Statement to Congressional Subcommittee on Crime, Terrorism, and Homeland Security Committee on House Judiciary

In recent testimony before Congress, Paul Ferrara, Director of the Virginia Division of Forensic Science Forensic and Distinguished Professor of Forensic Science at Virginia Commonwealth University, said that science laboratories face a crisis of increasing demands and reduced resources that threaten the full potential of DNA technology


Mr. Chairman, Members of this Subcommittee on Crime, Terrorism and Homeland Security:

As the director of a relatively large and comprehensive state forensic laboratory system, I have closely witnessed the burgeoning application of DNA technology to forensic science from its infancy to its present day adolescent stage. I use the term "adolescent" because we are far from a mature, fully utilized technology. Since its development less than two short decades ago, we have seen the influence of the DNA technology for identification purposes dramatically expand into all phases of our criminal justice system; from the initial stages of the investigative phase, through arrest and prosecution and finally into the post conviction arena. We have seen its successful application in all types of contemporary crimes as well as in "cold cases" dating back decades and, in a few cases, even centuries.

The technology has been refined again and again until it now provides an unparalleled degree of individualization, robustness and sensitivity without compromising personal medical information and privacy. Today, the courts, prosecutors and indeed the general public can feel more confident in our legal system because critical decisions of guilt or innocence are based, in large part, upon reliable, unbiased scientific evidence.

Yet for all of this promise, our forensic science laboratories today face a crisis of increasing demands and reduced resources that threatens the full potential of this marvelous technology. Physical evidence that potentially can quickly establish the identity and guilt or innocence of a suspect languishes in a police property room or in forensic laboratory evidence storage for months or years waiting until a highly trained and qualified analyst even opens the evidence containers. The old adage of "better late than never" does not provide much solace to Raymond Holder.

Raymond Holder was spotted on the street by a 12 year-old girl who identified him as the man who had held a jagged bottle to her neck as he raped and sodomized her. From the time Raymond Holder was arrested on August 25th, 1993, 10 days after the alleged rape, he claimed his innocence and asked for a DNA test while he remained in custody. Absent any knowledge of any special circumstances or a request for prioritization of this case, on April 30, 1994, almost 9 months after his arrest, our DNA tests exonerated Raymond Holder and he was finally released. Better late than never? Perhaps for Mr. Holder, but not for Jemma Saunders.

Following a rape/stabbing of a Virginia woman on August 31, 1998, the laboratories of the Virginia Division of Forensic Science received a victim's rape kit, physical evidence, and, by November 13, 1998, a suspect's blood sample. That suspect's name was Christopher Banks and on November 19, 1998 police told the laboratory that Banks could not be held on the rape charge and would probably be released after his upcoming court appearance on an unrelated shoplifting charge. Despite attempts to expedite this case, the analysis on this case was still not complete when Banks showed up in court on January 6, 1999 on the shoplifting charge; no police officer or prosecutor was present and Banks was released. Eleven days after he was released, Banks raped and killed 22-year old Jemma Saunders. When the results from the August 31, 1998 rape finally were completed by February 3, 1999 the results showed a match between the semen from the victim and Christopher Banks.

While we are not proud of these failings, it is critical to underscore the power of this technology and, conversely, the cost of incomplete or delayed implementation. These cases, and countless others like these which you may never hear of, point out the costs, in human terms, of individuals who should never have been victimized and innocent persons wrongfully accused or even convicted, all due to severely limited usage of the most remarkable forensic technology to come along in a century.

Recognizing this simple principle years ago, the Commonwealth of Virginia embraced this technology early and enthusiastically. In 1989, the Virginia General Assembly passed the nation's first DNA databank law, requiring convicted sex offenders to provide a blood sample upon conviction. Upon studying the very high recidivism rate across the whole gamut of felony criminal offenses, the State of Virginia expanded this statute to include all convicted felons in 1990, and in 1996, added juveniles age 14 or older convicted of a felony. During this period, the Virginia Division of Forensic Science worked closely with the FBI, primarily as a beta-test site for CODIS (Combined DNA Index System) software. Also, during this period, the backlogs of convicted felon samples grew as did the volume, variety and type of crime scene evidence being submitted.

