A recent economic analysis concluded that patients with metastatic cancer value their treatments significantly more than regulators recognize, with many expensive new therapies looking like veritable bargains to most patients. Yet the study ignored the values really driving oncology spending—the warped incentives oncologists have to promote their own bottom line by prescribing expensive treatments.

Let’s start with the economic analysis, another in a series of studies by a USC team of health economists funded by a large drug company. The USC team began its work with what seems like a reasonable assumption: “The amount someone is willing to pay for a good is the best measure of its social value.”

This assumption is the basis for much economic thinking, and in most market contexts, it is a reasonable take on the situation. When I refuse to buy a pair of bike shorts at $90, for example, but purchase them when the price drops to $70, my decision making says something about the value I place on those shorts. Based on this assumption, the USC team looked at what patients were willing to pay out of pocket for cancer treatments, and calculated what value they placed on these treatments. They concluded that most cancer therapies, by this measure, are a bargain.

But when trying to determine the value of expensive medications, this assumption—that “the amount someone is willing to pay for a good is the best measure of its social value”–is hard to swallow.

For starters, the price of most medical treatments is not transparent to patients. More importantly, patients’ decisions are often not made by patients themselves, but instead are made by their doctors. Patients don’t know enough about all their treatment options to be the savviest of consumers. In my book Critical Decisions coming out this fall (HarperOne, September 2012), I write about the challenge of getting patients to share in their healthcare choices. I’ll write more about this book in upcoming posts. But briefly, the challenge of shared decision making is made steep by the emotions that often surround healthcare decisions. A patient who has just learned that he has metastatic cancer is probably not going to be emotionally prepared to do comparison shopping!

Scared patients turn to their doctors for advice.

And that is where oncologic decision making gets really messy. Because in the United States, at least, many oncologists make a good deal of their income selling drugs to their patients. Here is how it works.

Oncologists purchase intravenous chemotherapy from pharmacies. Patients then receive these drugs in the oncologists’ offices, with outpatient chemotherapy being an increasingly common setting for cancer care. The oncologists then bill patients’ insurance companies for the treatments, including billing the payer for the cost of the chemotherapy PLUS a percentage based mark-up.

Medicare, for example, receives bills from oncologists that charge 106% of the cost of the chemotherapy. Many private insurers pay even larger mark-ups, especially from oncology practices that dominate their local markets and thus have pricing leverage.

This “buy and bill” practice creates an incentive for oncologists to prescribe expensive treatments.  After all, a $6 mark-up on a $100 treatment doesn’t do much for the bottom line. But that $600 mark-up on a $10,000 treatment? Give that treatment to enough patients and we’ll soon be talking about real money.

Many oncologists vehemently deny being influenced by this financial conflict of interest. But such denials defy both logic and data. Oncologists would have to be superhuman not to be influenced, at least unconsciously, by such strong incentives. After all, there is often no single “best” way to treat any given tumor, and there’s often good reason to believe that expensive new therapies might be better than older, cheaper treatments. In the face of such uncertainty, how could oncologists avoid being influenced by the knowledge that those promising expensive new treatments also help generate so much income?

Indeed, a team of Harvard researchers examined lung cancer treatments both before and after changes in Medicare reimbursement procedures, changes that led to the 6 percent rule I discussed above. (Believe it or not, prior to 2003, oncologists made even more than 6 percent profit on most of the treatments they prescribed.) Based on the 2003 law, some chemotherapies became far less profitable for oncologists to prescribe. For instance, reimbursement for paclitaxel dropped tenfold, from a little over $2,000 to $225 per month whereas a close cousin of this drug, docataxel, remained much more expensive at around $2,500 per month.

The Harvard team discovered that the percent of lung cancer patients receiving outpatient treatment grew dramatically after 2003. In other words, patients were less likely to receive such treatments in the hospital, and more likely to receive them in the oncologist’s office where the oncologist could buy and bill. In effect, since oncologists were making less profit on each treatment they gave, they tried to make up for this loss in revenue by increasing their volume.

The Harvard team discovered a second thing too—that oncologists were more likely to prescribe docataxel than they were before the reimbursement changes, at the expense (literally) of cheaper drugs like paclitaxel. They shifted to the more expensive and more profitable drug.

The USC team I described above did not cite this Harvard analysis. Perhaps they were so enamored of the idea that patients know what chemotherapy they want (because, you know, patients are so well versed on the relative pros and cons of docataxel and paclitaxel), that they overlooked the likelihood that chemotherapy decisions are primarily made by oncologists.

If we want to get better value out of medical care, we should pay for value. Giving physicians an incentive to prescribe expensive drugs is bad medicine.

Peter Ubel is a physician and behavioral scientist who blogs at his self-titled site, Peter Ubel and can be reached on Twitter @PeterUbel.  He is the author of Critical Decisions: How You and Your Doctor Can Make the Right Medical Choices Together.