Skip Navigation Archive: U.S. Department of Health and Human Services
Archive: Agency for Healthcare Research Quality
Archive print banner

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Stretching Scarce Resources: State Strategies to Design Effective, Affordable Benefit Packages (continued)

Cindy DiBiasi: So what do you see as the implications of what this research and [unclear] experiences are telling us about cost sharing?

Ann Markus: Well, I think in this area and many other areas that involve policy choices, for States it becomes really a balancing act between two things. First using cost sharing for any of the reasons that I touched upon earlier that might be important for the States for political or economic reasons, but also important to avoid the creation of the situation where the people targeted by their programs are subject to undue financial stress.

Given what the literature shows and the fact that it shows a disproportionate effect of cost sharing on low-income individuals, States that wish to impose premiums and cost sharing may want to consider an income-related sliding scale for their premium cost sharing schedule. They might want to use premiums that are fairly low, however that is defined. They may wish to use cost sharing that could include a deductible for inpatient care, maybe with exception for young children. Co-payments for certain types of services [unclear] brand versus generic drugs and for certain types of care for example, the emergency room versus other physician offices. They may want to avoid the use of co-insurance altogether and may want to use an annual cap as an additional protection. Finally, they may also want to broaden the definition of preventative services exempt from cost sharing.

Cindy DiBiasi: Ann, we are going to come back to you and talk more about this when our callers start calling in. So far we have spoken about approaches to stretch scarce resources by examining what is in a benefit package or through cost sharing provisions.

Another approach to providing cost-effective care for persons with chronic conditions is through the implementation of strategies such as disease management programs. These have been getting a lot of attention and have been implemented in a number of private and public sector programs including several State Medicaid programs.

Autumn Dawn Galbreath, who is the director of the University of Texas Disease Management Center, is with us now. Autumn Dawn, what are the principles upon which disease management programs are based?

Autumn Dawn Galbreath: Great question to start with because it is an often-misunderstood term. Cindy, I would like to give the down and dirty definition of disease management as managing a disease. That sounds too obvious to actually use as a definition, but it is the ongoing management of one particular disease in a given patient.

So it is a systematic approach but it also focuses not just on an individual patient but also on a population of patients who have that disease. It identifies the people who have the disease. It then intervenes to care for the disease in an ongoing fashion and then it measures outcomes. That is an integral part or one of the basic principles of disease management is that it is constantly looking at its own effectiveness.

It employs best practices of medical care so that it is always using published clinical guidelines or consensus of expert opinion in a region or in a locality so that it can reduce costs and improve quality is essentially the summary of the goals that it is trying to accomplish.

Cindy DiBiasi: I know the focus of your work is on disease management programs, but before we go too far I just want to get some definitions clarified for the audience. The term "disease management" is so often thrown around. Can you tell us how this is similar to or different from other interventions such as high-cost case management models?

Autumn Dawn Galbreath: Yes. I like to call this "management soup." There are so many terms that all have the word "management" in them and are frequently used and misused or substituted inappropriately for each other. Case management is sort of a generic term for giving attention to an individual patient for whatever a reason a patient who needs a little attention from the health care system. This is typically an allied health professional, a nurse or someone like that working individually with a patient in a long-term situation. But different programs have very different criteria for how they choose which patients are managed that way and for what that case management actually looks like.

High-cost case management is sort of a subset of case management. It is the management of individual patients who are high utilizers of the system. That is the criterion for choosing which patients get involved in a high cost case management program. They may have multiple, and usually do, have multiple diseases and multiple reasons for being high utilizers so they have a manager or case manager who works with them to try to put all those pieces together and manage the whole picture of their health care.

Disease management on the other hand is the management of a population with a given disease. So a person might have congestive heart failure and diabetes and COPD [chronic obstructive pulmonary disease], emphysema, and several other diagnoses, but their disease management program would focus only on their congestive heart failure or only on their diabetes.

Population management and managed care are other broader terms that employ some of the different principles of these other ideas. But disease management is distinct in that it focuses on a specific disease or a specific diagnosis within the patient population.

