Q&A with Jean Moore and Sandra McGinnis, SUNY Center for Health Workforce Studies

SUNY Center for Health Workforce Studies


By Alex McEllistrem-Evenson

The Center for Health Workforce Studies is located at the University at Albany, part of the State University of New York School for Public Health. One of six regional Health Workforce Centers funded by the Health Resources and Services Administration (HRSA) Bureau of Health Professions, SUNY’s center is situated at the front lines when it comes to managing health workforce data and dealing with data practices issues, producing an incredibly broad range of research publications and reports that have a direct impact felt by most everyone involved with the health workforce. Recently, Health Workforce News had the opportunity to talk to Jean Moore, Director of the Center, and Sandra McGinnis, Ph.D., Senior Research Associate, about some of the projects and initiatives which are commanding most of their attention right now.

Generally speaking, what are the most pressing issues facing those who depend on workforce data, in any field?

Jean MooreJean Moore (JM): It’s important to recognize how challenging it is to find really good data on most health professions. There are some states that are actively engaged in collecting data on their health professions, but others aren’t. I think that many states are beginning to recognize now that they need these data to inform policy decisions. There’s growing awareness that if they’re not currently collecting data, they should consider strategies to do so.

Sandra McGinnisDr. Sandra McGinnis (SM): One thing I want to bring up that is on everybody’s mind right now is health care reform. We’re seeing more and more physicians state publicly to the media that they’re concerned about the impacts of health care reform because they feel if you cover everybody, you’re not going to have enough providers to go around. That’s a frightening issue, but the fact is that we need to know before we make decisions about how the system’s going to be structured. We need to know what we have and we need to know what we’re going to have ten years from now. Otherwise you can’t make the appropriate policy decisions.

JM: I think that this issue is clearly on HRSA’s mind as well. They recognize that efforts to reform health care are more likely to succeed if we invest in research that can help us to better understand the health workforce we have now and and the workforce we need for the future.

Given the speed at which government is moving towards health care reform, are we going to have a situation where research needs to catch up to policy?

JM: Bear in mind that some states have already begun this work, and there is some active data collection. My suspicion is that those who can do the kind of analysis that will inform their policies will do so. One of the sessions at the recent health workforce summit convened by the HRSA was on state data collection and analysis. Given the level of interest and attendance, I think that many states want to get involved in this because they recognize the value of state-level health workforce monitoring. So you’re right, the move toward state-level data collection may seem pretty slow but I think there may be efforts to invest in it and to help states make good decisions about how to go about doing it. If that happens then I think it will be a valuable resource as we go forward with health care reform.

SM: I would qualify what Jean says about states. States are starting to collect the basic supply and demand data. There are issues where the surface is barely being scratched. We know where physicians are, we know where their practices are located and so forth, but relative to workforce, we don’t know who is getting care and who is not. That’s something that’s going to change with health care reform because people who have been outside the system are going to start being covered and looking for services. I think that there needs to be more connections between the workforce data that we have and the data on the patient end: how does this workforce affect patient access and patient outcome? I think that’s the next phase and we really haven’t done a good job of that.

JM: At times we’re too busy focusing on the supply side but we have to understand how it affects patients as well. It isn’t only about access, it’s about outcomes. What impact do different approaches to delivering services have on the health, well-being, and quality of care provided to patients?

I noticed that many of the SUNY Health Workforce Center’s projects relate to the nursing workforce. Do different Workforce Centers specialize in data on different fields?

JM: We conduct research in a lot of different fields. We’re not limited to any particular health profession. Our focus is health workforce research. We recently completed a physician supply/demand forecast study for New Jersey. We’re just wrapping up one for New York.

I was doing some research and came across a HRSA study from a few years ago that came out of your Center, about using facility-level data to predict nursing shortages. I think this is a project that provides a good indication about the complexity of workforce data and related issues.

SM: Basically we found out that predictors of shortages in one state [North Carolina] didn’t necessarily predict shortage as well in another state [North Dakota]. We didn’t have enough information to be able to say; we suspected it might have been because North Dakota was much more rural. And so these things predict shortage in rural states and these things predict shortage in less rural state but we really couldn’t take that any further.

JM: I think it’s important to point out that as the result of finding that we went in a different direction that actually did have some relatively positive results and we’re continuing to do that work.

The project shifted focus from facility-level data to making predictions based on geographic data, correct?

SM: There’s actually a journal article (PubMed abstract) that came out this month that summarizes the different iterations we went through. Basically the consequences were that we weren’t able to use facility-level predictors. We had to shift from trying to predict shortages in individual facilities to a county-level model that used county-level data. There’s so little comprehensive facility data collected that it really makes this work kind of difficult. We were able to do the work but not at the level of analysis that we had hoped.

Initially the first grant we had gotten was to try to come up with an alternative methodology to what the federal government had been using. They had been identifying shortage facilities simply by type – if you’re a disproportionate-share hospital you’re considered a shortage facility. They felt like this broad approach was very likely not the best that they could do. So part one of this project had been to explore different alternatives that we might have, instead of using the very basic  categorical definitions, and the result of our work on the first grant was to come up with what we felt was the best direction to go. With HRSA funding we are moving on to the next iteration of this work, making some modifications to the data we are using, and basically getting ready for it to be put into practice by the federal government.

The models based on geographic data are what is being implemented. We still feel like facility data have some promise but we need more of it. Basically the only two states that had these data comprehensively that we could find were North Dakota and North Carolina. If all of the states collected these data I think that this would be a workable strategy and probably a better strategy than geography. But that’s where the limitations of the data really made that an impractical approach.

So geographic-level workforce data is something that is collected adequately in most places?

JM: Not states. States aren’t necessarily collecting data.

