Transcripts of the Attorney General's Initiative on DNA Laboratory Backlogs (AGID-LAB) Working Group

Tuesday, October 22, 2002

NIJ-SUPPORTED RESEARCH

MR. SCHMITT: We're delighted today and very lucky to have Tim Schellberg and Lisa Hurst from the Smith Alling Lane law firm here in Washington to discuss the status of their research projects. As we discussed yesterday, Sarah has funded research projects with them on kind of the extent of the backlog here in the United States and some related issues, and I'll let them discuss those issues. So while you continue to eat we'll turn it over to Tim and Lisa.

MR. SCHELLBERG: Good afternoon. Again, Tim Schellberg with Smith Alling Lane and also Lisa Hurst from Smith Alling Lane. For those of you that have sat through many of our various presentations over the last few years, you will be happy to know that we're not going to be focusing on redundant data about the status of all the legislation around the country. We're going to merely talk about the projects that we have been awarded for NIJ regarding the assessment of issues that will help facilitate everything you have been working on.

The format is I'm going to explain briefly the United Kingdom project that was recently awarded to us just about a week ago. Then I'm going to go back to the main project that we are fully engaged in at the moment, which is the assessment here in the United States, and remind you how we got here, some of the issues that we're looking at, and then turn it over to Lisa that's going to walk you through the process, for example, how they developed the questions and the time lines.

To kind of start, to remind you - first of all, actually I'll start with the United Kingdom. The project, like I mentioned, was awarded to us about a week ago, and the goals of this project - and it's going to be conducted by Chris Asplen, who is an employee of Smith Alling Lane that resides in London - he is going to basically get a working document of how the United Kingdom's system works, what is their hit rate over there, how many casework samples they're doing.

We're hearing a lot of anecdotal stories about how successful they have been, but why is that and how does it compare to where we're at here in the United States at the moment. So hopefully that project should be done we believe in July and will be done at that time and hopefully address all of those questions that you're asking about.

The project we are completely underway with is the project here in the United States that the NIJ has put forward. To kind of walk you through how this developed and what we're after, if you remember, this slide is from our last presentation we gave to this group. Basically we identified some end goals that we believe this group was involved with; that is, we believe this group was looking at all felon databases in all but a few states in a few years, casework backlogs cleared as an objective, and finally routine casework at all relevant crime scenes.

In order to obtain all of those objectives let me remind you there are various decision makers that we identified that are involved. Of course, the state legislators are needed to pass the all felons legislation.The state agency responsible for the crime labs are in place to implement those laws. Congress and the Department of Justice, of course, are involved in helping facilitate the finances to get these programs off the ground, and, of course, law enforcement is in total control of doing the casework, which would include prosecutors, and also city and governments which are also in a position to fund a lot of this as perhaps the Federal Government may move out of funding eventually in the future.

To fund or to move those objectives considering these decision makers, what would a group like this need to perhaps educate those decision makers? Of course, a group like this it's best positioned to educate these decision makers through data. So what we thought about is what kind of data would be most appropriate for these decision makers?

I think you can look at data involved with forensic DNA in two ways. One is what I'll call political data, things such as what is the hit rate of having a large database in doing aggressive casework against it, easy, simple political data that you can see some of these decision makers moving forward with.

The second issue is practice data, things that we're talking today about the need to educate people in our labs, automation, things like that, things that, quite frankly, the decision makers here are not interested in on a simple level. So we're talking about political data that could be available for these decision makers, and that's what we're structuring this report after.

The questions as I see it break down into four major categories. One is what does passing DNA database expansion legislation do to the odds of solving a particular crime? We've all heard recently about the new phrase the hit rate. What is the hit rate involved in having larger databases and running casework against it? That's one of the questions that we're going to work to solve.

Number two, with funding to complete all DNA casework how many crimes will be solved? Basically this is a review of all of the backlog rape kits, homicides, where potential DNA could be yielded to run against the database. So given a certain hit rate, how many crimes potentially could be solved given that info?

Number three, what is the prevention angle of larger databases and aggressive case work? How many crimes based on criminal histories can we assume could be prevented if larger databases and casework was being completed?

