News & CommentaryArchive
Nov 19, 2009
More Evidence of Evidence Failure
I want to believe that arming people with good information will allow them to make better decisions, though I’ve seen plenty recently to suggest that belief is wrong. Books like Dan Ariely’s Predictably Irrational and Sunstein and Thaler’s Nudge have sold thousands on the premise that people do not often do what is in their best interest or make good decisions with the information they have. A recent article on the market failure of cancer prevention drugs offers another data point for their position. The piece outlines how doctors and patients alike have put great hope on diet, exercise and vitamin supplements as effective cancer preventatives. Yet research has produced only mixed results for the impact of diet and exercise, along with some damning evidence that vitamins may actually cause illness (check out the range of comments on this blog post for a taste of the response). The more hopeful news that treatment drugs Tamoxifen and Finasteride also work respectively as effective breast cancer and prostate cancer preventatives has been met with indifference.
Tim used the examples of hormone replacement therapy (HRT) and immunizations to articulate a similar point about behavior a few weeks ago—namely that good evidence on the increased risk of breast cancer caused by HRT caused almost immediate cessation of the use of hormones to alleviate menopausal symptoms, while very bad and well-contradicted evidence on the link between vaccines and autism has created a sometimes-violent anti-vaccine movement. It seems to me that the thread in all four examples (HRT, vaccines, vitamins and Tamoxifen/Finasteride) is human denial about the body’s limits. We are willing to believe that taking medication is bad for us; less willing to accept the opposite (and so many don’t believe that vitamins cause harm because they don’t see them as medication—they come from food!). The issue, in short, is nestled deep in human psychology. I don’t know how effective a drug has to be to change our minds, but if any measure can be gleaned by Tamoxifen, which cuts cancer risk by 50 percent, then the difference needs to be as close as science can get to a guarantee.
The relevance of these cases for philanthropy is that there are strong negative implications for prevention efforts related to any variety of illnesses—intestinal worms, measles, malaria, HIV. It is no secret that HIV prevention, for example, attracts only a fraction of the donations given for treatment programs. (This has always seemed illogical to me. What could be more effective at alleviating suffering than preventing an infection from occurring in the first place?) But if tests on the ground show that patients will not take drugs expressly for prevention than pharmaceutical companies will not research and test them, and donors will focus their efforts in areas where they can see the impact. This may result in a decrease in financing and research from private enterprise, and an increased burden for the nonprofit sector. So it is worth thinking about why pharmaceutical prevention efforts are sometimes unsuccessful, and finding ways for people to make good decisions with good information.