The following is a transcript of a technical assistance conference entitled Overview of State Snapshots held on August 10, 2010.
Select to access the full-length recording of the Web conference. (Flash File, 1 hour, 8 minutes)
Margie Shofer: Hello, I'm Margie Shofer in the Office of Communications and Knowledge Transfer at the Agency for Healthcare Research and Quality, otherwise known as A-H-R-Q or AHRQ. Thank you for joining us for this Web conference on the 2009 State Snapshots tool. Let me first say that if you haven't already downloaded the slides for today's presentations, please open the chat box for instructions on how to do so.
This Web conference on the State Snapshots is the third in a series of events highlighting several AHRQ tools created to help you identify and support areas for health care quality improvement. The first Web conference was on HCUP and HCUPnet, and the most recent was on MONAHRQ's Preventable Hospitalization Cost and Mapping Tool. If you're interested in learning more about these tools but missed the events, they will be posted on AHRQ's Web site soon.
Today we will be discussing the State Snapshots, which are dashboards that offer a visual presentation of health care quality measures. They provide State officials with a summary of their State's performance on types of care, settings of care, and care by clinical areas. Users can also drill down into the individual component measures of the summary to see how their performance compares to all other States, best performing States, and selected States in their region. New this year, users can look at data by payer. For example, you could review results for Medicaid.
We will begin our presentation today with an overview of the National Health Care Quality and Disparities Reports. Data gathered for these reports form the foundation for the State Snapshots. Next, we will talk about the development of the State Snapshots and provide a demo of the site. Following that, we will take some questions and then move onto a presentation of how Wisconsin officials have used the Snapshots in their quality improvement efforts. Finally, we will have another question-and-answer period.
Today's presentations on the State Snapshots will be given by Ernest Moy and Rosanna Coffey. Ernest is a medical officer in the Center for Quality Improvement and Patient Safety at AHRQ and leads the team that produces the National Health Care reports and the State Snapshots. Rosanna is a senior health services researcher and economist at Thomson Reuters, with over 30 years of experience. She has helped produce the National Health Care reports and State Snapshots since their start.
During the Q&A, you may hear from Marguerite Barrett, who is a consultant for Thomson Reuters and the technical brains behind the State Snapshots. The final presentation will be given by Judith Nugent, who is the chief of the Health Care Information Section at the Bureau of Health Informatics Division of Public Health at the Wisconsin Department of Health Services.
We would appreciate your active participation, as the primary purposes of today's Web conference are to introduce you to the State Snapshots and get your feedback on it, explore practical applications for the tool, and learn how you envision using the tool and for what purposes. Related to this last point, we hope you will tell us about the types of technical assistance that you might need in order to make full or better use of the State Snapshots tool.
There are two ways you can ask a question of our presenters today. You may submit a question at any time throughout this Web conference by typing your inquiries into the question-and-answer box located on the right-hand toolbar of your screen beneath the participant list. Or, during the question-and-answer period, we encourage you to ask live questions, which you may do by clicking on the "raise hand"button located at the bottom of the box containing the participants' list.
As a reminder, you can only ask a verbal question if your name is listed and if you have a phone icon next to your name. Raising your hand will place you in the queue of questions. We will try our best to get to all questions, but if we do not address your question, we will answer it within a few days after the Web conference. If you have remaining questions after the Web conference ends, please E-mail them to us. The E-mail address will be posted on the final slide, along with a link to access the 2009 State Snapshots. Without further ado, I am now going to turn this over to Rosanna.
Rosanna Coffey: Hello, everyone, this is Rosanna Coffey. Ernest Moy has asked me to kick this off, and he and I will be sharing the presentation. I would like to start with giving you a context in which we are talking about all of this with AHRQ's mission and strategy. In three words, AHRQ's mission is to improve health care, and that includes by addressing the efficiency, effectiveness, quality, safety, and equity of health care. AHRQ does this by supporting and conducting research, developing the knowledge, and then disseminating the evidence, helping to figure out how to measure quality, and then facilitating change and making health care better through that change.
