Genetic Justice
Page 29
This leads us to the third notion of privacy that is relevant to our DNA—familial or relational privacy. Because DNA is inherited, it can be examined to infer whether two individuals are related. In the case of parental linkages DNA testing is quite unambiguous, since 50 percent of our DNA was contributed by our biological father, and 50 percent by our biological mother. Sibling relationships can also be inferred, but beyond this, DNA testing is far more limited. Revealing an unsuspected personal biological relationship can have serious consequences for individuals or their families.
Similarly, DNA can reveal information about one’s ethnic origins. The Global Genome Project has introduced broad population groupings corresponding to a common set of genetic alleles called haplotypes. The extent to which our DNA contains these genetic loci gives population geneticists and anthropologists a clue to our geographical ancestry. Currently this information is of personal interest to some individuals who want to know what percentage of their inherited genome came from populations in Europe, Africa, Asia, North America, or Australia. However, this information can have significance beyond those interested in family genealogy. For example, American Indian heritage can sometimes be used in establishing tribal land rights. It can also be connected with social stigmas. When phenotype cannot reveal one’s ancestry, some would look to genotype.
Finally, DNA can provide information about whether an individual was physically present at a certain location. A person’s DNA found on a bedsheet at the scene of a crime is prima facie evidence that the person was at the location. In other words, DNA has implications for spatial or locational privacy. Traditionally, privacy has focused more on questions of who you are and what you are doing, and less on where you have been, where you presently are, or where you are going. What we do in our personal spaces, as well as whom we visit and where, should fall within our sphere of privacy. Increasing uses of radio-frequency ID tags (RFIDs), automated toll-payment systems, Global Positioning System (GPS) hardware and software, and surveillance cameras have directly challenged this notion of privacy. As an example, in 1996 the Federal Communications Commission (FCC) required that all cell-phone manufacturers equip their units with location technology. As a result, it is now possible to position a cell-phone caller in a geographical location.
DNA Privacy in the Context of Law Enforcement
How do the notions of genetic privacy just discussed play out in the context of forensic DNA techniques and practices? What specific privacy concerns come into play when law enforcement collects, analyzes, and retains our DNA? And what protections are currently in place?
First, there is the bodily intrusion associated with DNA collection. As mentioned earlier, when DNA was first introduced into the criminal justice system, DNA was collected by drawing a blood sample. Recently law-enforcement agencies have switched over to collection by way of a buccal swab. The physical intrusion associated with swabbing the inside of an individual’s mouth versus drawing blood is clearly lower. Nonetheless, the inside of one’s mouth is still an intimate space, a body cavity that is not generally made accessible to others. Certainly it would be wrong to say that no privacy interest at all is raised in circumstances where an individual is forced to open his or her mouth so that a police officer can obtain a saliva sample. Aakash Desai, a graduate student at the University of California, Berkeley, who was forced to give a DNA sample after being arrested at a demonstration on campus protesting custodial layoffs and furloughs, as well as tuition hikes, described the experience as follows:
I felt violated when the government took my DNA. I felt like I was being burglarized and not the other way around. I feel like the government now owns my genes. This hurts my sense of self, and makes me feel sick to my stomach. Each swab was like being coerced into giving up part of my being. The government had already taken my possessions and clothing, now they were taking the building blocks of my own body. It seems like some ownership of myself has been lost and my privacy violated.11
As discussed in chapter 2, when DNA data banks were first created in the United States, DNA collection was limited to individuals who had been convicted of serious, violent crimes. Otherwise, law enforcement could only either collect DNA “voluntarily” or require a DNA sample from individuals in cases where they had a warrant supported by probable cause. Over the last 15 years this standard has slipped dramatically. Today the overwhelming majority of states are collecting DNA from all felons, as well as those convicted of some misdemeanors, and 14 states have approved legislation to take DNA from individuals merely upon arrest. At the same time, a number of “suspect databases” have been created as a result of DNA dragnets, where thousands of individuals have had their DNA collected without probable cause, without any privacy protections for the subjects, and with little, if any, oversight. Similarly, as is discussed in chapter 6, there has also been a trend toward collecting DNA surreptitiously from individuals, without their knowledge or consent.
After DNA is collected, it is analyzed for information. The analysis of DNA raises additional privacy concerns. When the FBI published its Legislative Guidelines for DNA Databases in 1991, the agency was explicit that the DNA records held by the Combined DNA Index System (CODIS) should relate only to the identification of individuals, and that no records should be collected on physical characteristics, traits, or predisposition for disease. Currently all police departments that collect DNA samples create a forensic profile, containing information related to 13 loci. More recently some forensic laboratories have run Y-chromosomal short tandem repeats (Y-STR) typing on stored samples to help determine whether the source of that sample is likely to be a relative of the source of a crime-scene sample.
Law enforcement and other DNA database-expansion advocates have repeatedly characterized DNA profiles as nothing more than a “DNA fingerprint.” UCLA law professor Jennifer Mnookin has observed that early promoters of the use of DNA testing in the criminal justice system initially referred to DNA typing as “DNA fingerprinting” in an effort to enhance its appeal and to encourage judges and lawmakers to view DNA as nothing more than a more rigorous and precise technique than traditional fingerprints.12 For example, Assistant Attorney General William Moschella wrote in a letter to Senator Orrin Hatch, “By design, the information the system retains in the databased DNA profiles is the equivalent of a ‘genetic fingerprint’ that uniquely identifies an individual but does not disclose other facts about him.”13 DNA databasing advocates have underscored the DNA fingerprinting analogy with claims that the DNA profile used in the system consists merely of markers that correspond to repeating segments of so-called junk DNA. For example, in promoting the passage of the DNA Fingerprint Act of 2005, Senator Jon Kyl stated, “The sample of DNA that is kept in NDIS is what is called ‘junk DNA.’” 14 However, the “DNA fingerprinting” analogy is fundamentally flawed, most notably because DNA samples, which under current laboratory practices are permanently stored alongside the generated profiles, have the potential to reveal far more than a fingerprint. But the analogy is even mistaken as applied to the DNA profiles (see box 14.2). Although it is true that none of the CODIS loci have been found to date to be predictive of any physical or disease traits, this does not mean that such a correlation will not be found in the future. In addition, it is not necessary for any of the markers to correlate directly with any stigmatizing information for there to be a concern; the specter of discrimination and stigma could arise where one or more short tandem repeats (STRs) are found to correlate with another genetic marker whose function is known, so that the presence of the seemingly innocuous STR serves as a “flag” for that genetic predisposition or trait. A finding of this nature has already occurred; a study in England from 2000 found that one of the markers used in DNA identification is closely related to the gene that codes for insulin, which itself relates to diabetes.15
BOX 14.2 DNA as a “Fingerprint”: A Flawed Analogy
Perhaps the most common refrain provided by law enforcement in defense of the da
ta banking of DNA is that DNA is no different from a fingerprint. The fingerprint analogy is deeply flawed. First, fingerprints are two-dimensional representations of the physical attributes of our fingertips. They are useful only as a form of identification. Fingerprints cannot be analyzed to determine whether two individuals are related. They cannot tell you your likelihood of developing Alzheimer’s disease or breast cancer or whether you are a carrier for cystic fibrosis. Nor can they be read to determine which version of the monoamine oxidase A (MAOA) gene you have. There is no exponentially growing list of conditions that can be read from a fingerprint, or even significant research in this area.
Law-enforcement advocates underscore the fingerprint analogy by stating that the DNA profiles that are generated in DNA testing are merely “junk DNA”—that is, the 13 loci used in forensic analysis do not code for any phenotypic characteristics. But of course, they are focusing only on the DNA profile and quite blatantly ignoring the most significant privacy concern associated with DNA data banking, that of the DNA samples, which are stored indefinitely by forensic laboratories and have the potential to reveal almost unlimited information about ourselves.
Beyond this, even the reference to so-called junk DNA is misleading with regard to the DNA profile. Even noncoding regions of the DNA transmit more information than a standard fingerprint. Recent scientific studies have demonstrated that those regions are not devoid of biological function, as was once thought. And although they may never be found to have highly sensitive direct coding functions, they may very well be found to correlate with things we may care about and deem private. Since the completion of the Human Genome Project, no serious scientist refers to noncoding regions of DNA any longer as “junk.” In addition, simply by way of basic principles of inheritance, DNA profiles can be used to signal parent-child and other close family relationships. An examination of additional so-called junk DNA markers can provide more definitive analysis. Law enforcement’s recent ventures into familial searching are an indication that even criminal investigators recognize that DNA can provide far more information than a fingerprint (see chapter 4).
Source: Authors.
The analysis of crime-scene samples offers another example of how the privacy of DNA can be overlooked in the law-enforcement context. As discussed in chapter 5, there are currently no laws that we are aware of that prevent law enforcement from running any genetic tests of interest on DNA collected from the scene of a crime. Mining those samples for medical or other information that might be used to narrow a pool of suspects is certainly a possibility.
Finally, privacy concerns arise in association with the uses of the stored DNA samples and generated profiles. Those in the database are subjected to ongoing, repeated, suspicionless searches; every week the 7 million DNA profiles in CODIS are automatically searched against crime-scene profiles. In a sense, one’s inclusion in the database makes him or her an automatic suspect during his or her lifetime for any future crime. And as we have discussed in chapter 4, recent uses of forensic DNA databases to search for “partial matches” that may implicate a family member of an individual in the database make the privacy intrusion of repeated searches apply not only to the individuals in the database but also to all of their close relatives. There is also a real but neglected risk of being falsely accused of a crime that comes from being in the database. We discuss the potential sources of contamination and error in detail in chapter 16.
Perhaps the most significant privacy concerns arise about the long-term storage of the original DNA samples. Although the information contained in the DNA profile is necessarily limited, every forensic laboratory in the country currently holds on to the biological samples from which the DNA profile was generated, and, as discussed earlier, these can be mined for an increasing amount of highly personal information. Thus criminal justice agencies around the country have all the medically relevant information on a profiled individual that can be gleaned from one’s DNA. So long as the samples are retained, there exists the possibility that they could be disclosed to or accessed by third parties or used in ways that result in the disclosure of highly confidential information or for malicious or oppressive purposes. And because genetic information pertains not only to the individual whose DNA sample is stored but to others who share in that person’s bloodline, potential threats to genetic privacy posed by the samples’ long-term storage extend well beyond the millions of individuals who are currently in the system. Jean McEwen has warned: “The unique composition of forensic DNA banks . . . will make those repositories a nearly irresistible source of samples for behavioural genetics research or testing . . . without the informed consent of those from whom the samples were taken.”16
Federal policy limits disclosure of DNA information “for law enforcement identification purposes,” but of course there is nothing about “identification purposes” that prevents law enforcement from mining stored samples for information that is restricted to noncoding regions of the DNA. State laws provide little, if any, additional protection against unrelated use of DNA obtained for investigative purposes. Four states are silent on this issue, and most others parrot the federal standard. According to Mark Rothstein and Sandra Carnahan, “Many state statutes allow access to the samples for undefined law enforcement purposes and humanitarian identification purposes, or authorize the use of samples for assisting medical research.”17 The FBI has also conceded that most states do not have protections against the dissemination of DNA samples.18 Most states limit uses of the database to those that are “authorized,” but it appears to be left to the discretion of law enforcement to determine what is an “authorized use.” Only eight states expressly prohibit the use of a DNA database to obtain information about human physical traits, predisposition to disease, or medical or genetic disorders, and Alabama explicitly authorizes the use of DNA information for medical research. Some states explicitly prohibit the use of DNA samples and profiles for purposes other than law enforcement, but this, of course, does not appear to prevent law enforcement from mining the DNA for additional information. Other states are either vague or have laws that permit the restricted use of their DNA forensic data banks. About 15 states have statutes that expressly permit the use of data banks for medical research and humanitarian needs.19 Twenty-four states allow DNA samples collected for law-enforcement identification to be used for a variety of other non-law-enforcement purposes.20 Massachusetts’s law, for example, contains an openended authorization for any disclosure that is, or may be, required as a condition of federal funding and allows the disclosure of information, including personally identifiable information, for “advancing other humanitarian purposes.” Two states prohibit the use of DNA samples in medical research.21 Thirty-four states have statutory language authorizing the use of databases for statistical purposes related to improving law enforcement, such as developing better probability figures on false matches.
There is no national policy governing the retention of forensic DNA samples, and state laws vary on policies for destroying DNA samples and removing the profiles from the database in cases where convictions are overturned or charges are dismissed. Where sample and profile removal are possible, the burden is most often placed on the individual to have his or her records removed, rather than on the state. Wisconsin is the only state that requires that biological samples be destroyed after typing, but it appears that to date, no such destruction has occurred.
Some are dubious that forensic DNA data banks will ever be used by medical or behavioral scientists.22 David Kaye states unequivocally that behavioral genetics researchers who desire to use the samples in CODIS (and in state and local databases feeding CODIS) are excluded from doing so: “Behavioral genetics researchers who come knocking on the doors of state or federal administrators for the DNA of convicted offenders will find them locked, and the key cannot be located within the disclosure and usage provisions of the current database laws.”23 Their research would not qualify under the federal DNA database statute, whi
ch states that CODIS must be used for law-enforcement identification purposes only.24 This would include developing population statistics and validation criteria where all identifiers are removed, but not behavioral genetics or medical research. Thus, even if scientists had an interest in using the database, according to Kaye, in most cases they could not get access. Currently, no state explicitly allows DNA samples from the forensic database to be used in research on predisposition for disease or in behavioral research.25 The Alabama law states that the DNA population statistical database, which shall include “individually identifiable information,” can be used “to assist in other humanitarian endeavors including, but not limited to, educational research or medical research or development.”26 Davina Bressler implies that since the Alabama law prohibits research with DNA samples from the database of individually identified records, therefore, any such anonymized research would not be of any value for medical or behavioral science. In addition, some of those states that allow the DNA samples to be used for “other humanitarian purposes” refer more specifically to identifying human bodily remains or missing individuals, especially lost or kidnapped children. Therefore, these allowances were most likely not intended to open the database to broad medical or behavioral research.