Not all processes can be improved to this ideal, but the closer one gets to Six Sigma, the fewer the defects and the lower the cost of production. This is Deming's principle of quality improvement leading to higher productivity, which in turn leads to lower cost [6]. A better process means higher efficiency, savings of time, money, and effort, and better customer satisfaction.

This all makes sense - some say it's just common sense. However, businesses are often short-sighted and look only at the numbers for the next quarter. It's always quicker and easier to improve the short term figures by cutting costs, rather than improving quality and productivity to reduce costs. When cutting costs, they make their employees "do more with less." As we've said before, doing "more with less" eventually becomes "doing less with less." Six Sigma rejects the notion that you create savings with a slash-and-burn attitude. Process improvement is the approach. Six sigma performance is the goal.

Six Sigma provides a new methodology for measuring process performance and refines earlier methodologies for making process improvements. These process improvement methodologies are data driven and make use of statistics. There it is: the S-word - statistics - but we're only going to mention it for now. Just remember that statistics are part of the tools for making sense of data, measuring process performance, and helping you make improvements in your processes. For companies that commit to Six Sigma, there is an ongoing effort on process improvement - this was called continuous process improvement in earlier iterations of quality management. True Six Sigma companies are always improving their processes.

What's this about Black Belts?

Six Sigma has some strange terminology associated with it: Champions, Green Belts, Black Belts, Master Black Belts, etc. We're not sure these titles are necessary, but they do help delineate key roles in the implementation of Six Sigma.

Champion: As you might guess, this is someone, usually high up, who sponsors a Six Sigma project through the business. The first few Six Sigma projects may encounter some technical, political, or institutional obstacles in the business environment; it's critical that someone is there to help the project and the project members overcome any obstacles and resistance.

Green Belts: This is a title for someone who is involved with a Six Sigma project "part-time." That is, they have a job with normal duties, plus the added involvement in a Six Sigma project.

Black Belts: This indicates someone who does Six Sigma "full time." Their entire work effort is focused on finding defects, wherever they might be, and eliminating them from the business processes. They move from department to department, heading Six Sigma projects.

Master Black Belt: This indicates the highest level of skill and training, and sometimes designates an outside consultant who specializes in Six Sigma process improvements. These individuals provide training to others involved in process improvement, as well as developing an organizational culture that supports the implementation of Six Sigma.

The useful thing about these titles is their recognition of the business environment. Six Sigma doesn't assume that projects take place in an ideal world where everyone has time and is devoted to quality. It realizes that most businesses are structured in a way that will resist Six Sigma improvements. Accordingly, there is a need for individuals who will steer the Six Sigma projects through the business.

In companies that embraced Six Sigma, the belt designations have had more practical meaning. Black Belts get more Six Sigma training than Green Belts, different career paths, etc. But overall, we find the terminology is a bit silly. In TQM, we had organizational leaders (champions), team facilitators (master black belts), team leaders (black belts), and team members (green belts). Granted, "Master Black Belt" has a lot better ring to it than Team Facilitator. And if it helps business leaders feel good about Six Sigma by calling themselves Black Belts, that's fine!

Does Six Sigma apply to healthcare and laboratories?

After all this, you may still be tempted to reject Six Sigma. "Healthcare is different" is the common refrain whenever there is a new management technique or a new quality improvement methodology. That was also the argument against TQM! But those who want to assure the quality of healthcare would greatly benefit from applying Six Sigma to characterize process performance and gain a real understanding of the magnitude of defects in healthcare processes. In today's "outcome" oriented healthcare, we can't continue to ignore the bad outcomes or "defects" in our healthcare processes!

In healthcare laboratories, we're actually quite lucky. Other businesses and other departments in a healthcare organization often find it difficult to define their processes - they don't know how to break down what they're doing into processes. Furthermore, they're not sure how to measure performance and they're not as well-schooled in problem solving and making process improvements. In contrast, laboratories have readily identifiable testing processes, with established measures of performance (e.g., precision and accuracy), well-defined experiments for collecting performance data, and previous experience in using statistics to characterize that data. What most laboratories lack is an understanding of how to define the tolerance limits or quality requirements for their processes and what performance is needed, i.e., the goal of having six-sigmas fit within those tolerance limits or quality requirements.

Where Six Sigma gets a bit trickier is in the area of making improvements. How do you improve a test? Most laboratory tests are obtained from manufacturers. If you change the chemical or operating conditions of those tests, you actually change those tests to the "highly complex" category under CLIA, which in turn requires much more extensive validation and documentation by the laboratory. Therefore, it's not generally recommended that you modify a test method. Alternatives are to switch to a new method or a new instrument, though those improvement may be very time-consuming and expensive. An innovative approach in the laboratory is to make improvements to the quality control procedures to match their error detection capabilities to the performance of the testing process. That's right, improve the detection of errors by careful selection of quality control procedures, then you'll reduce the number of defective test results.

