Tuesday, July 21, 2015

Cite Your Sources!

Does my title bring back horrible memories of high school, when the teacher wrote "cite your sources" in big red pen on your page?

Well, he or she was right.

I saw this graphic on my facebook timeline.  I sighed.

It's getting worse.






THIS IS NOT OK!

First of all, the commonly used (and incorrect) statistic is 30%.  As in, 30% of women with early stage cancer will eventually have a metastatic relapse.  Although I suck at math, I did produce a child who is majoring in math at Caltech, and I don't think 30% is the same as 1 in 3.  So this number is getting worse as time goes by.

This bugs me because I understand how these things happen - somebody makes a mistake and it makes the rounds online and once it does - it never goes away.  Ever.

I'm still getting the Neiman Marcus Cookie Recipe meme first seen in the 1990s.

So what's the mistake, you ask?  You've seen this number everywhere and believe it's true.   For a while, I believed it too, so I don't blame you.  But I decided to look a bit further, and after doing a lot of research and listening to people who are smarter than me, I've come to believe this 30% number is false.

If not false, at least there are no legitimate sources that back it up.

So where did it come from?

It comes from the MBC Network which a lot of us rely on to get our stats, and which is considered a legitimate source by publications across the globe.

The MBC network is honest and they cited a source for their statistic.   I'm sure they used it with good intentions, but it's wrong and has been (mis)used so often people now take it as gospel.

First, the citation is from a CME test written by a Dr. O’Shaughnessy in 2005.    For those of you who don't know what a CME test is - it stands for Continuing Medical Education.  Doctors must keep up to date with their field, and so they take these Continuing Medical Education tests for credits. (I take them for fun too, and I pass more than I fail, which tells you how easy they are.)  They read a paper, answer some questions and get a credit.

This oft-cited test had a throwaway line in it that said,

"Despite advances in the treatment of breast cancer, approximately 30% of women initially diagnosed with earlier stages of breast cancer eventually develop recurrent advanced or metastatic disease."

That line has no citation with it.  Nobody knows where it comes from or where she got that number. She could have totally made it up.   A friend of mine had written her to find out, with no response.    I have not been able to find anything that confirms it, and I have looked extensively.

But if you read it carefully, it doesn't even say that 30% of women will have a metastatic relapse.  It says approximately 30% will develop recurrent advanced or metastatic disease.

Where is the punctuation?  Should it read "recurrent, advanced, or metastatic disease?"  Breast cancer can recur without being metastatic - women can have another incidence of local breast cancer.  Or they can have a relapse of locally advanced cancer, which means in the lymph nodes, but not necessarily a metastatic relapse.  (Metastatic is the Stage 4, bones and organs lethal disease we all know and love.) So did she mean 30% of all early stage women will have some sort of brush with cancer again, including metastatic cancer?  Then what is the percentage who will have a 2nd cancer with no mets?  Looked at that way, that 30% number goes down for metsters.

Who knows what she meant?  It's a badly written sentence and more important, she didn't back that number up with any study or anything.  It was a leadup to make a point, and she never expected that to be quoted and cited and used everywhere. (I hope.)

Anyway, that sentence had nothing to do with the reason she wrote the test.  It wasn't important and shouldn't have been cited by the Metastatic Breast Cancer Network since it had no citation or study behind it. (I have asked them to take another look.)  Plus, it was written in 2005, and while I don't know the numbers, I think medications like Herceptin and Perjeta are going to prevent some relapses and change some numbers.  So if nothing else, it's a bit old.  Still, that 30% number is prevalent everywhere, and I have heard oncologists such as Eric Winer dispute and qualify that number.

So what is the real number?

That is THE question. And I don't think there is an answer for it because I don't think any agency tracks relapse rates purely in this manner.  I don't think there is a motive to track this and keep these numbers.

If you lump all breast cancer patients together - if you have a 70 year old woman with Stage 1a, ER+ cancer, and a woman with Stage 3c Triple Neg who is age 25, and a woman who is 65 with Stage 2 HER2+ cancer - if you put them all in a bag and shake it out, you can come up with some metastatic relapse percentage (not 1/3, I'm positive).   But in what time frame do they relapse? 3 years? 5 years? 20 years? How do you get data like this? Who is tracking this in this fashion?  

Nobody, to my knowledge.   Government agencies  know how many women are diagnosed with breast cancer each year, and they know how many die of metastatic breast cancer each year.  Drug companies and research institutions do tests on treatments and track that.  There are survival statistics by stage, by subtype, histology,  etc, based on research and studies.  But do they lump it all together to make a generalized percentage of all women with BC who relapse over the course of their lives?

