DNA Fingerprinting and Civil Liberties Project
Report of Workshop 1
A. Overview of Project:
The American Society of Law, Medicine & Ethics held the first of a series of four workshops as part of the project DNA Fingerprinting and Civil Liberties on May 14 and 15, 2004. The project is funded by an ELSI grant from the National Institutes of Health, grant no. 1R01 HG002836 -01. The workshops are the centerpiece of the project and are designed to bring together a multi-disciplinary group of experts and stakeholders to explore various positions on new and controversial privacy and civil liberties issues involved in the application of DNA technology to the criminal justice system.
Each workshop consists of a half-day Public Forum, followed by a full day invitation-only meeting devoted to more in-depth exploration of the workshop theme. Through discussion sessions, viewpoints of the participants are sought on key issues consistent with the workshop theme. One of the grant goals is to record these viewpoints, to document consensus where it exists and, otherwise, to clarify positions on issues, noting advantages and disadvantages of each. These positions will be disseminated at a national symposium for law and policy makers so that their work in this area may be informed by this two-year, four-workshop process.
B. Overview of Workshop 1:
Workshop 1 explored the theme of identification technologies, history, and law. On the first day of the Workshop a public forum was held featuring talks on state, federal, and international laws governing DNA databanks; legal aspects of DNA profiling in relation to other types of law enforcement profiling; the experience of all-arrestee sampling in the United Kingdom; past, present and future of DNA technology; successes and pitfalls in DNA profiling. On the second day of Workshop 1, background talks and subsequent small group discussion sessions explored technical and policy issues surrounding (1) the efficacy of the current DNA databank system and policy issues implicated in its expansion; and (2) law enforcement use of DNA for phenotype prediction.
The discussions were conducted in three groups of eight or fewer participants, including the group discussion leader. Frederick R. Bieber, David Lazer, and Mark Rothstein, consultants on the project, served as discussion leaders. The discussions were summarized by rapporteurs. Also, prior to the conclusion of Workshop 1, a questionnaire was given to all invited participants. Fourteen completed questionnaires were received. The following is a summary of the discussion sessions and questionnaires.
II. Summary of Workshop 1 Breakout Discussions and Questionnaires
A. Breakout Session 1: Expansion of Offender Registries
1. Primary privacy interests implicated by DNA databases
Each group recognized that privacy interests are implicated by the collection of DNA for DNA databases, and that a balance must be struck between such individual interests and the collective security achieved through the use of DNA for criminal justice purposes. A distinction was made between the digitized DNA profile and the DNA sample. Given current technology, the sample itself, rather than the DNA profile, contains genetic information concerning the individual profiled. Thus, it was the collection and retention of DNA samples that was the focus of the privacy discussion. Consistent with that, a predominant view expressed was that the right to keep one's identity (who I am) secret from the government was considerably more difficult to justify than the right to be free from government surveillance (where I am).
It is this latter aspect of privacy, also referred to as "spatial [or positional] anonymity"-i.e., freedom to go from place to place without being tracked--that needs to be balanced against the need for and role of DNA within the criminal justice system. No consensus was reached on the appropriate balance. It was pointed out that the existence of DNA databanks and the fear of surveillance might have a chilling effect on political activity or lead to surreptitious searches.
Also, concerns were raised for potential abuses of genetic information through cross-referencing of a variety of databases, using DNA as the unique linking data point, or the creation of unofficial (private) databases. With regard to misuse of data, however, it was pointed out that actual cases of misuse of data from DNA databanks within the criminal justice system have not been documented.
Two groups identified the potential for using DNA for reasons other than for criminal investigations as disturbing. Participants cautioned against the use of DNA collected for criminal justice purposes for other purposes-no matter how worthwhile-- such as medical research, because public trust in the integrity of the DNA databank process may be undermined with this broader use.
