If I had chemio, my survival after 5 years would be 1.4%

Lucy2014 · May 2015

While collar butchers yet would be more than 1.4% richer

http://www.curenaturalicancro.com/en/aussie-oncolo...

Momine · May 2015

The reason the study has not gotten much attention is because the analysis is sketchy. You can easily find cogent and reasonable critiques of this paper, if you google.

Momine · May 2015

I am also wondering what on earth they mean by "absolute number of 5-year survivors due to chemotherapy."

leggo · May 2015

Absolute, as opposed to Relative.

leggo · May 2015

Absolute is a subtraction of percentages. Relative is a ratio. Absolute has a definitive starting point (zero). Relative begins with an amalgam of stats, like risk vs. benefit, odds, etc. The medical field interprets statistics in relative terms. This article interprets them in absolute terms and is pointing out the difference. In absolute terms the numbers suck. In relative terms, percentages "sound" great. It's all in the mathematical interpretation.

To me absolute is the only number that matters. Everything else is gambling. That's just me though.

leggo · May 2015

Here's a good example I found on google.

Consider the risk for blindness in a patient with diabetes over a 5-year period. If the risk for blindness is 2 in 100 (2%) in a group of patients treated conventionally and 1 in 100 (1%) in patients treated intensively, the absolute difference is derived by simply subtracting the two risks:

Expressed as an absolute difference, intensive therapy reduces the 5-year risk for blindness by 1%.

The relative difference is the ratio of the two risks. (NB: Relative risk, relative rate, rate ratios, and odds ratios are all examples of relative differences.) Given the data above, the relative difference is:

Expressed as a relative difference, intensive therapy reduces the risk for blindness by half.

juneping · May 2015

lol that's so misleading....thanks leggo!

Momine · May 2015

Leggo, what I meant is how they arrive at the number of survivors that can be attributed solely and specifically to the chemo. I am also wondering, in the case of the breast cancer stats, for example, how many patients actually got chemo. Given that they include cancers for which chemo is not typically a first line treatment (prostate and testes), it would stand to reason that the number of patients listed is the total number counted, not the number of patients that received chemo.

Put another way, take a bunch of diabetics, 100 diabetics. 95 of them are stage II with good glucose control, so they get metformin. 5 of the diabetics need insulin to survive. At the end of 5 years one diabetic has died, but 99 are still alive. So, 4 survived thanks to insulin, and therefore insulin only gives a 4% survival advantage, going by absolute numbers. Except not really, since 95 of the diabetics never needed insulin in the first place and the 4 who survived would most likely have been 100% dead without their insulin.

leggo · May 2015

Generally speaking, an absolute is a one-off. Relative has substantial mathematical and statistical variables. I think your example is relative, not absolute, so I'm not sure how to respond. I guess it all depends on whether or not one is willing to accept those variables, and we all know cancer has many, in a "statistic" as percentage of benefit/risk. I don't, when it comes to cancer. There's no grey area for me. Black or white, right or wrong, good or bad.....Absolute or relative.

Edited for clarity. Mind over keyboard.

leggo · May 2015

Hi June. Just read your appointment went well. Yay!

juneping · May 2015

thanks leggo!! It's interesting for sure ...

Momine · May 2015

No, Leggo, my example is not relative in the sense you mean. It is simply, as in the other study, including a load of irrelevant data into the starting sample. Anytime you calculate a percentage it is relative in some sense, however, because it is a ratio.

However, to go back to my diabetes example, if the benefit of insulin were calculated relatively, it would mean that insulin conferred a survival advantage of 80% (the 4 survivors relative to the 5 people needing insulin).

What the chemo study does is take those 4 people and turn them into a percentage relative to ALL the diabetics, in other words, they would say that the insulin only gave a 4% survival benefit.

