Media Home | Contact Us
CDC News Conference Transcript
Findings of Task Force on Community Preventive Services: Effectiveness of Strategies for Preventing Violence
October 2, 2003
DR. BINDER: I'm Dr. Sue Binder. I am the director of CDC's Injury Center. This is the center at CDC that focuses on unintentional injuries, like injuries from motor vehicle crashes, fires or drowning, and violent injuries, homicides and suicides, as well as the nonfatal consequences of violence.
Also with me today on the phone is Dr. Jonathan Fielding. He's the chair of the Task Force for Community Preventive Services, and he's also the director of the Los Angeles County health department, a professor at Drew University, and he has a list of responsibilities and prestigious awards that is way too long for me to go into today.
Violence is a terrible and tragic public health problem. Suicide and homicide are among the top four killers of children and young adults in America. The good news is that we can prevent violence. We have interventions that are proven to work.
Today, we're here today to discuss findings in the Task Force for Community Preventive Services that are being released in the MMWR Report and Recommendations.
The task force is chaired by Dr. Fielding, who you'll hear from in a few minutes.
They use a rigorous process to synthesize the scientific literature about key public health issues. The goal of their work is to identify which interventions work from a public health perspective, according to very strict criteria.
Today's reports are part of a series that will be published about the problem of violence (in MMWR).
The first report being released examines one particular intervention for reducing child maltreatment, which is home visitation programs, and the second examines firearms laws.
While the evaluation of the firearms laws concluded that there's insufficient evidence to determine whether they affect violence, the report on home visitations found that these programs, when they're conducted by professionals, may prevent up to 40 percent of child maltreatment among high-risk families.
This is a very important finding. Every day our hearts go out to the victims of child abuse and neglect. We see them in our newspapers, you hear about them on the news media, we may even know some of them personally.
No child should suffer at the hands of an adult. Around 900,000 children are abused and neglected each year, and about three or four of them will die every day. We can prevent some of this from happening. That's the key message here.
There are other societal impacts of child maltreatment in addition to the very unfortunate and immediate effects on the child.
Children who are maltreated have a higher risk of growing up to become victims of violence or perpetrators of violence. They have riskier behaviors. They develop more chronic diseases. They use more health care resources.
Thus, abused children suffer and we, as a society, pay for child maltreatment, for years and years after it happens, and we perpetuate the cycle of violence. We can prevent child abuse and prevent violence. That I think is the exciting take-home message of the work being released today, and I'd like to now turn this over to Dr. Fielding for additional remarks.
DR. FIELDING: To reiterate, the Community Preventive Services Task Force is an independent nonfederal panel of experts supported by CDC, and we conduct systematic reviews of published studies to identify and report on evidence of effectiveness.
Adding to Dr. Binder's comments about what strategies can prevent child maltreatment, we found that early childhood home visitation programs are indeed effective in reducing instances of child maltreatment among high-risk families, and by home visiting programs, we mean those that include visits to parents and children in their home by trained personnel.
They provide information, they provide training, and they provide support on issues of child development and health and health care.
Strikingly, there was a 40 percent reduction in child maltreatment episodes in the reviewed studies.
In general, the greater improvements were found in programs using professional visitors and those that lasted at least two years.
On the basis of this very strong evidence of effectiveness, the task force recommends that these programs be implemented, or if they're already in place, that they be continued.
And let me add as a pediatrician, it's very heartening to be able to recommend programs that can make such a big difference in the health and safety of our most precious assets, our children.
I also note that some of the best-designed studies suggested other positive impact on children and families. Improved school performance, reduced substance abuse and less contact with the juvenile justice or criminal justice systems.
We are also today releasing some of the results of our efforts to scientifically examine evidence on what can reduce the burden of injury.
The types of interventions we are assessing overall cover a wide-range, from therapeutic foster care for violent children to community policing, to development of pro-social skills in school.
One part of that inquiry was to systematically assess possible effects of legal strategies in preventing death and injury from firearms, including bans on specified firearms or ammunition, restrictions, waiting periods for acquiring firearms, firearm registration, licensing firearm owners, shall-issue concealed weapons carry laws, child access prevention laws, zero tolerance laws for firearms in schools, and combination of firearm laws.
Evidence regarding the effectiveness of these laws was carefully reviewed and found to be insufficient to discern any conclusions due to inappropriate numbers of studies, unreliable data, inappropriate analyses, and inconsistent result.
