AMCHP 2005 ANNUAL CONFERENCE
DELIVERING RESULTS, IMPROVING PREGNANCY & BIRTH
February 19-23, 2005
CAROLE LANNON: Thanks to Ed and Meg for inviting me to be a panelist this morning.
I work at the Center for Children's Healthcare Improvement, and our goal is to transform the way that children's healthcare is delivered in this country and to support and build the capacity of others to improve that care. We all have a big job ahead of us, and I think as Ed said, it's partnering and working together that we can get that done.
I also have a role directing the quality improvement committee at the American Academy of Pediatrics. Several years ago there was a landmark report by the Institute of Medicine , Crossing the Quality Chasm. Are people in the room familiar with that report? Some nods. This report said that there should be several aims that we all work together to achieve for the 21st century healthcare system. We should aim to have healthcare that's safe, timely, effective, efficient, patient‑centered and equitable. Things that I think all of us can agree upon as being important.
One of the challenges is that between those six aims and what we have now, and that was the title of the report. It said we have, lies not just a gap but a chasm.
So I think there's ‑‑ and Ed showed you several slides of here's where we'd like it to be. There's a gap between what we know how to do and what actually happens in practice. What the IOM report said is that to improve the care, to actually achieve those six aims, we need to think about working at four levels. And I think these are things that probably you all are very familiar with. This was for me a very helpful concrete tangible framework to use.
The IOM report suggested that there are four levels in the healthcare system, and it's not just the levels themselves that are important but the interactions between and among them. If you look at the arrows, they're pointing both ways. So let me just walk through this briefly. The environmental context, the policy area. For those of you working at states, at federal areas, this is a no-brainer. It makes a lot of sense. At the macro organization level, depending on where you are, it could mean that you're working at your state MCHB. You could be working in a large county level.
For those people who are in practice, it actually might mean working at a health system level. There's the micro system of care delivery. What happens in that small site, whether it's a practice, a health department, an early intervention program, that that is really is place where the child and family intersect with the healthcare system. And then there's the child family and community. How do we ensure that these groups work together and, as Ed said, really partner to improve care?
I'm going to come back to this a bit later and talk about how some of the work we've done set a framework for trying to put, have a way to make these connections actually happen.
Let me go back to that gap. And I'm going to focus on preventive services for children for a minute. I'm a pediatrician, so I tend to use this slide and think about pediatricians. You all can apply it to the areas in which you work. Pediatricians spend more time providing preventive care than any single activity, yet when we look at the performance of preventive services, we know that at best 75% of two‑year‑olds are fully immunized. Rates of other preventive services are less than optimal. As Ed noted, more than 50% of parents do not receive information about child rearing or development. And the identification of treatable psychosocial risk factors, such as maternal depression is low. This is a problem not only in private settings, but also in public settings.
And I think when I often give this talk to pediatricians, they might be feeling like they're doing well, it's the other guys that are the problem. And I think we know that for all of us there's a gap and we would like to be doing better.
Why is there this gap? When we look at ‑‑ this is from an AAP survey. And what it did was look at the length of the preventive care visit by patient age group. If you look at birth to two years, the child, they broke this out, interestingly, by male or female pediatrician. I didn't have a slide where it was clumped together, because nobody really spends I think as much time as we would like. This looks at the time that's given for all personnel. So if you look at birth to two years of age, you can see that the child is in the office, being seen by anyone for about 20 minutes, give or take.
So I think time might be a factor here and might make us think that it's important that we partner, because 20 minutes of a child's life and a family's life is not a lot of time to get the work done that's needed.
Let me just see if I ‑‑ okay. Looking at the length of ‑‑ we just looked at that. Sorry. The causes of variation in preventive services. There was a slide here, and that's what I was just looking for that we actually did looking at 44 different practices in North Carolina , urban and rural practices. And what we showed was that there was an incredible variation in care. And I think there were several causes for that. When we look at why there's a gap, we can say preventive care, although we think it's fairly simple, is actually more complex than we realize. The American Academy of Pediatrics recommends over 200 specific preventive services in the first five years of life.