Today we're going to talk about one of the products that AHRQ has developed to facilitate change. Standing behind that tool, which Ernest will speak about, are the National Health Care Quality Report and the National Health Care Disparities Report. Both of these reports were mandated by Congress. The National Healthcare Quality Report addresses national trends in health care quality. The Disparities Report addresses disparities in the delivery of health care, whether it be to racial groups, income groups, or other priority populations. With the 2009 report, AHRQ is doing its seventh annual report.
In developing this kind of report, there were tradeoffs and assumptions that AHRQ had to make. The primary audience for the reports is policymakers. The reports themselves are not a direct way to change quality or improve quality, but the reports are developing the knowledge that policymakers need to take action. The primary use of the reports is expected to be tracking. They were not developed to create public reports or to support pay for performance, and they primarily focus on geographic areas, not individual health care providers.
In terms of the measures, they are broad rather than narrow and deep. They are measures that have come together through a consensus process where AHRQ has vetted this very carefully with important stakeholders in the field. The measures generally are put together in composites rather than focusing on individual measures, although there are plenty of individual measures behind these composites. The reports are designed to be built on simple analyses, not complex statistical analyses, and the examples are typical examples of things that you would be worrying about in health care rather than the exceptional rare cases, and it really is a family of products that we're talking about today.
The reports are organized around four major chapters: effectiveness, patient safety, timeliness, and patient centeredness. The Quality Report also looks at efficiency, while the Disparities Report also looks at access to care and priority populations. Then, under effectiveness, we have seven different clinical conditions, as well as lifestyle modification, functional status, and supportive and palliative care.
The reports are built on 250 measures that have been culled from about 40 databases, and these databases are just some examples of those that are used to put the reports together. They fall into four categories: survey data collected from population, survey data collected from health care facilities, extracts from administrative data systems, and surveillance and vital statistics systems.
The two reports are really meant to be companions to each other. We've talked about the one being quality focused, the other being disparities focused and bringing in access. What we haven't talked about is how they look at these different populations. The Quality Report focuses on variations across States. The Disparities Report focuses on variation across populations.
I would like to share with you some of the findings from the 2009 report. These are no longer preliminary because the reports have been released. The first graphic shows the number of measures and the percentage of measures that are presented in the report and how those measures have changed over the previous measurement year.
If you start from the bottom, the white is showing measures that have become much better over the time period. Black shows those measures that have become better. Blue shows those measures that have not changed at all and green shows those measures that have actually gotten worse. From looking at this, you can see that over 60 percent of the measures are improving. But the pace is slow, especially for two types of measures that aren't shown on this graphic. One is preventive care measures, and the other is chronic disease measures.
Some areas merit special and perhaps urgent attention. This graphic shows hospital measures on the first column, those hospital measures that are related to safety. The second bar shows those hospital measures that are related to everything else that AHRQ is able to measure. The color scheme here is the same as on the previous slide, and what you see is that over 50 percent of the patient safety measures, including health care-associated infections, have improved; but they haven't improved as rapidly as other hospital measures, which are on the right.
One more view. This is a quick view of the findings from the report. This is from the Disparities Report, and here these two panels represent, on the left-hand side the static one-year view, the most recent year that AHRQ was able to measure for the 2009 reports, and the panel on the right represents the change over the measurement period from the previous time period. The color scheme, again, is the same, but we have reversed the order here just to keep you on your toes. Green is on the bottom this time. That represents a worsening of the situation or measures that were worse for this population, and blue, the same, and black, better.
If you look at the first bar, you see that over 50 percent of the measures for blacks are worse than they are for whites. If you jump over to the fourth bar, you will see that over 60 percent of the measures for Hispanics were worse than they were for whites. The last bar looks at income differences for the poorest group and the wealthiest group, and over 75 percent of those measures for the poorest group were worse than they were for the highest income group. Looking over time, you'll see that more than 50 percent of the measures are not improving, so this is why AHRQ feels that this is really an important area for everyone to be focusing on, disparities in health care.
