Questions and Answers

Click on a question to see the answer given at the
VINCENT video conference on June 12, 1997.

General questions

Data questions

Policy questions

Violence prevention questions

Injury prevention is not a new field -- why the emphasis on it now?

Dr. Runyan: Injury prevention has certainly been a problem for a long time, but it's just been in the last decade or two that people have recognized injury as a public health problem. I think it was all too easy before that to see injury as a problem in a variety of different ways and assume that someone else was going to take care of the problem. I think in recent years, because of the magnitude of the problem and the numbers involved, we have realized that this really requires and deserves public health attention.

What role do you see for students of injury prevention, and what is the best way that they can make an impact on injury as a public health problem?

Dr. Runyan: Students can work within thier institutions to address some of the injury problems that surround them. They can volunteer and work with various kinds of organizations to help develop programs. Perhaps most importantly, they can make a committment to continue with this field once they leave school and become employed.

Dr. Rosenberg: I agree with Dr. Runyan. I believe that an interest in and commitment to the field of injury prevention can yield a much bigger return than many other fields. Injury control is a new field with many opportunities to explore areas that no one else has investigated. The chance to make change in this field is tremendous.

If injury is both a big problem and a preventable one, why haven't we been able to do more about it before now?

Dr. Rosenberg: I think the biggest reason why people haven't taken it seriously is that people view injuries as just a fact of life and see accidents as things that just happen. People accepted injuries as a cost of living in a modern day world: kids getting killed by cars, babies falling out of windows if they built high rises, people getting shot if they lived in a modern day city. People did not understand that these events were preventable. But this is a public health problem -- one that we can solve and one that we can change. I think we need to totally transform the way we think about this. We need to let people know that the way they have been thinking about injuries is not really the best way.

Does the National Center for Injury Prevention and Control coordinate with the Office of Minority Health in order to examine the injury prevention efforts and any data collection on injury that disproportionately impacts people of color?

Dr. Rosenberg: Yes, we work very closely with offices of minority health at the level of our department at Health and Human Services and the level of CDC. So there is a lot of coordination that goes on. There probably are not enough resources to respond adequately, given the magnitude of some of these problems, so I think we could all do better if we start to pool these resources and work together.

Ms. Mallonee: I might also mention that most state injury programs also work very closely with the state office of minority health. It's very important.

How does the CDC support injury control and violence prevention activities at the local level?

Dr. Rosenberg: The CDC works in a number of ways to get to the local level. First we try to provide information and approaches that people can apply through programs such as VINCENT. We also work through state and local health departments trying to support them in their efforts to direct local resources. We try to be a resource to answer questions, to collect data, and to continue to try and get resources for local efforts that can make a difference. We also work with organizations outside of state and local government. We work with organizations such as Safe Kids, American Academy of Pediatrics, and the National Safety Council. We try to pull these various organizations together so that we can collaboratively provide the kind of support that you need to get started at the local level.

Ms. Mallonee: At the State Health Departments, our total role is to coordinate and assist local efforts at preventing injury. We do this through trainings, technical assistance, and sometimes with funding.

Dr. Rosenberg: I think, as Ms. Mallonee has stated, that if changes are going to be made, they're going to happen through interventions at the local level. That's where it's all got to happen.

Why aren't health insurance companies actively involved in injury prevention? Wouldn't it be cost effective for them?

Dr. Rosenberg: Some Health Care/Managed Care organizations that are getting involved in this area. For example, there are some in Minnesota that are looking at the value of bicycle helmet programs for two reasons. The first reason is that it pays. They might be able to reduce the number of very costly bicycle related head injuries. These types of injuries can result in lifetime disabilities.

The second reason is that they find that it adds value to their services and their programs. It lets their members and potential members know that they care not only about taking care of illness when it occurs, but also about keeping your family well and keeping your children healthy. I think one barrier to getting managed care organizations involved is that there are a number of interventions for which we don't have good cost effectiveness data. Also, people move around frequently from one managed care plan to another so that sometimes it may not pay for a managed care organization to institute a large scale program. They might figure that their clients will just move to another place. On the other hand, if organizations can get together within an area and decide that it pays to do this all together, it will ultimately save money and injuries for everyone.

