TALE OF TWO HEADLINES

I planned something else for today's blog but this is in keeping with my dislike of statistics.  I also tend to dislike headlines. Yesterday, the following headlines appeared in my medical news feed:

Lumpectomy Saves Lives In Early Breast Cancer

Lumpectomy May Have Better Survival Than Mastectomy

I can NOT be the only person on the planet who is thinking, "These headlines couldn't possibly be about the same data." And yet, they are.

SAVES LIVES is the way the headline appeared on MedPage Today.

MAY HAVE BETTER SURVIVAL is the way it appeared in Medscape.

You must sign up to access the pages.  It's free but you will need to set up an account and password.  

Bottom line?  Both articles are well written.  The Medscape one digs a bit deeper.  Some important, some VERY important pieces of information should be noted and based upon that information BOTH of those headlines are irresponsible.  They are misleading.  I suppose that will be this week's theme?  Skewing statistics, twisting the truth.... sensationalizing headlines.

Geez..... It's even on a .gov site, a service of the National Institutes of Health.  A direct quote:

"The results of the study: Women were more likely to survive breast cancer if they had the less invasive lumpectomy plus radiation, versus mastectomy and this was the case across all ages."

Ummmmmm.  NOT EXACTLY.....

I'm going to pull out the parts that I feel should be taken into consideration before such broad strokes are painted. Medscape starts one section of the article "Nice Study, But........"  And yes, BUT ..... Medscape reached out to two independent experts for their opinions.  While they were impressed with the study, they "expressed reservations about the findings."  Me, too.

One expert observed that the results seem "counterintuitive" and noted that they may be a reflection of unmeasured variables.  "Unmeasured variables include the use of adjuvant systemic therapy."  In English?  Chemotherapy.  When they analyzed the the data, they looked at the years involved for each of the groups.  It appears those in the mastectomy group were treated at a time when chemotherapy was primarily administered when there were lymph nodes involved.  Advances in chemotherapy are credited for a large part of the mortality reduction over the past two decades.  Another expert said that the absence of data regarding whether or not chemotherapy was administered was "especially concerning."  DAMN right.

Chemotherapy for the lumpectomy patients and not for the mastectomy patients?  Those of us who have had the pleasure of sitting in that chair can attest to the fact that CHEMO COUNTS.  Really?  These data were not included in the observational study??  Then, there was an issue regarding access to care among the mastectomy patients.  There are quite a number of glaring holes in this "observational study" with the splashy headlines that are ONCE AGAIN sending an overtly skewed message.

The researcher who conducted the study and her colleagues "admit that various unmeasured variables were 'likely' to have biased the findings toward higher mortality in women who underwent mastectomy."  And yes, "likely" is in quotes in the article.  REALLY???!!?!?  Ya THINK?????

Don't tell me half a story.  Put it ALL out there.  I'd love to know how much money was spent and the number of resource hours used to put this thing together.  Damn.  This feels like it wasn't even an observational study.  It feels like someone said, "Let's prove that lumpectomy is better than mastectomy and let's find a way to tease that information out of the statistics to show we are right."

If anyone can not log in and wishes to see copies of the articles, send me an email and I will be more than happy to share them with you.

anncicc@gmail.com

Just to pour some gasoline AND fan the flames, this comes on the heels of a study published in Medscape on January 15. The title:

Bias and Spin in Reporting of Breast Cancer Trials

The first sentence reads:

Clinical trial investigators are often biased in their reporting of results from breast cancer studies, and they use spin to emphasize secondary results when the primary end point is not significant.

Emphasis mine.  I can email that one, too.

This whole thing?

It's very disingenuous.  Again.