My last point is that AHRQ absolutely understands that reports on the shelf do not equate to better health care, so they have spent a lot of energy and time thinking about how do you take these reports and have them facilitate and move the system toward better care? While the Quality Report and the Disparities Report present the information that policymakers hopefully absorb and understand where there are problems, that needs to be combined with local input. That local input can be in the form of local data and local planning, and local leaders, champions, and stakeholders all need to be involved to actually have quality improvement take place. That actual quality improvement that occurs will be at the point of care between providers and consumers of care, and consumers also have an important role in keeping their health at its optimum and improving their interaction with their physicians so that they get better care.
With that, I'm going to turn it over to Ernest, who is going to share with us one of the products that begins this facilitation toward change and making our system better.
Ernest Moy: Thank you, Rosanna. One of the ways that we can try to drill down from the reports, looking at life at the national level to something that is a little bit closer to the providers that actually effect change is to drill down geographically. That in a nutshell is the genesis of the State Snapshots. This is a product that takes all the information that we have collected at the national level and for those measures where we have State data, cuts it and shows at the State level so that States can see how they're doing, as well as compare their performance with other States.
There are a number of ways we think that the State Snapshots can be put to use. One, I think it's useful for raising awareness and having information at the State level. Again, it is closer than at the national level for generating an understanding of the need for change. All the measures that we produce are standardized, such as we do in the Quality Report and the Disparities Report. If others will use these measures and apply them to their local data, they will now be able to have benchmarks against which to compare. These benchmarks can then be used to target actions to particular kinds of services.
One of the things that I will show you is that we have specific benchmarks that highlight the best performers. These aren't just national averages comparing against, but you can actually find out who is best in the Nation and try to benchmark against them. I will show you all this stuff in the demo, but I did want to kind of very quickly summarize some of the high points of the State Snapshots so that you will be prepared to see what you're going to see. This is just a little bit of our philosophy and organization.
The goal of this was to try to summarize for State-level policymakers what's going on, how's my State doing? Instead of showing a lot of individual stuff, we have a focus on these composites and these performance meters, so that's what you'll see primarily. Although underlying them are all the individual measures, and I'll show you how you can drill down from these high-level performance meters down to the individual measures.
There are several emphasis areas that we have introduced over the last couple of years. One is clinical care, so you can look at cancer or diabetes, et cetera. This is the second year, I believe, that we're introducing information related to disparities, race, ethnicity, and income. This is the first year that we're showing a focus on payers that allows people to look at their State's performance now for specific payer groups. We also have parts of the State Snapshots that are meant to provide context for the State so people could understand how, for instance, their payer mix may differ from other States and how that may influence their performance. We have a user guide that can walk people through the State Snapshots.
I mentioned these high-level performance meters. This is kind of what it will look like. You will see on these performance meters two arrows. You will see a solid arrow, which is current performance, and a dotted arrow, which is baseline performance. This is simply a summing up over all the different measure sets that go into a particular meter of their performance, relative to a comparison group, so it is designed for a State and comparison groups; common comparison groups are all the States or the region. Below is an example of how this metric is actually calculated in a nutshell. You get one point for being above average, half a point for being average, and zero points for being below average, and then you take the average of that performance to find your place on this given meter. I'll show you some examples when I do the demo.
This is the overall high-level organization of the Web site, the different kinds of meters that you can get. You can get an overall meter. You can get a meter divided by the different types of care, preventive services, acute care services, and chronic care services; a meter divided by settings of care, hospital, ambulatory care, nursing home, home health; and the clinical areas, which I had mentioned previously. Then there are a number of different kinds of foci, and I will pay particular attention to the ones at the bottom. We have an expanded disparity section this year, a new payer section this year, and a new section that focuses on variation.