I have read a lot of complaining here on BCO over the years about the "absolute" vs. "relative." It is not, in my opinion, something to be irate about. Relative risk reduction is the only practical way to calculate, but I do think it is important that doctors explain better exactly what that means for an individual patient. "50% risk reduction" sounds grand, but obviously isn't if the guy leaves out that my total risk is only 0.5% (or 2% or whatever, pick any low number). On the other hand, if your total risk is 50%, then a 50% reduction is pretty dandy.

suzieq60 · May 2015

What a load of crap - how could they know if a patient benefited from chemo or not. The best measure is if they develop mets or not. My husband had mets in the lung from bowel cancer - after surgery and chemo he is still with us 8 years later. I had chemo 6 years ago - I'm still here. I went into it knowing that there was no way to know if it would help but that is the whole point of it - it might just have helped prevent mets - that is what adjunctive chemo is for.

NancyHB · May 2015

The title of this post keeps catching my attention, so I know I need to read the article to understand what this poster is trying to say. Seriously - 1.4% chance of being alive at five years post-chemo? I'm quite certain that's not the intended message (...or is it?), and I see from responses that this is a mixed cancer sample so "apples and oranges" that can really skew results. I love a good discussion

Momine · May 2015

NancyHB, that is why it caught my eye too. I don't think the original poster meant it to come out that way, though, because even if you read the article, that is not what the paper says.

AmyQ · May 2015

So what's the survival rate with no chemo after 5 years? I'll take the chemo, thank you.

leggo · May 2015

I don't think anyone is getting "irate" about it. You stated you didn't know what "absolute number of 5-year survivors due to chemotherapy" meant. I responded with an explanation and an example, that's all.

Momine · May 2015

Leggo, I did not say that you, personally, were irate, only that I have read a fair amount of ire on the subject over the years. I should have been clearer. I do know the difference between "absolute" and "relative," as it applies to stats. I was more questioning how they arrived at the figure and the definition.

leggo · May 2015

Sorry, I misunderstood.

Momine · May 2015

No worries.

leggo · May 2015

Kayb, I just want to clarify something. Is this how you and your friend interpret and draw the conclusion of in the article?

"WHAT THIS MEANS IN NORMAL ENGLISH

Only 2.1 % of all the people who get chemo are alive 5 years later because the chemo saved their lives. The other 97.9% of the people who get chemo are dead. Many people died much earlier than they would have because the chemo killed them."

I understand you state that's it's misinterpreted on the internet that way, but I could use some clarification.

leggo · May 2015

Thanks for explaining.

leggo · May 2015

Unfortunately, that's how cancer, treatments and statistics are quantified after initial diagnosis....5 year survival. Everything we know is based on that starting point.

I think this is a very important study, albeit a little late showing up here. Not everyone is aware of the fact that the medical establishment bases their statistics in relative terms.

To me (and I'm only speaking for myself), starting at the 5 year mark and quoting relative statistics is a sketchy way of improving those numbers on paper. All it is is mathematical manipulation. I'm not saying that one is better than the other....they're both perfectly valid calculations, but not pointing out the difference to cancer patients is sort of playing a little dirty. Especially when we're glazed over from a cancer diagnosis.

suzieq60 · May 2015

I was told my risk of recurrence was 23% - chemo and Herceptin would halve that - good enough for me thanks.

Momine · May 2015

Leggo, I have to disagree about the importance of this study. The "everything and the kitchen sink" approach pretty much makes any analysis of the data pointless. It is not about "relative" risk in the normal sense. They even throw cancers in there for which chemo is not typically used. To go back to my little diabetes example, it is like calculating the efficacy of insulin not only relative to diabetics who don't need it or take it (similar to stage 0-1 BC patients, who make up the bulk of the BC patient population), but calculating the efficacy of insulin relative to diabetics who don't take it AND people with thyroid problems, say. Both are glandular illnesses after all.

Even the worst onc in the world is not going to tell a stage 0 BC patient that chemo will reduce her recurrence risk by 20%. He will only tell that to a patient for whom chemo is actually indicated and that 20% reduction is relative to that particular patient's total risk, which is a completely different "relative" than the one set up in that study.

If I had chemio, my survival after 5 years would be 1.4%

Comments

Categories