Wednesday, April 13, 2011

Health Myth #2: Correlation Equals Causation

Even though the axiom "correlation does not equal causation" is frequently parroted within the sciences, it is much easier to say a principle than it is to actually integrate it into one's structures of thought. Case in point, a lot of scientific thought is premised on the contrary notion that correlation equals causation. And nowhere is this more apparent than in the field of medicine.

Let's start with examples from Myth #1. The fields of medicine and public health are very concerned with how much people weigh, going so far as to decry an "obesity epidemic." What, exactly, is the problem with being fat? We are told that "excess" fat is associated with heart disease, high blood pressure, diabetes, and just about every other threatening condition you can think of. In a slippery way, "associated with" becomes "caused by," in that weight loss is recommended to prevent all of the latter conditions; but losing weight can only prevent these conditions if it is a cause.

Doctors are hard-pressed to address the inquiries of more critical patients who ask, "If the purpose of losing weight is to prevent high blood pressure and I have very low blood pressure, why are you telling me to lose a few pounds? If I eat right and exercise daily, should that not be enough?"

At the same time, other health professionals have noted that fat storage appears to be an inflammatory response that is co-present with other such effects of inflammation. According to this view, it is very probable that the association between obesity and things like heart disease and high blood pressure is due to another underlying cause. Furthermore, when it comes to inflammation, the probable causes start to become more nebulous, being attributable to things such as "stress." (But I will save this discussion for Myth #3.)

As with fat storage, the same goes for cholesterol. Both fat and cholesterol are vital substances that perform necessary bodily functions. However, in each case a correlation has been discovered between high amounts of these substances and undesirable health conditions, and in each case this correlation has been construed as causation. Now, both fat and cholesterol are often seen as "unhealthy." Yet, just as high fat storage may be viewed as an inflammatory response to some other (internal or external) condition, it has also been noted that the ill effects of certain forms of cholesterol only occur in the presence of.... you guessed it, inflammatory responses in the cardiovascular system that only implicate cholesterol in a tangential way.

These same cardiovascular inflammatory conditions may also have an indirect, non-causual relationship with salt (just like cholesterol). Although people now tend to view salt as "bad" and avoid it like the plague, the only direct effect of salt is that it increases water retention (which, in itself, is a neutral effect, and can, under certain circumstances be beneficial). However, in the presense of unfavorable conditions created by inflammation, the increased fluid volume can coincidentally contribute to high blood pressure.  Hence, decreasing salt intake does nothing to address the inflammatory causes of high blood pressure.

Of course, it is possible that "inflammation" may eventually replace "bad diet" in the simple causal chain employed by public health professionals (a move which would by no means lend to a more complex, holistic understanding of health); but, for this to happen, the idea of "inflammation" must be condensed into a simple, material entity that can be integrated into and reconstitute the web of market relations and commodity chains that comprise the practice of public health.

No comments:

Post a Comment