Often in a practice setting, perhaps in a health department setting, clinicians may have their own way of doing those preventive services. We often found that one clinician might choose to check the hematocrit at six months of age, someone else at nine months, someone else at 12 months. So when Ed talked about a system that works together, when there's a lot of variation just within one building or one site of how things are done, it can make it very difficult for things to be simple and move along efficiently.
And then a lack of systems. In this case, less than 20% of clinicians that we looked at used a prompting sheet at the time of the visit to help track what needed to be done and remind what needed to be done, and less than 20% used recall and reminder systems. We all go to the dentist. Some of you may have pets, and you know that veterinarians and dentists are very good in making sure you don't fall between the cracks. In healthcare, we don't tend to do as good a job of that.
So how do we close this gap? I've just laid out all the things that aren't being done well, and is there any hope here, how can we close this gap, and as Ed suggested, I think we all need to work together to help do this.
One of the models that I've used in my work is something called The Model For Improvement. And what it says is that you need to know what your aim is. You need to have some measures that let you know if you're getting there, and you need to have some ideas about how to test and change that.
So this says what are we trying to accomplish, how will we know the changes are an improvement, using some measures, and what changes can we make that will result in an improvement. Ed talked about agreed upon outcomes. And I think that's part of the aim.
When we're working with practices or with groups, maybe the aim might be improving the delivery of preventive services to children under five years of age. Maybe under two years of age. Maybe it's improving a specific set of preventive services, but it's by getting everyone to agree on the outcome then you can target and focus together on what you would like to do.
In public health, there's often a lot of statewide data that's collected. And I wanted to talk briefly about the use of data, getting down to the practice or site level. Here's the slide that I was looking for earlier. Each of these dots, let me explain this to you, represents one pediatric or family practice, or health department in North Carolina .
We did some baseline work with 44 different groups and plotted where each group was. Like immunizations, tuberculosis, anemia lead, blood pressure, vision screening and how many were up‑to‑date. And this black line is what the average or mean level was. What are your thoughts when you look at this slide?
UNIDENTIFIED SPEAKER: (Inaudible).
CAROLE LANNON: Right. So for tuberculosis, 0 to 100. This is right after a study had come out in North Carolina talking about the high rate of TB in rural areas. Not just after, a while after. So there's really quite a lot of variability among these. Any other thoughts when you see this?
One of the reasons that we like to use data like this is because when you look at this, you can say, you know, there's an incredible range. Some people aren't screening at all and other people are almost doing it on everyone. What can we learn from this person up here about how they got that to happen. What can we learn from lead screening about the people at the top so we can share it with those at the bottom?
UNIDENTIFIED SPEAKER: Do you know whether the dots that are in the top are in the top across all those categories?
CAROLE LANNON: You're like a great plant in the audience. That's about what I was just about to say. Interesting. And it's for me, as a facilitator for this, improvement was a great thing. No, nobody was tops across the board. And in a sense that says ‑‑ I think part of the reason is that there are different systems you need to put in place for immunizations that you need for vision screening, that you need for others, and the nice thing, it's as though you're in a classroom. It's nice that various people can be the stars, or be the teachers. And so you can P you will out and say what are the practices here doing with immunizations, what are these practices doing with lead. In general, some of the practices we're doing well across the board. There were a few we're doing well. Nobody was the best in every category, or even the top couple of best. Everyone had a gap, which, when you're working with a group, actually makes it a lot easier to get a sense of working together.
One of the other reasons that is important to measure is not just to find out who is doing it well but because we're all human and we overestimate our performance. Before we did that study we asked people to estimate how well they were doing. This Violet or blue line is what they thought in the marine and the pink is what actually was the case. Now, less you think that child health service providers have such big egos they think they're doing better than they do, this has shown up when you've looked at healthcare for adults, for mammograms, for any type of screening surgeons. So in fact the gap is much higher there. But I think one of the reasons we measure is to find out what it actually, what the gap actually is and to find out who is doing well so we can learn from them.
It's actually that last sentence, I think, is a very important piece of quality improvement. It's not to point out and say who are these people? They are really bad or bad apples, but let's learn from who is doing it well and share those learnings. So that's, I think, an important paradigm shift, particularly for me in medicine, where it's not always a supportive environment, if you're not doing well.
