Am. J. Hum. Genet. 58:393-397, 1996
`
`The Need for Anonymous Genetic Counseling and Testing
`Maxwell J. Mehlman,1 Eric D. Kodishj'5 Peter Whitehouse,7'8 Arthur B. Zinn,6'7
`Sharmon Sollitto,7 Joshua Berger,n Emmeline J. Chiao,3 Melissa S. Dosick,3 and
`Suzanne B. Cassidy6'7
`
`'The Law-Medicine Center, 2Consultant, The Law-Medicine Center, and 3School of Law, Case Western Reserve University School of Law;
`4Department of Pediatrics, Rainbow Babies and Children's Hospital; 'Center for Biomedical Ethics and 6Department of Genetics and Center
`for Human Genetics, Case Western Reserve University School of Medicine; 7University Hospitals of Cleveland; and 'Department of
`Neurology, Case Western Reserve University, Cleveland
`
`Summary
`Concerns are mounting about the risks of genetic dis-
`crimination resulting from the release of predictive and
`presymptomatic genetic test results to employers, insur-
`ers, and others. The ability to keep this information
`confidential is questionable, particularly in view of the
`expansion of electronic medical databases. One solution
`is to afford individuals access to anonymous genetic
`counseling and testing. Probands would be identified
`only by a code that would not reveal personal informa-
`tion, and test results would be stored, retrieved, and
`released solely on the basis of this code. The experience
`with anonymous HIV testing, while not completely anal-
`ogous, suggests that such an approach would be both
`practical and effective.
`
`Introduction
`As the ability to conduct predictive and presymptomatic
`genetic testing has increased, so have concerns about
`protecting the privacy of the proband. Numerous third
`parties might have an interest in obtaining genetic infor-
`mation: family members of the proband; potential
`spouses; health and life insurers, who could use the in-
`formation in reviewing policy applications or in setting
`premiums; employers, who could use the information in
`hiring or firing decisions; the proband's treating physi-
`cians; medical researchers who are investigating specific
`genetic disorders; public health officials gathering epide-
`miological data; and government and private entities
`seeking to influence individual lifestyle and reproductive
`choices.
`There appears to be general agreement among schol-
`
`Received September 9, 1994; accepted for publication November
`22, 1995.
`Address for correspondence and reprints: Professor Maxwell J.
`Mehlman, The Law-Medicine Center, Case Western Reserve Univer-
`sity School of Law, 11075 East Boulevard, Cleveland, OH 44106.
`© 1996 by The American Society of Human Genetics. All rights reserved.
`0002-9297/96/5802-0016$02.00
`
`ars of medicine, law, and ethics that genetic test results
`should be kept confidential and released only with the
`proband's consent or in narrow circumstances in which
`disclosure is necessary in order to prevent immediate
`harm to identifiable third parties (Andrews et al. 1994).
`It is not clear, however, that medical record storage and
`retrieval systems can be devised that will provide this
`degree of protection (Rothfelder 1992; Gostin et al.
`1994; Lebacqz 1994). Indeed, the risk of harmful disclo-
`sure may be increasing. President Clinton's Health Secu-
`rity Act, H.R. 3600, S.1757, would have authorized the
`creation of a national electronic medical data network.
`Medical information would have been entered into the
`system automatically, and there was no provision in the
`legislation that would have given patients a veto over
`the inclusion of sensitive information. (H.R. 4077). The
`Fair Health and Information Practices Act of 1995, H.R.
`435, which is the surviving piece of health-reform legis-
`lation dealing with privacy and disclosure of medical
`records, would permit medical records information to be
`disclosed to third-party payers-including an employer
`who operates a health benefit plan-without the express
`consent of the patient (H.R. 435).
`
`1. The Need for Anonymous Genetic Counseling and
`Testing
`
`A potential solution to the problem of protecting the
`privacy of genetic information would be to afford indi-
`viduals the opportunity to obtain anonymous genetic
`counseling and testing. Probands would be identified by
`a coded number or other identifier that does not reveal
`any personal information about them, including name,
`address, phone or social security number, employer, or
`insurer. Test results would be stored, retrieved, and re-
`leased solely on the basis of this code.