How do you build the best and most effective media partnerships possible?

Dr. Rosenberg: There are a number of ways we can work better with media, and there have been some striking examples of effective programs. In Boston there was a collaboration between local media and the Harvard Center for Health Communications at the Harvard School of Public Health. They decided that they wanted to provide better coverage of stories on domestic violence. There are a lot of news stations and in particular, local news stations, that carry only injury stories -- shootings, muggings, rapes, fires and drownings. The general rule of the news media is, "if it bleeds it leads." If you have something that is very violent and gory, it often gets top billing, but it gets a very short billing.

What media outlets failed to realize was that in the area of domestic violence there were stories that a man beat his wife or that a man murdered his girlfriend, but there was nothing behind it. They decided that they would try to work with the media to educate the writers and reporters on how to better cover stories. Instead of just giving the gory details they could go back over the history and talk about the factors that led up to the story, to make it clear that there were places where the public could have changed the outcome, where this could have been prevented, and how the resources in the community might be organized to operate to improve the outcome in the future. As a result, the average length of a story on domestic violence, when it was covered in Boston in the newspapers, changed from about 3 inches to about 13 inches. It really changed the expectations of the reader and it changed the ways they covered these stories. I would say that for almost any type of injury, we could benefit by getting the media to also cover why something happened, what lead into it, and whether we could change it. We talked before about what is keeping people from addressing injuries as a public health problem. Part of it is the perception that injuries are just accidents. Here's a chance for the media to get across a different notion -- the notion that injuries are preventable. It could make a big difference!

Do you have suggestions for encouraging collaboration and information sharing between agencies at the local level?

Ms. Yuwiler: I recommend bringing together representatives from the state health department, the law enforcement community, the mental health community, education, the public health community. I'd bring them together in a room and say, "Let's look at the data. What are our problems and what can we do about them?" By bringing them together you can begin to start that dialogue.

What do you do about the territorial qualities of competing agencies?

Ms. Yuwiler: I think you need to respect what each agency is doing. You need to listen to what programs they already have and what area of expertise they have. The attempt should not be to take away from each of them, but to enhance what they are doing. For example, the department of education may have some programs that other agencies can help with by providing data, information or personnel. You should look for ways to share resources. You have to start small. For example, start by just getting a sense of who the players are that should be involved. As you begin to work together, people will begin to build that trust and that sense of, "Hey you're not out to grab all of my funding and my program." You really want to operate from an air of respect.

Dr. Runyan: Another idea that I've tried is to work through Haddon's matrix together. This method allows you to identify a bunch of different intervention strategies and to demonstrate that there's plenty for everyone.

Dr. Rosenberg: I just wanted to add that institutional egotism is a big barrier and sometimes it's very hard to crack. Sometimes you can appeal to the fact that there are victims, perhaps children, who are dying and who are being injured. Sometimes, it is necessary to answer the question "What's in it for me?" If you can structure it so that there are rewards for all parties involved, this can help you overcome inter-agency barriers.

What is the difference between the medical model and the community model, particularly with respect to risk factors and resiliency factors?

Dr. Rosenberg: There are risk factors and reliance factors not only at the individual level but also at the community level. If there's a certain type of problem occurring in a community, the questions are: what are the risk factors that the community can modify and what are the resiliency factors that the community can strengthen. I think there's a very big difference between the medical model and the public health model here. I think the medical model tends to deal with individual patients and the diseases that they get. The two important differences with the public health model are that public health looks at community needs, future needs, and prevention for the whole community in an interdisciplinary approach. It does not just draw on the medical community, but it draws on law enforcement officers, on social services, on educators, on people who are in charge of public housing and on first responders. All of them come together in the public health approach. I think it's much more comprehensive. It includes the medical approach, but is certainly not defined by it.

What is the difference between the terms "prevention" and "control?"

Mr. Teret: People in the field often use these terms interchangeably, but there is a difference. Prevention, particularly primary prevention, is seen as a subset or one part of control. If you define injury control in its broadest sense it includes not only preventing the injury from happening in the first place but also to minimize the damage that the injury does if it occurs. So, for example, if somebody is already in a crash you may not have prevented the injury that resulted in the crash. But, by Emergency Medical System picking up the person, by getting the person to a hospital quickly, and by providing the proper medical care, you may change the severity of the injury. So, control refers not only to primary prevention but also to secondary and tertiary prevention that's sometimes called - that whole range of minimizing the likelihood of the injury or the extent of the injury if it occurs.