I think that's enough of talking about what you will see. Now let me just show you what you're going to see. Before I start the demonstration, I need to share some information for those who have closed captioning; that is, when I open the demo, the closed captioning window will drop out of view, but there will be a toolbar at the lower right corner of the screen. If you want to open the captioning window, you need to click on the icon farthest to the right on the toolbar and then there are two arrows. In the upper right corner of the captioning window you can use these arrows to increase the amount of text in the panel. Click the right button and then the left one.
With that, I think I can move over into the demo. I picked two States to highlight today. The first State I'm going to highlight is North Dakota. You go on the State selection map and you click North Dakota, the State I'm going to highlight today, and I will talk through the high-level information that you can get off of this version of the State Snapshots, as well as the State Snapshots in the past.
The first window that you come into is the State dashboard, and you can see summarizing across all the measures in our measure set, the performance of the State, in this particular case, North Dakota, both in the current year as well as the baseline year. Their performance is in that yellow bar, so about average. As you scroll down, you can see, using our rainbow pattern here, performance for the current year in the blue triangle, and in the baseline year, the open triangle, how they are doing across all the different dimensions that are summarized in the State Snapshots. You can look at types of care, and you can see that they are doing better on acute care and chronic care, a little bit worse in preventive services. You can look under settings of care, and you can see there might be some problems with ambulatory care measures. You can look at care by clinical area and you can see they might have some problems with cancer measures and with the heart disease measures. We think that's primarily how people would be interested in using this.
People usually are interested in areas where they could improve, and this is meant to quickly summarize some areas that they might want to focus on to investigate further about how they might be able to make improvements. Behind all these different metrics are tables, so I'll just take our overall quality summary metric. If I click on the meter there, you will get an underlying table, which shows how we got that particular performance on the meter. At the very top is a summary of the different measures that go into the meter, and you can see that for North Dakota they had a total of 77 measures: 19 were better than average, 42 were average, 16 were below average, and 32 were missing, and you got their overall average performance, and on the right you see how all the States in total performed.
If you scroll down, you will now see for individual measures how North Dakota performed. The first grouping is where North Dakota is better than the all-State average, better than the Nation as a whole. Then the measures where they're about average, and this is where people usually want to go, those measures where they are worse than average, and you can see information for the State, as well as a couple different comparison groups, the all-State average, all the other States, the regional average, just their particular geographic region. All the way here on the right, I won't click on it, but it's the NQR table number, and it might be a little bit obscure. You get these weird numbers. But if you click on it, you actually get a table from our Quality Report that shows for this measure how all the different States are doing. This is the source data that actually goes into supporting the production of the State Snapshots.
I'm going to just very quickly show some other features of this high-level view of North Dakota. Underneath the dashboard overall quality are strongest and weakest measures tables. This shows for North Dakota those measures that they're doing the best compared to the rest of the Nation and where they're performing worse compared to the rest of the Nation. Again, some places that people might want to go to get this high-level assessment and perhaps identify some targets for improvement.
I'll also point out one other thing that's available down here, which is an all-State data table for all measures. You can also see this on the left-hand bar here under other information, but this gives you the mega table that has for all the States how they're doing on all the measures. For those people who really want to drill down into everything, that's where it's located.
I also just want to show you an example of what one of these different more focal measures looks like. I'll pick in this particular case, preventive care, and you can see that the meter looks a little bit different. Earlier, we identified preventive care as an area that North Dakota seemed like they were having a little bit more difficulty with, and so they're on the lower average of that little yellow wedge. Again, I'll show you that behind this you can see the underlying table now just for the preventive services measures, and you'll see the better than the State average, about the same, and worse than the State average tables, allowing someone to drill into a particular kind of measure maybe to try to improve North Dakota's performance on preventive care.
I'll also show this feature in the upper right corner here, which is for someone who says, "Well in North Dakota it's not really fair to compare against all States. I'd rather compare against the States in my geographic region." You can pull up the West North Central States here for comparison, and you see that North Dakota does do a little bit more average for preventive services compared to the West North Central States.