`The attractiveness of this approach was apparent at
`a recent conference in Washington, D.C. on hereditary
`cancers: "Among the speakers at a conference last
`month on hereditary cancers were several people who
`had been identified as likely carriers of recently discov-
`ered cancer genes.... They freely shared their perspectives
`
`393
`
`Snap's Exhibit No. 1063
`Page 1
`
`

`

`394
`
`on genetic cancer testing. But they chose not to reveal
`one thing: their full names" [our emphasis] (Jones 1994).
`Our experience at University Hospitals of Cleveland
`confirms the potential attractiveness of anonymous ge-
`netic counseling and testing. One of five persons seeking
`to be tested for Huntington disease (HD) requests anon-
`ymous testing. So far, one patient who requested anony-
`mous testing has completed the testing and counseling
`process. Several other anonymous persons are in the
`process of being tested at this time. The person who
`completed testing asked to be referred to as "Captain
`James Kirk."
`Anonymous genetic counseling and testing would of-
`fer numerous benefits. It would encourage individuals
`to acquire important personal information at the same
`time that it safeguarded the information against disclo-
`sure. Anonymous testing would promote patient auton-
`omy by making the proband the only person who could
`reveal identifiable test results to third parties. It would
`effectively eliminate concerns over unauthorized dis-
`closure by the test provider. By avoiding automatic or
`unauthorized disclosure to insurers and employers,
`anonymous testing would reduce the potential for dis-
`crimination. It would be up to the proband to decide
`whether to reveal the fact that he or she had been coun-
`seled and tested, and the test results, to insurers, em-
`ployers, physicians, and family members. Although
`probands must be cautioned that, in some cases, with-
`holding this information would be fraudulent and
`could affect the likelihood of obtaining insurance, the
`law is evolving in terms of whether employers and in-
`surers even have the right to ask for this information.
`If they do not, then anonymous testing would give pro-
`bands greater assurance that their legal rights would
`be protected. The availability of anonymous testing is
`likely to encourage anonymous counseling and testing
`as well as greater numbers of persons to be tested who
`would not submit to testing if it presented a greater
`risk of unauthorized disclosure; this would enable more
`people to obtain preventive or treatment measures or
`to make better-informed lifestyle and reproductive de-
`cisions.
`
`I. The Experience with Anonymous HIV Testing
`While the concept of anonymous counseling and test-
`ing is new in the context of genetic testing, there is
`precedent for it in the context of testing for HIV anti-
`bodies. According to the latest data from the Centers
`for Disease Control and Prevention (CDC), almost a
`third of all HIV tests in the United States are provided
`anonymously. (CDC 1993b) In 1993, this comprised
`>55,000 tests. Nineteen states have enacted statutes
`that guarantee individuals the right to anonymous HIV
`testing (Arizona, Connecticut, Florida, Georgia, Hawaii,
`
`Am. J. Hum. Genet. 58:393-397, 1996
`
`Illinois, Iowa, Kansas, Kentucky, Louisiana, Maine,
`Mississippi, Missouri, Montana, Nevada, New York,
`Ohio, Virginia, and West Virginia). In Ohio, there are
`51 sites at which anonymous testing may be obtained.
`The federal government funds state testing programs
`under a statutory provision that authorizes the Secretary
`of Health and Human Services to "make grants to the
`states for the purpose of providing opportunities for
`individuals: (1) to undergo counseling and testing with
`respect to the etiologic agent for acquired immune defi-
`ciency syndrome without being required to provide any
`information relating to the identity of the individuals;
`and (2) to undergo such counseling and testing through
`the use of a pseudonym" (42 U.S.C. § 300dd-31). As a
`condition of receiving federal funds for anonymous as
`well as identifiable testing, the law stipulates that the
`state program must provide for both pre- and post-test
`counseling (42 U.S.C.A. § 300dd-32 [a]), and sets forth
`detailed requirements for post-test counseling, which are
`specified separately for individuals with negative and
`positive test results (42 U.S.C.A. S300dd-32[b] and
`[10]). The law is administered by the CDC, which has
`entered into 65 cooperative agreements with state and
`territorial officials to enable them to carry out confiden-
`tial and anonymous testing using federal funds (R.
`Gerber, personal communication). The CDC also issues
`guidelines for confidential and anonymous HIV testing
`(CDC 1994) and provides a standardized form that may
`be used by testing centers for recording test results. State
`health departments administer the federal testing grants,
`coordinate the testing programs so that they conform
`to CDC guidelines, and transmit statistical data to the
`CDC (F. Martich, personal communication).