SM: The geographic unit we used was counties. And really, there is not sufficient data out there. There are really no statistics available to tell us how many active nurses are working in every county in the United States. We’ve had to try to estimate that from various sources. Currently we’re looking at licensure lists. But first of all they tend to give the address of the nurse’s residence rather than her work. In some areas around major cities that can be a big difference. The other shortcoming is that probably twenty percent of people who are licensed as RNs aren’t in the labor force.

What expectations do you have with this second phase of the project?

dataSM: Phase two is well underway. The question is, “what are the best data sources to plug into the model to be able to do this work?” So we’re pretty close to figuring out what that is, coming to terms with what the limitations are. Once we do that, we want to run the model, produce county-level detail of where we think shortages are most severe, using a relative comparison of level-of-severity within a state, and we actually want to reach out to a small number of states and ask them to look at what we found and to help us understand what we may have missed. We want to understand what the flaws in the model are. Ultimately, the purpose of this is for the federal government to give out and make decisions on nurse scholarships and nurse loan repayment awards. They really want to target the neediest areas.

JM: I think we’re going to come away with a model that is going to be as close as you can get right now given available data. I think that the federal government would really like to use what we’ve learned to make better-informed decisions about the selection of placement sites for nurse scholars and loan repayment recipients. In our outreach to states for feedback on the model, we want regional representation and we  want to reach out to groups within those states that  have a good sense of what’s going on and can tell us  how close our model is to predicting where shortages are most severe.

We’ve talked a lot about this nursing shortage project because I brought it up — are there other specific projects you’d like to tell us about?

SM: One thing we’re doing in New York that’s really putting our state ahead of the curve, with support from our Departments of Health and Education, is the development of a health workforce tracking system for certain professions. We’re implementing this through re-registration surveys for certain professions which now include physicians, nurses, dental hygienists and dentists –

JM: — and we’re soon adding nurse practitioners, physician assistants , and midwives –

SM: — when they re-register with the Department of Education they get a survey with their re-registration packet, one page front and back. It allows us to keep tabs on those professions. Eventually we’d like to expand that even further, maybe to include pharmacists, social workers, physical therapists. With any of the licensed professions,there’s the potential to do this.

Why do we need these data?

JM: I think the short answer is that we can’t afford to be surprised by health workforce shortages and we need to be mindful of supply and distribution, and frankly, mal-distribution. One of the things that we do with our physician data is an analysis of primary care shortage areas. We’ve noted over time that the distribution of physicians in the state is changing. Using finely-grained data to do an analysis of shortages in the state is pretty valuable, and you can target resources in the directions you need.

In terms of additional projects, we’re also part of a larger group that is assessing the adoption of health information technology (HIT) in New York. The state has made some sizeable investments, trying to get providers moving in the direction of adopting HIT and they’re also looking at creating a statewide network to support health information exchange. We’re close to wrapping up a statewide survey of hospitals about their use of HIT and we’re about to launch a survey of a sample of ambulatory physicians about their use of HIT. Given the federal investment in HIT that’s in the pipeline, there is considerable interest in what we know about what’s going on, what’s changing and what the issues and barriers are.

When you say “pipeline,” are you referring to ARRA, the economic stimulus bill?

JM: Stimulus funding is what has everybody’s attention now. Is there some potential for more resources [for HIT] in the longer term? I would say possibly. I think that when you talk about health reform it’s very clear that this technology is going to be a cornerstone for creating more system efficiency.

What else are you and your staff focusing on right now?

JM: A couple of years back, HRSA talked about changing the methodology for HPSAs (Health Professional Shortage Areas). Because we do HPSA analysis, we actually looked at the impact of the change on currently designated HPSAs and MUAs/MUPs (Medically Underserved Areas / Populations) in New York. One of the things we concluded was that we don’t systematically assess the state for shortage designations. Typically, an area that is currently designated as a HPSA is likely to be reassessed to determine if it still qualifies. If there is a potential for a new HPSA, somebody has to take a lead in a community to try to find out if it qualifies or not. We think that lack of systematic approach is very troubling, particularly because we know that given the changing distribution of physicians in the state, there might be more areas that would qualify for shortage designation.

We secured a local health planning grant from our State Health Department to create a statewide set of primary care rational service areas. We want to base these areas on the commuting patterns of patients. We want to know where people live, compared to where their primary physician is located — how far they’re travelling. We want to use that as a basis for creating an initial set of rational service areas and then we want to reach out to local stakeholders in communities to help us understand what the data don’t tell us, related to language barriers, other sorts of barriers that we may not necessarily appreciate or capture in the data that we have.

What method are you going to use for that community outreach component?

JM: We actually identified a number of groups that have agreed to convene local advocacy groups, provider associations, consumer organizations, and others to help us understand what aspects of this we have to consider as we go forward. When we were thinking about the outreach to local stakeholders we built it in as we were writing the grant to identify the organizations we thought could help us get to the right groups to give us the feedback we need.

SM: That part will be qualitative. We’ll be talking to them, interviewing them, asking them what they think of this.

That’s interesting – the bad rap on research is that it’s impersonal, that the qualitative is often left out.

JM: I think qualitative is key. Quantitative and qualitative research can go hand in hand if you want to understand the landscape. I agree with you, I think research has gotten a bad rap because we fail to provide practical applications. And I think it’s important to not be too academic in our efforts but to really understand what information policymakers need to inform decisions and that research should be geared to do that. I think you’re right that we have to turn this around and find ways to really create more relevance for what we do.

Please note that the views expressed in this article are the opinions of the interviewee and do not reflect the official policies, positions, or opinions of the Health Workforce Information Center or its funder.

For more information about the six regional Workforce Centers: