Correlations vs Causation

You may have read that getting less sleep is associated with heart disease.  This one included a well-written description of how correlations are not causation. 

Senior author Diane S. Lauderdale cautioned that the new report does not prove a cause-and-effect relationship between a lack of sleep and heart disease.

“It’s important to say that this is the first report and this does not yet prove the association is causal,” said Dr. Lauderdale, an associate professor of health studies at the University of Chicago Medical Center. “Until we know what the mechanism is — that it’s really a direct or a causal relationship — there is no point in making recommendations based on this.”

Although a number of studies have suggested that people who sleep less are at greater risk of heart disease and death, this is the first investigation to measure how much its subjects actually are sleeping, said Dr. Sanjay Patel, assistant professor of medicine at Case Western Reserve University and expert in sleep medicine. Patients’ own self-assessments can be very inaccurate, he added.

What isn’t clear is whether reduced sleep triggers physiological changes that increase heart disease risk, or whether a third, unrelated factor causes both changes, he said.

“It’s possible, for example, that people who are under more stress may be both sleeping less and at higher risk of heart disease,” Dr. Patel said.

If so, he added, “If we got those people to sleep more but they still were under a lot of stress, it wouldn’t change their risk of heart disease.” 

Pity we don’t see more of this sort of clarifying about, oh, being fat.  When you look up, oh, “obesity” on the CDC website,  it’s “Obesity increases the risk of many diseases and health conditions.”   When you do see it, it’s “[Is] Obesity a Disease or Just a Symptom”?   Suggesting that someone with a BMI of 27 – considered “normal weight” in the early 90s – could be healthy without losing weight is heresy. 

Or maybe I’m just feeling gruntled today :)   Am I overreacting?

2 thoughts on “Correlations vs Causation

  1. Nice to see a study author saying so!

    I can think of a lot of reasons why poor sleep could be correlated with heart disease. Hypothyroidism often causes poor quality sleep and can mimic or cause heart disease. Sleep apnea often goes undiagnosed or untreated and can lead to heart damage. Stress, like they said. And dozens of other diseases and illnesses, both physiological and psychological.

  2. People always jump to conclusions so quickly when it comes to so-called “risk factors” … I’m still trying to come up with an easy way to explain the difference between a risk factor and an actual risk. How come when a study shows a correlation between fatness and diabetes, nobody exclaims, “Oh, you mean diabetes makes people fat?!” It’s always the opposite.

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 )

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