The task force therefore recommends that high-quality studies of these and other firearm laws be undertaken, along with additional surveillance of violent injury.
I want to point out that, when we conclude that there's insufficient evidence of effectiveness of a given intervention, in this case firearm laws, with respect to a desired outcome, such as violent injury this means we simply do not know what effects, if any, the law has on that outcome.
We don't mean that the law has no effect on the outcome, which is why it's very important and we recommend additional high-quality research.
With that, let me turn this back to Dr. Binder.
DR. BINDER: Okay. Well, we're ready to take questions, and we'll take our first question from the phone lines. We'll take the first question from the phone, from Associated Press.
QUESTION: Yes, ma'am. This is Kristin Wyatt calling from Atlanta. My question is about the second half, about the firearm studies. Can you tell us a little bit about where these studies came from. Are they just a survey of--you said 51 different studies.
Were these CDC studies, or where do they come from?
DR. BINDER: Dr. Fielding, can you answer that one, please?
DR. FIELDING: We systematically surveyed the published literature, all peer-reviewed studies that appear in established market economies during a defined period of time, which I believe in this case was over a 20-plus year period, ending in 2002, some time in 2002. We then rigorously go through these and look for the qualities of studies, the execution of studies, the kinds of outcomes they've looked at; we abstract them, we then look to make sure that we're talking about independence of studies, that they're not looking at in fact the same thing using the same kind of data, and then based on all these, we look at that body of evidence and decide is the evidence sufficient or insufficient, and if so, what kind of recommendations might we make based on those.
QUESTION: And if I could have just one follow-up. The report calls for more study. My question to you is the CDC saying that they want more funding for study of firearms? I'm referring to kind of the argument the CDC had with the NRA in the 1990s about whether the CDC should even be studying gun violence or should limit itself to disease.
If anybody could speak to that. Is this calling for more funding for that kind of stuff, or are you saying, well, we don't know, so there's no conclusion to be had, I guess?
DR. BINDER: I'll answer that question. This is Sue Binder.
QUESTION: Yes, ma'am.
DR. BINDER: The task force makes its recommendations, they're independent from us, and they are not sending their messages directly, their recommendations directly to CDC. We're hopeful that these are useful to a broader community.
We, at CDC, have spent a lot of time thinking about what the research priorities are for us related to violence. We had a long process to develop a research agenda related to violence and have identified many areas where we think there needs to be resources placed in order to move very, very promising interventions forward, so that they can ultimately be used to reach people who can benefit from them.
And that research agenda, which was developed really independently of this effort, is available on the Web and is available from us and our highest priority, with our relatively limited resources, is to implement our research agenda.
So I think that the recommendations are potentially useful to researchers around the nation.
Our priorities are public and are available from us.
Do we have a question from the room?
QUESTION: Hi. David Wahlberg from the Atlanta Journal-Constitution. I guess I'll sort of ask the same question but a different way.
How might more research on firearms laws continue given the restrictions the CDC has on not funding research that promotes gun control?
DR. BINDER: Well, the language, what Mr. Wahlberg is referring to is language that we get in our appropriations, and the language does actually say that CDC has responsibilities related to data collection and dissemination of information and research in this area that would be objective and with grants awarded to an impartial scientific peer review process.
We are prohibited from using any of our funds to advocate or promote gun control. So I think that the restrictions are really about advocacy for laws or policies.
QUESTION: A quick follow-up. Do you know how many of the 51 studies were funded by the CDC?
DR. BINDER: Dr. Fielding, can you answer that one, please?
DR. FIELDING: I don't know how many studies. When we look at these studies, we don't look specifically or track who funded, and we simply look where they were published in the peer-reviewed literature.
I wonder, Dr. Binder, if you have any information on the number of studies that might have been included which were funded partially or fully by CDC?
DR. BINDER: Yeah. I think that some of them were, and I'm a little hesitant to give you an exact number but I would guess that it is, you know, well less than 20 or 25 percent and we are very happy to provide you that information separately, if you'd like.
We'll take another question from the room.
MODERATOR: Cory Thompson from Fox 5 News in Atlanta. You said that you haven't come to any conclusions as far as its effectiveness but what conclusions have you come to by looking at all of these studies?
DR. BINDER: You're talking about the firearms legislation studies.
DR. FIELDING: As I indicated, what we found was that the evidence, which we carefully reviewed, was insufficient to really discern any conclusions, and what we recommend is that there be additional high-quality studies so that we can come to a clear understanding of what various interventions can or can't do. I don't think we can take it any further than that.
When we conclude there's insufficient evidence of effectiveness, for example, of a given firearm law, with respect to an outcome such as violent injury, this means we simply don't know what effects, if any, it has on the outcome.
We don't mean it has no effect, and that's why it's important to do more studies.
DR. BINDER: We'll take a question from Health Day on the phone.
MODERATOR: Thank you, ma'am. Sandy Gardner, your line is now open.
QUESTION: Thank you for taking the question. I'm curious about data collection for injuries and deaths relating to firearms, and how we collect the data, and is there any initiative to make that more comprehensive and centralized in how that might affect the research?
DR. BINDER: Thank you for that question. That's actually a very exciting question to us.
Most of the data that we use about deaths from firearms comes from vital statistics. It comes from the death certificate and the national collection and reporting of death certificate data about all causes of death.
Recently, CDC, working with many partners, has been developing a system that we're calling (the) National Violence Death Reporting System.
So if you think about what happens when somebody dies, let's say from a violent death, they come to an emergency department, for example, let's make up a hypothetical person--they might come to an emergency department and if they die there, then a physician, will fill out the death certificate, which is usually the information we have.
Well, if the person was shot, there also was probably a police report done. They probably also will have a medical examiner or coroner evaluation and that's a separate information set that's sitting in the medical examiner or coroner's office, and usually we aren't able to connect that data.
The concept behind national violent death reporting is let's put this data together, so that we can get the information from a police report. For example, was the man maybe involved in some other kind of activity that was of concern?
Was there another crime involved, prior to his death? Take the information from the medical examiner report. Was there alcohol involved? Were there drugs involved? And the information from the death certificate. Put it all together to answer questions like the role of alcohol and drugs in certain kinds of violent deaths. The percentage of times that a homicide is followed by a suicide. Things like this ultimately we hope will lead to prevention. So that's a very exciting effort that is really pretty recent, building on work that has been done, funded by foundations in the past.
We think that the information will be helpful at the national level.
We also think it's going to be extremely helpful at the local level, to help states and cities understand what their problems are really all about.
We'll take another question from the room.
QUESTION: Another firearm question, though I have one about home visits, too, if there's time.
I know that the reasons why the 51 studies aren't enough evidence are somewhat complex, but on the surface, having 51 studies not being sufficient seems a bit odd. So I'm just wondering how many studies would we need to have conclusive evidence about firearm laws?
DR. BINDER: Dr. Fielding?
DR. FIELDING: It's not so much an issue of the number of studies. It's how the studies are designed. It's how the studies were conducted. It's the independence of the studies and it's the kind of results they find.
So if you have--we have published criteria for how we assess the strength of a body of evidence on the effectiveness of population-based interventions and it depends on the quality of execution, the design, the number of studies.
In some cases, a single study may be sufficient to talk about impact and to feel that there's a coherent body of evidence.
In other cases, it may take as few as two or as many as five, or it could take a much larger number, if in fact you have inconsistent results. So there's no--we have a translation table, if you will, but it includes a lot of variables, including the effect size. If you have a very, very strong effect size with a good design and good execution, then it really doesn't take very many. It may take one or two depending upon other factors.
But in this case, what we have is, first of all, we have a lot of different kinds of interventions, waiting periods of firearm registration, shall issue laws, child access prevention, zero tolerance laws.
So it's not all about one type of law, and we find in a number of cases that there not--are just question of number of studies but there are methodological problems and there are inconsistent results.
DR. BINDER: We'll take a call from NPR on the phone line.
MODERATOR: Thank you. Ted Vigosky, you may ask your question.
QUESTION: Thank you. Dr. Binder, let's move back over to home visitation a little bit because that's something you have a lot of concrete results, and I want, if you could, to talk about the stakeholders or partners.
I understand in Metro Atlanta, for instance, Grady Hospital has a program and there's another one for Hispanic people, and one in Cobb County for indigent children.
How do you help the stakeholders or partners in the home visitation area do what they do better?
DR. BINDER: That's a great question.
First of all, I'd like to make a distinction between what was found here and what we usually think of when we think about addressing problems of child maltreatment, because I think there's a lot of focus these days, and appropriately, on children who are in homes, who are already being abused, and how they fare in the long run.
We're talking here about primary prevention. We're actually preventing that pattern of abuse from starting in families with the home visitation program which is very, very exciting.
