pretty good. There are special populations that don't do nearly as well—the elderly and the poor in particular. However, both the Medical Outcomes Study and Ed Yelin's study, as well as Hal Holman's very nice studies on utilization and cost for osteoarthritis all show that overall, outcomes are pretty good in prepaid group/staff HMO settings.
However, these special populations—my fourth point—are very, very important. There seem to be two different parts to this and I think both are worth emphasizing. We heard from Karen Davis and again from Mick McGarvey that these populations have access problems, first in getting into and staying in any system, and then, after they are in, they have access problems related to the process of obtaining care within the system. I think those are two different things. Getting in and staying in the health care system is difficult for some individuals, but equally important is getting access to care if, for example, you don't have a telephone or transportation, a network of friends with these things, and so forth.
My fifth point is that, as Brad Gray, Karen Davis, and others pointed out, managed care entities are very variable. They are variable with respect to their content, their organization, their incentives, their training, and their results. I think that we should not even talk about managed care entities in general. Those of us in geriatric medicine like to say that the elderly aren't all the same, and that if you have seen one old person, you have seen one old person. We might want to think about managed care programs and entities in that same way, and try not to paint with such a broad brush. I suspect that new models will evolve as well.
It seemed to me that some of the data that were presented, and many of the concepts discussed, coalesced around four structural issues, my items six through nine, that are relevant to chronic disease and managed care.
The first of these, from what I heard, was practice guidelines. It seems to me that it is important for us to understand that there are at least three different kinds of practice guidelines. There are behavioral practice guidelines, which really were not mentioned but are really important in chronic disease. These are guidelines like "You should exercise," and getting you to understand that you should exercise. Then there are medical practice guidelines: "This is how you should exercise." The third type of practice guidelines are psychosocial, helping people deal with adaptation to chronic illness. These are very specific, and very different, and nowhere more important than in chronic disease.
The second structural aspect of managed care I heard about today that is important to chronic disease has to do with information systems. I think these are very important. That should be the front page, because one of the real advantages of managed care is the ability to follow patients. Providers can track medication use, for example, to see if their patients are using narcotics for pain. That should be routine in a good managed care entity with a good