Can new QC procedures have an impact on the performance of your laboratory? You bet! With properly designed control procedures, you provide the right amount of error detection and minimize bad test results. With properly designed control procedures, you also minimize false rejections, which reduces the number of repeat tests, saves time in trouble-shooting, or minimizes interruptions or delays in the delivery of test results. Your customers will be happier with fast and accurate test results. Your staff will be happier, too! There is a huge financial impact in the laboratory from both inadequate and excessive quality control. Six Sigma will help you manage your testing processes to improve quality and productivity and reduce costs.

So do I really have to learn Six Sigma?

After reading through all the arguments, we hope you'll agree that the answer is YES (and we hope you didn't skip to the bottom to find out.) Six Sigma is worth learning about - it will improve laboratories and healthcare if implemented properly. It's easy to do for laboratories and testing processes because it's a logical data-driven or scientific management technique. Not only is it worth learning about, it's worth doing well.

Our Six Sigma series of lessons and essays will describe how to apply Six Sigma to your laboratory. It will involve some use of statistics, mostly looking at graphs, but it won't hurt much - really. With this new series, we'll show you how easy it is to implement Six Sigma Quality Design and Control in your laboratory.

Six Sigma Quality Management

Six Sigma and Requisite Laboratory QC


  1. Westgard JO, Burnett RW. Precision requirements for cost-effective operation of analytical processes. Clin Chem 1990;36:1629-1632.
  2. Pande PS, Neuman RP, Cabanagh RR. The Six Sigma Way: How GE, Motorola, and Other Top Companies are Honing Their Performance. McGraw-Hill, New York, 2000, page 5.
  3. Ibid, page 7.
  4. Harry M, Schroeder R. Six Sigma: The Breakthrough Management Strategy Revolutionizing the World's Top Corporations. Currency, New York, 2000.
  5. Juran JM, Endres A. Quality Improvement for Services. Juran Institute, Wilton, CT, 1986.
  6. Deming WE. Out of the Crisis. MIT Center for Advanced Study, Cambridge, MA, 1987.

Books on Six Sigma Quality Management

Breyfogle FW. Implementing Six Sigma: Smarter Solutions Using Statistical Methods. John Wiley & Sons, New York, 1999, 791 pages. A heavy weight text, with a high level of equations and statistics. Not for the faint at heart.

Eckes G. The Six sigma Revolution - How General Electric and Others Turned Process into Profits. New York:John Wiley & Sons, 2001, 274 pages. One of the most readable books, with minimum statistics, but still treats process capability in a quantitative manner. Clearly distinguishes between the indirect method of estimating the process sigma-metric from counting defects and the direct method of calculating the sigma-metric from the observed standard deviation of the process. My favorite book so far, even though it's written by a psychologist.

Harry M, Schroeder R. Six Sigma: The Breakthrough Management Strategy Revolutionizing the World's Top Corporations. Currency, New York, 2000. The most popular of the current books, it mixes concepts and ideas with success stories from well-known companies. No equations and little discussion of process capability and quality control.

Naumann E, Hoisington SH. Customer Center Six Sigma: Linking Customers, Process Improvements, and Financial Results. ASQ Press, Milwaukee, 2001, 340 pages. Half of the book is about customer satisfaction and surveys; the other half is about doing something to make improvements. Doesn't work for me, but might be appealing to the marketing oriented person.

Pande PS, Neuman RP, Cabanagh RR. The Six Sigma Way: How GE, Motorola, and Other Top Companies are Honing Their Performance. McGraw-Hill, New York, 2000, 422 pages. Organized around the themes of I. Executive Overview, II. Gearing up and adapting Six Sigma to your organization, and III. Implementing Six Sigma - Roadman and tools. Good overall treatment of quality management and improvement. Mainly focuses on process yield and defects, with little discussion of process capability and control charts.

Pyzdek T. The Six Sigma Handbook: A Complete Guide for Greenbelts, Blackbelts, & Managers at All Levels. McGraw-Hill, New York, 2001, 711 pages. This handbook weighs in as one of the biggest books available. It provides a thorough coverage project improvement methodology and group problem-solving tools, management tools, control charts, and design of experiments. It takes Six Sigma as the theme and unites the quality improvement methodology behind that theme.

Smith GM. Statistical Process Control and Quality Improvement. Prentice Hall, 2000, 650 pages. Focuses on statistical process control (SPC), but provides a broad perspective on how SPC fits into the overall management of quality. Chapter 1 provides a good introduction to statistical tools for process control. Chapter 2 reviews TQM, Deming, Crosby, Malcolm Baldridge award criteria, and ISO-9000. Chapter 6 integrates process capability indices with six-sigma quality. The rest of the book is about statistical QC, different types of charts and techniques.