Skip to content Skip to navigation

The problem with root cause analysis

February 1, 2008
by Brian Garavaglia, PhD
| Reprints
It's a buzzword that can very easily become meaningless because not all problems have a single cause


Root cause analysis has become an important quality assurance method for identifying problems encountered in long-term care. Without a doubt, it is critical to identify and isolate issues that can lead to problems in nursing home care. Root cause analysis can help you identify and correct issues, leading to higher quality care.

Unfortunately, however, root cause analysis has become a “buzzword” in long-term care, conveying the impression that asking questions and examining “the roots” of an issue, often in desultory fashion, will produce a solution. Also implicit in the advocacy of this approach is that it can be completed by anyone, regardless of skill level. All of this is far from the truth.

Nothing New

Root cause analysis is not new. Quality assurance and investigatory methods using it have been around for a considerable time. Moreover, the use of these methods is quite variable. J. Toyoda's five “Why's,” K. Ishikawa's “fishbone diagram,” V. Pareto's chart and analysis, and even more complex methods such as Bayesian analysis, all help to identify the cause of certain problems. However, regardless of the sophistication or complexity of the root cause analysis methodology used, certain considerations must always be kept in mind.

One thing that needs to be understood is that not all problems can always be reduced to a single root cause. In life, the complexity of many problems is such that they are multicausal or may even lack any true, discoverable root cause. There will be times when identifying a single cause will be impossible. This is not a shortcoming of root cause analysis; it is a reality of the world in which we live. Isolating a cause when dealing with human beings is always much more difficult than isolating physical forces acting upon physical objects.

Major Problem

Second, one of the major problems that practitioners of root cause analysis face is isolating the problem.1,2 Doing so can be much more difficult than initially thought. As John Chaffee writes, “The first step to solving problems is to determine exactly what the central issues of the problem are. If you do not clearly understand what the problem really is, then your chances of solving it are considerably reduced.”3

In teaching problem solving and analysis, I have personally come to find that people often have a great deal of difficulty isolating the problem and defining the starting point for any investigation. Consider this scenario: A resident falls out of bed and fractures her hip. The fall clearly led to the fracture. Add, however, to this scenario the following fact: The resident was in bed and suffered from compromised bone density. She was often quite restless in bed and suffered from dementia, but the family members who were the responsible parties chose not to consent to the use of side rails, even though they had been made aware that their relative was a fall risk and, in fact, had fallen earlier that day on the same hip. So, what is “the problem” in this situation? You begin to perceive the difficulty.

Larger Context

When many people think of root cause analysis, the example that comes to mind is a linear, straightforward “A” leading to “B.” In reality, the scenario I've just described is more common. Notice the larger amount of context that exists in that scenario. In most cases this larger qualitative context is what most healthcare practitioners have to deal with daily. As is evident, when the number of variables increases, the complexity of extrapolating to the true cause becomes increasingly difficult.

It is often taught that healthcare professionals can achieve a successful root cause analysis by asking questions and continuing to probe. However, as can be seen, if one starts off with the wrong question due to not properly isolating the problem, then the inevitable result will lead to a wrong answer.

This leads to the confusion that people experience involving facts, validity and truth.4,5 Often people feel that root cause analysis can stop at a fact since a fact is equivalent to a truth. Once, when doing a financial analysis within a long-term care setting, someone said to me that the numbers in a particular part of the expense area were “fact,” and therefore I was well over budget. Does this fact mean that it conveys truth? In this scenario, the numbers were inappropriately placed in my expense column and did not reflect the true costs. Although the facts demonstrated excessive expenses, the facts themselves were not true, and therefore would mislead someone if they took these facts as starting points to isolating the true problem.

Logic of Problem-Solving

Another distinction has to be made between what is valid and what is true. Here again, many individuals use the terms interchangeably, but they are not interchangeable. Let us take a brief excursion into the logic of problem solving. Using a syllogism (a method of deductive reasoning that is often used in root cause analysis in healthcare,) we can demonstrate the lack of interchangeability by using an absurd example:

All dogs have three legs.

Spot is a dog.

Therefore, Spot has three legs.

Pages

Topics