`There is some opposition to anonymous HIV testing.
`Several states appear to be considering phasing out their
`anonymous testing programs, under pressure to take
`more aggressive public health actions to prevent the
`spread of AIDS (Doughty 1994). In 1991, for example,
`North Carolina reduced the number of anonymous test
`sites from 100 to 18 (Kassler et al. 1992). However,
`individuals are reported to be traveling to neighboring
`states to obtain anonymous testing, and AIDS activist
`groups have filed lawsuits against the state challenging
`the cut-backs (CDC 1994; R. Gerber, personal commu-
`nication). In May of 1995, the CDC suspended a pro-
`gram under which it had funded anonymous HIV testing
`of newborns for statistical purposes, which had been
`attacked in New York State in particular because posi-
`tive results were not disclosed to the infected mothers
`(Wall Street Journal 1995). But this type of anonymous
`testing for statistical purposes is completely different
`from the CDC's continuing effort to provide anonymous
`testing so that probands can maintain their privacy.
`Moreover, numerous studies have shown that the
`availability of anonymous HIV testing significantly in-
`
`Snap's Exhibit No. 1063
`Page 2
`
`

`

`MehIman et al.: The Need for Anonymous Genetic Testing
`
`creases the number of persons being tested (Fehrs et al.
`1988; Kegeles et al. 1990; Paringer et al. 1991; Browne
`et al. 1993; CDC 1993a; Meyer et al. 1994). Judging
`from the support that anonymous HIV testing receives
`from the AIDS activist community (J. Carroccio [Act
`Up Cleveland], personal communication), the approach
`seems to work well in terms of preserving anonymity
`and providing counseling. Anonymous HIV testing also
`appears to fall within federal and state budgetary con-
`straints (CDC 1994).
`The analogy between genetic and HIV testing is not
`perfect (H. Edgar, personal communication). HIV infec-
`tion is communicable. The fact that it is not curable may
`diminish an individual's need to know whether infection
`is present, but it enhances the public health interest in
`testing in order to encourage affected individuals to
`modify their behavior and reduce the risk of transmis-
`sion. HIV infection also is "reportable": state laws man-
`date that a diagnosis of AIDS and a positive test for
`HIV infection be reported in identifiable fashion to pub-
`lic health officials. In contrast, only a few genetic tests
`are reportable at present under state law (e.g., newborn
`screening for sickle cell, phenylketonuria, galactosemia,
`homocysteinuria, and hypothyroidism). This makes the
`specter of government involvement, and the threat of
`unauthorized government disclosure of test results,
`greater in the case of identifiable HIV testing. Finally,
`testing for HIV is relatively inexpensive on a per unit
`basis (costing -$4 and $35 for the ELISA and western
`blot confirmatory tests, respectively). Genetic testing is
`more complex. Costs for direct DNA counseling and
`testing easily can exceed $500 per test (Andrews et al.
`1994).
`Nevertheless, testing for HIV and genetic testing share
`important similarities (H. Edgar, personal communica-
`tion; Macklin 1992). Both create the risk of stigma and
`discrimination. Both have been the focus of concerns
`over lack of privacy and confidentiality. Both have stim-
`ulated efforts to design records systems to safeguard test
`results from unauthorized disclosure. While a genetic
`characteristic, unlike HIV infection, is not horizontally
`transmitted, it may be vertically transmitted. It thus
`raises some of the same issues as HIV testing with regard
`to the rights of others "at risk" to gain access to test
`information.
`
`Ill. Appropriate Diseases for Anonymous Genetic
`Testing
`The apparent success of the anonymous HIV testing
`program suggests that anonymous genetic counseling
`and testing may be worthwhile to consider. However,
`designing a protocol presents a number of challenges.