The key to a successful use of DNA databanks, particularly in the critical early stages of a criminal investigation, is three-fold:
a) the presence of a very sophisticated and secure data/processing system that allows searching of DNA profiles within a state or locality and between states and localities (and the FBI);
b) large databases of DNA profiles from persons previously convicted (or arrested) for certain crimes, and
c) having the capability and capacity to process all crime scene evidence from all types of criminal cases and develop any foreign DNA profiles present.
Without all three of these elements in place, the huge investigative value of a DNA databank program will be lost.

Fortunately for the forensic science community (and the criminal justice system as a whole), the FBI has developed and provided states and localities with a most effective system (CODIS) which allows accurate and rapid searching of DNA profiles between convicted offenders and crime scenes. A "hit" (or match) occurs when the DNA profile, e.g. of a seminal fluid donor developed from the vaginal swab of a rape victim, matches the DNA profile of a convicted offender, even if the person was convicted in another state. Thus, even absent any suspects, leads or eye-witness identifications, the name of the putative perpetrator of this rape can be given to the investigating agency as soon as the laboratory has processed the evidence. But therein lies the rub. When it takes a laboratory months or years to enter convicted offender DNA profiles or process evidence from a rape kit (because a thousand other cases are already in the queue), then that rapist has that much larger of a window of opportunity during which he may commit additional rapes. By the same token, the police will be unnecessarily expending six man-months of investigative resources on this case, during which investigation, innocent persons may become suspects.

Recognizing that additional months or years may lapse from the time of an arrest until conviction and sentencing, on January 1, 2003, a new Virginia law went into effect which requires persons arrested for a violent crime to be included in Virginia's DNA databank, but only after a finding of probable cause and issuance of an arrest warrant by a magistrate. In only the first 6+ months of this new law, some 23 additional DNA databank "hits" have been made against these arrestees, four of them for unsolved rape cases and three related to unsolved homicides. This translates to 23 investigations assisted/solved months or years earlier than if we waited until these individuals were convicted and sentenced and then sampled.

But unfortunately, due to the language of the DNA Identification Act of 1994, the DNA profiles of these arrestees and those of juveniles juvenile cannot be searched outside of Virginia. Furthermore, Virginia cannot have these juvenile juveniles and arrestee samples tested and funded under NIJ grants for convicted felons. The irony is that as Virginia tests samples from persons upon arrest, there is no need to retest these individuals upon their conviction, under which circumstance we would be eligible for these federal grants all because the existing and out-dated federal law limits CODIS to "convicted offenders". Arguably, if a state chose to sample convicted misdemeanants, it would be eligible to receive federal money to test them and enter the profiles into CODIS. Clearly CODIS should be able to allow any sample taken legally by state law to be entered into CODIS for searching nationally and these samples should also be eligible for federal funding. To do so will significantly increase the number of inter-state "hits" and the overall performance of CODIS.

B. REDUCTION OF CONVICTED OFFENDER BACKLOG The elimination of backlogs of convicted offenders and crime scene evidence nationally is more problematic. The reduction of convicted offender samples is well on its way, thanks to federal funding to the states administered through NIJ. In 1998, when Virginia's backlog of convicted felons reached 160,000 samples, Virginia took the unusual step (at that time) of contracting with a private laboratory (The Bode Technology Group, Inc., Springfield, Virginia) to process these samples by the same rigorous standards established by the FBI's DNA Advisory Board and followed by our laboratory. The State of Virginia committed over $9,000,000 of state funds over a four-year period for this purpose supplemented by federal funds since 2001. Thus, by 2002, the Virginia Division of Forensic Science had a DNA database of approximately 190,000 convicted felons. Since then, the proliferation of private, accredited DNA laboratories adhering to FBI Quality Assurance Standards has provided an invaluable resource to supplement the public forensic laboratories. Automated and dedicated to the processing of large volumes of relatively uniform and pristine known samples of convicted offenders, the private sector laboratories have provided sufficient capacity to reduce or eliminate the backlogs of convicted offender samples in a few short years. This vastly improved capacity in the private sector laboratories is critical as states inexorably move toward expanding the qualifying offenses for inclusion in a databank. Most importantly, the "DNA Analysis Backlog Elimination Act of 2000" provided the federal funds to pay the private labs for their work. The National Institute of Justice has worked tirelessly to administrate the equitable distribution of these funds to the states while requiring strict compliance of the labs receiving this funding to national standards. Thus, we can see now a light at the end of the tunnel with respect to the backlog elimination of convicted offender samples. That now leaves the hardest nut to crack, i.e., the reduction of backlogs of crime scene evidence.

