Thank you to Kaytee RR, Max RR, Sophie L, and my Sweetie for workshopping some of these ideas and encouraging me to write this piece.  Updated 3/27 thanks to generous feedback from Rachel W.

CW: Topics in this post include discussion of doctors and the medical industrial complex, body weight/dieting, a broken bone (no graphic descriptions) and frank discussion of racism.  If its not a good time for you to engage with these themes, I invite you to exit out and do something nice for yourself!  We’ll be back with more spreadsheet tips and tutorials soon.  For folks reading on, I look forward to your thoughts and feedback!

humpty-dumpty-256x300Humpty Dumpty had a great fall…

…leaving this sassy, feminist, tech blogger with a broken arm and a fierce critique.  (Yes, I broke my arm last weekend.  I am typing with one hand, fueled by how infuriated and indignant I feel about my experience at the Urgent Care center and what that has to do with data and justice.)

When I left the Urgent Care facility after I fell, staff members handed me a wad of receipts including my diagnosis, my visit summary and notices about my BMI and blood pressure.  In this blog post: why these data are NOT alright.

BMI?  More like Bad Measurement Initiative!

BMI (aka Body Mass Index) is a very common ratio (dating back to the 1800s) to indicate if a person is “over or under-weight.”  To calculate BMI, you would use this expression:

Weight in pounds/(height in inches x height in inches) x 703

(The 703 is to convert the index from the original metric version of the formula)

Then, you would match your answer against numeric ranges that represent weight/body type categories (underweight, healthy, overweight, obese) (without taking sex, age, & environmental factors into account).  Also, let’s recognize that these term and concepts that describe a “standard, healthy” body deserve further scrutiny.  Read more here about an alternative approach to health and wellness.

You could figure out you own BMI using arithmetic OR you could go to Urgent Care for a broken arm and get your BMI handed to you in your take home receipts!  Note: this is a pattern heavier women face all the time when navigating doctor visits (go to the doctor about something completely unrelated to weight and next thing you know…)  [angry face]

There’s a big problem here:  As a health predictor, BMI is notoriously inaccurate.

  • Propensity for density:  Since BMI is essentially a density formula, it mistakenly classifies extremely tall, short or muscular people (especially professional athletes) as overweight or obese.  P!nk also has something to say about this.
  • squarehead_sh-960-x-960
    Who am I? What am I?

    FALLacy:  The math backing up the BMI formula is a logical fallacy.  Sure, people who are obese have high BMI.  But all people with high BMI are not obese.  Its the good ‘ole “a square is a rectangle, but a rectangle is not a square” trick.  We know better than to fall for this one!  #PunsIntended #Ouch

  • Racism is no joke – so no punny headline.  As if BMI wasn’t bad enough already, new research suggests that BMI is extra, extra bad at predicting chronic illness across different races and ethnic heritage.   Turns out there’s no such thing as a “standard” weight or size, (newsflash!) especially not across hundreds of years and different cultures and climates.  And yet medical professionals STILL compare everyone to a so-called “standard” (White) body, and fail to identify health risk factors, especially for people of Chinese and South Asian descent.  Read more at the links above.

Ok so we’ve established that BMI is a) ineffective; b) based in bad logic; and c) racist.  Do you still need convincing that BMI is terrible?  Check out 10 reasons why BMI is bogus.

Why the data are NOT alright

I was shocked when I read the receipt about my weight and blood pressure.  I know that my blood pressure is healthy, having just completed a battery of testing a few months ago.  It was no surprise that I had elevated BP … since I had just broken my arm and had severe pain!  As for the BMI results, they went on to suggest that I cut 500 calories per day from my diet.  Really?  25% of the “recommended” calorie intake?  That seems extreme – and reeks of bad advice — especially when we didn’t even have a face to face conversation.

If I was re-designing the health metrics system at the urgent care office, I would want to show all of the major health alerts to the health care provider, and allow that person to print and discuss whatever was necessary with the patient.  Some things, of course, would be mandatory like contagious or life threatening diagnoses.  But other things (that are contextually irrelevant) would need to be excluded from the report.  For the most part, only a person can make that call.

Metrics and algorithms aren’t alright, either

The metrics and algorithms that we build and use are a direct reflection of us and our culture and politics – biases and all.  To learn more, take a look at my former professors’ work on this topic here and a peak at Hannah Sassaman’s hard-hitting research and activism here – two different angles on discriminatory algorithms in the criminal justice system.  And many more examples here.  I think BMI is one more example – linked with discriminatory insurance premiums.  Even though we know now that BMI is bad at BOTH monitoring “healthy” weight AND predicting health problems), it is still popular and widely used in medical and fitness circles.  It’s really just a vestige of our culture’s obsession with thin-ness (and literally punishing fat people) that it is still used as a “vital” medical sign at all.  This is bad data – and it does real damage.

So why am I writing about this on The Data Are Alright?  Good data is not just about getting the “right” answer or automating human processes (step on a scale, print out a ticket).  Do I believe that each and every one of you can be an #automationchampion?  Heck ya I do!  But what makes you a champion is approaching the sweet spot where “efficient number crunching” meets “making meaning out of data” meets “authentic person-to-person dialogue” and ALWAYS bending the arc toward justice.

Sticks and stones (and falling off of a stoop) can break my bones (ouch)

Words (and numbers) absolutely CAN hurt me

But I believe our bodies & our data can be alright

This post is dedicated to all of y’all who are #PleasantlyPlump.  But no matter where you fall on the lousy BMI spectrum, remember BMI is a load of trash, the diet industry is a dumpster fire and you cannot be reduced to a number – and that’s coming from ME! the cape-wearing Queen of Metrics.  xo  Also, consider bringing this to your dr appointments if you want to push them on how they measure, use and discuss weight.  (h/t Rachel W.)

4 thoughts on “Sticks and stones, broken bones, bad data

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s