You can also pull up a comparison group of the best performing States, which now pulls up a table that shows you how North Dakota does, kind of in the middle there. But on the top, you see the top five best States, and what their performance looks like, as well as the different quartile cut points. Then further down, you can see for each State alphabetically what overall performance is, so you can actually see how your State is doing compared to any other States that you want to compare against.
The other little thing I'm going to show about point of care is that you can scroll down and you see some links here. Some people said, "Well, okay, you've convinced me now that I have a problem with preventive services. What can I do about it?" What we tried to do is put in a link to AHRQ's Innovations Exchange, which is a listing of different kinds of stories about quality improvement. A person could get on there, click on it, and they would be taken over to the Innovations Exchange site, which would show success stories related to quality improvement. They're organized around different kinds of things, so you could find specific improvement stories related to prevention or you could find specific improvement stories related to North Dakota. Those are two different options for looking through that particular site.
What else related to North Dakota should I show you, settings of care? This is organized very similar to types of care. You can click on hospital care and you can see that they are doing pretty well on this particular metric, and all the same kinds of features are available for this.
I'm now going to switch over and talk about the new features that are contained in the State Snapshots this year, and for this, I'm going to change States to a State that has a little bit more ability to flesh out information about disparities by payer than North Dakota, and I picked Florida to show this. You can get the same, of course, overall dashboard and summary measures, but I'm going to drill down into some of the specific features where Florida is a good illustration. We have a couple of these focus areas that focus on conditions, and I'll show you one that focuses on the issue of diabetes. We've had this now for a couple of years, and I will just try to summarize in a little bit more detail information related to diabetes.
It's not just one giant meter. We provide a little bit of information about the prevalence of diabetes in different States. We provide the context for understanding what these metrics look like. Under quality of care we divided it into processes of care, where you can look at the processes of care related to diabetes versus the outcomes of care, in this particular case, hospitalizations for diabetes-related complications.
We have some links here for quality improvement related to issues, and in particular, we developed a cost calculator related to diabetes. On this you can see how much it costs the Florida government versus other States versus all the States in the Nation in expenditures for diabetes for State government employees. It's meant to try to motivate the State to act in the area of diabetes. From that cost calculator, we can also get some estimates of potential cost savings if diabetes control were improved by .5 percent or 1 percent in terms of hemoglobin A1c levels, how that would translate into savings to the State in terms of their expenditures on State government employees with diabetes.
We also have a focus. This is our first venture into the area of disparities. We looked at disparities in diabetes treatment by income. You can see a map about how the different States performed not in overall diabetes care, but in the gap between low-income and high-income-residents. You can see that Florida is about average in terms of this overall gap. At the bottom you can see the actual levels, in this particular case, receipt of hemoglobin A1c testing in Florida between the low income in blue and the high income in green for Florida, the South Atlantic States, and all States overall. In this kind of focus area on diabetes (we also have a focus on asthma), we're trying to summarize and wrap together lots of information that might be helpful for any State that is interested in pursuing some kind of quality improvement activity in the area of diabetes.
Another feature that we've had in the past is this focus on disparities. What's new this year is the expansion of measures that we have included in the focus on disparities. Under this, the first one called "all races" allows the State to look at its performance across the different racial groups. On the meter on the top here, what you see is the relationship between performance of care received by whites versus Hispanics and whites versus blacks and whites versus Asian/Pacific Islanders. There is not enough information to inform the white-Asian/Pacific Islander disparity, but there is for the white-Hispanic and white-Black disparities. In this case, those two racial groups do pretty well compared to whites in the State of Florida.
If you scroll down, you can see what the actual values are for our different sets of measures. Here we have on the top a grouping of ambulatory care measures, which are the avoidable hospitalization types of measures, and you can see the performance across the different racial and ethnic groups. Then the bottom is new this year; it is a performance meter on a package of hospital care measures. These are measures related to inpatient mortality when people are hospitalized for different conditions, as well as performance on different patient safety measures. So that's the quick summary.