Cindy DiBiasi: What do we know about how well these models work?

Autumn Dawn Galbreath: Well, I have a slide that has a quote that says, "The literature is now replete with titles in disease management, quality improvement, and clinical practice improvement, but there is not much evidence that these improve efficiency in the care process for the diseased." That pretty well summarizes the literature. I might be jumping ahead of you a little bit, but the literature is full of trials in disease management and many of them are maybe not very methodologically sound from my academic standpoint. There are some, which show very good results that disease management works well and there are some that show that disease management doesn't work well. There are many that fall somewhere in between. So actually we end up, if you do a thorough review of the literature, we end up sort of at the same place where we start which is that we have a lot of questions about the effectiveness of disease management. We don't know for sure if it is cost effective. We don't know for sure if it improves clinical outcomes. We don't know for sure if it improves the quality of life or patient or provider satisfaction. We don't know for sure it is the best way to administer a disease management program.

That is exactly what the focus of my research is. To give the little plug for what we are doing in south Texas, we actually are doing a large-scale, randomized clinical trial that happens to deal with congestive heart failure but we are trying to very methodically answer some of those questions that are still out there and the literature so that States and academic health centers and sort of public health in general will have a little bit more rigorous information from a scientific standpoint when they make decisions based on implementation of disease management programs.

Cindy DiBiasi: Given all those caveats and those questions, what do you think makes a disease optimal for a disease management program?

Autumn Dawn Galbreath: That is a great question. If you are managing a population of people who have a certain diagnosis, then you need to have a population. You can't just have one or two patients or four or five patients with that diagnosis. You need to have a high volume of patients who have that disease and it needs to be a disease that has a significant impact on the economics of your health care system. Otherwise it is not worth the upfront investment of developing or implementing a disease management program. So high volume and high cost are the two fundamental characteristics that a disease would need to have.

The third thing, though, is that it needs to be a disease which is routinely managed in different ways by different providers so that if I gave five doctors each a patient with congestive heart failure, they would all do something a little bit different. The idea is that if I can convince them of the best way or if we can figure out the best way to do things and have them all do it the best way, we would reduce costs and improve outcomes.

Cindy DiBiasi: So given your experience implementing this project, what are some of the lessons you have learned and the implications that exist for designing this type of service delivery program among indigent populations?

Autumn Dawn Galbreath: We have certainly run into a number of barriers. We could spend the whole rest of this call giving people advice about barriers to implementing a disease management program. The biggest barrier, sadly enough, is our health care system itself. The way that our health care system in the United States is structured, we work on an acute model of care. All of us who deal with uninsured patients know that those patients are well aware that if they really want to be seen, they show up at the emergency department and somebody will eventually see them. So the entire system is built on this acute care model where we give the most acute patients the most attention and in between acute episodes we sort of let them slip through the cracks. That system is what works against us so much in managing chronic diseases.

Another huge barrier is information systems. Every doctor's office, every hospital, every public health clinic has its own little information system which is not compatible with those of anybody else so it is very difficult to work together as a multi-disciplinary team within a health care setting because our information systems don't work together well.

A third barrier—and we probably have some physicians listening to this call and I am sure that I will have comments from them when the time comes—is, frankly, lack of physician support. They perceive these types of interventions as taking away some of their autonomy. Actually, a well-implemented disease management system shouldn't do that, but many disease management systems are not well implemented and do interfere with the provider's autonomy.

Resources, that is always a barrier to anything we want to do in a public health setting. The last thing that I would mention as a barrier is the danger of fragmentation of care. This is a concern to many physicians out there that if you have one person managing or working with a physician to manage a patient, diabetes, and another person maybe helping the patient with their emphysema, that you can sort of end up carving the patient up so that you have got the little toe specialist and the right ear specialist and ne'er between shall communicate. There is definitely the danger of running into fragmentation.

There are some lessons that have been learned by ourselves and by States that have implemented disease management programs. I don't know if we might want to discuss those when people start calling in?

