An Interview with Ira Moscovice, Ph.D.
Ira Moscovice is the Mayo Professor and Head of the Division of Health Policy and Management, School of Public Health, University of Minnesota. He has been the director of the University of Minnesota Rural Health Research Center since its inception in 1992.
Moscovice has more than 30 years experience conducting rural health research and has served as the principal investigator for numerous rural health projects funded by federal and state agencies and private foundations. Moscovice currently serves as the Principal Investigator for the Flex Program Monitoring Team. Moscovice also has served on numerous research and policy advisory committees for federal agencies and private foundations, including the Committee on the Future of Rural Health Care. He was the first recipient of the Distinguished Researcher Award from the National Rural Health Association in 1992, and received a Robert Wood Johnson Foundation Investigator Award in Health Policy Research in 2002. Moscovice received his Ph.D. in Administrative Sciences from Yale University in 1976.
In his spare time, Moscovice enjoys photography, travel and restoring antique clocks. He and his wife, Sarah, live in the Twin Cities and have two daughters and one grandson. For more on Moscovice, and for a selected list of his publications, see his School of Public Health Faculty Profile page.
Much of the focus of your research center is on quality measurement. Why are quality measures so important? And why do they pose such a challenge for rural providers?
We’re now on the verge, with health care reform, of linking payment to quality and outcomes. So it’s essential that rural providers are able to participate in health reform activities.
I believe all rural hospitals, including Critical Access Hospitals (CAHs), need to report on relevant quality measures and measure the quality of care being provided in their facilities. They need to be willing, in a transparent way, to share that information with the public, with payers and others.
That’s been a challenge, particularly for CAHs, which are cost-reimbursed. Right now they report on a voluntary basis to Hospital Compare, but often they don’t have a large enough sample size for a particular measure for a particular condition to have information that we’re statistically confident about.
A big challenge for many rural facilities is staffing. Larger hospitals may have staff that deal just with quality issues. In a rural hospital you’ll have one staff member for this and it may be only one part of their responsibilities. The other challenge is that many rural hospitals believe that the quality measures being used are not relevant for them. For instance, there might be quality measures for surgery but many rural hospitals don’t do inpatient surgery.
Can you explain how quality measurement will be tied to health care reform?
We believe there are relevant quality measures for rural hospitals, and that this information should be collected and reported publicly. That will help lead to improvement in the quality of care in rural hospitals and also show larger facilities that rural hospitals can be good partners under health care reform. Quality measurement, and being transparent with the outcomes of that measurement, is relevant for rural hospital and other providers’ negotiations under health care reform. We have developed a comprehensive set of quality measures that are relevant for small rural hospitals and CAHs.
How will health care reform affect rural hospitals beyond quality measurement?
The ACA (Affordable Care Act) is focused on implementing health reform nationally but there has been only a modest amount of effort in thinking about how this is going to work in rural areas, where you have almost 20 percent of the U.S. population.
First, let’s discuss Accountable Care Organizations (ACOs). An ACO is defined as a set of providers that are responsible for the cost and quality of health care for a defined population (for example, Medicare beneficiaries). The goal is to contain cost, improve quality and coordinate care. ACOs need a formal organization and structure. The real concern is for providers in rural areas, particularly those who are not part of a health care system. How are they going to participate in ACOs?
If we think about the characteristics of how we want health care to be delivered, they include: clinically relevant information available, care coordination among multiple providers, clear accountability for total care of patients, easy access to appropriate information, and a system that is continuously innovating and learning. You need some form of organization to achieve this, with that organization having established mechanisms for working across providers and settings. The reality is that doesn’t exist right now in most rural communities.
To become an ACO, CMS requires an organization to serve at least 5,000 patients. Even then, 5,000 seem like a small population for an organization to be able to take on the financial risk that’s involved. The rural challenge is achieving this minimum patient base of 5,000, and rural hospitals don’t have a lot of experience with formal integration with physicians, FQHCs, and rural health clinics.
The recent final ACO rule states that CAHs (that bill for both facility and professional services), FQHCs and rural health clinics are eligible entities to form an ACO. I don’t know how much of that is actually going to happen. CMS set up a program called an advanced payment ACO model to help support the development of the infrastructure that’s necessary to become an ACO. So there are going to be opportunities and we’ll have to assess carefully how rural providers can get engaged in ACOs.
Are all rural hospitals going to be required to adopt the ACO model when health care reform is fully implemented? Will there be exceptions?
