Skip to content
Back to search results

Six Sigma in Healthcare

Introduced by
Motorola, Bill Smith, Mikel Harry
Year
1986
Type
Statistical quality improvement methodology
Family
Quality management, Process optimization
Font size

What does it do?

Six Sigma is a data-driven quality improvement methodology originating at Motorola in 1986 that aims to reduce process variation and defects to achieve near-perfect quality (3.4 defects per million opportunities).

In healthcare, Six Sigma uses statistical analysis and structured project methodology (DMAIC: Define-Measure-Analyze-Improve-Control) to reduce errors, improve safety, and enhance patient outcomes.

The intuition behind the method

Healthcare processes have natural variation: not every nurse takes the same time to administer medication, not every patient responds identically to treatment. Six Sigma assumes that processes can be measured and optimized by reducing variation. The approach differs from Lean: Lean focuses on eliminating waste (doing the wrong things less often); Six Sigma focuses on reducing variation (doing the right things consistently). The two are complementary: Lean identifies which processes matter; Six Sigma optimizes them.

History

Bill Smith at Motorola introduced Six Sigma in 1986, named for the statistical measure (six standard deviations from mean to specification limit). The goal was to reduce manufacturing defects dramatically. General Electric, under Jack Welch (1995–2001), popularized Six Sigma across all business functions, achieving reported savings of $10+ billion. Healthcare applications expanded in the 2000s. The methodology evolved to include design-for-six-sigma (DFSS) for new processes. Debate continues about whether Six Sigma's benefits are overstated and whether it inhibits innovation.

When to use it

Use Six Sigma when a well-defined process has unacceptable defect rates or variation, when process data can be collected reliably, or when improvement requires statistically rigorous analysis. Most applicable to high-volume, repetitive processes (medication dispensing, laboratory testing, scheduling) where variation data are abundant. Avoid Six Sigma if the problem is novel (no historical data), if the process is unmeasurable, or if qualitative improvements (clinical judgment) dominate.

Limitations and considerations

  • Requires substantial statistical expertise; not all healthcare leaders and staff have training in hypothesis testing and design of experiments
  • Time-intensive: a Six Sigma project typically takes 4–6 months; not suitable for urgent problems
  • Focus on variation reduction can miss opportunities for process redesign or innovation
  • Statistical significance does not guarantee clinical significance; a measurable improvement in defect rate may not affect patient outcomes
  • Can be perceived as burdensome; staff may resist formal measurement and control if not managed with engagement and transparency

Source conditions do not replace independent scientific review or validation by an analysis engine.

Strengths

  • Rigorous statistical approach quantifies baseline variation and improvement, reducing subjective claims
  • Highly structured methodology (DMAIC) ensures disciplined problem-solving and completion
  • Certification programs (Black Belt) develop expertise and leadership in quality improvement
  • Produces measurable financial returns: documented savings at organizations like General Electric and manufacturing partners
  • Applicable to both manufacturing and service processes, including healthcare

Applications

  • Reducing medication errors by standardizing medication verification and dispensing procedures, measuring defect rates before and after
  • Improving laboratory turnaround time by analyzing variation in each step (specimen collection, transport, processing, reporting) and eliminating delays
  • Enhancing surgical quality by tracking complication rates and implementing standardized checklists to reduce variation in technique
  • Optimizing patient scheduling to reduce no-shows and overbooking by analyzing historical patterns and implementing reminders or confirmations
  • Improving patient safety in intensive care by reducing catheter-associated infections through standardized insertion and maintenance protocols

How it works

How it works

1

Define the process and the critical-to-quality (CTQ) characteristic that matters to customers (patients, clinicians). Collect baseline data on current performance: process times, error rates, variation metrics (standard deviation).

2

Measure capability: how consistent is the process? Calculate process sigma level, histogram variation, and identify early warning signals of deterioration.

3

Use statistical tools to identify root causes of variation: design of experiments, regression analysis, multivariate analysis. Distinguish special cause variation (assignable, fixable) from common cause variation (inherent to the process).

4

Implement changes to reduce variation: standardize procedures, eliminate special causes, improve control systems. Redesign process if necessary.

5

Monitor improved process to confirm gains persist. Establish control limits and monitoring plan to prevent regression. Train staff on new procedure and sustain improvements.

Common pitfalls

  • Applying Six Sigma to low-volume processes where statistical power is inadequate; need sufficient data to distinguish signal from noise
  • Focusing on easily measurable metrics that do not correlate with desired outcomes; e.g., reducing turnaround time for lab results does not improve patient care if results are not acted upon
  • Ignoring human factors in process design; standardizing a process without considering how staff actually work leads to poor adoption
  • Misinterpreting control charts; a point outside control limits signals special cause, not necessarily a problem to fix
  • Declaring victory prematurely; improvements often backslide if control systems are not sustained

Frequently asked questions

What is the sigma level and what does 'six sigma' mean?

Sigma is the Greek letter used to denote standard deviation, a measure of variation. Six sigma means the process mean is six standard deviations away from the nearest specification limit. At this level, defect rates are approximately 3.4 per million opportunities. Most processes operate at three to four sigma (66,000–6,200 defects per million).

What is DMAIC and when do I use it vs. DFSS?

DMAIC (Define-Measure-Analyze-Improve-Control) is used for improving existing processes. DFSS (Design-for-Six-Sigma) is used when designing new processes from scratch. Use DMAIC for incremental improvement; DFSS for new technologies or major redesigns.

What is a Black Belt and what training is needed?

A Six Sigma Black Belt is a certified expert trained in statistical methods, project management, and change leadership (typically 4–6 weeks of training). Green Belts have less training (1–2 weeks) and support Black Belt projects. Certification demonstrates mastery of DMAIC and statistical tools.

How do I choose what process to improve with Six Sigma?

Prioritize by impact: processes with high defect rates, high cost impact, and high frequency. Start with a pilot project where success is likely and benefits are clear. Communicate results to build momentum for larger initiatives.

How do I sustain Six Sigma improvements?

Establish control charts and visual management to monitor the key metric. Assign ownership and responsibility for sustaining the improvement. Include the metric in performance reviews and organizational dashboards. Periodically review whether the improvement persists.

Related methods

Sources

  1. Harry, M. J., & Schroeder, R. (2000). Six Sigma: The Breakthrough Management Strategy. Currency.

    ISBN: 9780385494015
    Copied.
  2. Pyzdek, T. (2003). The Six Sigma Handbook (2nd ed.). McGraw-Hill.

    ISBN: 9780071418935
    Copied.
  3. Tennant, G., & Field, A. (2007). Six Sigma and Lean: Convergence and Implications. The Quality Engineer, 17(4), 327–337.

    Open publication reference
    Copied.