FANTASIES OF COST CONTROL by David Himmelstein, M.D.

With spiraling costs threatening to derail Massachusetts’ health reform, politicians and health policy wonks are rounding up the usual cost-control suspects. Unfortunately, the tired ideas they’re trotting out have virtually no chance of success. Here’s a quick rundown of some things we know don’t work, and a few that do.

1- Computerization – In the 1960s, Lockheed marketed a hospital computing system that was first installed at the Mayo Clinic and then at El Camino Hospital. A 16mm film from that era proclaimed with great fanfare that this system would improve efficiency, eliminate paperwork, improve accuracy, speed communication etc. It proved a disappointment and was quietly abandoned at both hospitals.

Similar rosy claims for electronic medical records have appeared regularly ever since. But despite the fact that virtually all hospital billing and most physician billing is now computerized, the cost savings have never materialized. In fact, paperwork consumes a much higher proportion of health spending now than it did at the dawn of the computer age.

When it’s done right (which it usually isn’t) computerization can help improve the quality of care. But there’s no credible evidence that electronic medical records will lead to substantial cost savings.

2- Prevention

Back in 1986, Louise Russell wrote a book called “Is Prevention Better than Cure.” She concluded that almost all medical prevention efforts actually increased costs, though many saved lives. More recent work has reached the same conclusion. It turns out that smoking and obesity actually save money on health care in the long run, because their victims die younger and hence avoid years of costly medical care. And a review of a comprehensive database of cost effectiveness studies found that 80% of preventive measures actually increased costs. So prevention is worthwhile because it can save lives – but not money.

3- Disease Management – It’s an attractive notion that if we took meticulous care of chronic conditions we’d save money. Unfortunately, the data says otherwise. A Congressional Budget Office review found little evidence that disease management programs save money, and Medicare recently abandoned its disease management demonstration project because it has failed to realize any savings. Like prevention, such programs may improve care, but there’s zero evidence of cost reductions.

4- Higher Co-payments and Deductibles – The theory here is that when people have to reach into their own pocket to pay for care, they’ll use less of it. The Rand Experiment, a randomized trial that assigned some people to high deductible insurance plans and others to a plan that offered full coverage, seemed to bear this out – the high deductible group used less care.

But real life is more complicated than the Rand study suggests. In that study, no doctor or hospital had more than a few patients in the high deductible group. So cuts in the number of visits had virtually no impact on any doctor’s or hospital’s revenues. But a raft of other research shows that doctors and hospitals keep themselves busy, even when co-payments go up. So a doctor with empty appointment slots will tell their diabetic patients to come back sooner than one who’s fully booked. When there are more urologists in a region, more prostates get removed than in a region with few such surgeons. And people spend more days in the hospital when they live in an area with more hospital beds – e.g. Bostonians used about twice as much hospital care as similar people in New Haven. Years ago Milton Roemer, a distinguished public health leader recognized this fact when he proposed Roemer’s Law – an empty hospital bed gets filled

In Quebec, when national health insurance (NHI) came in and all co-payments were abolished, the total number of doctor visits didn’t change at all. Doctors were working, on average, 50 hours a week before NHI and kept working 50 hours per week after NHI. But the abolition of co-payments shifted care: the wealthy visited the doctor less often and the poor more often. Moreover, abolishing co-payments in Quebec led to a big increase in the proportion of patients with serious symptoms who actually got care. Basically, when you eliminate co-payments for everyone, poor people who need care are more likely to get it, and the wealthy cut back a little on unnecessary visits. Conversely, studies in Manitoba showed that when a conservative government came in and boosted co-payments, there was no overall decrease in visits or in costs; but care shifted to the wealthy.What Works?


1- A Streamlined Financing System – Eliminate the middle men (insurance companies and The Connector); pay hospitals, clinics and nursing homes a lump sum budget, like a fire department; and cover everyone under a single public program. These measures could save at least $5 billion annually on useless health care bureaucracy in Massachusetts alone.

2- Limit the Profusion of Expensive High Tech Facilities – As discussed above, in health care, if you build it the doctors make sure the patients come. So our excess of CAT scanners has resulted in massive overuse of CAT scans – which can deliver radiation doses equivalent to 500 chest x-rays. Right now, hospitals and clinics across the nation are investing in hugely expensive linear accelerator machines which will send the costs of prostate cancer treatment skyrocketing; but there’s no evidence it’s any better than older, less expensive treatment. So long as we leave health planning to the market, the expensive medical arms race will continue.

David Himmelstein
Associate Professor of Medicine at Harvard Medical School
and Co-Founder of Physicians for a National Health Program

This program aired on April 2, 2008. The audio for this program is not available.


More from WBUR

Listen Live