The RAREEVENTS Procedure(Experimental)

The RAREEVENTS procedure produces control charts for rare events. A control chart is a graphical and analytical tool for detecting unusual variation in a process and deciding whether the process is stable and predictable. A rare event is one that occurs infrequently, with a low probability.

In this chapter, a control chart for rare events is referred to as a rare events chart. The data that are plotted in a rare events chart represent the times between successive events. Usually these are adverse events that represent unwanted outcomes in a process, such as an incorrectly recorded bank deposit, a patient falling in a hospital, or a chemical spill. Rare events charts have gained acceptance in health care quality improvement applications because of their ease of use and suitability to processes that have low defect rates (Benneyan 1999).

An important assumption for a rare events chart is that the events are independent. The occurrence of one event does not affect the probability that another will occur, and the probability of an occurrence is approximately constant over time. Rare events charts should not be used to monitor clusters of events, such as cases of a contagious disease, which violate this assumption. See Woodall (2006) for a thorough discussion of different control charts that are applicable to health care quality improvement.

The data for a rare events chart are often the times between consecutive events, such as the intervals between accidental needle sticks in a hospital. The intervals can be recorded as integer or continuous values. The opportunities for events to occur must be approximately constant over time. For example, the number of times that needles are handled should be about the same each day if you are monitoring the number of days between accidental sticks. Alternatively, the data can be explicit counts of opportunities for occurrence that come between events, such as the number of surgeries performed between occurrences of postsurgical infection. These kinds of data are preferable but often are not available.

A rare events chart has two decision limits: an upper probability limit (UPL) and a lower probability limit (LPL). By default, these are based on a geometric distribution for integer data and an exponential distribution for continuous data. A data value that is greater than the UPL or less than the LPL signals unusual variation in the process. A value that is greater than the UPL indicates that the time between events might be increasing, in which case the events are occurring less frequently. Because the events of interest are usually adverse, this can signal an improvement in the process. Conversely, a value less than the LPL indicates that events are occurring more frequently, which can signal a decline in the process.

You can use the RAREEVENTS procedure to do the following: