Friday, June 24, 2005
Correlation and Causation
It was either a Professor of Statistics or a Professor of Logic who taught me that if a correlation is established between A and B, there are three possible explanations, none of which, on the face of it, can be said to be more likely than another. They are:
A is causing B,
B is causing A, or
A third factor C is causing both A and B.
The June 13, 2005 issue of Modern Healthcare reported a study showing that hospitals in financial difficulty make more medical errors.
Based on these findings, the researchers (two economists with the U.S. Agency for Healthcare Research and Quality) expressed concern that “cost-cutting initiatives might erode patient safety and the safety culture within hospitals.”
In other words, if A is financial difficulty and B is medical errors, they jumped to the conclusion that A is causing B. As might be expected, Richard Wade, speaking for the American Hospital Association, readily agreed. After all, AHA’s main job as a trade association is to get more money for its members.
But what if B is causing A; i.e., making lots of medical errors causes financial difficulty?
Or what if some C is causing both A and B?
If C is management, my inclination would be to opt for the third alternative; i.e., that financial difficulty and medical errors are both being caused by inadequate management.
If I am right, providing more money won’t help. The remedy is to improve the management.
It was either a Professor of Statistics or a Professor of Logic who taught me that if a correlation is established between A and B, there are three possible explanations, none of which, on the face of it, can be said to be more likely than another. They are:
A is causing B,
B is causing A, or
A third factor C is causing both A and B.
The June 13, 2005 issue of Modern Healthcare reported a study showing that hospitals in financial difficulty make more medical errors.
Based on these findings, the researchers (two economists with the U.S. Agency for Healthcare Research and Quality) expressed concern that “cost-cutting initiatives might erode patient safety and the safety culture within hospitals.”
In other words, if A is financial difficulty and B is medical errors, they jumped to the conclusion that A is causing B. As might be expected, Richard Wade, speaking for the American Hospital Association, readily agreed. After all, AHA’s main job as a trade association is to get more money for its members.
But what if B is causing A; i.e., making lots of medical errors causes financial difficulty?
Or what if some C is causing both A and B?
If C is management, my inclination would be to opt for the third alternative; i.e., that financial difficulty and medical errors are both being caused by inadequate management.
If I am right, providing more money won’t help. The remedy is to improve the management.