How good are studies?

Tall Allen said...
Billybob-

I have tried to explain to you before that there is a standard that medical science sets for levels of evidence. I know you are as disinclined to learn about it as I am to read long, meandering posts with extensive quotes in the body of it. But I did actually look at the study you are talking about - it was very small (n=44) and non-randomized. How would any ever know what outcomes would have been if they had not taken the supplement?

In short, the study that you think is so important, actually shows nothing. It does suggest that if there is an effect, there cannot be very much of an effect. The hypothesis -- is there a benefit or a risk of supplementing Vitamin D in patients on active surveillance? -- cannot be answered by that study design. That question can only be answered by a large randomized clinical trial. Because there are much more important questions to answer, I doubt that such an RCT will be conducted anytime soon.

Indeed, they do not KNOW with out a placebo, they can only make a reasonable conclusion based on the historical norms at their institution. Which they supplied. Do you know of any follow ups of men on AS where 55% actually improve on their Bx and Gleason scores and even # of positive cores? But, even then, the numbers as you say are small. This calls for a follow up study, with placebo this time, to reproduce. Or not, however it goes. In the meantime, 6 years go by and no other similar studies to confirm or not.

I totally agree with you that no such RCT will be conducted anytime soon. You say because there are much more important things to be looked into, maybe so. Also, there are no billions to be made off of a vitamin D patent.

In the mean time, without the actual RCTs, I guess we can just go by the studies you have linked to previously declaring that Vitamin D is helpful for nothing. But, a declaration made on nothing like this study, where follow up Bx results are looked at for men after 1 year of 4000 IU/day. BTW, I'm sure you remember the warnings about the old U shaped curve? And those studies were based on relatively low levels of blood vitamin D even for the highest groupings. Well, did you notice any U shaped curve in this study with whopping doses of vitamin D administered? I don't see much but improvements on he whole, and especially compared to historical norms. You would certainly expect any U shaped curve to appear after an intervention of 4000 IU per day X 1 year, wouldn't you?

Even with such a small study and no actual placebo (only a comparison to their historical norms) many would be interested in another study that might fail to confirm it. I would. But not you apparently.

Why do you object so consistently to my long quotes from studies? You don't seem to like these studies for some reason. I'm suspecting that you have no trouble reading through long studies for patent medicines or RT treatments. Why the problem with these? Apologies in advance if this response has been too long and meandering.

Post Edited (BillyBob@388) : 4/19/2018 7:12:32 PM (GMT-6)

Tall Allen said...

BB said...
Do you know of any follow ups of men on AS where 55% actually improve on their Bx and Gleason scores and even # of positive cores?

Yes. In fact, that's quite common. Remember these AS patients were only given TRUS biopsies on f/u. To be clear... The median # of positive cores at baseline was about 1.1 and the median after treatment was 0.8 -- not a big difference. The percent where they found no cancer on a 12 core f/u biopsy was 34%, which is exactly the same (32%) as in this study among 657 men:

www.europeanurology.com/article/S0302-2838(18)30025-3/fulltext

You will also notice that the percent who stayed at the same grade was 52%, not significantly different from the 43% in the above study (p=0.23).

As I said, nothing to get excited over, and it isn't surprising there was no f/u.

If you want to get excited, look at this study (n=76) where a 5-ARi was used among men on AS. 54% had no cancer found on follow-up. And that difference is statistically significant.

I often read posts on my iphones, and it makes it difficult if a post goes on for multiple pages. I have no problem clicking on a link and don't need the info in both places. It actually detracts from what ever point you are trying to make rather than adding to it. Imagine if in those studies you are fond of quoting, if the authors had printed the full text study for every study they footnoted! If you would just pull out the excerpt from the study that you think is important, it would make your point a LOT more strongly. You may have noticed that you are the only one on this site who does this.

OK, point to you, at least for now as I don't have the time right now to go over that study in detail. So it seems you have a study where similar improvements were made with no intervention. Where as the study out of USC was comparing to their own " historical control subjects" in which case there was quite a difference.

Then apparently, it is normal per studies you have seen for most or many men on AS to actually improve on f/U bx? Or not? If so, all the more reason for low risk guys to go AS.

I'm going to have to look look closer at what you are saying about the mean # of positive cores, as in the study I quoted those #s(total #s of positive cores) dropped about 57% in the intervention group but would be expected to increase about 38% based on their historical control subjects. Hence their conclusion "Patients with low-risk prostate cancer under active surveillance may benefit from vitamin D3 supplementation at 4000 IU/d.".

Also, looking quickly through the euro study linked above, I do see that 25% of men were reclassified "increase in Gleason grade and/or an increase in the ratio of biopsy cores to cancer to
> or = to 34%.". I don't know about the ratio of Bx scores in the USC study(you probably do?), I only know about the total positive cores improvement mentioned above. But only 14% had an increased Gleason, compared to their norm of 21%, closer to the 25% "reclassified" I don't know how much of that was Gleason increase.

Post Edited (BillyBob@388) : 4/20/2018 9:59:21 AM (GMT-6)

All of your questions are answered in the full text of the study you cited, if you trouble yourself to read it.

BB said...
But what I never understand is: why do you even read my posts? Just skip over or, as you are so fond of telling others, there is always the "ignore button".

The danger of sites like this is that people will spout misinformation and others will believe them and be harmed as a result. That is doubly a problem when people quote studies that they don't really understand or bother to read in full (as you just demonstrated). They seem more knowledgeable that they are. I have tried to steer you in the right direction - take a class in statistics, read about levels of evidence, don't cherry pick studies, etc. But, thus far, you have made no attempt to heed that advice - I honestly hope you will - you might actually enjoy it! Frankly, I do mostly ignore your posts - I read the first few lines only, just to make sure you don't say anything that in my opinion may endanger others.

Looks like this thread has run its course...

Jim