Where can I find national E-code data for the Hispanic population?

Dr. Rosenberg: Unfortunately we don't have data broken down by ethnic group the way we would like. We are starting to collect data by ethnicity. There are some national level data available. I think if you called the National Center for Injury Prevention and Control we could give you access to these data. Even through our web site you may have access to some data broken down by ethnicity and, specifically, Hispanics. One of the underlying problems here is how we define Hispanic and how we start to collect it. A lot of data may come from surveys, so at a local level a survey may one way to collect thoese data, but you'd have to know how you want to define Hispanic. Again there are sources of these data that are available at the national level and there are data that you can collect at the local level.

We work on an Indian Reservation and our injury tracking efforts differ from the national statistics. From what data sources have you developed your data?

Dr. Rosenberg: You're right in that depending on where you look and what population you look at you'll get different pictures. I've used a number of different sources in our data. When I used the chart on the ten leading causes of death by age group that was for the US as a whole. That uses vital statistics, deaths that are reported to each state and then by each state to the national center for health statistics. I suspect that if you look at Native Americans, or other particular subgroups, that you will find patterns that differ by circumstances and by location and also by culture.

Are alcohol, cocaine opiates and other substances included under the category of poisoning in your statistics?

Dr. Rosenberg: Yes. These are generally included under poisoning. Some poisonings are unintentional drug overdoses. Some are intentional poisonings, others are unintentional poisoning by young children who get into substances. The substances you mentioned could fall into all of these categories. We're trying to get much better coverage of substance abuse related problems through poison and control networks. Ideally we'll find a way to include all of these in the category of poisoning.

How can somebody at the local level determine where to go to start finding data on their community?

Ms. Mallonee: Start by calling around they community -- call the local hospital, call the ambulance service. Call and see if there is a trauma registry. See if there is a Safe Kids Coalition. Call the coroner or the local medical examiner. They need to determine what data is available. Call the National Center for Injury Prevention and Control or use some of their documents and see what local data is available there. Call their health department or their local university. Get on the internet to see if there are rich sources of info there. If one must start from scratch, try organizing other leaders within the community to determine what data is wanted, what the purpose is, what data exists and then start going after it. It requires a lot of energy and a lot of commitment.

Most communities don't have a lot of expertise in terms of evaluation design, data collection, and data analysis. Can you give us some suggestions of where community groups can go for help with evaluation design, data collection and data analysis?

Ms. Yuwiler: For evaluation design, I would send you first to a children's safety network guidebook for evaluation of local programs. It's very straight forward. It gives you an outline and some exercises that are designed for community based programs. You might be able to go to a university, your state health department, injury prevention resource centers, children's safety network, or the National Center for Injury Control and Prevention. All of these organizations have folks that can provide technical assistance.

Dr. Rosenberg: Another thing I'd add is that increasingly people are beginning to realize that the community needs to be a partner in these evaluation processes. I think frequently we've looked at the community as providing the data for where the problem is and how to address it, and helping to mount the evaluation. Then we've gone to outside experts -- frequently academics -- to help with the design of the evaluation because we've thought that the evaluation is very difficult. It is very difficult, but we're realizing that this is a phase where the community needs to have some input. The people affected - the people within that geographical area who constitute the target area frequently have very valuable things to say about the evaluation and it makes the evaluation all that more valuable.

Despite overwhelming data supporting the effectiveness of motorcycle helmets, the state of Texas is poised to repeal its helmet law. What can we, as health care providers, do if politicians fail to recognize what is truly in the best public health interest?

Mr. Teret: If you're trying to convince legislators, it has been my experience that it's important to focus on who pays the cost of injuries to motorcyclists. I mentioned earlier that there have been cases about upholding the constitutionality of motorcycle helmet laws where the court pointed out that it is the public that pays that cost. There have also been studies done since these court decisions that have confirmed the court's understanding. There are studies that look at head injuries to motorcyclists and other injuries that show that roughly 80 - 85% of the cost for their injuries are paid by the public rather than the motorcyclists or their insurance. I think ultimately we all have to realize that changing the legal environment and changing people's behaviors for the purpose of reducing the likelihood of injuries is a long term process. It's not something that's done overnight.