Finally these decision makers want to know what is the cost value? What is the cost-benefit analysis? If you have a hit rate of 50% and it costs you $3,000 or three DNA tests in a homicide weighed against a typical amount of investigative time to investigate that crime, what is the cost-benefit analysis? So those are kind of broad categories that we are looking at, and all of the questions that we have derived start with these four areas as a premise.

Again, we're focusing more on this beefy political data without getting into a lot of minutia about things that may not be interesting to a member of Congress, a state legislator, those folks.

Before I turn it over to Lisa, Lisa is going to discuss basically the process and the questions, but let me kind of fill you in on where we have been so far. We were funded in July. We spent most of July working with a university, Washington State University division of governmental studies. Dr. Nick Leverage is the director. We've identified them to help us. They're local to us in our Washington state office. It made a lot of sense. They also have a history doing similar studies. For the U.S. Department of Justice they've done some, a lot of projects for the Washington State criminal justice entities, such as the Washington State Patrol, Washington State Criminal Training Commission. So they fit into what we were doing and have been very willing to work with us.

So we set that relationship up, and they will be doing all of the statistical review. They have consulted with us on the creation of the survey instruments that Lisa will describe. Then August, September, and October is when Lisa has been working with them in a working group which many of you are involved with to actually develop the assessment instruments.

So now before I turn it over to Lisa I wanted to mention while I can there is another issue that seems to be coming up that I would recommend that this group take a look at, although it may not be appropriate - it probably isn't - to blend into this assessment. It's the issue of collections. The collection issue, our guess is that the state legislatures have authorized just this year alone probably 300,000 samples to be collected, but our guess is less than half of them will actually be collected this year.

What has happened is the state has - typically the collection entity has been the state penitentiaries. They're set up to take the DNA blood or swab or whatever it is, but as the state legislatures move into crimes outside of homicide, rape, serious crimes, these felons are doing their time, their felon time in the local jails. The sheriffs in the states are not set up, nor do they know how to take the collection, and many of them are balking calling it unfunded mandates even though it could simply be blended into the intake process in the jail. As you know, the state budgets are all struggling, and they're using any activity placed upon them to complain it's an unfunded mandate.

So I suspect that while we believe through the passage of those laws that we're rapidly growing the database, we're probably operating at about half capacity because of this collection problem. I guess I'm telling you there seems to be a very real problem based on what we're hearing out there, and this could be another assessment that you take a look at in the future.

With that I will turn it over to Lisa, who will describe the process that's going on with the existing survey.

MS. HURST: As Tim just explained, we were awarded the assessment project back in early July working in partnership with Washington State University to answer the questions you just saw on the previous slide. The assessment form will be sent to three separate groups. It's actually three separate forms, each one kind of tailored to the entity that we're trying to reach. There is a state lab assessment, which asks a lot of questions about backlogs, of course, your offender index, that sort of thing. This is going again to all local labs that perform DNA testing. Local labs receive something slightly different. Of course, we don't ask the local labs too much information about offender backlogs. Then this is also going to local law enforcement agencies, not to all local law enforcement agencies; there simply isn't enough time or willpower to get that done.

We are assessing all agencies with 100 or more officers. That's approximately 1,000 agencies in the United States. Then we're also giving to another 2,000 agencies - this is going to be a statistically valid sampling. This is a number that we worked through with Washington State University that they assured us we could make some good data interpretations by including this other sampling, and then, of course, the Indian tribes, we don't want to leave anyone out. I'm not really sure what kind of data at all. I haven't heard too much information about what the tribes might be doing or where that type of evidence is even going, but we will be including them in the assessment.

The time line right now, the assessment tool is to be mailed on November 1, so we're actually right now in the process of getting this to the printers, to the mail house. It should be going out very shortly. It's going to be sent out in three waves. The first one goes out November 1 with a November 30 due date. We'll do another one December 1 with a December 30 due date. It's going to be the same assessment tool, and, of course, we will be filtering out those people who respond so they don't get the subsequent ones, but we're trying to keep a very narrow time frame so that we can get some accurate results very quickly.

We hope to have a preliminary report to NIJ by the end of January. This timing, of course, hopefully will be keyed to potentially bringing some of this information to the federal legislatures and state legislatures and even your local county officials who might be thinking about budgets, give them some preliminary indication of what kind of real numbers we're coming up with. We do also hope to continue to give reports on the updates of what we're finding until the report is finalized. We hope to have this finalized by June, July of this coming here.

