Monday, August 3, 2009

Obesity and Health Care Costs

One can perform an interesting study to see what effect certain diseases have on health care costs. Cancer incidence has not changed dramatically, other than being able to detect it sooner, such as in prostate cancer, and other non-obesity related diseases have also not changed greatly in incidence.

The word from the pundits is that we are spending more on health care, delivering too much to people. They never ask if the "people" have increased the demand as a result of some controllable exogenous factor, separate and apart from how health care works.

Well lets do something new, let us look at the data. First a chart from the CDC on obesity and overweight ("O&O") prevalence in the US. We show it below.



















This shows a growth in the O&O factors tremendously over the past fifty years. Perhaps there is a correlation with costs of health care and perhaps, just perhaps the O&O is causative. And then perhaps we can see how large it will become if O&O reaches 100%! And even then we can ask what can we do about it? After all we could do the same thing if we calculated what would happen is we had every driver drive drunk! And at 90 mph on top of that!

Now let us plot the O&O and the Health Care Costs as a percent of GDP (HC/GDP) for the periods.



















A pattern begins to form. Now we can plot the ratio of HC/GDP to O&O, to see if that ratio has remained somewhat constant, implying a possible causal relationship. It appears below:



















The result is interesting, it is somewhat constant. Thus we then plot a scatter plot of these two variable as below:



















And behold, we see that the growth in HC/GDP correlates very well with O&O prevalence! Perhaps the main driver, as we have argued in more details elsewhere, is the explosion in O&O? If that is what it seems to be, then the current Administration's approach to control HC/GDP growth will never work other than by rationing and letting old people just die! Perhaps the problem is less the old folks but more the fat young folks who will cause the major chronic health care loads in the future. If we let O&O go to 100%, the HC/GDP ratio goes to 30%! Close to the number we have all seen in the scare tactics of the current Administration. Yet the problem is NOT the health care system it is the people who allow themselves to explode on unhealthy diets and become morbidly obese! They are the ones who are costing the system. They are the ones who can be controlled!

Thus what can be done. Despite my dislike of the Mankiw idea of the Pigou Club of taxing to reduce demand, here I believe it works and works well for beneficial purposes. If you cannot tax excess BMI, a fee say $1,000 per year per 1.0 in excess of 25.0, then tax the foods which get there, processed and fast foods!

Also having posted this I read Posner's blog and suggest it is worth the read on this same topic. He does not deal with data and facts but he is persuasive none the less.