`The first is to identify the genetic tests to be offered on
`an anonymous basis. We believe that anonymous testing
`
`395
`
`is most appropriate for largely nontreatable, delayed-
`onset diseases such as HD and familial Alzheimer disease
`(FAD) associated with defects on chromosomes 1, 14, or
`21. Persons with delayed-onset genetic diseases arguably
`have less legal protection from discrimination. Although
`the
`Equal Employment Opportunity Commission
`(EEOC) in March of 1995 stated that persons who were
`presymptomatic for or predisposed to a disabling disease
`rather than actually disabled could be considered disa-
`bled under the Americans with Disabilities Act (EEOC
`1995), it is unclear whether this interpretation will be
`sustained by the courts (Doe v. Kahn, Nast, & Graf
`1994) and whether it will provide protection from dis-
`crimination in insurance. Moreover, persons with de-
`layed-onset disease may be subjected to many years of
`employment and insurance discrimination during which
`they are free of the symptoms and debilities of their
`ailment. More importantly, in the case of adult-onset
`diseases such as familial cancers that are treatable, espe-
`cially with early detection, the proband is unlikely to
`benefit greatly by anonymous testing. A proband whose
`genetic test result is negative is likely to want to disclose
`this information to an insurer or employer in order to
`decrease the risk of discrimination on the basis of family
`history. Probands with a positive test result may fear
`discrimination, but they are likely to conclude that the
`benefits of being identifiable outweigh the risks. For ex-
`ample, they need an ongoing patient-physician relation-
`ship in order to be assured of obtaining necessary screen-
`ing or treatment. In contrast to a counseling and testing
`relationship, such a relationship makes it unlikely that
`anonymity could be preserved. Moreover, even if pro-
`bands could bear the costs of anonymous counseling
`and testing, the costs of treatment or repeated screening
`are likely to be prohibitive without third-party payment.
`Yet, the need to submit claims information to payers
`would effectively preclude anonymity.
`A further prerequisite for anonymous genetic testing
`is that an accurate single-gene test be available for the
`disease. This would avoid the anonymity problems that
`would be created by the need to test relatives for linkage
`analysis. A single-gene test is available commercially for
`HD. A single-gene test is not available commercially for
`FAD associated with defects on chromosomes 1, 14 or
`21. A test for the apolipoprotein gene on chromosome
`19 is commercially available, but its benefits are uncer-
`tain at this time (Roses 1995).
`The possibility of offering anonymous genetic coun-
`seling and testing to minors raises difficult issues, such
`as whether to permit minors to obtain genetic testing
`without notifying parents. Yet, these questions also arise
`in the case of identifiable genetic testing. (Wertz et al.
`1994) The major difference between anonymous and
`identifiable testing in this regard would seem to be that,
`in the case of identifiable testing, health care providers
`
`Snap's Exhibit No. 1063
`Page 3
`
`

`

`396
`
`could contact parents in a crisis, such as if the minor
`appeared unable to cope emotionally with the test re-
`sults. For this reason, providers may prefer to offer
`anonymous counseling and testing only to adults. Fur-
`ther debate on this issue is needed, however. For exam-
`ple, the Ohio Department of Health reports that, despite
`comparable concerns, anonymous HIV testing generally
`is available to minors (S. Egan, personal communica-
`tion).
`
`IV. Maintaining Anonymity
`Anonymous HIV testing typically employs a CDC-
`approved protocol that assigns random numbers to pro-
`bands; the numbers are also used to label and keep track
`of test samples during laboratory analysis and to identify
`the results when they are presented to the proband and
`during follow-up counseling. A similar system could be
`implemented for anonymous genetic counseling and
`testing, as evidenced by our experience with "Captain
`Kirk." One problem that would have to be solved would
`be how to create anonymity when a proband was re-
`ferred to the genetic counselor in an identifiable man-
`ner-for example, with accompanying identifiable med-
`ical or genetic records. Copies of the records could be
`made and retained without the identifying information,
`and the originals returned to the proband or her refer-
`ring physician, but care would have to be taken to make
`sure that the new version of the records could not be
`linked to the proband. It also would be necessary to
`establish a method by which genetic counselors could
`get in touch with anonymous probands to give them
`follow-up counseling and psychological support or in-
`formation about new treatments for their diseases. One
`approach would be to suggest to probands that they
`rent a post office box under a pseudonym.
`Probands seeking counseling and testing for diseases
`for which anonymous testing was available would need
`to be told about the alternatives of anonymous and iden-
`tifiable counseling and testing. Furthermore, they would
`have to be informed of the risks and benefits of anonym-
`ity versus identifiability in order to give their informed
`consent to whichever approach they chose. An im-
`portant aspect of the informed consent process would
`be to make it clear to anonymous probands that they
`could change their minds at any time and proceed to be
`counseled and tested in an identifiable manner.
