Thursday, October 13, 2011

Incentive distortions in cancer treatments

I have blogged recently about market failures associated with health insurance which is a consequence of the inherent problems with health care. Nowhere is this more evident than in the case of cancer treatment.

Cancer screenings - mammograms for women and prostate cancer screening for men - have emerged as among the most controversial areas in healthcare, representative of the information asymmetry related market failures that characterize the sector. NYT writes about the difficulty of making screening calls,

"When doctors screen for early cancer, all the incentives — cultural, financial, professional and legal — line up in one direction: Don’t miss it. As a result, doctors overreact to even the tiniest abnormalities, which leads to the two basic harms of screening: false-positive tests and overdiagnoses.

False positives are really common in both breast and prostate cancer screening... Approximately 15 to 20 percent of women and men who are screened annually over a 10-year period will have to undergo at least one biopsy because of a false-positive mammogram or PSA — prostate-specific antigen — test...

Overdiagnosis is less common, but much more consequential because it leads to unnecessary treatment. Screening finds abnormalities that meet the pathological definition of cancer, yet will never go on to grow or cause any symptoms, let alone death. Sometimes patients choose to wait and see if the cancer grows, but most opt to treat it; once you’re told you have cancer, it’s difficult to wait and see what happens next.

Patients who are overdiagnosed are the big losers here. They undergo surgery, radiation and chemotherapy unnecessarily. And then there are the associated complications: chemotherapy can cause nausea and radiation can burn normal tissue; breast surgery can be disfiguring, and prostate surgery can lead to bladder and sexual dysfunction.

Doctors don’t know which patients they are treating unnecessarily, but they know how the unnecessarily treated patients got there in the first place — because they were screened for cancer.

Now let’s consider the winners — those who have avoided dying from breast or prostate cancer by getting screened. While there is some debate about whether they really exist, my reading of the data is that they do, but they are few and far between — on the order of less than 1 breast or prostate cancer death averted per 1,000 people screened over 10 years. That’s less than 0.1 percent.

Overall, in breast cancer screening, for every big winner whose life is saved, there are about 5 to 15 losers who are overdiagnosed. In prostate cancer screening, for every big winner there are about 30 to 100 losers. "

The incentives of all sides are aligned towards overtreatment. At the slightest suspicion of cancer, the patient is subjected to screening by a risk averse and liability claim fearing doctor. This impulse gets amplified if the doctors gets an incentive for every diagnostic test done, as is most often the case. The anxious patient is ever eager to get himself cleared off any suspicions and therefore willing, even demanding, that all tests be done. In any case, since he is insured, he bears no or little expenditures on those tests. The inusrers too prefer an early diagnosis, since they believe it would save them of the much greater expenses from the treatment once the case turns malignant.

1 comment:

sai prasad said...

It would be difficult to convince people to not take up screening for reasons of human psyche. The cases of over-treatment would appear very distant to the target group, while the death of cancer sufferers is very vivid.

No amount of 'rational' arguments would succeed.