I have been thinking a lot about cancer screening tests. It seems that there has been a constant stream of articles about screening in both the lay press and professional journals — as well as the inevitable stories in the lay press about the reports in professional journals — but this is more personal. I have had two recent experiences that I can’t get out of my mind.

A few weeks ago my wife and I went to Boston for a bittersweet reunion with a tight-knit group of my former medical school classmates. It was a bit of a last minute thing, triggered by a classmate’s business trip to Boston and our learning that one of “the gang” had advanced metastatic prostate cancer. Ironically, the one with prostate cancer is an oncologist. Amidst the merriment of our first hour together, between updates on our careers and stories of our kids, he addressed the “elephant in the room” by speaking frankly about his condition and prognosis. I won’t share details, but he did mention in passing that the problem with screening is that it “sometimes finds the small ones and sometimes misses the bad ones,” which I took to mean that whatever screening he had had, it failed to find his disease at a more treatable stage.

The comment probably stood out because of the conversation I had had with another doctor friend a week before. We were out to dinner with our wives, and I mentioned our upcoming trip and how a man our age was dealing with metastatic prostate cancer. I guess I must have said something about my own decision to not have a PSA done when I last saw my internist over the summer, to which he nearly erupted at the table. It was some variation on: “Are you nuts? You just told me that you have a friend with metastatic prostate cancer and you don’t want to get your PSA checked?” It is fair to say that the conversation got a bit strained after that. I tried to go with the “poor specificity of the test leading to unnecessary treatment” line of reasoning, but it was clear that there was an unbridgeable gap in our thinking on the subject, and I wanted to bury the conversation before it ruined everybody’s evening, so we dropped it and moved on to dessert.

I think these two related conversations have stuck with me for a few reasons.

First, I am sad about my old classmate, and of course his illness is an unwelcome reminder of my own mortality. Second, they remind me that it is incredibly difficult to separate the “anecdote” from the “evidence” in our clinical practice, and in our own lives. Third, thinking clearly about the risks and benefits of screening is hard: there is a lot of uncertainty in the evidence.  And finally, even if we could all agree on the facts, there would still be no single right answer about what to do.

Ultimately, decisions about screening, like so many other decisions about testing and treatment, come down to personal preferences — how we value certain states of health, how we feel about risk, and how we approach uncertainty.

Ira Nash is a cardiologist who blogs at Auscultation.