Epidemiology and Elephants

Epidemiology and Elephants

Preventive Medicine Column

Dr. David L. Katz

Apparently, around about the time we lost the ability to see the forest for the trees (which we are incessantly cutting down, by the way), we also confronted the threat- according to poet John Godfrey Saxe- of missing the elephant for its parts.  In Saxe’ poem, The Blind Men and the Elephant, six blind and learned men set off to “satisfy” their minds about the nature of elephants by reaching out to touch one.

As fate would have it, each takes hold of a different part- from trunk to tail, leg to ear- and so each reaches a totally different conclusion about what the whole elephant must be like.  And of course, all are wrong.

Perhaps the wise, blind men and their elephant offer up a precautionary tale as we consider the modern challenges of preventive medicine.  On July 28th, the New York Times cited a dramatic trade-off in research funding between tobacco control, and obesity control.  As spending by large foundations and the federal government on obesity has risen, spending on tobacco control has plunged.  Not parts of an elephant, in our case, but parts of modern epidemiology.  Each of us who grasps a part perceives that, it seems, as the thing that matters most.

As an obesity researcher, I might simply applaud the shift in funding to my area of expertise and work.  But my work does not take place in a vacuum, and the formula goes in the other direction entirely: the reason I AM an obesity researcher is because obesity was clearly a burgeoning public health threat just as I did my training in Preventive Medicine in the late 1980s, and early 1990s.  Gertrude Stein wisely advised us that “a difference, to be a difference, must make a difference.”  Wanting to make a difference, I directed whatever talents and energy I have to the rising tide of epidemiologic trouble.  That meant a preferential focus on obesity.

But that does not make one part of epidemiology more important than another, any more than one part of an elephant is more important than another.  The true goal not only of Preventive Medicine, but the full array of disciplines housed within biomedicine, is to advance the human condition.  It thus won’t do, for instance, to cut off the trunk as a tourniquet for a bleeding leg. Healthy elephant parts require a healthy elephant.

But since research dollars are finite, how should we spend them?  Is it right or wrong to spend more on obesity, less on tobacco?

In my view- which seems to differ from the author of the article in the Times- that’s not the right question.   The right question is: given the perils we face and the limits to our resources, what particular use of those resources produces the greatest net advance in the human condition?  In other words, how do we best promote health overall?

That tobacco and obesity should be contenders for marquis status is clear enough.  We have known since 1993 at least, with the publication in JAMA of a seminal paper entitled ‘Actual Causes of Death in the United States,’ that premature death and chronic disease are overwhelmingly the result of a short list of behaviors we control, and that list, in turn, is dominated by just three: tobacco use, dietary pattern, and physical activity.

Every examination of the data since 1993, including a follow-up review paper in JAMA by CDC scientists in 2004, has reaffirmed the finding.  Along with establishing feet, forks, and fingers as the indisputable master levers of medical destiny, this research has also suggested that as the toll of tobacco use has been declining over recent years, the toll of eating badly and lack of activity – represented in obesity among other adverse outcomes- has been rising.  Concentrating the levee building where the flood waters are waxing certainly makes sense.

But not if it means the levees we’ve already built are allowed to fall down.  Flooded is flooded, from whatever direction.

While it does not yet seem that the allocation of funds to support obesity has undone any of the progress against tobacco, it is clear that the rate of tobacco use in the U.S.- about 20% of the population- has stopped falling.  Perhaps we could get below this threshold with more dedicated attention.  Perhaps the lack of on-going progress is a result of neglect, and a portent of worse to come.

I am by no means prepared to say we have shifted too much attention from tobacco to obesity; honestly, I don’t know.  And neither does anyone else.  It is in response to the hazard of this ignorance that I have two suggestions to offer.

First, we can and should conduct modeling exercises to determine what general allocation of research and policy dollars- across an array of conditions, behaviors, and even types of research- would most improve our health over a defined period of time.  While individual research grants get intense scrutiny, there is altogether too little scrutiny from altitude to determine the best overall pattern of research grants.  Colleagues and I have designed and proposed just such a modeling exercise, although it has not been funded to date.

Second, if you can forgive the New-Age feel of the term, we need to get a bit more…holistic.  A healthy person is healthy- in whole, not in part.  Recent studies have shown that people who don’t smoke, eat well, are active, and control their weight are roughly 80% less likely to get ANY major chronic disease than their counterparts who do the converse in each case.

So, a healthy person doesn’t smoke.  A healthy person eats well.  A healthy person is physically active.  The question we should be asking is: what interventions for individuals, families, schools, worksites, communities, and more will encourage, promote, and empower the adoption and maintenance not of some single preventive strategy, but of healthful living?

It will not do- for epidemiology any more than elephants -to define the soundness of the whole by any one part.

 

-fin

 

Dr. David L. Katz; www.davidkatzmd.com

By | 2016-10-18T13:55:41+00:00 July 30th, 2010|Categories: Blog, DNSFP, Dr. Katz Blog|0 Comments