JS: Generally, you want diseases that are common. Diabetes, congestive heart failure and hypertension are common illnesses. You need diseases in which interventions make a difference within a relatively short period of time. It is possible to control people’s blood pressure, to improve the level of diabetes control, to get the right medications to patients with congestive heart failure and lower their admission rate–all within one year. You want diseases in which interventions make a clear difference and in which usual physician care could be improved by the addition of disease management services. Patients with diabetes, congestive heart failure or hypertension don’t have as good control of their illness when simply relying on a primary care physician and we can add to the resources that primary care physicians by themselves are providing.

JS: We’d like to. As other vendors and other HMOs get into the disease management, we’re seeing that it is possible to do disease management for example, in coronary artery disease, chronic renal failure, and the like. We plan to develop an internal training program and prepare the nurses to offer services in another disease. One of the areas we’re looking at now is coronary artery disease.

JS: We ask the nurses to partner with local primary care physicians right in their offices. The nurse can see the patient for one-on-one education or case management, depending on the disease, the patient’s needs, the physician’s preference and the culture of the town. For example, if you get patients with diabetes to be more effective monitors of what their blood sugars are doing and you show patients how to do it and you give them tools to do it, patients have a big stake in keeping their blood sugar under control and avoiding or managing the complications. It’s one thing for me as a physician to personally arrange an eye appointment for a patient with diabetes once a year, but that will happen a lot easier if the patient has the wherewithal and the resources to say to the doctor, "It’s been a year since I last saw the eye physician. How about if you arrange that for me?" So, you can get the patient to prompt the physician to do the right thing if the patient understands why they need to do something like that. You can instruct patients with diabetes involving their feet to physically inspect their feet every day and avoid complications in terms of amputations. When patients with congestive heart failure are getting into trouble—retaining fluid and getting short of breath—there’s a toolbox that patients can use on themselves to get out of trouble, including taking additional doses of diuretics as letting them know that, when they are getting into trouble that’s a reason for them to contact their physician right away instead of waiting until three o’clock in the morning and going to an emergency room in a semi-crisis mode.

JS: The more doors you have into your disease management program the better. Patients can refer themselves, so we spend some degree of effort in our member newsletters to let patients know that, if they have certain targeted diseases, to please give us a call at our 800-number. We work very closely with physicians. Each one of our nurses is an ambassador for our disease management programs in these primary care sites. Once they establish a good relationship with the docs and they locally prove themselves and develop a level of comfort with the physicians, we find the docs begin to refer patients to us. Physician office staff—nurses and office managers—are very powerful sources of patients for us. Through our claims system we can identify certain ICD9 codes and other patterns of health care utilization that signal that a patient has a particular illness and that we can help that patient. Also certain types of emergency room visits—when patients are admitted to a hospital with a disease as part of the discharge planning, we want to be there when the patient gets out of the hospital to keep him out and avoid unnecessary future hospitalizations.

JS: No. Every primary care site participating in the Geisinger Health Plan has a nurse assigned to it. Each individual nurse devotes a part of her time during the course of the week to each of these sites. One typical arrangement for a lot of our nurses is to devote one half-day a week at a particular site, and it really doesn’t make any difference if it’s a medical group owned by the Geisinger Health System or if it’s an independent primary care site that isn’t owned by the Geisinger Health System. There’s no difference in protocol. We try to make it as transparent as possible. Our interest is the Geisinger Health Plan membership independent of where the patient is specifically located. It is not at all unusual or uncommon for patients to move between primary care sites due to a number of factors. Also, a majority of the Geisinger Health Plan business right now is outside of the Geisinger Medical Groups. More of our membership is in panel sites.

JS: We touch 16,000 Geisinger Health Plan members among all seven of the programs.

JS: It depends on the disease and its severity, as well as the patient’s and the physician’s preference. For example, if a patient with diabetes is very knowledgeable about his disease, is early in the course of it and knows what to do, it may be possible for a nurse in just one or two visits to set things up and enable the patient to be a self-sufficient participant in their own disease for a year. For other patients, if it takes once a month, that’s what the nurses do. We have some patients who are very ill with congestive heart failure and, despite our best work, still have to go to the emergency room. Patients like that may be seen by the nurse once every two weeks. In some instances we have nurses calling patients up to twice a week, if that’s what it takes, to make sure they’re doing okay and helping them through the process.

