Posted 7/6/2012 5:17 AM (GMT 0)
I second (ok, Raddad beat me to it, so I'll "third") Goodlife. Those prediction tools describe outcomes from men diagnosed in previous years. Statistical models predict outcomes well on average, when the world you're predicting (what will happen to your husband in the next few years) is consistent with the world the model was built on (what happened to men in the past).
There has been dramatic progress just recently and there's more in the pipeline, so much so that I believe that the averages that were calculated based on men diagnosed years ago probably won't apply to those of us using those tools today to try to make the best decisions about our tomorrows.
And I say this not just as a wife who hates the predictions those tools make for my husband, but also as a researcher who spent most of my career building predictive statistical models. I did that for manufacturers, so they could predict how many cookies, bottles of detergent, etc. they'd sell. The hardest category to predict was prescription drugs, because the history, the old results we wanted to use to predict the future, would become obsolete every so often when a new drug or diagnostic test changed things up. I feel like I just took a crash course in prostate cancer over the past month, and wow, it is evolving really quickly.
My heart breaks for what you and he are going through, and I absolutely understand his thinking. But please don't let those predictions steal your hope -- there are a lot of reasons to believe he'll do better.