A recent article in USA Today posits that doctors perform thousands of unnecessary surgeries every year.  It claims that unnecessary surgeries might account for 10% to 20% of all operations in some specialties, including a wide range of cardiac procedures.  It generally cites three reasons as the cause of the problem:

– doctors who enrich themselves by bilking insurers for operations that are not medically justified;

– doctors who simply lack the competence or training to recognize when a surgical procedure can be avoided;

– and doctors who recognize that the need for surgery may be questionable, but have sufficient medical findings to proceed.

There is really no way to determine what extent of surgeries are related to poor training or fraud, beyond anecdotal examples.  However, according to National Practitioner Data Bank (which we discussed in my last post), over the past decade, approximately 1,000 doctors have made payments to settle malpractice claims or lawsuits in surgical cases involving claims of unnecessary procedures.

Proving a surgery was unnecessary in court can be problematic.  The issue of medical necessity will often devolve into a battle of the experts, with each side’s expert disagreeing with the other.  One indication that surgery is being unnecessarily performed is based on an examination of records and claims data.  When this indicates that a doctor’s surgery count is so misaligned to a doctor’s peer group, it is strong evidence that surgeries may be unnecessarily performed.  Consider this example provided by the USA Today article:

Medicare patients in Lincoln, Neb., were nearly four times more likely to have knee replacement surgery than those in Honolulu.  Lincoln’s rate for the surgeries was the nation’s highest, nearly 55% above the national average; Honolulu’s was lowest, less than half the average.

Under the fee-for-service model, it is generally in the doctor’s best financial interest to perform a surgery.  The 2010 health care law promises changes in the payment system, but only time will tell if and how any changes impact the current system.