If so, how?   What are the methods? Some women live 15 years, some 3. Some are survivors of 14 years who have a late metastatic relapse.  Some are like me and had a relapse within 2 years.  How would they factor into any tracking?   A woman diagnosed Stage 4 from the start is easy to track but what about those of us who started out early stage?  When I die, metastatic breast cancer will be put on my death certificate but until that day, I don't think anybody is interested in the fact that I relapsed.

Think about it - to get to that number, the government would have to take all the women diagnosed with breast cancer say, in 1990.  That would be your study population. Fast forward to 2015.  How many of those women relapsed?  How many died?  How long did they live? Then you could kind of extrapolate that out to get the yearly number.  But as far as I can tell, it's not done that way and of course, treatments have changed over the past decades.

I might be wrong and I would love it if somebody could give me an accurate picture of this and a couple of sources (that are cited).    But to my knowledge, this overall stat doesn't exist and that throwaway line in a test was a guess.

Another way to look at it:  if 250,000 women per year are diagnosed with early stage invasive breast cancer  (which isn't the case) and 40,000 women die of metastatic breast cancer every year (another fluctuating number) then you can say generally that maybe a little over a sixth of women will relapse.  (That is not 60% either!) But that's pretty flaky as some of those 40,000 were diagnosed Stage 4 from the start and there are a million variables that I haven't thought of because...?  You guessed it.  I suck at math.

The real truth?  None of this matters for an individual.  You will relapse or you won't.  And it's more likely if you have a later stage, a more dangerous histology, if you aren't a chemo responder, etc.  But it's not a certainty.  I have a friend with 3c breast cancer, diagnosed at 22.  She is healthy and cancer free today, at age 30.  Me, Stage 2a diagnosed at 51, I am the metster.  You just don't know.

As human beings, we want concrete answers. Those of us further along the line of metastatic cancer come to realize that stats are not important.  Sure, at first we try to squeeze ourselves into those boxes, as if they could foretell our future.  But we learn that they are for populations, not people. Stats say I should never have had a metastatic relapse.  Stats say that since I did, I should already be dead.  It is human nature to want to apply those numbers to us. That isn't the way it works though.

I don't think it helps our cause as advocates for women with metastatic cancer to repeat misinformation.  I also think that certain groups (K*ough*omen) deliberately misuse statistics for their own purposes, stats that were never meant to be used the way they are and which are now misunderstood by everybody.  To me, these groups look foolish.

Do we -  as metastatic advocates and patients - want to join them?

I don't.  I think it is powerful enough to say 40,000 women a year die of metastatic cancer, without having to pretend 1/3 of the women with early stage breast cancer will be one of them.

Most people have no understanding of statistical analysis and why it's done. So these kinds of misunderstandings happen, even to me as like I said - math is hard.  But I think it's harmful to misstate things, use scare tactics and otherwise try to make a bad thing worse.

Trust me, it's already bad enough.

Bottom line:  Don't believe everything you read online, even from me.  And don't forget - check your sources!




18 comments:

  1. If the statistic came about in 2005 it is inevitable inaccurate because of the advances recently made with HERS+ cancer. Pre-Herceptin HERS+ had recurrence/relapse % of 91%. It is much lower now due to Herceptin and it's progeny. Herceptin began to be used in 2000 or thereabouts and survival rates are calculated by 3, 5 and 10 years. Even if the 30% was accurate at some point in time it is no longer accurate today.

    ReplyDelete
    Replies
    1. I'm sure that's true too. Thank goodness!

      But that statistic isn't really a statistic, it was just a statement. I'm pretty sure they don't track it this way.

      Delete
  2. Thank you, thank you, thank you. I had searched the medical literature ( I am a doctor with breast cancer) and I could find no basis. You summed up everything beautifully. Again, I say, thank you.

    ReplyDelete
    Replies
    1. Thank YOU. Coming from a doctor and somebody trained to look into this, it means a lot. I'm glad you looked for it too and came up with the same thing I did - nothing.

      FYI: I've been looking into this for 7 months. Just can't find it.

      Delete
  3. Thank you for summarizing this information so well. I have always questioned the 30% and I fear the numbers may be higher. But we are also individuals with our own DNA makeup and our own unique response to treatments.