Now the studies that were done I think are really excellent but I think one of the key things that has to happen now--not only do we need to get this information out, which we're really pleased you're all helping us to do, we also have some need for further research.
We need to know how to deliver these services in the most cost-effective way and we need to know how to deliver them in different contexts, in different kinds of communities.
You mentioned Hispanic communities. The studies that were done are very good studies but they're in much focused populations and they've been done in very rigorous ways.
We need to make sure we know how to get these out under the circumstances that agencies and hospitals and community organizations, and other organizations that serve these families need to operate under, that they still work in these situations, and this is one area where we do want to invest, to try to make these interventions available and still be effective to all those groups that are serving families.
Are there any other questions? Did you have another one? Well, I'll take another question from the room.
QUESTION: Regarding home visits, do we know how many home visits are going on now, and sort of what percentage of the potential, you know, market is being met right now?
And also could you clarify the cost-benefit figures that are talked about in the report. It mentioned $3,000 and $350 and I'm not quite sure what the overall cost benefit is.
DR. BINDER: Well, I'll take the first part of the question and then hand it over to Dr. Fielding.
My sense, and we can probably try to find out more for you, and I'll ask John Lutzker if he has any additional information, is that home visitation is not a widespread practice, so we really think very few children are receiving--or families, I really should say families at risk are receiving this service.
Was there another part to that question, cause I'll ask Dr. Fielding to talk about the cost benefit, measuring the [inaudible]. Okay. And then Dr. Fielding, if you want to take the second part of the question, please.
DR. FIELDING: Yes. One of the things we try and do in every one of our studies to look at the evidence is to understand if there is economic data to help us look at cost benefit, cost effectiveness, and in too few cases do we find it. But in one of these cases there was information. There was the careful study of a nurse home-visiting program.
It was based on the costs incurred and paid for by the government. So that's one important caveat.
But in this study sample, the costs exceeded the benefits directly attributable to reduced child maltreatment services by about $3,000 per family. The program cost estimates, which are largely dependent on the frequency of visits and how long, and whether you have nurses or paraprofessionals, in this case it was nurses, ranged from about $958 to $8,000 per family.
So, again, it's quite a variation. But in looking at low-income mothers, the one study we looked at carefully, it showed a net benefit of $350 per family based on 1997 dollars.
That is, when you took the costs and the benefits, the benefits, in terms of government payments, outweighed the cost.
I would just add a couple of points to what Dr. Binder has said in terms of what we know. There is a very small percentage of children who could benefit, are being reached by these programs, but some of the findings we had I think are helpful in that the effect size of programs delivered by nurses and also by mental health workers was very substantial, both in the high forties percent range.
For paraprofessional visitors, effects were mixed and the median effect size was less than 20 percent. It was still important but it was significantly less.
Also in programs using paraprofessionals, we found that beneficial effects were consistently evident only when the programs were carried out for greater than two years.
So this gives some information that can help policy makers, decision makers at the local level decide what kind of program they're interested in and to look at the cost for them and the expected benefits.
And I would point out that we were narrow in this because we were looking at outcomes related to violence.
There are a lot of other outcomes that I suggested, that are very possible in these studies, and they suggested things like improved education of mothers, less dropout, less contact of mothers with the criminal justice system, even better outcomes of the children who are the subject of these home visiting programs through adolescence.
DR. BINDER: Are there any more questions?
Well, I'd like to just end with a short statement, because it really strikes me that it wasn't that long ago, and we still get questions today, “Is violence a public health problem?” and our answer to that is to look at the causes of death, and point out that, you know, there are about 30,000, almost 30,000 people who die every year from suicide and there are about 17,000 who die from homicide, and there are lots and lots of people who suffer or may have even long-term disabilities as a result of violence.
But I think what this report shows, and also some other data that have not been through this particular process but are really very promising show, is that public health approaches can be extremely helpful to violence prevention.
We have today the very strong evidence that home visitation programs work.
We have other kinds of programs for child maltreatment that we also have good effectiveness data on, that we are not evaluating through this process. We have data on preventing youth violence. We have data on preventing dating violence. We're beginning to understand how to prevent violence and I think that is incredibly, incredibly powerful and very, very exciting.
So thank you all for being here and if you have any follow-up questions, we'll be glad to answer them later on.
MODERATOR: This does conclude today's conference call. We thank you for your participation.
This page last updated October 2, 2003
United States Department of Health and Human Services