A related Questionnaire item also speaks to the view of the participants on this topic. The participants were asked to assume that a DNA sample has been collected in accordance with local law from someone arrested and charged with, but not yet convicted of, a serious crime. They were then asked to respond to various scenario related to inclusion of DNA profiles of arrestees in the databank. (The inclusion aspect of this question is discussed in a later section.) With regard to uses of DNA profiles beyond the criminal justice needs, the participants were asked their reaction to the following proposition:
Arrestee's DNA profile should be anonymized if charges are
dropped or if person is acquitted, but the DNA sample should
remain available for construction of population databases or for
other approved research.
Ten individuals disagreed with this statement, eight of these ten "strongly" disagreed. Three respondents did not have a strong opinion; and one "mostly" agreed with the statement.
Consensus: DNA databank collections implicate privacy interests, specifically one's interest in "spatial or positional anonymity"-the right to be free from government tracking or surveillance.
Consensus: Research concerning the administrative processes currently in place to collect, analyze and store DNA samples and data is necessary to better understand the risks of misuse of information and to prevent misuse.
Consensus: Limiting the use of forensic DNA databanks to criminal justice purposes is necessary to preserve public trust.
2. Expansion of sample collection for DNA databanks
The discussion concerning expansion of sample collection for DNA databanks demonstrated agreement as to the need for balancing collective security and individual rights. However, there was substantial disagreement about what the balancing point should be. This disagreement manifested itself most clearly with respect to two issues: (1) who should be included in the database; and (2) retention of biological samples collected from convicts.
The group discussions of inclusion focused primarily on the expansion of current systems to include all arrestees and on the creation of a national data bank. With regard to sampling arrestees, reservations expressed included the diminished expectation of privacy for the arrestee and the practical concern of insufficient resources to run the influx of additional samples. Also, concern was expressed that the creation of a "slippery slope" of expanding databanks may compromise public trust. (Discussion in one group cited similar reasons for not including individuals convicted of misdemeanors in the database.) Discussion in two groups focused on the potential impact of arrestee inclusion in the database on law enforcement behavior, expressing concern that police officers may overstate or even fabricate reasons for arrest to get a DNA sample.
Another issue concerning arrestees that was raised was the need for a policy concerning the performance of a keyboard search upon arrest, without then entering the individual's data into the databank system.
Some support for a national database containing every citizen was in evidence. While some participants expressed doubt that this approach is politically viable, others saw such a system as a means to reverse the tendency to include a disproportionate number of people who are poor and of color in current DNA databases.
A related questionnaire item revealed that all respondents agreed with inclusion of violent felons in a database, and 12 of 14 favored inclusion of all felons, breaking and entering convicts, sexual crime convicts, and non-felony lewd and lascivious conduct convicts. Four favored the inclusion of DNA samples from all arrestees for qualifying offenses; and two respondents favored the inclusion of all arrestees. Four respondents favored inclusion DNA profiles of all adult citizens and newborns in a national database.
Consensus/Disagreement: All or most agreed with current statutory criteria for inclusion in DNA databases and that arrestees should not be included in the DNA databanks. There was some, though limited, support for establishing a national databank of DNA from all newborns and adults.
3. Retention of profiles or samples from the databank
The consensus of two groups was that from a scientific point of view samples should be retained once a profile is entered into the database for possible re-examination with new technologies. Also, it was noted that retaining samples would mitigate lab errors by allowing for reanalysis, and would be needed when seeking to exonerate deceased suspects or convicts. Sample retention was considered wise given concern about the reliability of laboratories responsible for testing, and the possible adverse consequences of incorrect records, particularly due to a sample switch in the laboratories. Members of these two groups recognized that the retention of samples may lead to an inappropriate use, such as framing suspects for other crimes, or using the sample for other purposes, such as for research, health tests, profiling, or even fraud.
One of the three groups expressed the unanimous opinion that technology has developed to the point where there is no longer a need to retain samples. Current technology will be compatible with future advances. The overriding concern of that group was that keeping samples creates a risk that public trust will erode due to fear that law enforcement will use the samples for purposes unrelated to identifying criminals. This group also cited this issue of sample retention as one of the most pressing policy issues today. The privacy interests are different if samples are retained or destroyed. Thus, other issues, such as inclusion in the database and under what circumstances biological material can be collected from an individual, hinge on the answer to this question.
Consensus/Disagreement: Although a majority of participants were in favor of sample retention in the event of technological advances, a consensus was not reached on this issue.
4). The use of DNA obtained from voluntary" exclusion samples
Discussion focused on the "voluntary" nature of obtaining exclusion samples. The participants expressed reservations as to whether consent is truly "voluntary" in such cases because individuals may feel pressure to cooperate. Thus, there is a need for informed consent prior to obtaining DNA samples from such volunteers.
Consensus: Truly informed consent is needed when collecting voluntary exclusion DNA samples.
5. Effectiveness and efficiency of current DNA databank systems for crime fighting
The inadequacy of current measures to assess the effectiveness and efficiency of DNA databases was recognized by all groups. One group acknowledged that the current system is a very valuable investigative technique and that the existing parameters are useful, but not adequate. One difficulty pointed out is the segregation of databases, which prevents the current system from being used as an integrated investigative tool.
Participants proposed a variety of other metrics that would be useful for policy evaluation, if somewhat difficult to obtain. The most direct would be how many crimes were solved by DNA evidence, how many convictions were obtained, or the number of warrants obtained. Also useful would be speed of resolution and number of guilty pleas, as well as the speed and efficiency with which innocent suspects are cleared, as well as a reduction in intrusiveness for suspects.
One group pointed out the need to assess the impact on both solving crimes and deterring crimes. Although difficult to evaluate, the participants opined that DNA databases are a potential deterrent, with the logic that potential criminals who are in the database would avoid any crime that might leave a biological trace that could guarantee their apprehension and conviction.
Consensus: The number of "hits" and "investigations aided" are inadequate measures of the effectiveness and efficiency of the use of DNA databases for crime fighting.
Consensus: Outcomes of interest as potential measures of effectiveness and efficiency include: crimes solved that otherwise would not have been solved; deterrence of crime; speedier and cheaper resolution of investigations; avoidance of investigations of individuals who otherwise would have been suspects.
B. Breakout Session 2: DNA-Based Phenotype Prediction
1. Ethical issues that arise from DNA-based predictions of physical characteristics or of race/ethnicity
All groups were fairly unanimous in expressing the need for caution in using DNA evidence from a crime scene to identify racial and phenotype profiles of the perpetrator. Participants identified the need for a better understanding of the error rates, and for in depth studies of just what useful information could be gleaned from a genetic sample.
Generally, given some degree of reliability, there were far fewer concerns regarding the use of DNA to develop a profile for appearance than for using DNA to predict race or disease states. With respect to appearance, most felt that there was little qualitative difference between developing a profile from an eyewitness account and from DNA With respect to race, many were of the opinion that independent of predictors of appearance, predictors of biogeographical ancestry would generally not be useful to investigators. However, when asked on the questionnaire which personal attributes they would "favor determining (if it were possible from study of biomarkers) from a biological sample left at the scene of a serious crime with no known suspects", half the participants selected race.
Participants expressed concern that DNA information might be seen as "more scientific" or "infallible" by law enforcement and the public, and thus weighted too heavily. With respect to race, there was also concern about reintroducing race as a dominant factor in finding suspects in the criminal justice system, although others pointed out that law enforcement already pays substantial attention to race.
There was also some hesitation about using biological samples to predict disease status. Underlying the opposition to the use of DNA to predict disease status was the fear of potential abuses by law enforcement-e.g., a finding that someone had a particular disease might lead law enforcement to intrude on the privacy of a large set of people with that disease. For example, in pursuing a suspect, concern was expressed that law enforcement may seek access to medical records of people with particular medical conditions.
2. Privacy expectations of abandoned crime scene DNA
With regard to suspects and the privacy expectations of abandoned crime scene DNA, some participants noted that information about an individual's disease status that is not relevant to the investigation or to the crime, and not likely to be known by others, is more problematic in terms of privacy. One individual noted that testing for an array of medical conditions may create an ethical duty to convey bad news to the individual from whom the sample was taken. Also, it was noted that there may be an expectation to be protected against unnecessary social stigmatization.
Consensus: Caution is needed in using DNA evidence from a crime scene to identify racial and phenotype profiles of the perpetrator. A better understanding of the error rates, and in depth studies of just what useful information could be gleaned from a genetic sample are needed.
Consensus: Phenotype prediction of characteristics related to physical appearance is less problematic than predictions of biogeographic ancestry.
Consensus: With the possible exception of disease status that is not related to the crime or its investigation and that has a tendency to stigmatize, no privacy interest in DNA samples left at the crime scene was identified.
III. A Summary of Salient Points from the Workshop 1 Questionnaire
The following summarizes the salient aspects of the questionnaire questions as they tracked the discussion sessions summarized above.
Question A, as noted above, asked respondents to circle any number of seventeen "categories of data you would personally advocate including in a centralized government operated database." In addition to the categories capturing convicts currently included in most state DNA databank statutes, a majority of respondents would also include military enlistees and officers, all individuals with government security clearance. Half of respondents would advocate including DNA from all sworn civilian law enforcement personnel. Categories of all non-citizens immigrants, all those on extended US visas were circled by five. As noted earlier, four individuals would include DNA from all adult citizens and all US newborns. "Suspects in qualifying crimes" and "all arrestees" were selected by only one and two respondents, respectively.
Across all categories, the category of DNA profiles was chosen the least, while name/address was selected the most for inclusion in a centralized government operated database. Next to DNA, passport photos were circled the least, followed by fingerprints, while social security numbers were the second most selected item for inclusion in the database.
A. For each class of individuals listed below, circle the number representing the categories of data you would personally advocate including in a centralized government operated database.
Responses (14 Per Class of Individual)
|Class of Individual||Name / Address||SS#||Fingerprints||DNA Profiles||Passport Photo|
|Breaking and Entering Convicts||10||10||11||12||11|
|Sexual Crime Convicts||12||11||12||12||12|
|Non-Felony Lewd and Lascivious Conduct Convicts||9||9||9||10||9|
|All Arrestees for Qualifying Offenses||7||7||7||4||5|
|Suspects of Qualifying Crimes||5||6||3||1||1|
|All Non-Citizen Immigrants||7||6||6||5||6|
|All Those Holding Extended US Visas||8||7||7||5||7|
|All Military Enlistees||8||8||8||8||7|
|All Military Officers||8||8||8||8||7|
|All Sworn Civilian Law Enforcement Personnel||9||8||9||7||8|
|All With Government Security Clearance||9||8||9||8||8|
|All Commercial Airline Pilots||7||7||6||5||5|
|All Adult Citizens||6||6||2||4||1|
|All US Newborns||5||5||3||4||1|
- Until we get a uniform definition of "felony" and "breaking and entry" and "non-violent sexual crimes" I would limit the collectible information; my opinion might change once the definitions became available.
Question B demonstrated a difference of opinion as to the handling of a DNA sample from an arrestee who has been charged with a serious crime, but not convicted. A majority of ten respondents strongly agreed that the DNA profile should be searched against crime scene samples in both local and national DNA databases, while three strongly disagreed. Just half favored destruction of the sample and elimination of the DNA profile from the databank if charges are dropped, or if the individual is acquitted, while five strongly or mostly agreed with the retention of the sample and profile. A majority of ten respondents mostly or strongly disagreed with the retention of the anonymized profile under these same circumstances for the construction of population databases or for other approved research.
B. In a hypothetical situation, assume that a DNA sample has been collected in accordance with local law from someone arrested and charged, but not yet convicted, for a serious crime.
Responses (14 Per Question)
|Statement||Agree Strongly||Mostly Agree||No Strong Opinion||Mostly Agree||Strongly Agree|
|Arrestee's DNA profile should be searched ASAP against all unsolved crime scene samples in both local and national DNA databases.||10||1||0||0||3|
|Arrestee's DNA sample should be destroyed and the DNA profile eliminated from the local and national databases if the charges are dropped or if the person is acquitted.||7||2||2||2||1|
|Arrestee's DNA sample should be destroyed if charges are dropped or if person is acquitted, but the DNA profile should be retained and searched against future unsolved crimes.||3||2||2||3||4|
|Arrestee's DNA profile should be anonymized if charges are dropped or if person is acquitted, but the DNA sample should remain available for construction of population databases or for other approved research.||0||1||3||2||8|
- Charges are "dropped" for a wide variety of reasons. The reasons matter!
- A person's sample should only be kept for research, etc., with their consent.
- Once we acquire information we should use it.
In Question C, respondents were asked to circle the personal attributes "you would personally favor determining (if it were possible from study of biomarkers) from a biological sample left at the scene of a serious crime with no known suspect." Gender was unanimously selected, while skin complexion, eye color, hair color, and age were selected by ten or more respondents. Half of the respondents selected HIV status, ethnic group/race, non-genetic disease states, and genetic disease states. While there was little agreement on most categories, the only attributes that received more "no votes" than votes to determine were obesity (7 no; 6 yes; 1 uncertain) and carrier status for genetic disease (7 yes; 4 no; 3 uncertain).
C. Circle each of the personal attributes, which you would personally favor determining (if it were possible from study of biomarkers) from a biological sample left at the scene of a serious crime with no known suspect.
Responses (14 Per Attribute)
|Genetic Disease States||7||6||1|
|Non-Genetic Disease States||7||6||1|
|Carrier Status for Genetic Disease||4||7||3|
- Only traits which can be definitively determined.
- I doubt that we will get accurate genetic prediction of skin color in the next 50 years, if ever.
- I'm saying "yes" but I have grave concerns and would impose many, many conditions, e.g., the police could be advised of the result only with judicial approval.
- Yes for genetic and non-genetic disease states if observable and useful in the investigation.
- Hard physical characteristics have evidentiary value and are less subject to abuse than probabilistic predictors of characteristics.
Question D posits a dragnet collection of voluntary exclusion samples following a series of unsolved crimes. The question posed a hypothetical situation where investigators are seeking to obtain voluntary biological exclusion samples for DNA profiling from 125 individuals living in an area where an ongoing series of unsolved serious crimes was committed. In response to the question should "individualized voluntary informed consent agreements" be required before DNA samples are collected, thirteen respondents agreed, and one respondent strongly disagreed. In response to questions concerning voluntary sample retention, ten strongly or mostly agreed that the samples should be compared to unknown samples from the specific series of unsolved crimes; and the same number of respondents mostly or strongly disagreed that DNA profiles derived from these voluntary samples should not be compared to any unsolved crime scene evidence in the local and national databases. All fourteen respondents mostly or strongly disagreed with retaining DNA profiles and samples of those individuals eliminated as suspects. Twelve disagreed with the retention of anonymized DNA profiles and samples for approved research was, with 2 indicated "no strong opinion."
D. In a hypothetical situation, imagine that investigators wish to obtain voluntary biological exclusion samples (for DNA profiling) from 125 individuals living in an area where an ongoing series of unsolved serious crimes were committed. For each question below, circle the single choice closest to your own personal opinion.
Responses (14 Per Question)
|Question||Agree Strongly||Mostly Agree||No Strong Opinion||Mostly Disagree||Strongly Disagree|
|Individualized voluntary informed consent agreements should be required before sample collection||12||1||0||0||1|
|DNA profiles obtained from voluntary samples should only be compared to unknown samples from the specific series of unsolved crimes||8||2||2||1||1|
|DNA profiles from the volunteers should be compared to any unsolved crime scene evidence in the local and national databases.||1||3||0||4||6|
|DNA profiles (and the samples) should be retained by police even from those eliminated as suspects.||0||0||0||6||8|
|DNA profiles and samples should be anonymized and used for approved research.||0||0||2||2||10|
- The answers to these questions depend upon the scope of any informed consent obtained.
- [Some answers presuppose (informed) consent.]
- Research approved by whom?
- DNA profiles/samples anonymized and used for research only with consent specifically for that purpose.
Question E asks respondents to select from a list of possible measures of the effectiveness of DNA offender registries those measures for which "good," "poor," or "no" data exists. Each respondent had a choice of three responses for twenty-four possible measures. The number of responses over all the categories and levels of data quality was forty-one. Thus the response rate for the question was markedly lower than for the other questions.
E. In your own personal opinion, list the most important criteria to measure the effectiveness of offender DNA registries. For each of the criteria you list, what is your opinion of the quality of existing data on such measures?
No. of Responses re: Quality of Data
|Measure of Success||Good Data Exist||Poor Data Exist||No Data Exist|
|Hit Rate / # Hits||5||0||0|
|Exonerations/Exclusions of Suspects||2||2||2|
|Prevention (of crimes)||0||1||1|
|Crimes Definitely Solved||0||0||1|
|cost savings to police||0||0||1|
|cost savings to justice, courts||0||0||1|
|Effect on Allocation of Funds||0||0||1|
|Length of Investigation||0||0||1|
|Time Between Hit and Apprehension||0||0||1|
|Expectation of judicial process (follow through for final court decision)||0||1||0|
|Public Expectations (safety to public)||0||1||0|
- Number of Hits is a law priority
- I am not well-informed on the issue of "effectiveness". Back logs remain a serious problem in many jurisdictions that need attention.
- My opinion is based upon a hasty reading of the materials. I don't think re-arrests or hits or investigations aided are reliable or precise measures.
The success of this first Workshop was due in large part to the efforts and expertise of the project consultants, Drs. Frederick R. Bieber, David M. Lazer, Philip R. Reilly, and Mark Rothstein, and to the distinguished participants who attended and contributed so generously of their insights and perspectives. A listing of all Workshop #1 invitees follows as "Appendix A." The Workshop #1 Agenda is attached as "Appendix B." A description of future workshop topics is attached as "Appendix C."
Workshop 1 Participants
Frederick R. Bieber
Honorable Andre Davis
David H. Kaye
Benjamin W. Moulton
Alice A. Noble
In future Workshops, the following topics will be explored:
DNA Forensics and Civil Liberties
Workshop 2: Policy Issues: Privacy and Justice
- Partial matches and family relationships
- Resource allocation and equity
- Role of federal and state governments
- Autonomy of science in criminal justice
Workshop 3: Research and Ownership of Samples
- Who owns the samples?
- Medical research
- Behavioral research
- Epidemiological research
- Obligations to those who provide samples
- Collections of DNA for identification in mass disasters, such as TWA flight 800 and the World Trade Center
- Role of medical personnel
- Role of federal and state governments
Workshop 4: Views from the Other End of the Needle: Present and Future Policies
- Exonerating innocents
- Families of prisoners (indirect DNA information about family members and families' rights to know medical information)
- Effects on family relationships
- Racial identification using DNA
- National DNA identification cards
DNA ELSI GRANT HOME
Workshop #1, May 14-15, 2004
Workshop #2, September 17-18, 2004
Workshop #3, May 13-14, 2005
Workshop #4, September 16-17, 2005
- The Non-Forensic Use of Biological Samples Taken for Forensic Purposes: An International Perspective
- Summary of California Proposition 69
- Summary of the Justice for All Act
- Grid: Survey of DNA Database Statutes
- Users Guide to Grid: Survey of DNA Database Statutes
- Grid: Survey of Post-Conviction DNA Testing Statutes
- Users Guide to Grid: Survey of Post-Conviction DNA Testing Statutes
Summaries of Current News