JS: Everything that the nurses do was proven to work by health services researchers in academic medical centers that were reported in peer-reviewed medical literature. So, if the nurses are promoting the use of a certain medication for a certain illness—for example, Ace inhibitors have been found to be a very important drug that battles congestive heart failure—our nurses will educate the patient about the availability of that particular medication, will go to the doctor and ask if the patient is on the medicine, or if they have they have thought about it. The nurse protocols facilitate the clinical guidelines that we also developed internally with input from physicians who are in the network of the Geisinger Health Plan. As we collect data and, if for one reason or another we are not happy with what the data are looking like, we’ll tweak the guideline or the protocol and try to improve it.

JS: That question is best considered in the context of team-based collaborative care. If you’ve got a patient, a nurse and a physician all working together, that delineation between what part of the pie is owned by each person can be very fluid and flexible. Ultimately, the doc is the quarterback and is responsible for the content of care for each individual patient, and our nurses recognize and respect that boundary. Some physicians have a practice style that is very tight in terms of what they’re comfortable with, as far as the patient and nurse go. Other physicians are somewhat more flexible and open and accommodating to nurse suggestions.

JS: We regularly survey physicians and ask them how we’re doing and what they think we can do to improve things. What we’ve generally found is that the numbers go 40-40-20. Forty percent of physicians right away embrace and welcome the idea of an extra pair of hands helping them take care of their patients, especially when it’s one of our nurses who can come right to their practice setting and work side-by-side with them. Another 40 percent are physicians who are more reserved and take a "show-me" attitude. They will allow the nurse to come in and they need several weeks or months to develop a comfort level and to get used to the idea that this nurse is in their office doing this kind of work. The nurses, I think, prove themselves under those circumstances. And then there’s about 10 or 20 percent of physicians who, no matter what we do and no matter how accommodating we are, decide that this is just something that’s not for them. Even among those physicians, however, there’s still an aura that our program casts off to their patients. They still receive our newsletters. They still might be telephoned by our nurses, not to interfere with the content of clinical practice of the doc, but in terms of education about the nature of the disease and things that they should be aware of. We don’t have any data to prove this, but we think it’s possible or even likely that the patients of those kinds of physicians are still walking into physician offices and bringing things up like, "Hey doctor, I hear that this particular medicine is good for my disease," or "I understand that I should get an eye exam done," or "What is it about my feet that I should be aware of?" We take a long-term view on this. We’re optimistic that the holdouts, if you will, will come around. Even with a minority of physicians who choose not to participate in our programs, our data is still showing very solid improvements in the outcomes of care for our disease management program participants.

JS: Not within the disease management programs themselves, not within our protocols. Our health plan, however, does have incentives for physicians for quality of care based on patient satisfaction data as well as HEDIS measures. We argue to physicians that having patients participate in the disease management programs will drive HEDIS and, as HEDIS score improve, there’s an opportunity there for the physician. But, within our disease management program alone we don’t have any incentives. For physicians who choose not to follow a guideline, ultimately what we do is we say, "You know doctor, we think this is good. We think this is something you should pay attention to. If there is anything else we can do to help you, please let us know." And then we give him or her a wide berth.

JS: We think so. When we compare follow-up data to baseline data we’re showing consistent improvement in all measures. The Geisinger Health Plan claims database enables us to see what the cost of health care is down to the individual ICD9 code for each patient, down to an individual claim level. So, we can see what the claims patterns look like for patients prior to entering the disease management program and after entering the program. The claims data consistently show increases in the use of primary care services. We think that’s because we teach patients to more effectively use primary care. We’re seeing less in the way of inpatient charges for patients in our disease management programs. And we’re seeing increases in the quality data—the HbA1c of patients with diabetes, use of steroid inhalers among patients with asthma, level of blood pressure control among patients with hypertension. As all our patients’ claims mature we see what the claims patterns are looking like and we come up with an average per-member-per-month cost for patients after the date that they enter the disease management program and we compare that to the baseline. In our diabetes program we managed to find a control group and we found that the per-member-per-month charges among patients with diabetes who weren’t in our disease management program were substantially higher than patients who were, everything else being equal.