The Combined DNA Index System
Report No. 01-26
September 17, 2001
Office of the Inspector General
FACTORS AFFECTING FORENSIC DNA TESTING LABORATORIES
There are a wide variety of factors, often outside the control of any one agency or organization, that can affect the extent to which a laboratory or a state has utilized CODIS and contributed DNA profiles to the national index. Because these factors affect the success of CODIS at the national level, a brief overview is included here. Some of the factors are interrelated and can include very complex issues; therefore, our discussion is only an overview and does not include all aspects of this subject.
Currently, DNA laboratories use one of three methods to develop DNA profiles from crime-scene evidence or convicted offender samples: the Dot Blot method, the STR method, or the RFLP method. All of these methods focus on areas of DNA that vary widely from one person to the next. These areas are considered junk DNA because they do not "code" for anything (i.e., the DNA does not translate into a personal identifying characteristic like "blue eyes" or into a genetic predisposition for disease). Unfortunately, the DNA profiles produced using one method are not comparable to the DNA profiles produced using a different method.
The three methods currently used to produce DNA profiles have both advantages and drawbacks. The Dot Blot method is the least discriminating (able to distinguish one person's DNA from that of another person) of the three methods but it is fast and can be performed using small amounts of DNA and degraded DNA. DNA profiles developed using the Dot Blot method are not accepted at the national index because they are not discriminating enough to be useful with extremely large populations. The RFLP method is very discriminating but requires large amounts of good quality DNA and is the most time-consuming method. As of February 2001, RFLP profiles accounted for 34 percent of the profiles in the national index. The STR method is also very discriminating when 13 loci are tested, but unlike RFLP, small amounts of DNA and degraded DNA can be used. In addition, the STR testing can be completed in a matter of days. As of February 2001, STR profiles comprised 66 percent of the profiles in the national index. Since STR analysis is both fast and discriminating, many laboratories are in the process of switching to this method from either RFLP or Dot Blot.
Laboratories that have a well-established RFLP program face a time-consuming process with many obstacles as they switch to an STR program. The laboratory has to obtain different equipment and supplies, train its staff, and may need to change the physical configuration of the laboratory. It generally takes months to train analysts in STR analysis. The analysts spend most of their time performing training-related activities so they have little time to perform RFLP analysis on DNA samples. As a result, a laboratory's backlog may increase as it switches to STR testing. In addition, many laboratories use STRs to retest all of the DNA samples previously tested using RFLP. This is a time, labor, and material-intensive process, and also increases a laboratory's backlog since the RFLP-tested samples become part of the STR-testing backlog.
The incompatibility between RFLP and STR results can greatly impact how many matches are made in a state if the convicted offender samples are tested with STRs but forensic samples are tested using RFLP. When two different methods are used the number of matches will be reduced. Thus, the method(s) a laboratory has chosen to use can impact the effectiveness of the DNA indexes in solving crimes. The method(s) used affect how many matches occur and how many samples are waiting to be analyzed (in a given time period a laboratory using RFLP will not be able to process as many cases as a laboratory using STRs, all other factors being equal). Those laboratories using Dot-Blot are not able to contribute DNA profiles to the national index so their work benefits only state and local laboratories.
A few examples of the type of resource issues encountered by the laboratories are discussed below.
All 50 states have added a DNA collection statute to their legislation. These statutes require that offenders convicted of specific crimes provide DNA samples for testing. The DNA profiles developed from the samples will be added to the state's convicted offender database. Many of the state statutes did not contain appropriations to cover the cost of collecting, analyzing, and entering all of these DNA samples into CODIS. When this factor is combined with the slow development of laboratory technology and facilities and the limited availability of DNA analysts, the end result is a growing backlog of samples waiting to be processed.
Further complicating convicted offender sample processing is the fact that some state statutes divide up the responsibility for sample collection among several different organizations such as the state's Department of Corrections and the Office of Probation and Parole. The amount of time that a laboratory spends coordinating, instructing, and corresponding with the agencies collecting the samples reduces the amount of time laboratory personnel have to perform DNA analysis, which ultimately increases a laboratory's sample backlog.
Role of Sample Collection Agencies
As mentioned in the previous section on state statutes, state laws often require agencies that are not connected to the laboratories to oversee the collection of convicted offender samples and the safe transfer of those samples to the laboratory. These agencies face the same hurdles as the laboratories including limited resources and unfunded legislation. The way the process is set up makes the laboratories dependent on the accuracy and thoroughness of the collection agencies. The collection agencies must make sure that the correct individuals are providing samples and that complete and accurate identification and criminal history information accompany the sample to the laboratory. A breakdown in internal controls at the collection agencies can result in a laboratory having unallowable profiles in its convicted offender database, inaccurate information associated with a profile, or a convicted offender database that is not as effective as it would be were all the authorized samples actually collected.