About Statistical Process Control

To help you understand the intent of the Statistical Process Control section, we have created this documentation to walk you through some key concepts of statistical process control as they apply to medicine.

It's about the process

A patient's journey through your facility can be thought of as a process. Individual measures are indicators of that process. For each cohort, we have selected 3-4 key measures for you to track so you can focus on the most telling indicators of your process for those particular patients. Once you have identified abnormal measure values, you can drill into your data and use other Clinical Analytics analytic tools to investigate what occurred in your process during that time. We have also selected relevant adverse events for each cohort, so you have a drillable at-a-glance view of any incidents.

Benchmarks are not the focus

The SPC section does not allow you to select a benchmark. The intent of SPC is to analyze the current processes within your facility, so we intentionally disregard all benchmarks (internal and external) in this section. This allows us to track the consistency of our process without comparing to benchmarks, targets, or other goals. Once we have established consistency, we can focus on overall improvement, such as lowering Length of Stay or increasing HCAHPS scores.

Types of variation matter

Variation is a general term used to describe changes in we see in a process or in the process outcomes. In any process, there are three types of variation we may see. It is important to understand the sources of variation unique to your facility so you can address their impact on the process.

Common Cause variation

Common cause variation is inherent to the process. These are differences in measures values that can be attributable to the standard practice at the facility, such as holiday scheduling, Operating Room policies or discharge routines. Holding individual employees accountable for common cause variation is generally not appropriate, as this variation can only be addressed by management making interventions to improve the process.

Special Cause variation

Special cause variation is variation that is assignable to a specific and preventable cause and is not inherent to the process. By drilling into the SPC charts, you may find variations in measure values to be attributable to a particular unit, a certain physician, or a unique event. Special cause variation can generally be addressed by holding individuals accountable.

External Cause variation

External cause variation, as its name implies, is change that is not part of the process nor assignable to any correctable special cause. External variation is a concept not traditionally addressed in Statistical Process Control. It is inescapable in medicine, however, since a hospital must treat the patients as they present. Without any changes in process or mistakes by hospital personnel, a change in case mix severity or volume can cause changes in outcome measures.

NOTE: Clinical Analytics accounts for most external variation in the SPC section with risk and volume adjustments to the measure values in the control charts. By statistically accounting for the diagnosis severity, age, facility, payer, and HCUP's Clinical Classification Software (CCS) categories for each encounter within the cohort, we can be reasonably certain that the variation you see in your control charts is assignable to common or special causes and is not due to external causes. See SPC calculations for more details.