AMCHP 2005 ANNUAL CONFERENCE
DELIVERING RESULTS, IMPROVING PREGNANCY & BIRTH
February 19-23, 2005
JEFFREY GOULD: Thank you. Good to see some old friend here. When I started this project, I think, in a certain way its development was informed by three experiences. One was the perinatal profiles where we do risk adjusted neonatal mortality for all 355 of delivery hospitals in California , and the feeling that it just wasn’t very sensitive as a way to do quality improvement. The second was my experience with Storks. Does anyone remember Storks? Okay. So Storks was the Cadillac of collaborative projects up in the state of Washington . It was the hottest thing going. It was the most dynamic database, the most dynamic buy in and when the funding ended after three years, no one, you are probably the last rememberers of Storks in the whole world is right in this room. And I thought to myself, gees, what a shame, you know, that here is something that was so hot and dynamic, as soon as the money went away, it went away and that said something about, you know, what was wrong? You know, at public health we always talk about sustainability but what was wrong?
And then I had a job as a consultant to a private corporation that was starting a health management thing and all they ever talked about was value. You know, we have a bright idea and they’d say, well, what is the value? What does that bring to the customer? What does that bring to the patient? What does that bring to the bookkeeper, you know? And everyday anytime we would propose something it was value, value, value. And then I said, well, gee, maybe what we want to do is if we’re going to do something, we really have to think about value and sustainability. And maybe that Storks fell apart when it was no longer externally funded is it hadn’t built an intrinsic internal value base. So that’s a long introduction but I think it’s extremely important because what we’ve been trying to do is to use a bottom up value oriented perspective to build CPQCC. The worst thing about this organization is that no one can figure out a great acronym and we ended up calling it CPQCC.
We’re a group of public and private leaders in healthcare committed to improving care and outcomes for state pregnant mothers and newborns. As we show you the list, it’s pretty clear that a lot of these folks are enemies. California business group is an enemy of the practitioner who sees him as kind of pushing a different agenda. Even within the state there are various agencies that are, that have a little bit of rivalry. The thing that weds this whole collaborative together is the notion that everybody although they have different views on the world, believes that it’s important to improve the quality of care for, of mothers and their infants. We were organized in 1998, really with the coming together of MCH Grants, CCS, California Association of Neonatologist of the Nonoxed Network in the Packard Foundation. And our major goal was to improve the quality and outcomes of perinatal care in California . I learned that you always had to have a goal in Public Health School. I didn’t realize how important it was when you’re working with disparity groups.
The only thing that holds you together is having a goal and I certainly learned that lesson. So we wanted to develop a collaborative network really to support a system of benchmarking and performance improvement activities and that’s really what we’ve been able to achieve and we wanted to be value oriented. We wanted to bring value to the stakeholders. We wanted to have a value driven bottom up orientation so that practitioners rather than seeing this as a burdensome thing to do, really felt that it was contributing to the quality of their practice and had a good feeling about what was going on. And then we wanted to have value driven fiscal sustainability so that we wouldn’t go the way of Storks and so many other big projects.
Stakeholder value, to the California Association of Neonatologist, the value was that there were a lot of funding restrictions and people wanted to be able to point to how well they were doing or if they were starting to slide down to say, we told you, you couldn’t cut our budget as much as you cut it. There is inevitable report carding it was very hard in the late ‘90’s. It died out for a while and now it’s HUD again. And they really wanted to have input on report carding. They wanted to make sure that risk adjustment was fair. They wanted to make sure that people weren’t gaining their results as long as they had to do this. And they also wanted to make sure that when a report card was released, the context and the limitations of the report was clearly presented. People in California had been burned by all three of these concerns. And they also wanted to look at organized quality improvement as a possibility, but initially it wasn’t very exciting. The Neonatologists got into this initially really because they wanted numbers to support what they were doing.
