Understanding variation in cardiopulmonary bypass: Statistical Process Control Theory.

Cardiopulmonary bypass (CPB) is a complex clinical process that takes place within an extremely complex care deliver system. During the past 50 years, many technicaland device-related improvements have been implemented. However, there remains enormous opportunity to improve. As has been shown in other industries, understanding variation is the key to improvement of any process, and Statistical Process Control Theory (SPCT) is a powerful method for studying variation. SPCT was first described and used by Walter Shewart in the 1930s to improve telephone component production at Bell Laboratories (1). Shewhart’s hypothesis was that production quality improves when front-line workers are informed and empowered with information necessary for that improvement. He invented control charts as a method for transforming data concerning variation into understandable information. With the application of this theory in the form of control charts, Shewhart provided production workers with a deeper understanding of the variation in their production system. Control charts distinguish random variation that exists in a system from variation that has a specific cause. Shewhart noted, “While every process displays variation, some processes display controlled variation whereas others display uncontrolled variation.” He called the former “chance cause” and the latter “assignable cause,” whereas today these types of variation are referred to as “commoncause” and “special-cause” variation, respectively. Common-cause variation requires no immediate action or intervention, whereas special-cause variation should be studied and perhaps acted upon. Acting on common-cause variation is wasted effort and, in many cases, may worsen circumstances because these actions often lead to making a process less stable than it was formerly. Shewhart referred to acting on common cause variation as “tampering” and considered it counterproductive (2). Stable processes are desirable because their outcome are likely predictable in their content and quality. Armed with this Address correspondence to: Robert Groom, Cardiac Surgery, Maine Medical Center, 22 Bramhall Street Portland, ME 04102. E-mail: [email protected] Robert Groom Donald S. Likosky Hans Rutberg JECT. 2004;36:224–230 The Journal of The American Society of Extra-Corporeal Technology