During the President's symposium on Youth Crime, President Clinton ordered new warnings to be posted about the dangers of letting guns fall into the hands of underage people. Essentially, his suggestion was that all gun dealers would have to display these signs and distribute written warnings every time they sold a gun. How effective is an intervention of this kind?

Mr. Teret: I would say that it's a step in the right direction. I think, however, that although it's a step in the right direction, it's a baby step and too ineffectual a step. I say it's in the right direction because it's dealing with the product itself. It's not looking only at who's pulling the trigger but it's looking also at the trigger itself. But, it's a baby step because we can't rely only on a posted warning. We live our lives in an environment of posted warnings. There are endless stories about well meaning and even very intelligent people who have disobeyed warnings.

I can give an example from my own experience. I know that there's a rule that you never run into a building that has smoke or fire in it. One day I was home with my kids and started to smell smoke. I said to my oldest son, "Go out into the street and go down to the corner and wait there for me." I then went up to get my youngest child and ran outside. At this point I saw that the house next to ours was engulfed in flames. This was an attached house. So, after I got my kids safe at the corner with neighbors watching, I said, "I'm going in for the valuables." I ran into the house filled with black smoke. I'm in the house choking when I realize that I'm a graduate student, which means I have no valuables! I came out with the toothbrushes. I had to come out with something in order to save face. But the point is that I'm trained in injury prevention and, yet, I still behaved in this way. The point is that you can't always rely on people to react in an intelligent fashion. We need something more than a label that's sold with a gun that says, 'watch out you could get hurt by this.'

What is the political feasibility of obtaining some sort of improved gun control legislation in our country or at least in our local communities?

Mr. Teret: We did a nationwide poll of 1200 people that represent the population as a whole. We asked them many questions about gun policy. One of those questions was, "Do you want to see handguns banned?" The percent of people who want to see handguns banned is actually quite small. It's far below fifty percent of the population. So as a political realist I don't believe that that's going to happen whether you or I want it to happen or regardless of what we think it's going to do to the incidence and severity of gun injuries. So, in the absence of public support for such a ban, there are many things we can do in the absence of taking guns away from people. For example, we can let people have guns but just alter the products so that the guns are safer.

How would you effectively and positively address the issue of unintentional injury from guns?

Mr. Teret: I think that your question is a very important one. We're talking about unintentional injuries -- not injuries that are committed with the intent to harm somebody. Again, we can look back on the experience that we just had with motor vehicles. We can ask people to behave in safer manners, we can train people like we do with training people to drive. We license people to drive. Some people might suggest that there be required safety training for people who are licensed to have a gun where licensing exists.

We can also look at the product itself. For example, I'm aware of a situation in Massachusetts that happened not long ago where two twelve-year-old boys were playing with a gun. It was a semi-automatic pistol. They took the magazine out of the gun after they had racked a cartridge into the chamber. One of the boys pointed the gun at the other boy. The bullet entered on the left side of his nose and exited from the rear of his head. He was rushed to the hospital and by the time his parents got to the hospital all that was left to do was to provide the consent to harvest his organs so that other people could have the benefit of organs in his body.

That kind of tragedy doesn't have to exist because you can make a gun that has a magazine disconnect. What that means is that if the magazine is out of the gun, the clip is out of the gun, and the gun won't fire even though there's a bullet in the chamber. You can make a gun that has a load indicator. If you went out and bought a camera now the camera would tell you not only if you had film in it but also how much film you have left. The technology has existed for years to make guns that have load indicators. So you'd never again hear the common story, "I didn't know it was loaded." We can make those kinds of changes that would address the likelihood of unintentional gun injuries.

How have gun manufacturers responded to proposed policy changes?

Mr. Teret: Some manufacturers have been quite receptive to discussing the idea of 'how can they change the design of guns so that fewer people are unnecessarily injured. I know people who work at high level jobs in gun manufacturing companies. When I think of those people I don't think for one moment that those people in any way want to see injuries occur. I know that those people want to prevent unintentional injuries. Some manufacturers have been very receptive in discussing ways that we might go about changing the product itself to make the product safer.

