The Importance of Measurement
VISA International administers and operates the global network that handles VISA credit card transactions. On many days, the transaction volume on the network exceeds $1 billion. When considering potential investments in new fraud controls, VISA, like many other credit card operations, considers its "confirmed fraud rate" (which indicates the proportion of VISA card transaction dollars lost to fraud and never recovered.) They like to keep that rate below ten basis points (or 0.1 percent), a level the credit card industry generally regards as satisfactory for fraud in the system, the "acceptable price of doing business."
The health care industry, where the fraud rate might be 10 percent or higher--at least one hundred times worse than in the credit card industry--lacks precise instrumentation on which to base its control investment decisions and generally makes no serious attempt to measure the problem. Only in the last three years has the Office of Inspector General (OIG) instituted its annual audit program for Medicare, and that with a weak methodology that produces misleadingly low loss estimates. Only two states ( Texas and Illinois) have so far attempted to measure the losses due to fraud, abuse, and errors in their Medicaid programs. Other major public programs do no measurement of the losses, nor do most commercial insurers. Given the general lack of such measurement efforts within the industry and the painfully slow development of them within the major public programs (even when they are required by law), it's worth pausing to stress the importance and effects of measurement; many believe measurement studies are a distraction, a waste of time, a research luxury with no practical benefit.
Basic decision theory teaches the value of information when choosing between alternative courses of action. Presumably the health care indus-