Once you've measured and you know where your gaps are, one of the things that we talk about doing is having a change package. A little bit jargony, but what that means is that you've got some specific aims you would like to improve. You've got some key concepts that you would like to involve. You have ways to measure, to know if you're improving, that are simple and can be done on site. There's ways you can train people both in if content and in improving you have some tools or supports to do that, things as simple as here's some examples of post‑it notes that you can put on a chart if the child is behind. Here's some examples of postcards that you could send out, and then providing coaching and support to help people make some changes. I'm involved in a project with the academy and many folks here. Chris Cross is on the steering committee for that looking at implementing bright futures. And what we've tried to do is say if you want a system for implementing bright futures, what are some of the key components to improving preventive and developmental care? And let me just tell you, out of the kitchen sink of things, we found things that have a proven track record of working and are pulling these six together as sort of the linking the systems framework, using a preventive services prompting sheet, so that when someone is seen, you've got a way of tracking what they're up‑to‑date on and where the gaps are, and using that at every visit that they come in for.
A use of a structured developmental assessment, a way to evaluate parents' needs and involvement. And this could be a range of needs and strengths. It could be things like they need to be connected to the WIC program. They need some Social Services support, some home involvement, or it could be focusing on what are their strengths as a parent? How can we emphasize this. It really is trying to be parent and family‑centered and meet the family where they are.
Using recall and reminder systems. Having ways to link to communicate resources, and when I say this, it's not just saying it but it's actually having a system in place in your office or site, having someone tagged with updating those list of referrals, calling those numbers, maybe once a month, to make sure those are still in place. Finding out who those are and what your population of children and families need.
And then considering, and by that I mean identifying children with special healthcare needs and helping them get connected in this process. So these are some ways that we've tried to take what's a big area and make it very tangible and concrete with matching tools and strategies.
You can come up with, and in a group of practice is in health departments that we're working with now, we're working with 15 to pilot those, that grouping of Bright Futures components. We're trying to say these are some things that we would like to target, how can we get there? And it's first by identifying your gaps, and then trying to put in place some things to help that improve. Some of these look at preventive services. The first three. One looks at a structured developmental assessment and the last one is on counseling.
The last thing we do then when we've identified some gaps, we've decided what we want to work on. We've got a package that can be helpful. One of the things we want to do is try what are called tests of change, and this is something called plan, do, study, act. It's a way of saying, when I first started working in this about 15 years ago, we'd say, well, we've got this great package, okay, March 1st we're going to start this clinic wide, health department wide and get everything going. We brought in donuts, tried to do it. We made a lot of mistakes early on and learned from them. And the tests of change approach says when you're wanting to make some changes, test it with one nurse. Test it with a doctor or nurse Dyad. Change the form. There's no magic about some of these things you need to adapt them to your local environment. So how can we work on this and make things work where we are?
What I wanted to do, and you'll hear some great examples of people who have done this in Rhode Island and North Carolina , is show you where we've tracked some of these things in two different projects. One was in New York City , when we were doing a project with the health departments there on preventive services and lead screening. And this is a way of showing just how we ask people to measure five or ten charts each month to let us know how they're doing and if they're making progress. So when the project started in April 30% of the sites were doing pay risk assessment for lead. Started to improve. Maybe the nurse went on vacation here in the summer. But something went down, and then things started to go back up. So it's by keeping some constant attention on this and focus and trying to work on making improvements.
Here's an example of a very dramatic improvement that happened in North Carolina, when we had 15 practices there and a pilot project looking at, using a structured developmental assessment, not just eyeballing, this child looks like they're developing well. What you'll see is a really dramatic improvement at the beginning, this was last June 10 to 12% were doing a structured assessment. It went up to 75% in a month. And one of the important things, and back to this slide, the reason I think that that changed and I think this graph really illustrates why you need to take this four level framework approach is what happened in this part here was that these groups were brought together. They were shown some examples of structured assessments, given some strategies about how to implement this, and July 1st, 2004 , is when there was reimbursement for specifically doing a developmental assessment. So you can see that that policy change really can help what's going on at the other levels.
Thanks for the opportunity to give you an incredibly brief overview of how we work on making changes. I'm going to turn this over to Ed, who is going to introduce the next folks, and I think you'll hear some very specific examples of how this happened. Thank you.