State MCH needed morbidity assessment to push its agenda for quality because clearly mortality wasn’t sensitive enough. State CCS pays for a lot of care. You saw that on one of Shebeer’s slides and they have a mandate to assure quality insurance and they weren’t getting their yearly reports in a real usable format, so the idea of having a more user friendly report was important. Specific Business Group on Health was consumer oriented. It was kind of interesting working with them. They said that we had to go public with results, we had to post results on the Web, which everybody agreed to do. Packard and Van were interesting. Van is the Vermont Auction Network. It’s now about 400 neonatal intensive care units around the country and the world that voluntarily put in data. This data is then presented in the form of confidential reports that allow you to tell how well you’re doing. Packard funded this and it had absolutely no activity in California . So they wanted to see a local application in California .
The big three in CPQCC are the Executive Committee, the Database Abstractors Advisory Group and the Perinatal Quality Improvement Panel, and I’ll talk about each as we go along. Here’s the Executive Committee and the interesting thing is that in the first several years of the organization, the leadership of these folks sat on the committee. We didn’t get, it was very, very important that the head of MCHB, that the head of (inaudible) be there, that the head of, the CEO for PBGH be there. Now we have people that match the, because we’re now rolling along. We don’t often have, you know, the high level of involvement. But I think it was really important to get this off the ground by having support from the leadership and not someone three or four tiers down the rank. Our membership has risen from 25, yesterday we were at 98 hospitals with about 50 and on board coming up.
The interesting thing is that we have a mixture of regional community intermediate and undesignated nurseries, so this is one of the biggest networks. We’re really much population based and we’re recovering about 70 percent of all kids that end up in intensive care in California are cared for in these hospitals. And these hospitals are at various levels, so it’s good for modeling change. You can see that there is a plateau and then there was a burst. The burst came when CCS decided that in order to be a CCS reimbursed nursery you had to join CPQCC and report through CPQCC and that boosted our membership. Well, the quality improvement has a challenge. There are three components and if you don’t really focus on all three you don’t get anywhere. One is data, the second is taking that data and turning it into information, and then the third is take that information and doing something with it so it’s translated into action.
Our database, we started with Van’s small baby database, under 1500 grams. We expanded that to babies over 1500 grams who were on the ventilator, had surgery sepsis transport death, so we have a high risk of big babies. And then we do topical supplements. And we did a supplement for postnatal steroid uses. As you know, it’s very commonly used for chronic lung disease and yet it has very bad neurological outcomes so for a year we collected on postnatal steroids. We’ll probably be doing several yearly topical supplements to look at specific issues. Again topical supplements come from the membership not from the leadership so, you know, we have a lot of Neonatologists say we’re really concerned about what’s going on with steroids and that’s why we did the topical supplement. In a sense topical supplement is important because it gives us a closer tie with the bottom up approach.
We have a Data Abstractors Advisory Group that are incredible and we’re, you know, very few database organizations actually have an Advisory Group of women and men who sit in the cellar of the hospitals or the institutions extracting data and this has been extremely successful. Vital Stats is now starting to do that around the country. Actually meet with their Vital Stats registrars after years and years and years and years. I have to say that anyone doing a big database group, if you don’t have as one of your prime advisory groups, an advisory group of data abstractors, it’s just amazing. When we did the postnatal steroids, the doctors told us what they wanted to collect. We showed these to the data abstractors and some of the things they said, well really they just couldn’t collect. It was just ridiculous. But they did mentions something that the doctors hadn’t. They said well you know, you’re looking at steroids for chronic lung disease and, you know, they use a lot of steroids for low blood pressure.
And we brought that back to the doctors and they said, oh yeah, we forgot about that or we didn’t realize that. It turns out the use of steroids for low blood pressure is almost twice as great as the use of postnatal steroids for lung disease, and that was brought to us from a collaboration with the actual data abstractors who are in the cellars of the hospital. So this is a very, very important part of our work. Our database reports are quarterly and yearly. The big deal is that we have a confidential web base report. There’s a sample of it and people go on and take a look at that if they want. And here’s our website, pretty easy to remember. There’s of lot of interesting things on it as I’ll mention. Here’s a copy of a page from our 2001 report. We’re up to 2003 now. You’ll notice it has some unique features. One is that if you don’t like numbers, you can look at the little cartoons on the right and those lines are interquadal ranges and the dot is where you sit. So if you don’t, if you hate numbers, you can just run down those interquadal ranges and see where you are.
