The problem with choice

What ends mean

This principle doesn’t just apply to mechanics. It applies to all experts. The whole point of consulting an expert is to leverage their expertise. We form a partnership of experts. I am an expert on what I want and they are experts on how to get it. I tell them the ends and they tell me the means. It’s an ongoing discussion because choosing different means sometimes changes the ends in unexpected ways—like side effects or complications. Communicating ends includes discussing the value placed on particular goals, the likelihood of achieving those goals, and the likelihood and negative value of undesirable outcomes. My mechanic knows that I value saving money. And she does not believe that any other ends will be risked in choosing to save money on Brand B. Her question to me was entirely about means. To make it purely an ends question she would have had to ask whether I wanted to spend extra money for an equivalent product.

When I try to influence the means I generally cause problems. You might think it couldn’t hurt to ask my mechanic about Brand C that I had seen in a billboard ad. But it might. Maybe my mechanic would lean toward Brand C just because I asked about it and not because she thought it was a good choice. I don’t really want her to do that. I care about making my car run well and saving money. I don’t care which fuel filter will achieve that end.

On the other hand, I would be concerned if my mechanic installed seat warmers without asking me. Warm buttocks are an end (sorry, I couldn’t resist). When experts start answering ends questions on their own, it creates problems.

You might think that all of this only applies if I have confidence in my expert, and that is partially true. I need to feel confident that my expert both listens to my goals and puts my interests first. But my expert doesn’t have to know everything. She doesn’t even need to know more than other members in her field. My mechanic knows more than I do and that is what matters. If I doubt her knowledge and abilities, my best strategy is to find a new expert, not to start telling her how to achieve my goals.

The end of means

Back in my office my first patient is in for a periodic health examination. I am about to ask him whether he wants a prostate-specific antigen (PSA) screening test when I pause. I know his goals. He is a healthy 50-year-old man. He wants to live as long as possible and has a very dim view of urinary incontinence and impotence. Why am I asking him about PSA screening? I don’t know whether a PSA screening test will increase his chances of living longer or better. I am just following the American Cancer Society guidelines which say: “Information should be provided to patients about benefits and limitations of [PSA] testing.”1

On the other hand, the United States Preventative Services Task Force “found good evidence that PSA screening can detect early-stage prostate cancer but mixed and inconclusive evidence that early detection improves health outcomes … evidence is insufficient to determine whether the benefits outweigh the harms for a screened population.”2

Prostate-specific antigen screening is not controversial because of conflicting ends. It is a promising but unproven means.

Most of my patients choose not to have PSA screening. I attempt to present the pros and cons in a neutral fashion because of the American Cancer Society guidelines, but I think my patients catch on to my opinion even if I don’t say it openly.

My anxious patients insist on screening anyway. I cringe before I ask them about it; PSA testing is very bad treatment for anxiety. They are hoping a negative test will reassure them that they won’t die yet. After years of experience I have learned to anticipate that if the test is negative, this specific anxiety will quickly be replaced by another one. Once I tried telling a patient that even if the test were negative he could still get hit by a bus. (I only tried that once.)

Some patients come to me specifically asking for the test after the personal recommendation of a highly qualified television talk show host. But others have not even considered the PSA choice until I bring it up. They must wonder why I raise the subject.

Of course, that is what brought to mind the bewildering discussion I had with my mechanic this morning. Upon reflection, I don’t think it is a good idea for experts to ask their clients or patients to make choices about means. If thousands of doctors can’t agree on whether PSA screening results in any benefit, then it makes no sense to ask the patient to settle the dispute. I am perfectly willing to discuss the pros and cons of PSA screening if my patients raise the subject (and order the test if desired); but it is a big leap to suggest that I should raise the subject myself and then subliminally discourage the test—or attempt to be neutral when I am not. There are many unproven interventions out there, and many interested parties who want me to discuss the pros and cons of their favourite interventions when the evidence is inconclusive. I have decided to stop doing this. I don’t want the experts in my life to ask me means questions and I am not going to ask them of my patients.