The status right now, I think, as I mentioned, the survey is in final review. We have included several of the forensic and law enforcement community in the review process, several who are in this room right now. It has included obviously the National Institute of Justice, our colleagues at the Washington State University. The state laboratories have given us some input, local laboratories have given us some input, and some of the local agencies have also reviewed this and helped us to write an assessment tool that hopefully most people can answer.

Insuring results. In some cases this assessment - it is asking for a lot of information. As I mentioned, we will be sending it out with two subsequent followups. We're going to be focusing on personal calls to all of the forensic labs to identify - it is going to the crime lab director, but in most cases it likely will maybe be handed off to the person who oversees the DNA section of the lab. We will be following up with all of those people to make sure that we answer all of the questions and that there is not a lot of confusion on what we're looking for.

We're also making personal calls to all of those 1,000 agencies, the local law enforcement agencies that do not have labs. This is going to be a little bit more - this is going to be key because I'm concerned in a lot of these local law enforcement agencies the kind of questions we're asking is going to take some coordination between various units. We are asking for what kind of backlog they have not just on rapes, but also on homicides. So especially when you get into the larger jurisdictions, it may fall on us personally to try to coordinate some of that data collection. Then as time permits and as the remaining local agencies that are smaller and the Indian tribes, we'll also try to contact them personally as we can.

Just a few things about what the project goal is not. We are not trying to obtain a statistical survey of all the labs. There again is not enough time. The kinds of information that we would have to ask you to come up with I'm sure you probably don't have the time to come it with it yourself. We're asking for educated guesses in a lot of these instances. We're also not trying to complete a complete census of all the untested DNA evidence that's out there in the local agencies. Again, there are just too many. They're too diverse. They have too many different systems. We would never be able to complete that in a timely manner.

The project goal is, though, to provide a general assessment, to give you an overview of how law enforcement is using DNA right now, what kind of crime scenes are they collecting it from. We want to come up with a base from which we can make some extrapolations about how it might be growing, what the use might be in the future. Again, we're not surveying everyone. Some of this information we're coming up with we're going to be making just projections from, but it should give us a good, solid statistical base from which we can come up with some of these projections.

The project goal is eventually to have a report that we can take to legislatures, to the policymakers, and that will include not just a big problem of how this backlog is so big and how we need so much money to get to where we want to be, but a very big portion of this assessment is going to include a review of the successes of CODIS and a little bit more in detail we hope than some of the general numbers that you see from the FBI's Web page, but maybe into a little bit more detail on what kind of offenders are these, what do their criminal histories look like in some cases, maybe some case studies of how people have been caught and the benefits that this has had in the communities.

So the first question that Tim mentioned, what does passing DNA database expansion legislation do to the possibilities of solving crimes? Again, an extensive review of CODIS is going to be included here. One of the things we're asking for is criminal history records of your offender hits. This we understand in some cases will be - in some states will be very easy to come by. In other states it's a very tightly guarded secret.

Again, I will be working with each of the state labs and local labs to hopefully find a way around some of the privacy issues if that is a concern. We have filed and do have on record a privacy certificate that we had to go through to get this federal grant, so that's also available.

As I said, case studies. Obviously Virginia and Florida have come up with some great numbers doing their own assessments. We want to use that information. New York recently - the first 75% of their first 500 hits was from their 1999 expansion. So we're going to be looking for any kind of anecdotal or other types of studies that we can tie in to create a document again that legislatures can use to see the value of this, the value of their investment.

The second question, if we give you the money, how many crimes are going to be solved? Again, this is the assessment of the backlog, not just rape kits. We want to see if we can get some kind of handle around cold cases, some kind of handle around who is doing the property crimes, which labs are, and what does their backlog look like, how many cases do they handle on property crimes.

Report on the state of DNA casework, what types of crimes are coming in for analysis. That's a little bit of the same question, but from the local perspective is it on their radar to collect from property crimes? Again, comparisons of all felon states compared to successes in states that aren't doing all felons, probably a pretty good guess of what you will find there, but again we want to pull that out into the final report that we do with some case studies.

