Data Visualization Double Take

You know those times when you glance at something and it hits your brain in a way that causes you to do a double take?

Just the other day I actually did a triple take when I stumbled across a visual display of mammography data in a patient education publication.

First look — “interesting.”

Second look — “hold on a minute.”

Third look — “not so fast — this isn’t the whole story.”

The title of the visual display is: “Are Mammograms Important?” Underneath the heading are two black circles — the circle on the left is significantly larger than the one on the right. Underneath the large circle is a caption that reads: “average size of breast tumor at diagnosis in the early 1980’s when only 13% of women were getting regular mammograms.” The caption on the smaller circle says: “average size of breast tumor at diagnosis in late 1990’s when 60% of women were getting regular mammograms.”

data visualization double take

Frankly, this is a terrifically effective visualization. It is simple and conveys the authors intended message — “Mammograms are Important.”

But here’s the problem: There’s an assumption built in that more is better and some important and new information is missing.

What this display leads the viewer to believe is that mammograms are important for all women. After all –detecting and diagnosing breast tumors earlier is better, right? Well, maybe not.

Does the data support that? Does screening more woman regardless of their age or health status really result in better outcomes — specifically fewer deaths from breast cancer?

And, has the increase in women receiving annual mammograms had any unintended consequences — has it caused harm to women?

Bottom line — is more better? (I know the answer is yes if we are talking about chocolate — but I digress.)

The U.S. Preventive Services Task Force (PSTF) researchers studied mammography screening data in an effort to understand just these questions. The result of their study found that of women aged 40 to 49 and aged 70 and older:

  • There is virtually no difference in mortality rates between those who received a screening mammography versus those who did not.
  • Screenings included many false positives resulting in over-diagnosis.
  • As a result of over-diagnosis women are at risk of additional testing, biopsies, radiation exposure, and pain during procedures, anxiety and distress.
  • Full Article

And so, while the visual display described previously is quite compelling it has oversimplified and ignored information that women and their doctors need in order to make a fully informed decision about screening mammograms.

This visual display is a perfect example of the warning of English Mathematician and philosopher Alfred North Whitehead issued when he admonished us to:

“Seek simplicity and distrust it.”

Share
This entry was posted in Communicating Data to the Public, Data Analysis, Etcetera, In the News, Newsletters. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *


one × 2 =