Cindy DiBiasi: Yes, let's do that because we are going to open up the phones for discussion soon and we are going to get back to you, Autumn Dawn. In a moment we are going to open up the discussion for your questions.

Before going to questions, however, I would like to say a few words about our sponsors. The mission of AHRQ is to develop and disseminate research-based information that will help clinicians and other health care stakeholders make decisions to improve health care quality and promote efficiency in the way that health care is delivered. AHRQ's User Liaison Program serves as a bridge between researchers and State and local policymakers. We not only take research information to policymakers so they are better informed, we take the policymaker's questions back to researchers so they are aware of the priorities. Hundreds of State and local officials participate in ULP workshops each year.

The Council of State Governments, CSG, is a multi-branch State association that serves elected and appointed officials in the executive, judicial and legislative branches of State government. CSG's mission is to champion excellence in State government through advocating multi-State problem solving, tracking trends and innovations and providing education, research and leadership training services.

Both sponsors hope today's Web-assisted audio conference and the other two events in this series will provide a forum for a productive discussion between our audience of policymakers and researchers. We'd appreciate any feedback you have on these audio conferences. Please E-E-mail your comments to the AHRQ User Liaison Program at

Now to your questions. We have had a lot coming in. John Santa, you get the first one. What kinds of services are explicitly excluded under the Oregon Health Plan and they would like you to give examples of what the 3 percent of excluded services are.

John Santa: Well, on our prioritized list, we have excluded a number of futile services, for example. Treatment of infants with anencephaly, treatment is futile in those circumstances. Treatment that is ineffective, treatment of viral infections, for example. The ones that really we work with a lot is we exclude for example, treatment for temporal mandibular joint syndrome. We are not paying for treatment related to sexual dysfunction so we don't cover Viagra, for example. We don't cover some of the medications, some of the treatments for hay fever, for allergic disorders, the ineffective or the conditions that really aren't proven to be effective or the burden of disease is not high.

Cindy DiBiasi: A question from Christy Goodwin: How much financial relief does cost sharing provide States?

Ann Markus: That is a good question. I am not sure how much information there is on that particular issue. States obviously expect to raise a certain amount of revenues from the imposition of premiums and cost sharing. But as I said earlier, there is some anecdotal evidence that the effort might not be worth the cost. For example, in New York I remember we talked to an official there who said that the estimated $3 million per year that they were collecting from cost sharing was well below what the State's administrative cost related to collection of premiums and cost sharing was. So they decided to drop that completely. As far as the exact impact or the exact relief that States can expect to get from cost sharing, I don't believe there is a whole lot of information on that.

Cindy DiBiasi: We have a few callers on the line. First from the Pennsylvania Life Insurance Company. Jeff from Pennsylvania, hello?

Jeff: Hello. I am with the Pennsylvania Insurance Department. The reason I was calling has to do with the concept of limited benefits packages. Pennsylvania is putting together a program for uninsured adults. We are using tobacco settlement money for it. It actually is going into effect in a few months. The issue that came up when we were dealing with it, one of the issues had to do with benefits. Specifically things such has potentially higher cost benefits, mental health and substance abuse, prescription drugs. We decided to go, we got those benefits to have more money to cover more people. But in listening to Mr. Santa's comments, I was wondering, feedback, experience from the panelists or their perspectives on tradeoffs. When you are making those tradeoffs between a program or when you are trying to cover as many people as you can, you have a limited budget versus a situation where you would be putting more money out there, not more money but more benefits and fewer people. We are not drawing down any Federal money with this, of course since it is tobacco dollars.

John Santos: Maybe I can just give you a couple of examples. Our focus is all on our adult population here in Oregon that are eligible by income. We have tried to put aside the disabled and Medicare and children. No change in those populations. So we are specifically focusing on 18-65-year-old adults. We have 100,000 people. We provide a very rich benefit plan with virtually no cost sharing at this point in time and premiums that are from $6-23 depending on the sliding scale. It is a very rich benefit plan. Our proposal is to reduce that benefit plan by 20 percent, but with that and with some other flexibility and leverage issues and with some employer sponsored-contribution, we could cover 50,000 more people. So the question is, would we be better off with 150,000 people that may only have an 80 percent benefit package or 100,000 people with a 99 percent benefit package?