The third question, how many crimes, particularly rapes, homicides, your violent crimes - how many of these could be prevented if legislation was passed? Crime rates, this is going to the local law enforcement agencies, asking them what their crime rates are, asking them what their comparative is then for all of the homicides that you have in a year, how many do you collect DNA from. If they're rapes, how many do you collect DNA from. Of property crimes how many do you collect DNA from.

Then again case studies, we want to see if we can find a few issues to bring out with serial offenders who had criminal histories where had they been in another state, maybe they would have been on the database a lot sooner. We possibly could have prevented a few crimes that way, cases again perhaps like Debbie Smith's case that we can bring out where a backlog actually led to a delay in an arrest.

This is duplicative, but again comparing how is this working in states who have the all felons who have a mature database versus states where this just hasn't been going on as long, how do the law enforcement agencies differ?

Will expanded databases in casework funding create other efficiencies in the criminal justice system? This is something that I don't know that we really hope to get a firm answer to. This might be something that points to further studies down the road. We do want to look at what the cost is, what are the needs of labs. We have several questions in there that we'll ask. If you were given a pot of money, how would you choose to spend it? We will be asking that kind of question and trying to come up with some answers to that, to what labs really feel they need.

Historic levels of funding, where have you been getting your money from? Have you been getting it all from your state? I know there are some labs that have been getting all of their money from the feds, and local law enforcement agencies, where have they been getting their money?

Again, a few case studies. I know that a lot of you have some cases that you could probably tell us of today where you have been able to exclude and quickly exclude potential suspects, saving that person a lot of personal anxiety, possibly saving them their job, making sure that we get the right offender.

Prosecuting DNA cases, how is this shaking out in the courts? Are they more likely to plead guilty to some of these crimes? Are prosecutors in states where we do have a lot of DNA cold hits - are we seeing more cases in courts? Are prosecutors getting overburdened? Again, this isn't a question that we really want - we are not interviewing prosecutors, but we do hope to have a few case studies where at least we can bring this issue up and maybe point to something that needs to be studied a little bit more in the future.

That's pretty much an overview. We would be happy to answer any questions.

MR. COFFMAN: Are you asking the members of law enforcement questions like do you utilize your database to help you in your investigation in other ways other than just making the hits and things like that? There is a metric that I don't think is being captured, and that is when we get a call from a law enforcement official and he says, look, I have these three really good suspects and I noticed on your criminal history web site that they're all in there, can you make sure that they're analyzed, and if they don't hit, it helps them get their investigation back on track, like maybe that's not the right offender or anything. Is there a question like that in there?

MS. HURST: There is not. I think that we were probably operating more from the premise that it's such an underutilized thing right now that most of them may not even know CODIS. Perhaps Florida and Virginia might be in particular some exceptions where law enforcement might know a little bit more. In fact, actually we do have a question in there that is a question of if you had a good expectation of quickly identifying a suspect through the DNA database, would you be more likely to use it? It's a leading question. We hope we'll get 100% yes. I think that's where our basis of where we kind of felt maybe some of this misunderstanding was, but I would be happy to talk with you a little bit about that. I could certainly look into - I could add that one right after that, maybe something to that extent.

MR. COFFMAN: Because I know before we got rid of our backlog in our offender database - thank you, NIJ - before we got rid of our backlog it was phenomenal how many times they called and said, hey, we've got a good suspect; can you work this guy out of order, get him in the database so we can search, and he never matched. It was very rare that the guy that they had matched at that level of the investigation.

We kind of put our database on the Criminal Justice Network in our state, so anybody in criminal justice, defense, prosecution, anybody, can access it and see if we have someone in the database and if he's worked or not. So we're not getting those calls as much anymore, but we have been told that they have been utilizing that service. So I just think that's something that we haven't really captured that we're helping the investigation very early on when we know they're in the database and they've been worked.

MS. HURST: I think that's something we could certainly look into adding and at the very least that might be something I would like to bring out in the final report. If we don't find any good data that people are using it that way, maybe we can talk to a few officers in your jurisdiction who have used that it way and bring that out as something that could be done if only.