    ReplyDelete
    Replies
    1. I don't think anything would points to a higher number. But it doesn't matter much if it's higher or lower - if it's .0001% it would be too many!

      Delete
  4. Hi Ann,
    I try to stick to the facts, too, because I agree, we need to be credible. I have used this 30% figure because I assumed it was accurate. I must admit, I'm confused. I am looking forward to more discussions on this. If it's an inaccurate figure, we need to adjust. Regardless, 40,000 women and men dying every year from this wretched disease is a horrible number and that's the bottom line. Thanks for the post.

    ReplyDelete
  5. Thank you for this. The problem I have with the wrong statistic is that it adds a horrible amount of unnecessary stress to someone who is trying to survive ... to the point where I think it actually does harm ... http://bcbecky.com/2015/07/there-is-harm-in-giving-false-numbers/ ...

    ReplyDelete
  6. Look slike MBCN has responded to questions about their statistics here: https://mbcnbuzz.wordpress.com/2015/07/26/mbcn-responds-metastatic-recurrence-rate-statistics/

    ReplyDelete
    Replies
    1. So I've been reduced to a "cancer blogger?" They won't even say my name?

      The blogosphere used to be about sharing and debate and linking. They refuse to even allow people to see what I said. Disappointing.

      On the other hand, I must have more power than I thought I had to scare them enough to write a rebuttal without even mentioning me. They completely misunderstood my point, got defensive but ultimately agreed with me, so I guess I don't get why it threatened them.

      Delete
  7. one of the best blog posts I've ever read. period.

    ReplyDelete
  8. The Metastatic Breast Cancer Alliance was formed to help correct the issues with data and MBC, as well as fight for needed research in MBC. We MBC patients are a few treatments away from "we have nothing else for you." Without the research, I will run out of treatment options and die. Folks cite the 30% with hopes it will put the problem in perspective and feed public support. Let me use another number here. Almost 600,000 people in USA die of Stage IV cancer. That's impressive and certainly points to the urgent need for research in metastatic cancer research. Read the MBCA report. It tells you how little money is spent on MBC research http://www.mbcalliance.org/. WE all need to work together to focus on getting research funding money, so all of MBC patients can have the hope of years to come. And, btw, here's my source for the almost half million. http://www.cdc.gov/nchs/fastats/deaths.htm

    ReplyDelete
  9. Hi Ann,

    Thanks for getting the ball rolling...it led to this recent Medscape article:

    The Mystery of a Common Breast Cancer Statistic
    Nick Mulcahy
    August 18, 2015
    (free registration required)
    http://www.medscape.com/viewarticle/849644

    You're right, the same old supposition keeps getting passed around as valid information, but is it?

    Thanks for the food for thought!

    Healing regards, Stephanie

    ReplyDelete
  10. Hi Ann,

    Have you read this Medscape response to your above blog post?

    http://www.medscape.com/viewarticle/849644#vp_1

    best wishes, Stephanie

    ReplyDelete
  11. HI Ann,
    I am with you on citation. I have came across blogs, and articles that are just write ups with no citation. These blogs are just out there hoping to make some advertising money. The bloggers do not claim anything and just copy and paste literature from which they just read somewhere.

    ReplyDelete
  12. According to Dr. Friedman, Collins was unusual in that, when her breast cancer was diagnosed, it was already stage 4, which is the most advanced type of the disease. “Only about 4 percent of U.S. women are diagnosed at stage 4,” he says.

    According to the American Cancer Society, only 22 percent of women survive five years after being diagnosed with such an advanced stage of cancer.

    Read Latest Breaking News from Newsmax.com http://www.newsmax.com/Health/Health-News/jackie-collins-breast-cancer/2015/09/20/id/692464/#ixzz3mPrsd2C4
    Urgent: Rate Obama on His Job Performance. Vote Here Now!

    Try to ignore the Newsmax self promotion. I read these numbers today and had to wonder 4% of what? All the women in the U.S., all the women with breast cancer? More meaning less numbers but I thought the rest of the info was informative.

    ReplyDelete
  13. My oncologist is a Dr. O'Shaughnessy. I suppose it could be someone other than her but that would be a wicked coincidence. I have an appointment tomorrow; I may ask her. :)
    BTW I am a 32 year old stage iv metastatic patient, and Dr. O is amazing!! She is one of the top researchers in her field, especially in regards to young women with metastatic breast cancer.

    ReplyDelete

Thank you for commenting. If the post is over 14 days old, the comment will be moderated and will approved later. This is a spam prevention technique - but I love to hear from you!