I am going to take a break from the Mandate Wars... for now.
Today's New York Times and Washington Post report on some unexpected results in an important study of diabetes treatment.
Like the recent Avandia and Vytorin scandals, this highlights some of the problems in our health care system, and some confusion that doctors and the public have about what is meant by an outcome.
The initial studies on Vytorin and Advandia that suggested they were beneficial were based on PROXY or INTERMEDIATE outcomes or just MARKERS for disease.
This means they were looking at things like blood sugar and HbA1c (indicative your average blood sugar over about 3 month period) or serum cholesterol levels. But of course what we are really interested in is actual disease. And in fact the real outcome of interest is death. That is death from ALL causes (not just the one the drug supposedly treats).
When the first drug treatments for lipid lowering or sugar lowering drugs were studies, the association between lowering blood sugar for diabetes, or lowering bad LDL cholesterol for cardiovascular disease, and the linkage of these outcomes to reduced death from diabetes or cardiovascular diseases was made.
We know having normal glucose or normal cholesterol is an indicator of health associated with, and somewhere in the causal pathway, of not having diabetes or heart disease, and prevention the real outcomes of these diseases and ultimately less mortality. But in the actual human body there are many complex physiologic pathways that are involved in the normal process to achieve this.
For instance, Statins have positive effects not only by lowering bad cholesterol, but also probably have other positive effects in other causal pathways for reducing cardiovascular disease, such as anti-inflamotory.
In addition, a drug may lower your lipids or blood sugar, but raise your risk for something else, leaving you no better, or even worse off, overall.
Diseases typically have their effects through many different physiologic pathways.
Most drugs have effect on many different physiologic pathways.
Figuring out the real longterm benefits is complicated, and too often doctors and patients are fooled by looking only at one proxy or marker outcome or just an intermediate outcome.
Now, most FDA approved drugs do work. Getting your bad cholesterol down with a statin (but not a combination drug like Vytorin) and moderately tight glycemic control is probably remains the goal for persons with elevated LDL cholesterol or diabetes.
However, using a one or pharmaceutical agents via indirect, after-the-fact and non-normal pathways to achieve this misses the point.
A normal environment, healthy living, good diet and physical activity really are the ways to prevent and even to treat this.
There is still a real advantage of taking diet and exercise seriously. Not only after the fact to treat obesity, impaired glucose tolerance and cardiovascular risk factors, but especially as lifelong and society-wide prevention.
Alas, our health care system and society-at-large, reward drug development and drug sales, and spends way too little money and does not take seriously non-drug therapy, we are all losers.
It would be nice if HHS, FDA, NIH and the other agencies were funded to:
- do long-term post-marketing follow-up on all drugs.
- discourage too many me-too look-alike drugs.
- do head to head comparison of me-too look-alike drugs.
- Always use real outcomes, as well as proxy and intermediate outcomes.
- do more, bigger, better studies of lifestyle as prevention and treatment.