MR. KRESBACH: Similar to what Dave was getting at, our system is a little bit different clearly, as most states are different from one another, but we've experienced where, one example, we have a serial rapist that a crime scene profile has been characterized, it's in the database, it's searched against all of our offenders, and we have a couple of very good detectives that have latched onto the idea as they generate additional suspects in their investigation of this case, the first thing they do is call me and say is Joe Blow already in the database, and I can say yes. Of course, if he is, he has been searched against the case. They can drop that angle of the investigation and not even contact this individual, whereas if he's not, then they can pursue it, ask for a standard to be compared or do the dumpster diving thing where they pick up his trash and such, which I wholeheartedly disagree with; it's a real mess, but that's one of the things.

It's similar, but maybe not quite exactly the same as what Dave has, but it characterizes in a sense a way of aiding an investigation without necessarily having to do additional work. It's already there. We can direct the detectives down a better path sometimes.

I had one other thing, though. I was curious. The report that you're going generate for NIJ, will it be split or categorized, if you will, so that you can get the data not only on a national level, but also state by state by state so the data that you get say from New Mexico, I can utilize that for my needs, you know, that I need to approach this law enforcement agency or whatnot versus a broad national report?

MS. HURST: That's something I hadn't quite tackled fully in my mind yet. My only concern is I would have to take a little bit of a closer look. We've tried to be careful to tell agencies that we're not going to publicly point to them, and if there is a case where I want to point to someone, I'm going to call them in advance and talk about it and get their permission probably in writing before I do that.

Again, it does get back to our privacy certificate, and we want to make sure that people are answering us truthfully without having to feel like they need to say the politically correct answer. There might be maybe some broad answers we can give you for jurisdictions in your state as long as we're maybe not pointing at somebody.

MS. HART: I think the answer to this is we're going to if we can. We're going to try and make sure you have the information you need unless there is some really compelling reason that we can't do it, and then we would let you know.

MR. SCHELLBERG: As long as it's in compliance with whatever we signed when we signed that 100-page contract.

MS. HART: I wanted Tim to follow up a little bit. I didn't know if you were planning to talk a little bit more about the project that was going on in Great Britain or you just wanted to touch on it since that is very new.

MR. SCHELLBERG: Quite frankly, it's very new. As you know, we just got your letter last week, and we signed it and sent it in the end of last week. So we are prepared to start. Our employee in Great Britain is setting up his meetings, getting it all organized, and we believe we can deliver it to you by the deadline, which is in July.

MS. HART: Since this is something that Tim is at a bit of disadvantage because he wasn't there when we were all talking about the need for this, I thought it would be helpful if I just described a little bit about what we're looking at here.

We've actually spent a fair amount of time dealing with Great Britain and getting results from them. They've done a very extensive evaluation on the cost-benefits of using DNA evidence, and they have made a very substantial investment in the use of forensics and timely testing there and are showing not only significant public safety benefit, but significant cost effectiveness, financial benefits from investing in this technology.

One of the things that we really don't have a good handle on is the front end of this. They have also made a substantial investment in the evidence collection and training officers who arrive at the scene to collect evidence in burglary cases, in car theft cases, and stuff that's so low down because you've got these backlogs that you really can't focus your time on that.

So what we want to try and capture here is the whole question of - also on the collection end of this, what kind of training do they have for their officers? Do they have specialized collection teams? How often are they able to generate usable DNA evidence in these cases and what are the costs and benefits of investing in the collection end of this?

So because they've already started it, we want to try and capture some of the information from them so we can make some informed decisions and recommendations on the evidence collection end and the timely analysis of it and to help state and local law enforcement make informed decisions about where to invest their limited tax dollars in this. So I'm really looking forward to this report.

MR. SCHELLBERG: It should be consistent with our report here in that we plan to get preliminarydata, although it will not be in the form of a final report, sometime towards the end of February. So if that matches with the data you may want to present to Congress as they prepare their '04 budget, that's one of our goals is to get you preliminary data by February, and I'll do what we can to make sure the U.K. is on track with that as well.

MS. HART: Great.

MR. SCHMITT: Tim and Lisa, thank you very much for your participation here today. It's fascinating as always. I have 12:45 on my clock. Let's take ten minutes just to clear your plates and get refills, and truly in ten minutes we will reconvene with the next area of discussion.