Now we have had discussions similar in some degree to the approach that Utah has just launched. Should we really go with a very basic primary care package? A package in Utah's case for example, is a physician, dental, some prescription drug, no hospital, and no mental health. A very basic package, but you can give it to more people. I think those are the questions we should be asking and as tough as it is, I think those are the things that some States should be trying as admittedly, these are demonstrations. These are trials to see how can we make our resources as effective as possible.

Cindy DiBiasi: Thank you, John. Our next caller is from the Ottawa County Health Department in Michigan. Brenda is on the line. Hello?

Brenda: Hello, this is Brenda. My question is for the Oregon plan. Are birth control pills also covered in that?

John Santa: Contraception is actually a high priority. It is covered. Contraceptives, basically all services related to pregnancy and maternal care are covered in the list have a high priority. We have also, by the way, in our latest effort, pregnant women would continue to get the current 100 percent benefit. It would not be part of this tradeoff.

Cindy DiBiasi: A question for Autumn Dawn. More of a clarification the caller wants. Earlier on the call they say you mentioned an awful lot of things that we are not sure about. Are you saying that States shouldn't move ahead with implementing a disease management program?

Autumn Dawn: Not at all. In fact, all I am saying is it is not a magic bullet. If you, and I am sure many of our audience members are familiar with the Florida disease management programs that they have put in place over the last few years. If you know their history, they started out by making an approach to their legislature and telling them that disease management is the most wonderful thing since sliced bread or something like that. Told them a lot of things that they anticipated would happen with the implementation of a disease management program. Those things might have been anecdotally, there might have been anecdotal evidence that those things had happened in other places, but they weren't things that were scientifically, predictably proven. But their legislature's response was to say well, since this is such a great program we are going to take $5 million out of your budget now in anticipation of how much you are going to save us in the next fiscal year. So my cautions are not to deter people from implementing disease management programs. But to advise you not to approach them with unrealistic expectations, not to promise your legislatures or other folks in your State governments, things that disease management can't reliably deliver. It is a great tool and it has great potential but it is not the end-all, be-all.

Cindy DiBiasi: So what are the key questions a policymaker should ask when considering the possibility of implementing a disease management program for their Medicaid, SCHIP, or State employee health benefit populations?

Autumn Dawn Galbreath: Well, let me clarify your question. What are the questions they should ask of themselves as a legislature? The first thing that they need to know absolutely, unequivocally, is what their bottom line is at the outset, where their starting point is. Which diseases really are problematic in their State, which may be extremely different from what diseases are problematic in another State? How much those diseases are costing them, how many people they have in their populations with that disease because these are decisions that have to be made extremely systematically because typically a State is going to hire a disease management company, a vendor company to come in and provide these services and they are going to have to make financial arrangements with that vendor company. There is going to have to be an outlay of money upfront, usually. You have absolutely got to be certain of where you stand before you begin that negotiation process. Does that answer your question?

Cindy DiBiasi: Yes. Ann, a caller's question. Can you explain more about the impact on health outcomes of requiring cost sharing for the very low income?

Ann Markus: Yes. Actually this is a very good question because the evidence that is out there in terms of the impact of cost sharing on health outcomes is not very comprehensive. The little information we have on that again comes from the RAND Health Insurance Experiment which did find deleterious impact of cost sharing on lower income populations which were either defined by the way as people with incomes below 100 percent or 200 percent, but the results were not necessarily broken down by those brackets. So there was some impact found but because of the caps that the experiment imposed on these lower income people, they couldn't get a true picture of how big that impact was.

A number of the other studies have also found by proxy using proxy measures of lower income people some effect, but it is really not clear how much of an effect it has.

Cindy DiBiasi: John, to explore that question a little more, what do we know about resources available to poor and low-income people?