More and more, it looks likely that reimbursement from Medicare and from the private sector is going to be linked to a new kind of delivery model. Rural providers would be foolish not to pay attention to this. There are going to be some rural providers who will participate in demonstrations and some will take a wait and see approach, and delay change as long as possible. I think that latter group is postponing the inevitable and missing out on an opportunity.
The second piece of this is the issue of bundled payments. That goes hand-in-hand with ACOs. What the government is trying to do is remove the silo structure of medical payments, with care coordination accomplished across provider types and different settings. What bundling does is provide specific payments for a set of services for a defined episode of care. For example, hospital admissions are being linked with a specified number of days post-discharge, so we’re coordinating acute care and post-acute care. It is likely that physician services will also become part of the bundle.
If payments are bundled, key questions include who is going to receive the payment, who is going to take the risk, and how will payment be distributed to the providers involved from the initial admission, through rehab, home health, and other services?
The ACOs that we discussed could take this risk, receive bundled payments and distribute them to various providers. The concern is if the ACOs are mainly located in large, urban integrated delivery systems, are they going to be willing to negotiate agreements with rural providers who are not part of their formal delivery system? My answer is yes if you can show you are a low-cost rural provider who is providing high quality care. Then you have a much better chance of negotiating contracts with larger urban systems and receiving appropriate reimbursement.
What could happen to those rural providers that don’t become part of an ACO or a bundled payment structure?
If that doesn’t happen, you may well see rural residents being treated in urban hospitals and then getting their post-acute care in non-local facilities. That has significant implications for rural residents, for their families who want to be close by, and for rural providers who want to provide services in rural areas. The thing to understand is that it’s much more complicated, time-consuming and expensive to manage and implement multiple contractual arrangements with independent providers then it is to just restructure and reorganize internally within a health care system.
I think concerns about bundled payments could lead to increased provider consolidation and fewer provider options and margins. Providers may react to bundled payments by integrating and consolidating. A concern is protecting rural consumer choice and patient provider relationships. That’s why it’s important that rural providers participate in the CMS demonstration projects that have just begun.
Getting back to the issue of quality, what are some the factors that affect quality of care in a health care facility?
An important factor is linkages outside an institution (e.g., whether a provider is part of a formal system or network, or working with a state quality improvement organization). Linkages within an institution are important also since teamwork and communication are parts of effective care coordination. Other factors that affect quality include the financial health of the institution, physician leadership, nurse staffing, availability of technology, and an organization’s culture.
How can rural hospitals measure quality if they don’t have an adequate size?
There are several ways to address this issue. One is to use global measures across conditions. For example, with surgery measures, you can combine the same surgeries performed in-patient and out-patient. It’s important to use some quality measures even if you don’t have adequate patient sample size. You can use these measures internally, as compared to using them externally with payers. Internally you should always be providing the components of care that have a strong evidence base.
Could a rural hospital measure quality over time? For example, could they look at pneumonia cases over five years since otherwise they might not have a big enough sample size?
It’s preferable to measure quality annually. Virtually every urban hospital has enough sample size to do this every year. There are payment incentives geared to how well a hospital does in a given year. A three-year or five-year timeframe is not relevant for payment updates. As discussed earlier, global measures across conditions may help address this issue.
How did you move from an undergraduate degree in electrical engineering to health policy?
My training and background provided me with a very strong empirical, analytic way of looking at issues. I also wanted to work on issues that directly influence policy in the U.S. Yale had a graduate program on health policy that was linked to a strong analytic background. I wanted to do my doctoral research on health provider efficiency issues using a large computerized database. Back in the mid-70s there were just a few institutions that had this type of computerized information. An opportunity with an urban health system fell through at the last moment. One of my advisers had an MPH student working with the Frontier Nursing Service (FNS), which provided midwifery care in the hills of Appalachia. They were interested in someone assessing the cost and quality of the care they provided using the computerized patient encounter data they had collected for several years.
I went to Appalachia for the first time in the mid-70s and was really impressed with what FNS was doing, the tremendous health needs in the region, and the barriers that had to be overcome including geography, economics, and the lack of health professional supply. From that point on, I was interested in rural health care. It’s been three decades now. I never have looked back or asked if I made the right decision. It wasn’t planned. But sometimes the most fortunate things happen that way. It’s been my privilege to be able to contribute to improving the impact of health policy on rural populations and health professionals and institutions.