UNFORTUNATELY FOR those who hoped to get the government to pay for their Weight Watchers clinics and Atkins diet books, the recent decision enabling Medicare to treat obesity as a disease does not -- yet -- have such broad (or astronomically expensive) implications. The change was in fact a narrow one: It will allow Medicare to start gathering evidence to determine whether particular treatments, including diet programs, behavior therapy and surgery, are effective in weight control. In that sense, it is a useful regulatory change, because this is precisely the sort of evidence gathering that Medicare, with its financial clout, ought to be doing, and that the diet industry, long a haven for quacks and cranks, desperately needs.

At the same time, there is something absurd about a small group of government officials trying to decide whether obesity is a disease. That absurdity goes to the heart of what is peculiar about Medicare, and indeed medicine as it is practiced in this country. Clearly, obesity is a health problem: It can lead to diabetes, high blood pressure, heart disease and many other health problems. Recent surveys suggest that obesity is more directly associated with illness than alcohol, cigarettes or poverty.

Yet obesity is also part of a problem that is far easier to deal with in much earlier stages, before it becomes debilitating. Preventive medicine of the kind that would catch weight problems early on isn't necessarily built into most health care plans, including Medicare, which are designed to deal with the consequences rather than the causes of diseases. Some health plans are beginning to experiment with various forms of preventive medicine. While Medicare officials are looking at the success rates of some weight-loss programs, they should consider these too.