C. REDUCTION OF CRIME SCENE EVIDENCE BACKLOGS The ability to develop DNA profiles from an infinite variety of objects (cigarette butts, chewing gum, hat bands, telephones, half-eaten bagels, breast swabs, etc.) containing minute amounts of various body fluids and tissue, is both a blessing and a curse; the properly trained, DNA-savvy crime scene investigator will collect many more items of evidence from more crime scenes which may potentially contain some probative biological material. It should be noted here that the term "crime scene investigator" in reality refers to all types of specialists including, but not limited to, criminal investigators, forensic pathologists and medical investigators, forensic nurses, EMTs, ER physicians, first responders and forensic technicians. With all these well trained experts, the effective use of DNA technology in criminal investigations will continue to give rise to increased case submissions consisting of dozens or hundreds of individual items of evidence from a single crime scene. In the Virginia Division of Forensic Science, we have seen an average increase of 30% per year in requests for DNA testing. Furthermore, the DNA technology has been found to be very effective in solving or assisting in the investigation of property crimes as well as violent crime. Thus, efforts to reduce the backlogs of physical evidence (all kinds of physical evidence, not just DNA) will continue to pre-occupy forensic science laboratories for years to come. The only long term solution to backlogs of crime scene evidence is to fund the education, training and employment of the estimated 5,000 to 10,000 new forensic scientists that will be needed in the coming years to meet this growing demand. While our focus here is understandably on the use of DNA technology, we must also consider that analysis of crime scene evidence often crosses many of the other forensic disciplines, each of them not without the same issues we face with DNA. DNA evidence does not exist in a vacuum and many different forensic disciplines with their own databases can be brought to bear on a single or multiple pieces of evidence. For latent fingerprints we have the FBI's IAFIS (Integrated Automated Fingerprint Identification System); for firearms we have ATF's NIBIN (National Integrated Ballistic Information Network); for paints we have the RCMP's (Royal Canadian Mounted Police) PDQ (Paint Data Query); for footwear we have a proprietary database, "SoleMate"; for tires we have another proprietary database called "Tread Assistant". This list is not all inclusive, but is presented to illustrate that new technologies impact other forensic science disciplines in much the same way as does the DNA technology, i.e., as a valuable source of investigative information for the criminal justice agencies we serve.

As a member of former Attorney General Reno's Commission on the use of DNA technology in Forensic Science, and most recently as a member of Attorney General Ashcroft's Initiative on DNA Backlog Reduction, we grappled with this issue of the backlog build-up of crime scene evidence at length. In assessing possible strategies to cope with this situation, it is important to point out that forensic science laboratories are very unlike analytical and clinical laboratories where high volumes of identical samples are tested for some analyte(s) of interest. Instead, each forensic science case presents a unique set of circumstances, forms, amounts and conditions of physical evidence and testing methods. The public forensic science laboratories have no control over the form, quality, amount or condition of the physical evidence. Therefore, the forensic scientist must carefully and laboriously scrutinize every piece of evidence to determine the appropriate type and order of testing so as to provide the most useful information without compromising the integrity of the evidence for other forms of forensic testing. While automation and robotics can and do assist in facilitating the intermediate stages of testing, ultimately it is the individual scientist(s) who must assess the evidence, isolate a probative portion, conduct an analysis, interpret the results and provide an accurate report and unbiased testimony. Thus, any solution to the backlogs of physical evidence in forensic laboratories must include growth in the education, training and staffing of public, and for that matter, private forensic laboratories.

I mention private DNA laboratories because they too play an important role in assisting the reduction of backlogs of crime scene evidence, particularly rape kits. They have already demonstrated their ability to screen large numbers of old unworked rape kits for foreign DNA profiles. They have assisted the NYC OCME in identifying the remains from the WTC. They are a valuable resource for independent examinations or re-examinations of DNA evidence for the defense. However, the private sector cannot, in my opinion, have as dramatic an impact on crime scene evidence in general as they have had on convicted offender samples.

