Burton on the VSD

One of the primary topics to be discussed at this hearing is access to the Vaccine Safety Datalink. To help fill scientific gaps, the CDC formed partnerships with eight large health maintenance organizations, through an agreement with the American Association of Health Plans, to continually evaluate vaccine safety. This project is known as the Vaccine Safety Datalink or VSD and includes medical records on millions of children and adults.

Up until this year, access to data from the VSD has been limited to researchers affiliated with the CDC and a few of their handpicked friends. This "good old boy's network" practice has predictably led to questions about the objectivity of the research and the fairness of the results.

The VSD data should be made available to all legitimate scientific researchers so that independent studies can be conducted and results verified. This database contains a wealth of data involving millions of patients over a 10-year period. If properly utilized, it can help researchers study vitally important questions about the safety of vaccines, the effects of mercury-based preservatives in childhood vaccines and many other questions.

The committee first raised this issue with the CDC two years ago. For two years, the CDC delayed. Six months ago, we were informed that the CDC was developing a plan to expand access to the database. Finally, in February of this year, after a great deal of prompting from the committee, Dr. Robert Chen, Chief of Vaccine Safety and Development at the National Immunization Program, informed our committee staff that the CDC had finalized its plan and that it was poised to put it into effect.

Under this plan, any legitimate scientist could submit a proposal to the CDC to conduct research using VSD data and access to the data would be provided along with some scientific or with some basic safeguards.

In preparation for today's hearing, committee staff asked the CDC why the plan described to us in February had not been put into effect. And the staff was informed that the plan had been put into effect. However, there had been no public announcement. They put it into effect, but they didn't tell anybody.

How are researchers supposed to know about the availability of the data if there is no announcement? It took two years of prodding by this committee to get the CDC to open up access to the database. For four months, it appears that the CDC didn't inform anybody but this committee of the data's availability.

That doesn't make it appear that the CDC is making a good faith effort to open up this database. It looks to me like the CDC is trying to do the bare minimum that they have to do to get us off their backs. And that's not acceptable.

That's why I insisted that Dr. Chen be here today. I just wanted to ask him why they didn't tell anybody about the database being available. I'd like to know how he expects researchers to use this data if nobody tells them it's available.

Dr. Roger Bernier is here from the CDC to testify about these issues. He is accompanied by both Dr. Chen, the creator of the VSD Project and Dr. Frank DeStefano, the CDC official who is also a co-author of the MMR-IBD study. They are here to address our questions on the VSD project and the vaccine-autism research. The CDC employees are accompanied by Dr. Stephen Foote from the National Institutes of Health and Dr. William Egan of the FDA.

As representatives of the people, we have a responsibility to ensure that our public health officials are adequately and honestly addressing this epidemic and its possible links to vaccine injury.

BURTON: And now you're staring to look into it. And I want to tell you, we appreciate that. And I'm sorry it took so much prodding to get it started.  We were talking about the Vaccine Safety Datalink. For two years now, we've tried to get that information so that other doctors and scientists who are not connected to our health agencies who have credentials could start using that information to do studies on their own. We were told in -- when was it? January or February? In January or February that that was going to be made public. Before this hearing, we asked why it had not yet been made available to responsible people in the scientific community. And we were told, "Oh, it has been made available."

I didn't know it. Did you make any kind of report to the public? Did you announce this in a press release or anything?

CHEN: Off-mike.

BURTON: We can't hear you. Can you pull the microphone closer, Dr. Chen? Have you got it turned on? There you go.

CHEN: I think several members of the audience were present at the meeting. And we discussed several issues. The VSD Project is a very important and unusual project that contains about 7.5 million persons in the United States and their personal medical records.

And so with all the public concern in terms of data privacy, it is very important to work out a process in which we could balance the privacy of these individuals' medical records on the one hand, as well as secondly, the desire for us to be able to look at independent data.

Now the two years that it has taken us to develop a process, in fact I think we, number one, when we first approached the HMOs, there were severe concerns on all of them that, in fact, they would not agree to this and that they would, in fact, withdraw from the project. And so we've had to take the time to work out a compromise in which they would still be willing to participate in this partnership with the government with our ability to look at data safety issues, as well as meet the needs of the HMOs, in terms of protecting their privacy.

So I think that's the question in terms of why it has taken time. And so we have, in fact, come from where each of the HMOs, not only the principal investigators, but also their governing bodies, were opposed to this idea. And we've kind of gone through each of them and, in fact, convinced them to come around to the other way, to this research data center.

So that's what has taken considerable amount of time and convincing.  Now in terms . . .

BURTON: Let me pursue this. So in February, you had a meeting and other CDC employees were involved, with committee staff. And they discussed the release of the Vaccine Safety Datalink raw data to researchers. That was in February. And at that meeting, CDC provided a draft proposal. Have we got that exhibit? Exhibit number one is -- can we put that up on the screen?

Well, never mind. It's in your file there, exhibit number one for researchers to access to the VSD data. At that time, the staff was told that the project was ready to go.

Isn't that what we were told? It was ready to go at that time in February?

CHEN: That's correct.

BURTON: And we did not receive, up to this meeting today, a press release or an advertisement in any medical journal was seen or on any CDC website regarding this new program. Now if you're going to make an announcement, how do you propose to let anybody know unless you tell us?

CHEN: As I mentioned at the meeting to the people that were present, that this in fact is the first time that we have tried to develop this mechanism with the National Center for Health Statistics. It's a pilot project, using their research data center, which historically have not put this type of personal medical records available for public use. It's really for kind of national health interview survey, where people over the telephone are willing to answer those type of questions about their health status. Those are the type of things that are put on there.

So this is in fact a pilot process. And so until we work out all the potential concerns through the first couple of test projects, if you will, it is our sense that it will be premature to widely advertise it.

BURTON: But I think, with the quantum leaps that we've seen in technology, there's not any real risk if you don't want the researchers from the outside to know who the individuals are that is on the data.

You can do that. You can protect the privacy of those individuals. You can make sure that there is no public announcement about that.

CHEN: Unfortunately, that turns out not to be really feasible in this database. If you could imagine that for any vaccine safety study, you need several parameters that are key to be able to make the analysis.  You need to know the date of birth of that individual. You need to know the date of vaccination of that person and any medical visits and whatdiagnoses they had.

So you need all of those three elements in order to be able to do your analysis. And it turns out that with the key variable on date of birth-- so for example, this was actually one of the major concerns expressed by one of HMOs in Colorado, the principal investigators, his daughter in fact recently had a sprained ankle.

So he kind of posed hypothetically to his analyst that if you attended a birthday party and knew my daughter's date of birth and you also happened to find out that the child had a sprained ankle the previous day or the previous week, can you find this child? And in fact, he was very easily able to find that.

BURTON: I see where you're going. We're talking about how many people? Six million?

CHEN: 7.5 million.

BURTON: 7.5 million. And you're concerned because there is a sprained ankle and somebody goes to a party that they might be able to tell by using the birth date who this person was.

BERNIER: Mr. Chairman, may I interject if I may? I want to put on the record very clearly that CDC does support sharing information and trying to work transparently, which I think is where you've been trying to get us to go. So let me make clear . . .

BURTON: What I'm trying to find out right now is why, when we were told in February they were going to release this -- you know, every day is important to people who are going through these problems. My grandson, my granddaughter, all these people out here who have kids who are autistic, the people whose kids are becoming autistic today. Every day is important to them.

And when we're told in February we're going to get information and the information -- here we are, at the end of June, and we haven't received it. And yet, it was supposed -- we've been told that, "Oh yeah, it was made public a long time ago." But nobody knew it. That's important. And that's what I'm trying to get at here. Why if you made a decision, why didn't you tell us? Why didn't we know about it? Why didn't all these people in the scientific community that wanted to get started on this, why weren't they told about it?

BERNIER: Well, first of all, we have been trying to strike the right balance between the interests of all of the concerned parties, so that's part of the reason. The other thing is that this is new for us. We're not interested in highly publicizing something where it is a pilot type of project.

When we can iron out the wrinkles, we potentially will be in a position to make this more available. So part of this is that this is a new pilot project and there have been efforts to try, as Dr. Chen alluded to, we have to try to protect the cooperation of the HMOs. We have the proprietary interest of the HMOs and the privacy rights of the patients.

So we are trying to strike a balance. And we are trying to make this work as smoothly as possible. We don't know all of the issues that we will confront when we do bring in these researchers to reanalyze some of the studies that we have done.

So we are trying to move cautiously so that we can do so. But we will get to where you're going for people who want to reanalyze studies that CDC has done in the VSD.

BURTON: I have more questions. But I'll yield to my colleagues. As I said before, as I yield to Dr. Weldon, we all want you to be cautious.

We don't want to make mistakes. We all support vaccinations that's done in a responsible way because it has protected the health of this country.

But you've got people every day that are starting to suffer. There's huge quantities of people who have children now that are suffering from these diseases.

And the quicker we move, the better. And the more people that get involved in the research, the better. So having outside responsible scientists having this data so they can get started on it quickly is very, very important.