John Santa: Well, we know some in Oregon, but we would like to know more. Again, our focus has been on adults so that is who we have tried to find out more about. While it is clear that we are talking about people who have very little or no discretionary income, our sense from talking to them is that many of them would like to have the choice rather than having a benefit eliminated to in some cases, have the option of co-pays, premiums, etc.

We have looked specifically at, we have a large population within this adult population of people who have zero income. We have found a high proportion of those people do have resources in terms of family and friends and other sources that assist them. We have also been intrigued, and this is in large part on the community basis, by financing options that emerge in those communities that are more oriented to providing universal coverage, or getting to universal coverage, or providing continuous coverage. We are particularly intrigued by some communities in Washington that have responded to premiums and co-pays in the basic health plan in very interesting ways.

We have also, by the way, found that oftentimes this population has not maximally leveraged, if you will, the resources available to them. They are not getting the assistance they are eligible for in terms of food or housing, which might bring more resources to bear for them in general.

Cindy DiBiasi: We have a suggestion and a question from Carol from Maryland. She would like to know, this is for John and company, could you comment on why more States don't make use of community health workers as a means of helping low-income and culturally-diverse populations make optimal use of the medical system: keeping appointments, having appropriate followup and the like, and thus stretching State resources?

John Santa: Well, what a wonderful question. Actually I think that gets it perhaps next phases and next issues. I think many of us are interested in looking at different, if you will, primary care models. I think in some of these models we are just going to have to admit that we need to look at options that are community health worker-based to do some of the work, the primary care work that needs to get done here. Particularly around prevention and key information, especially information to pregnant women, to children at particular phases, to portions of the Medicare populations or people with chronic diseases. I think we are going to soon see a phase where it will make more sense for some of this work to be done by community health workers in very different settings.

Autumn Dawn Galbreath: This is Autumn Dawn. I would like to just add to that and maybe broaden the focus of the question a little bit. All of the different modalities that I mentioned in my "management soup" discussion all involves the use of multidisciplinary teams. Multidisciplinary health care teams, be they community health workers, nurses, physician's assistants, a variety of allied health care personnel really, in my opinion, are crucial to the process of extending the reach of our health care programs. It is just a matter of what format and what type of team structure you establish within your particular health care delivery setting.

Cindy DiBiasi: Ann, a question for you. Is there any evidence that targeted cost sharing can reduce inappropriate utilization and save States a lot of money, for example, on emergency room visits or expensive brand name drugs?

Ann Markus: I would a have a question about the question. What is meant by "targeted" cost sharing?

Cindy DiBiasi: I guess that would be the benefit of the phone [laughter].

Ann Markus: I can probably already answer the question in terms of; I don't think there is a whole lot of evidence on that. That is the short answer to it.

John Santa: Maybe if I could just add here. We had a lot of discussion, for example, around prescription drugs. We ended up actually with an interesting benefit approach. Two sets of tiered co-pays. One for people below 100 percent and one for people above 100 percent of the Federal poverty level, three tiers in each. Below 100 percent, $2 for generics, $3 for mental health, cancer and HIV drugs because those have been drugs of particular importance. Then a much more substantial co-payment for a brand drug, $15—pretty high, maybe too high. That is one of the things that we are looking at. While for people above 100 percent, the co-payments would be a couple of dollars higher at each level.

We know from some research we have done that above 100 percent of the Federal poverty level those co-pays are OK. That is what most insurance companies are at now. It will be an obstacle for some people, but it is probably a reasonable compromise. The question really is, well, can we do a better job for poor Medicaid folks, particularly for brand drugs? That is where this evidence-based approach at, evidence-based center, I think, is so important. We need information about which brand drugs are really worth the price differential and in which cases can we tell doctors and Medicaid patients, you can be very comfortable using something much less expensive.

Cindy DiBiasi: We have a caller on the phone, Katherine, from the New Mexico legislature.

Katherine: Yes, there is discussion on the problem with multiple disease management entities and individuals. I was just wondering if any State has really worked out how to consolidate disease and case management problems?

Return to Teleconference
Proceed to Next Section

The information on this page is archived and provided for reference purposes only.

AHRQ Advancing Excellence in Health Care