For this reason, Virginia elected to outsource convicted felon samples and to train additional DNA examiners (forensic biologists) in-house through the Virginia Institute of Forensic Science and Medicine and then hire them in the Division of Forensic Science. Thus, we concentrated all our forensic biologists efforts on crime scene evidence in-house while the Bode Technology Group, Inc. tested the convicted felon samples. This strategy resulted in a greater amount of crime scene evidence being processed and the resultant profiles entered into CODIS. The number of DNA cases processed went from 37 in 1989 to 2,284 in 2002.

As a result of these steps, on November 21, 2002, the Virginia Division of Forensic Science "scored" its 1000th "hit ", 894 of which identified the putative perpetrator and the remaining 106 associated crimes committed by the same (still unidentified) perpetrators.

The types of crimes assisted or solved by these first 1000 "hits" include:
244 rape/sodomy 111 homicide 12 rape/homicide 58 robbery 14 car jacking 9 malicious wounding 1 abduction/kidnapping 1 arson (occupied dwelling) 471 property crimes (B&Es, burglary, etc.) 85 Miscellaneous (indecent exposure, aggravated assault, eluding police, peeping toms, Project Exile case, etc.) In considering other strategies for backlog reductions of crime scene evidence, some may look at Virginia's "hit" data and question why so many property crime cases are processed in Virginia. It is tempting to suggest limiting use of DNA testing to violent crimes and only collect samples from violent felons. That is an approach used in other states. While on the surface this seems like a reasonable strategy, I submit to you that to do so is short-sighted and will dramatically reduce the efficacy of DNA databanks. By analyzing carefully these first 1000 "hits", our research has revealed the following:

_ Of the 894 case to offender "hits", i.e., where a match between crime scene evidence and a convicted offender occurred, 344 "hits" (38.5%) were to offenders in our database for prior felony property crime convictions. The crimes assisted/solved by these 344 "hits" included 54 sex offenses, 27 homicides, 6 assaults, 10 robberies, one rape/homicide, 2 abduction/car jackings and 214 burglaries/ B&Es/ larcenies.

_ Another 172 of the 894 case to offender "hits" (19.2%) were to offenders in our database for prior felony drug convictions. The crimes assisted/solved by these 172 "hits" include 35 sex offenses, 42 homicides (including two double homicides), 3 assaults, 18 robberies, 13 abduction/car jackings and 41 property crimes.

_ Even 47 "hits" were to offenders in our database for prior felony forgery/uttering convictions. The crimes assisted/solved by these 47 "hits" include 12 sex offenses, 8 homicides, 4 other violent crimes against persons and 22 B&Es/burglaries/larcenies.

_ Our data shows similar trends with respect to juveniles. Eighty (80) of the 894 case to offender "hits" were to juveniles in our database. The crimes assisted/solved by these 80 "hits" include 12 sex offenses, 8 homicides, 2 assaults, 7 robberies, one rape/homicide, 3 abduction/car jackings and 41 property crimes.

To summarize, 37% of violent crimes solved or assisted by a DNA databank hit were perpetrated by individuals with only prior felony property crime convictions as their most serious qualifying offense. Looked at in another way, 82% of these case to offender "hits" would have been completely missed if our databank was limited to only violent offenders. Clearly, DNA databanks are most effective with inclusion of all felons and DNA technology applied to all forms of criminal cases. IMPROVING THE DNA IDENTIFICATION SYSTEM We have in place today a very well established quality assurance (QA) program promulgated by the FBI's DNA Advisory Board as required by the DNA Identification Act of 1994. All CODIS laboratories, including the FBI, must adhere to these standards and must be accredited. The Office of the Inspector General oversees the FBI and, by extension, audits all CODIS labs as well as contract labs.