You can also drill down for particular racial groups. I'm going to click on the blacks' link here. Now we're no longer, again, on this meter looking at overall performance for the State of Florida, but we're looking specifically at the gap between blacks and whites. What is the black-white difference? We see on this particular one, Florida is on the high end of average, meaning that overall it's a gap. This black-white difference tends to be smaller than what we see for the Nation as a whole. This is then illustrated on the table below as well for individual measures. The money is over here on the right where we see the little yellow square indicates that the gap in Florida is about the same as the gap in the United States. The little red triangle pointing down indicates that the gap in Florida is bigger than the United States, and the little triangle, the green triangle pointing up, means that the gap in Florida is smaller than the United States. What you see is there are more green than yellow, and that's why they get this above-average overall rating on this black-white disparity.
I'm now going to switch over to a new feature, focus on payer. The focus on payer was modeled to some degree off of our disparities, now looking at payer differences instead, but actually has a number of new features as well. The first one you see is the distribution of payers in the State of Florida, this little pie chart, and then you can drill down to the individual payers. This is an all-payer summary. The all-payer summary, like the disparities, now looks at how the different payer groups in Florida are doing compared to the privately insured. In this particular case, it looks like all the groups are doing fairly well compared to the privately insured.
On the table below, again, it summarizes the performance for the different payer groups for the different measures. In this particular case, only the inpatient mortality measures and the patient safety measures could be used for examining across payers. If we look on the privately insured link, now we see the performance among privately insured individuals, and this is a unique perspective. When we look at the overall dashboard, we're looking at the overall performance of Florida compared to the United States overall, covering all different payer groups. But, of course, not all States have the same distributions of payers. What this does is makes a comparison of privately insured people in Florida compared to privately insured people in the United States. So you're comparing apples with apples from a payer perspective, here just privately insured people.
Below it you can then look to see how the performance in Florida compares to the United States as a whole, and similar to previous views, the little yellow square indicates performance is roughly the same, the red triangle pointing down means the performance is worse in Florida, and the green triangle pointing up means the performance is better in the State of Florida. There's about an equal number of greens and reds, and that kind of explains the overall average performance of Florida, looking at the privately insured.
I'm now going to switch over to Medicaid to show yet another perspective. The top part is still the same, so this is now comparing Florida Medicaid with U.S. Medicaid overall. You see that Florida Medicaid does pretty well compared to U.S. Medicaid overall, the top part of the yellow average wedge there. Looking down on the left-hand side, you'll first see on the first series, the third column is the first series of comparisons, which compares Florida Medicaid with U.S. Medicaid. You see that there are more green triangles than there are red triangles, suggesting and explaining why Florida Medicaid looks better than U.S. Medicaid as a whole.
I also want to point you to this column all the way on the right. Here we're now looking at the gap between Medicaid and privately insured in Florida versus the gap between privately insured and Medicaid in the United States as a whole. The comparison now is no longer looking at Medicaid versus Medicaid but looking at the gap for Florida versus the gap for the United States, using the same kind of color scheme. Here we see a lot of greens, which indicates that the gap between Medicaid in Florida and privately insured in Florida is smaller than the gap between Medicaid and privately insured in the United States, and that partly explains their good performance in this particular metric.
The last thing I'm going to show you as a new feature is a focus on variation in time. We saw that there are all these differences across all these different groups, and we had different ways of comparing it. One of the things that was lost was the huge amounts of variation across States on any given measure, and that was meant to be captured here. What we see in this particular figure is how Florida is doing in red and how the United States overall is doing in blue.
We also have all the other States, not labeled, but you see all the other States represented there, just to see how much variation there is across different States on some of these metrics. This is something that we felt was lost in other kinds of displays and was meant to be illustrated in this particular kind of view. If I click on it, you can see the actual tables that underlie this stuff for Florida and the United States.
I think that's probably as much of the demo as I want to show you, and I'm glad to talk about it more, but let me turn it over to Margie, who can take questions.
Return to Contents
Proceed to Next Section