Is there some way to retrofit guns to make them safer? Is the technology out there to do that?

Mr. Teret: Regarding retrofitting guns with the kinds of devices that we've already mentioned like magazine disconnects or load indicators, I don't really think that that is feasible. Maybe in some small examples it may be feasible to change a gun but I don't think so with regard to guns that are already out there.

A survey was done recently with the police foundation that found that there are between 190 and 200 million that are privately owned in the United States. Between 65 and 70 million of those are handguns. Many of the people who own guns own multiple guns. So it's not the fact that almost every one in the country owns a gun, rather about 25% of the people own handguns but they may own quite a number of them. So there are a lot of guns out there.

With regard to guns that are being used in crime now, when there are traces done on these guns by the bureau of Alcohol, Tobacco, and Fire Arms or other police departments, those traces show that the majority of guns that are being used in crime are about 3 1/2 years of age or newer. So it may be that the guns that are used in crime are not the guns that comprise the 190 some million that are stuck in someone's attic or basement. I don't think that there's a whole lot that we can do about the guns that are out there but I also don't know what size problem those guns pose.

We don't have good information about shelf life of guns and what's done with old guns. But when people ask me what to do about close to 200 million guns out there, it seems to me that if we want to change the design of guns, now is the time to do it. And it's not an excuse to say there are already 200 million guns out there. If somebody said, for a new school that was being built, 'Why go to all the pains to make sure that you're not putting asbestos in that school because there're already so many schools out there that have asbestos?' Nobody would accept that as a rational for not making sure that the new school is asbestos free. The fact that we have all those guns out there now is not a rationale for failing to use innovative policies that address, among other things, the design of the product itself that might ultimately reduce the incidence of the 38,500 gun deaths that occur in the United States per year.

Are the same methods and approaches used to understand and identify injury prevention as are used for violence control?

Dr. Rosenberg: The beauty of this simplified approach is that the very same process that you use to understand unintentional injuries -- such as bicycle related head injuries in kids or deaths from fires and burns -- can be applied to understanding problems of violence. For example, lots of people are paying attention to the problem of youth violence. President Clinton recently had a meeting addressing youth violence during which they celebrated the fact that there has been a 25% reduction in homicide rates in this country in the last several years. The reduction came about through the exact same process: by starting first to describe the problem. If you look at the problem of youth violence what are the injuries that result? How many deaths are there? To whom do they occur? How old are they? Under what circumstances? What's the relationship between perpetrator and victim? So step one is to use data to define the problem.

The second step is to look at the risk factors. What are the causes? What's changed? Why did the rates go up? Why was this young boy at risk? Why was this older person shot? What are the causes? Are they related to alcohol? Are they related to substance abuse? Are they related to socioeconomic status? The next step is to look at the picture of your problem. Look at the causes and ask: What might work? What do you want to try? What could make a difference? And then evaluate it. Does it work? The very same process that we applied to problems of violence is applied to problems of unintentional injury. What President Clinton was talking about with regard to youth violence is that when you start to analyze your data you see that for most youth the problem has been that violence has increased in terms of fatality. There are many more fatalities than there were 15 years ago. There's not a lot more fighting, but more frequently the fights are fatal. They occur between people who know each other, are usually of the same race, and most of these fatalities occur with a firearm, with a gun. Once you understand the problem, it suggests that you can start to move toward a solution. Solutions be things like community policing and providing alternative activities, or arresting serious offenders and letting people know you are serious. Whatever the interventions are they can be applied and directed based on your understanding of the problem.

Some traffic injuries may be intentional. Is there a way to estimate suicide hidden within traffic injury?

Dr. Rosenberg: It's a good question and one that has intrigued investigators for many years. I think the best studies suggest that there are some suicides where people intentionally crash their cars into an abutment or go off a bridge or off a road, but these studies also suggest that these incidents are rare. I would refer you to the literature that is out there, where several people have systematically attempted to say, "How could we best describe these?" Usually they look at single passenger crashes and then they take a sample of those and by going back, collecting information about the person, and reconstructing the person's life and circumstances they come up with an estimate of how large that number is. So they do occur, but it's probably a pretty small number.