Once you see where you are you might want to see where you are with relation to the rest of the network and so there you are in that colored bar and there’s the rest of the network. You can adjust this so that you can only look at hospitals of your same level. You can look at in borns versus out borns, or kids born elsewhere, you can look at various strata of birth weight all interactively on the web. Each hospital has approximately one gigabytes worth of information. We spent a lot of time working with the hospitals to show them how to extract useful stuff from the one gigabyte. One of the things that we do on all of the important outcomes is we also give them a standardized report and you can see this is standardized by gestational age and (inaudible) and birth weight and in born, out born. And this is fairly sophisticated and then this year out of light, when you look at this chart it glows a different color. I’m colorblind but it glows in different colors.
So again, you don’t even have to read the numbers. You can just tab through these things and when you see something that’s glowing this awful color then you can actually say ut oh, and look at the numbers. Well, so that’s the data. You’ve got to have data if you’re going to do any quality improving, you need data to see where you are. You need data to track where you’re going. But what we have to do is transform that data into information. So we have the perinatal quality improvement panel, Dave Werchafter is the chairperson who had done an incredible job. And what it does is it looks at the data to identify topics with improvement potential. We call them TWIPS. And once we have a topic that seems to have improvement potential, that means that as a state we’re doing worse than the literature says we should do and there’s a lot of variability from, from unit to unit. We then have to find out whether there is any strategies that have been proven to improve things. So if we have don’t have the strategy, we just don’t take it on. We love to take on inter cerebral bleeds but there’s no strategy so we don’t.
We did take on those (inaudible) infection because there are many good strategies that have been proven to reduce infection rates. So once we’ve assessed the validity of improvement strategy, we develop tool kits and then we have statewide education lectures, workshops. And the cycle we use, is we’ll pick up the topic, we’ll build a tool kit, we’ll present it at the yearly meeting of the association of Neonatologists along with a workshop. The workshop is now attended by teams from hospitals rather than individuals. They’ve got a lot of homework, so that by the time they come to the workshop they’ve already probably put in 10 to 12 hours as a team analyzing what’s going on in their hospital. So the workshop then becomes a workshop to solve some of the problems that they’ve discovered. It’s kind of like, you know, you’re already one leg up by the time they get there. And then we have a public release cycle for certain issues. This is just to show that we have a lot of Neonatologists, these are very prominent folks in California . We have Perinatologists, we have state people, we have our regional perinatal system folks in staff. And so it’s a very, very broad based committee and it’s really, and they really do meet.
The room is pretty full when we meet every two to three months. Here’s anti natal steroids. It was our first project that, you know, anti natal steroids should be around 85, 95 percent so we’re, none of our units were meeting that criteria. There’s tremendous variability and the literature said that with education you could really improve it. And so had 1999 as a baseline year. We developed our tool kit. We did a new convention in 2000. We did an evaluation in 2001. We fought about public release for a couple of years, how we were going to do it, the word snipping et cetera. And then in 2003 we went live on the web with the public release of every hospital in California and it was part of this and how they did. Two hospitals dropped out and didn’t want to report and their report is dropped out, they didn’t want to report. All right, here is the change. We went from I don’t know 76 percent to about 86 percent. And that’s not very exciting to actually, it’s pretty but it’s not very exciting.
But this is exciting. These were the seven worst hospitals at the stat. One had, was doing anti natal steroids for 46 percent, 50 percent, 51 percent, 60, this is really bad. This is really bad. And at the end we had gone from 46 to 88, 50 to 87, 51 to 87. So if like to shoot the outlier approach, here are the outliers and we really, we really gone to moving. If you like the bring everybody up approach, what had been our, what had been our median, what had been our median level became our lower quadtile level at the end of the intervention. So, we had shifted everybody up as well as these outlying folks.
