Deciding between radiation and prostatectomy

I used these charts to make my decision to go for LDR brachytherapy when I was first diagnosed with low risk (Gleason 6) cancer in 2020 as it showed a better outcome to remain disease free than surgery. Four years later with a recurrence of high risk Gleason 9 Stage 3 N1 disease I wish I would’ve opted initially for surgery.

As someone that used data extensively in an arguably successful engineering career I can safely state that data always tells a story however the story it tells may be based on an amalgam that includes data that may be irrelevant to a given and specific situation. Caution is always warranted to make the best possible decision.

Best wishes to everyone on their journey with PCa.

Yes, I understand what you mean. We have the same problem in my professional field (I don't want to be too specific, because it might make it easy to identify me). There's a big push for "data-driven decision making", but data alone gives a false sense of confidence that you're taking a factual/neutral approach, when it's really just a very-low-resolution, pixelated snapshot of a complex, messy reality.

These days, there's a big push to combine quantitative and qualitative data (including feedback from the people we're supposed to be serving and the experience of practitioners) to get a better-rounded picture of what's happening and where we can fit in. The data still matters a lot — especially for things like drug trials — but it's just the start of the journey, not the end.

(This a general reminder for everyone, including myself, not a criticism of any specific forum participant).

I suggest that we laypeople be very careful about making our own charts and doing our own data comparisons from different studies with different methodologies, because we don't know what we don't know.

For example, most studies track *overall* survival (OS). Since the majority of prostate-cancer cases (not mine, sadly) occur late in life and develop slowly, many of the people in advanced stages might not have had 10 years left in any case, so the OS numbers are going to skew low.

Now, what if aggressive treatments like surgery are more common for younger patients than for elderly ones? That would mean that their OS numbers might come in higher even if the two treatments are equally effective, because the younger person is less likely to die from other causes in the next 10 years.

Researchers probably apply different statistical models (and some WAGs) to try to adjust for this, but if we laypeople just pull numbers haphazardly from different studies into a single table or graph, we won't have that benefit, and the results might be very misleading (as we saw with much of the amateur research during the COVID pandemic).