When is suicide considered preventable in spite of the fact that it is intentional injury and when do you separate a suicide prevention program from a mental health issue?

Dr. Rosenberg: I think we would consider almost all suicides as preventable. I think all intentional issues are preventable, so the intentionality of it does not speak to the preventability of the problem. I think the real question is, what can we do about the suicide problem? Last week there was an article that came out about suicide in Washington and what it showed was a marked increase in suicide among the elderly. What that kind of analysis does is direct you again to a specific part of the problem and lets you focus your efforts. I think there are a lot of people who commit suicide because they are depressed. For example, when addressing elderly depressed people, you might start by identifying them, screening them and treating their depression. For young people who are impulsive you might find ways to screen and identify them and provide resources to them so when they are impulsive they don't have only a single option available to them. I think the approach to take whenlooking at suicides is to look at the patterns, separate them by age, by cause, find out the risk factors, and then find out a whole range of interventions that could be applied. There's a useful publication that you might want to write for which is focused on youth suicide. It is a resource guide for prevention that talks about strategies and prevention programs that seem to be promising. That's available from the National Center.

What kind of reliable data sources are available to identify perpetrator-victim relationships in circumstances in which violence occurs? How could this information be helpful in understanding the increase in fatalities resulting from violent situations?

Dr. Rosenberg: Information about perpetrator-victim relationships is very important. Unfortunately, our data sources on that are limited. The FBI collects information in the supplemental homicide reports. This information is collected from police departments for every fatality. There are also periodic victim-crime surveys done at the local and national levels that ask about victim-perpetrator relationships. Ideally, you would like to collect information from police reports and combine that with information from medical reports about the outcome of the injury. The next important step for all of us is to begin to link data from different sources. For example, we should begin to think about how to link data from the EMS to data from police reports. That will give us a much better view of what's happening.

In response to the other question about the change in lethality of the problem of violence, that information also comes from a variety of sources. Ms. Mallonee mentioned the Youth Risk Behavior survey earlier. What we've seen from this survey about youth fighting is that there's been an increase in fighting, but this increase has been small. But we can see from information collected from police files that there's been a very large increase in the number of homicides among young people. When we put those two pieces of information together, we can see that there's not a big increase in the number of fights that go on now, but a much higher rate of fatalities. When we look at this information, the difference seems to be that when kids fight today they use guns and that turn these fights into fatalities.

Does the procedure mentioned for identifying and analyzing data work with domestic violence. Can we normally access that much data or is there another system you suggest we use?

Ms. Mallonee: I think that is a bit more complicated than other injury problems. There may be additional data sources that weren't described in some of the handouts and some of the slides -- for example: data from womens' shelters, police department data, emergency department data. I think it is more complicated but it can be done. Often multiple sources are needed. Sometimes three or four or five sources must be accessed.

What if the agent of injury can not be modified -- if it's the human fist, for example? How do you prevent injury resulting from assault when the weapon is the human body?

Ms. Yuwiler: Look at your situation in which the fist is being used. Can you modify that situation, perhaps through conflict resolution? Can you teach people to manage their anger differently? Can you change societal norms so that it is not acceptable to hit people? You're not going to change that fist, but if you can change the circumstances and scenario around which that fist is used, you may have a chance to change what's going on.

Dr. Rosenberg: Sometimes you can modify the environment. Using Haddon's matrix might be useful. There are different types of assault. Break it down by going back to the beginning principles, collecting your data and looking for patterns. If the assault is that of a parent against a child there are a number of interventions that could be done, such as trying to relieve the stress of new parents in dealing with their children, giving them training, having people visit them at their home to see what's happening, or having more frequent examinations of the child.

If these assaults are occuring in the context of youth violence then separating kids may be helpful. The same thing is true with partners. If it's a domestic violence assault, sometimes what you need to do to prevent it is to physically separate the people. Sometimes giving a woman a safe place to be in a shelter may prevent some of that assault in the long run. There are a number of things that can be done to prevent assaults based on what's behind it, who's involved, and what the circumstances and reasons are behind it.

Table of Contents · Q&A · Workbook · Resources · Exit
Questions or comments: