AMCHP 2005 ANNUAL CONFERENCE
DELIVERING RESULTS, IMPROVING PREGNANCY & BIRTH
February 19-23, 2005
ALAN HINMAN: Thank you very much, Henry. It’s a pleasure to be here and before I get started I just ought to explain a little bit about the Public Health Informatics Institute, which is a not for profit organization that has been funded by the Robert Wood Johnson Foundation and we have also been working closely with HRSA over the course of the last five years on topics of integrated child health information systems. We started out as All Kids Count, which many of you may have heard of, a program on immunization registries about 13 years ago. And one of the things that I would also like to mention is you’ve heard a number of important points being made, issues of partnership or issues of trying to assess what the systems need to do. Some of these have been encapsulated into what we call a Source Book on Integrating Newborn Screening and Genetic Services with other Maternal and Child Health Systems. And we claim there are five important lessons that were learned from, in setting up these systems. The first is that data are for sharing.
Pat Nolan in Rhode Island says that information is not a possession. It is, in fact, it’s one of the few things that gains in value as it is passed along. So I think one of the real concerns we have with information systems is that people are not using them. They’re not using the information within them. The second point is listen-up. This is your partner’s needs, what the systems are, need to be able to do. The third is that changes is hard and in Utah they’ve tried to address this by changing things as little as possible for the individual programs. But even so, it is difficult to change and having as Sherry pointed out, having a good information system, in fact, will change the way you do things. And you need to be ready for that. The fourth is to let Public Health Program needs drive technology. And this gets into some of the different language stuff that Sherry was talking about. A sometimes coarse way of putting this, is don’t put the geeks in charge. Decide what it is you need the system to be able to do and then demand that the system be developed to do that.
And the fifth is to stay the course. This is a long-term process. Rhode Island ’s been working on it for 10 years. It ain’t perfected yet. It’ll never be perfect but it’s getting pretty darn good and the more you use it the better it will get. So that said, let me move into what actually is my presentation and start by giving you a brief outline of what it is I intend to talk about, which is to talk about an activity we undertook a couple of years ago to develop statements of principles and core functions. What is it our Integrated Child Health Information System needs to be able to do? And we looked at four systems for starters in cooperation with Health Resources and Service Administration. These four are vital registration, newborn dried blood spot screening, early hearing detection and intervention, and immunizations. These are four services that are provided to every newborn. They are primarily provided with the exception of vital registration in the private sector, but Public Health has a very important function. They’re also time sensitive. If they don’t get done on time, you can have a bad outcome.
And finally, they’re mandated in every state with the exception of 13 states don’t yet mandate early hearing detection intervention but they will over the course of the next few years. So, it seemed like a convenient package. We’re going to talk about how we developed the principles and core functions. Then, how we developed some performance measures. How are you going to know if the systems are working? Then we are going to go into specifics and this will be some interactive stuff. And then if there’s time at the end we’ll talk about some other ways, attributes of information systems. So a couple of years ago, nearly two years ago now, we convened a small work group to develop sort of a statement of what it is and Integrated Child Health Information System would have to be able to do. And we pulled together a small group of folks none of whom are in the room at the moment, but you’ve already heard about Amy Zimmerman. Utah was represented by John *Ikewald who’s now at CDC, and then representatives also from the Maternal and Child Health Bureau and the Institute. And over the course of a couple of days we came up with a list of 19 principles, which we consider to be sort of fundamental characteristics of Integrated Child Health Information Systems.
As you can see, the majority of them dealt with security, confidentiality, and how the technology needed to serve program stakeholder needs. Then 22 core functions reasonably evenly divided between confidentiality, security, getting records established, service, and technical functionality, and then eight desirable functions. And we then circulated this draft list to an external review committee, talked about it with MCH grantees and a variety of other things, other folks, and then actually published it in an article in the Journal of Public Health Management and Practice, which is back at the back. And just as a plug, there is a supplement to the Journal of Public Health Management and Practice from November of last year, which contains 16 articles on Integrated Child Health Information Systems. Dr. *Hollinshead is an article, an author of one of these articles, Ellen and Dave are authors on several of them. And there are copies of this available at the Public Health and Dramatics Institute exhibit in the Exhibition Hall.
We also have a lot of other goodies, handouts, you know, give me,s, that kind of stuff. Please take them. We do not intend to carry them back. I think you’ll find that many of them are quite useful for you. There are a series of materials that I put as handouts back in the back. One of them is the slide presentation, then the article on Principles and Core Functions. So we sent out these Principles and Core Functions. We then did a survey of SPRANS grantees, talked further with folks, made some site visits. The article was published and we have redrafted now at a meeting reasonably recently the Principles and Core Functions and that is now one of the handouts that’s in the back entitled Principles and Core Functions, January, 2005. Some of the changes were that in sober contemplation of the list of Principles and Core Functions folks thought that there was a mix of functions, requirements, performance measures. And so a lot of things were re-categorized. There was nothing that was felt to be wrong. There were some additional things brought in. And we then came up with a revised list, which you have. We won’t go into it in detail, but it’s important to bring it up because this is then the next question.
Okay. If this is what a system has to be able to do, how do you know if it’s doing it? How are you going to be able to measure its performance? And here we’re talking about the performance of the Information System, the Integrated Information System. Not the individual program or the individual program information system, which the program as in Utah is going to maintain its own information system and so we’re trying to get at the functions that are added by integrating systems. So we convened a year ago, a work group to draft a set of performance measures and based on the list of principles and core functions, which we had at that time, which is not this revised list, but is very, very similar. And so the notion is, okay. If these are the important functionalities of an Integrated Information System, what measure can we derive, can we think of that will tell us if it’s doing it? And the participants in this work group included several people who had been in the Principles and Core Functions Group, Sherry Spence was one of them.
Utah was again represented this time by Richard *Harward. And the folks included genetics counselors, hearing screening people, immunization program managers, laboratory, private practitioner Bob *Kosack is a practicing Pediatrician in Massachusetts , and parents as well of children with special healthcare needs. So we tried to get the major players involved in designing these performance measures. Also had really very significant representation from the Maternal and Child Health Bureau and from the institute. And we drew up a set of performance measures, which included some quantitative measures as well as some qualitative indicators. And the qualitative indicators we are calling functionality statements and we have a checklist of them. So after we drew this up a year ago, we sent it out for people to look at and we presented it at a variety of different sorts.
We made a number of site visits around the country to see if these passed the *laugh test and some of them didn’t. And some of them people thought, needed to be changed in some way or another. So we then proposed some revisions and convened a work group last month to review both the principles, the revisions to the Principles and Core Functions as well as the proposed revisions to the performance measures. And here we had again some of the same suspects from the previous groups as well as some new folks. And this has turned out to a very, very effective format for us to bring together this diverse group of people representing different kinds of perspectives and expertise. But by having some folks carry over from one section to another, it really leads to very good working together and folks worked together very well from the start but I think there’s something added by having some continuity in there. So again we had substantial representation form MCHB and from the institute.
Okay, now we can get into what they really are. And you have, one of the handouts is a list of performance measures and I’m going to go through them in some detail. The first of which is, the percent of newborns who have a record in the integrated system. And we say that will be calculated by determining the total number of live births that occurred in a jurisdiction in a specified time period as the denominator and the numerator is the number of live births that occurred in the jurisdiction that have a record in the system. This is, is it there or not? Quite related to that is how quickly did it get there? And so the next performance measure, 1B, is the percent of records that were established within two days, more than two days but within a week, more than a week but less than two weeks, more than two weeks but less than a month, and more than a month after birth. And the denominator here is the number of live births that have a record, so that’s the numerator from the previous measure. And the numerator is the number that have a record established within a given time period.
Then the next indicator is how many of these records have data on all four, program elements? And again we’re talking now only about these four, program elements because those were the group we started with. As things expand, clearly you would add additional program elements. And the calculation for that will be the same using the denominator, is the same as the numerator in the first one, that is how many live births that occurred in the jurisdiction have had a record established in a specified time interval. And the numerator is the number of records for resident children who are in the denominator. And here you get into the problem of some kids who are born in a state live in another state. And some kids who live in this state were born in another state. So we have here a slight change in the numerator. So, it’s the number that, have all four, program elements.
Then the question is how fast did the information get into the system? And here, although it seems like quite a ways out, we established that you had to have information within 90 days of birth. And the denominator here is, the numerator from the preceding one, that is the number of records that have all this information and the numerator is the number that include these data within 90 days of birth. So this is information on all four, program elements. Now lets look at individual program elements. How many of these records have immunization information available? And here the denominator is the same as the denominator in 2A, which is the number that have all four, all four elements.
And the numerator is the number of children who have immunizations other than a first dose of Hepatitis B given in the hospital. So because that is typically part of the initial establishment of the record, it doesn’t show that the record has had something added to it. So, it gets complicated and we spent a lot of time talking about numerators and denominators. And then the next question is okay, so it’s in the system, how fast did it get there? And we picked 30 days of administration. It would be nice to have it faster but it turns out that some places do batch entries or batch submission of information and 30 days seemed like a reasonable period. And the calculation here is even more complicated. It’s not based on children. It’s based on immunization events. So, because every child is supposed to getting, a whole bunch of immunizations. And you want to know that each of those is getting in, in a timely way. So the denominator is all immunization events recorded in a specified time interval and the numerator is the number of those that were recorded in the system within 30 days of administration. The next indicator is in the Integrated Information System, how many of the records have newborn dried blood spot screening information available. And here we have the denominator is the total number of live births occurring in the jurisdiction that have a record in the system and the numerator is the number of those that have--