The ASCLD-LAB accreditation process (which includes a committee review of mandatory proficiency tests) combined with the FBI QA Standards audits and the OIG oversight is more than sufficient to assure, to the extent ever possible, quality forensic work. No additional federal oversight is going to eliminate these statistically rare problems we hear of from time to time. Forensic science is unlike any other type of analytical/clinical testing. We have no control over the form, quality, amount, or condition of the physical evidence and have to do the best we can with whatever materials present themselves. The important factor is that QA mechanisms are in place allow us to determine when a deviant or discrepant result is obtained (usually due to human error), document and correct it. Fortunately, our technologies and the criminal justice system provide opportunities for complete disclosure and review of all laboratory work and sufficient sample for re-testing by the defense. To quote the 1996 NRC report, "no amount of attention to detail, auditing and proficiency testing can completely eliminate the risk of error. There is a better approach, one that is in general agreement with the 1992 NRC Report: wherever feasible, evidence material should be separated into two or more portions, with one or more portions reserved for possible duplicate tests". Any further layer of governmental regulation is only going to exacerbate an already complex oversight process and increase the backlog situation. The National Academy of Sciences, National Research Council, Committee on DNA Technology in Forensic Science, addressed this issue in its two reports (1992 and 1996) which provided that "one of the best guarantees of high quality is the presence of an active professional-organization committee that is able to enforce standards. Although professional societies in forensic science have historically not played an active role, the American Society of Crime Laboratory Directors (ASCLD) and the American Society of Crime Laboratory Directors-Laboratory Accreditation Board (ASCLD-LAB) recently have shown substantial interest in enforcing quality by expanding mandatory proficiency testing . . . Because private professional organizations lack the regulatory authority to require accreditation, further means are needed to ensure compliance with appropriate standards." Since that was written, of course, the 1994 DNA ID Act which I referenced earlier resulted in the promulgation of mandatory FBI QA standards, which were in turn endorsed by ASCLD-LAB. I should point out also that the FBI lab is also accredited by ASCLD-LAB.
We have always recognized that a major shortcoming of ASCLD-LAB was its voluntary nature (except as noted previously). Perhaps a federal mandate for this accreditation by ASCLD-LAB is an alternative worth consideration. Some of these labs with recent problems are neither ASCLD-LAB accredited nor, therefore, CODIS labs. It is also worth emphasizing that ASCLD-LAB accreditation covers all forensic disciplines and not just DNA. Like the ABA and AMA regulate their respective professions, forensic science must regulate its own.

About Paul B. Ferrara, Ph.D.
Dr. Ferrara is the Director of the Virginia Division of Forensic Science , Co-Director of the Virginia Institute of Forensic Science and Medicine and a Distinguished Professor of Forensic Science at Virginia Commonwealth University. He holds Ph.D. degrees in Organic Chemistry from Syracuse University and from the State University of New York, College of Environmental Science and Forestry.

The Virginia Division of Forensic Science is one of the larger full service state-operated forensic science laboratory systems in the country, serving all law enforcement agencies (state and local), Medical Examiners and Prosecutors in the Commonwealth of Virginia. It should be noted that the Virginia Division of Forensic Science is not part of any law enforcement agency but is organized under the State Department of Criminal Justice Services.

Dr. Ferrara was a member, co-author and consultant to the National Academy of Sciences, National Research Council Reports on the Use of DNA Technology in Forensic Science (1992 and 1996). He is a former member and Chairman of the American Society of Crime Laboratory Directors-Laboratory Accreditation Board (1992-1996), which is the only international accrediting body for forensic science laboratories.

In 1989, the Virginia Division of Forensic Science, under the leadership of Dr. Ferrara, became the first state forensic science laboratory in the US to provide DNA testing in criminal cases. That same year, Virginia enacted the first DNA Databank law in the country, requiring all convicted felons to provide blood samples for inclusion in a state operated electronic database. Since that time, Virginia's DNA databank has been directly responsible for the identification of the perpetrators in hundreds of rapes, homicides and other violent and property crimes in Virginia, as well as cases in Florida, New York, Georgia, Louisiana, and the District of Columbia. Currently, Virginia's DNA databank is making an average of nearly 2 such "cold hits" every day.

In 1994, Dr Ferrara was appointed by FBI Director Louis Freeh as a member of the FBI's DNA Advisory Board, representing state-operated forensic science laboratory operations. In 1998 he was named to then Attorney General Janet Reno's NIJ Commission on the Future of DNA Evidence. In 2001, Dr. Ferrara was the recipient of the Briggs White Award, the highest honor presented annually by American Society of Crime Laboratory Directors in recognition of excellence through leadership in Forensic Science management. In 2002, Dr. Ferrara was named to U.S. Attorney General Ashcroft's Working Group on DNA Laboratory Analysis Backlog Reduction.