UNKNOWN SPEAKER: This is the first question. Have you continued to track data for that to make sure those (inaudible).
JEFFREY GOULD: Well, that’s an interesting thing and we’re going to do that this year because people have been, you know, people have been wondering about that. One of the problems we’ve gotten into is that you’ll see how much work we’ve done. We’ve been moving so fast that we haven’t done as much evaluation as I had wanted. I actually try to put the brakes on this, and said, lets put some of our talent into evaluation and they just want to keep moving and I’ll tell you, when you have a room full of 15 or 20 Neonatologists that are really high end directors, and you’re all working on a toolkit, you know, you don’t want to disrupt that. I mean, there’s a, I mean, it really is. There’s an energy there. And I was torn because I really wanted to be able, who’s doing it, you know, lets, but these guys just want to keep meeting and developing toolkits, and having lectures and seminars at there hospitals. So I, you know, that’s kind of, that’s the long and the short of it.
We have cycles on anti natal steroid public release in 2003. We’re in a nose (inaudible) infection and it will be targeted for public release. We have product line prevention. We have breast milk for pre terms. We now have, we’re releasing protocols for normal baby hyperbilirubinemia to go along with Academy of Pediatrics guidelines. And we also have a hyper out guidelines as well. So, we have a lot of these. And these toolkits are available for free on the Internet. They are huge. They’re really quite large. It’s a one size fits none philosophy where if you’re really advanced there’s stuff you can do. If you’re really primitive, there’s stuff you can do. And it starts off, each toolkit starts off with the evidence for, yeah this is a big problem, why this is a big problem, and how it can be changed and it has been changed so, just the kind of stuff that you need to motivate folks within your institutions. And we invite people to come to our website and we had I think in the last three months, 800 downloads. And they’re downloading from all over the world.
The toolkits are really quite spectacular actually. And then we’ve been engaging packets from outside of a provider circle with specific business group on health. We’ve been working on cesarean section report cards and risk adjustment with the State Office of Health Planning and Development. Again, we’ve been working on cesarean sections and the deliveries we’ve found that hospitals with most cesarean section rates, sometimes have higher morbidity than expected, which has been problematic. We’ve been working on children’s health service to develop a specialized reporting process that will more easily direct intervention. California NCH we’ve been working with CQI and American Academy of Pediatrics on bilirubin guidelines. Quality has become very big in clinical medicine.
The American Board of Pediatrics is making quality in Neonatology part of the reapplication or relicenser of things that you have to do and we’re trying to work on guideline recertification by pushing the idea that rather than using synthetic measures and synthetic tests that people that are really involved in this kind of collaborative should get credit for recertification license or recertification. We have research. We’re doing research on C-sections, on postnatal steroid use, on data quality. And we have some future activities in the obstetric quality initiative I’ll talk a bit at the end. We’re starting to roll a perinatal transport database into this area. We’ll not only know morbidity in the infants but we can look at where they were born and issues about their transport and maternal transport. And we’re trying the rural hospital initiative, which will probably rollout in about a year.
Everybody thinks of California as L.A. and San Francisco and these big populated areas. It turns out that a lot of California is extremely rural. There are places in California you can drive for 45 minutes and not see another anything. And then you’ll get to a hospital that has an intermediate care nursery because in the winter with the fog and snow it’s completely isolated. So we have a huge rural population in California . And one of the questions that we’re trying to answer is what does quality mean to these rural hospitals and how do you improve care in rural hospitals and what are there needs et cetera. And we’re completely at a loss because everything I’ve talked to you has been really coastal folks and mid California folks that are big population University folks and so our rural initiative has really just gone off to focus groups of people that work out in rural hospitals to see what kind of database could be built that would facilitate improving things in that situation. Here are a couple of reading. Neo Reviews is on the web and if you’re interested in quality improvement and perinatal medicine, these are three pretty decent readings if folks are interested. And I think I’ll end at this point. My 20 minutes are over.