A saline inkblot

Anecdotes are funny things: they often come to us with no context, so we slip them easily into preconstructed narratives and learn nothing from them.  For example, this one lends itself to a pair of opposed conclusions:

[Nutrition expert Marion] Nestle adds, “Campbell’s soup, for example, just announced yesterday that [they] can’t sell low-sodium soups and so they’re adding salt back.”

One preëxisting narrative will produce a reaction like this one:

Oh my god, we — as a society — are so addicted to salt that even after decades of public service announcements about its terrible dangers people still crave the evil stuff!  There Oughta Be A Law!

While another will instead lead to something like this:

No shit, Sherlock; people tend to autoregulate the amount of salt in their diets, and if eating Campbell’s no longer does it for them they’ll switch over to something else.

If you believe that high salt intake leads to cardiovascular disease and early mortality, the anecdote that low-salt soup doesn’t sell is evidence of market failure, epidemic noncompliance, and other symptoms of technocratic priapism.  If you don’t accept the salt-hypertension-CVD-stroke hypothesis, though, the anecdote simply corroborates the notion that people tend to eat roughly as much salt as they need, and those who used to get it from Campbell’s soups are getting it somewhere else instead.

Similarly, when the same authority says in an interview with Scientific American:

[I]t’s impossible to put a population of people on a low-salt diet. Roughly 80 percent of the salt in the American food supply is in foods before people eat them—either in processed food or in restaurant food. Because so much salt is added to the food supply and because so many people eat out, it’s impossible to find a population of people who are eating a low-salt diet. They basically don’t exist. In the one comparative epidemiological study they did some years ago—the Intersalt study—they managed to find two populations of people in remote areas of the jungle someplace who weren’t eating a lot of processed foods and who weren’t eating in restaurants.

…your reaction might be “Oh, god, we are so totally screwed” and a wail of despair so heartwrenching that passers-by will think you’re reading a Greek credit forecast.  Or, on the other side of the coin, you might roll your eyes so hard you sprain something and growl “No shit, cupcake; maybe there’s a reason that the only people who don’t eat high-salt diets are the ones too geographically isolated from the opportunity to do so!”  It rather depends on whether you see your fellow H. sap. sap. as a bunch of habitual fuckups — well, except (of course) for you, and the people who write the New Yorker articles you like — who need constant hand-holding to prevent them from jumping feet-first into the nearest convenient wood-chipper because it makes a funny noise.

Even with a bit of data, we don’t necessarily escape the narrative’s stranglehold.  Consider this graph:

If you adhere to the hypothesis that hypertension increases stroke risk — the root of the “salt is bad for you” crusade — you might be a bit surprised that hypertension incidence increased steadily over the 12 years represented, while the other forms of CVD (strokes included) stayed pretty much constant.  On the other hand, you might object to the relatively short run of the study, expecting a corresponding increase in stroke rates when the obvious epidemic of hypertension starts killing people, and point to the rising hypertension rates as clear evidence that people just aren’t listening to the experts.

Then, when you see this chart…

…you might first point to the steadily-decreasing death rates as evidence that the anti-salt movement — in major effect since the ’70s — is doing good and saving lives.  You might then wonder why this interpretation clashes with your pervasive-noncompliance narrative from the past two anecdotes, and why the death rate from hypertension seems to have remained unchanged while general CVD death rates have dropped dramatically.

And then you might find yourself here.  Or here.


Leave a reply; use raw HTML for markup. Please blockquote quotations from the post or other comments.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

anarchocapitalist agitprop

Be advised

I say fuck a lot



Statistics FTW

%d bloggers like this: