At least, that's what this intriguing op-ed in today's New York Times said. Jon Gabel of the University of Chicago's National Opinion Research Center reviewed the Congressional Budget Office's record over the past several decades in predicting the eventual costs and savings from Medicare legislation. And what he found was that the well-respected agency time after time overestimated the costs of legislation (the prescription drug benefit, for instance) or underestimated savings from cost-cutting measures (moving to the episode payment system for hospitals in the early 1980s, for instance).

His analysis has major implications for the health care reform legislation that Congress will take up when it returns after Labor Day. Gabel notes:

The budget office has particular difficulty estimating savings when it considers more than one change at once. For example, last December the office reported that it found no consistent evidence that changes in medical malpractice laws would have a measurable effect on health care spending. It also reported that increased spending on studies comparing the effectiveness of different drugs and medical treatments would yield no net savings for 10 years. Yet if both malpractice reform and comparative effectiveness studies were instituted simultaneously, they might work together to yield substantial savings; doctors would gain more confidence in the effectiveness of less aggressive treatments and, at the same time, could use those treatments with less to fear from lawsuits.

I complained last December when the CBO "options" report suggested comparative effectiveness research was not a money-saver. Gabel understands the mindset that led to such a conclusion. "Too often, 'unknown' becomes zero - even though zero is not a logical estimate," he writes.