`A significant issue that would arise in the case of anon-
`ymous counseling and testing is how it would be paid
`for. Initially, the costs would have to be borne by the
`probands. Claims for anonymous counseling and testing
`could not be submitted to third-party payers because
`this would compromise the anonymity of the probands
`and could lead to precisely the types of discrimination
`in employment and insurance that anonymous testing is
`
`Am. J. Hum. Genet. 58:393-397, 1996
`
`intended to prevent. In any event, third-party payers
`often refuse to pay for genetic counseling or testing (An-
`drews et al. 1994). If anonymous genetic counseling and
`testing were available only to those who could pay for
`it, low-income individuals would have difficulty ob-
`taining it. As noted earlier, the federal government gives
`the states funds to finance anonymous HIV testing pro-
`grams (42 U.S.C. §300dd-3 1). A similar approach might
`be adopted to finance widespread access to anonymous
`genetic counseling and testing, at least for selected dis-
`eases. Another approach would be to expect genetic test-
`ing laboratories to absorb some of the costs of anony-
`mous testing in return for being certified by various
`professional societies or patient advocacy groups.
`Anonymous genetic counseling and testing, like anon-
`ymous HIV testing, precludes the possibility of disclos-
`ing test results to third parties, including third parties
`who may be at risk. It is noteworthy that anonymous
`HIV testing is widely available, despite the possibility
`that disclosing positive results to third-parties could pre-
`vent the transmission of the virus. The prevailing view
`in the genetics, bioethics, and health law communities
`is that it is not appropriate under any circumstances to
`disclose genetic test information to third parties without
`the proband's consent (American Medical Assocation
`[AMA] 1991; Suter 1993). This would be especially true
`for anonymous genetic testing, at least if it were only
`offered for nontreatable diseases. While genetic counsel-
`ors should identify other at-risk individuals in probands'
`families and strongly recommend to the probands that
`they be informed and offered the opportunity to obtain
`genetic counseling, it should be the responsibility of the
`proband, not of the genetic testing professionals, to no-
`tify these individuals (Suter 1993).
`
`V. Conclusion
`The principles of privacy and confidentiality have a
`long and respected history as a fundamental precept of
`medical ethics. Yet real confidentiality often seems in-
`compatible with modern medicine. One writer even sug-
`gests that the concept of confidentiality is "decrepit."
`(Siegler 1982)
`The development of testing to identify the probability
`of late-onset genetic disease poses a serious challenge to
`individuals' privacy interests. Insurers, employers, and
`others will seek access to test results for economic rea-
`sons. A positive result may lead to the denial of health
`insurance or employment, even though the proband may
`never get the disease or may become symptomatic only
`after many disease-free years. One option is to refrain
`from offering individuals access to genetic testing that
`presents these risks. We believe that a better approach is
`to offer these tests to those for whom they are medically
`appropriate and to protect privacy by giving the pro-
`
`Snap's Exhibit No. 1063
`Page 4
`
`

`

`Mehiman et al.: The Need for Anonymous Genetic Testing
`
`bands absolute control over the disclosure of their ge-
`netic information.
`
`References
`AMA (American Medical Association) (1991) Code of Medi-
`cal Ethics Reports of the Council on Ethical and Judicial
`Affairs of the American Medical Association. Report 36:
`ethical issues in carrier screening of cystic fibrosis and other
`genetic disorders. American Medical Association, Chicago
`Andrews L, Fullerton J, Holtzman N, Motulsky A (eds) (1994)
`Assessing genetic risks: implications for health and social
`policy. Institute of Medicine, National Academy Press,
`Washington DC
`Browne Major C, Galli R. Fearon M, Chang CH (1993) HIV
`antibody testing: does it make a differences? Int Cong AIDS
`9:758
`CDC (Centers for Disease Control and Prevention) (1993a)
`Comparison of anonymous and confidential registrants for
`HIV testing. MMWR Morb Mortal Wkly Rep (Jan 29): 53-
`56
`
`(1993b) CTS client record database, US Total 1993
`Annual Report, US Department of Health and Human Ser-
`vices, Washington, DC
`(1994) HIV counseling, testing, and referral stan-
`dards & guidelines. US Department of Health and Human
`Services, Washington, DC
`Doe v Kahn, Nast, & Graf (1994) No 93-4510 (ED PA, August
`8)
`Doughty R (1994) The confidentiality of HIV-related informa-
`tion: responding to the resurgence of aggressive public
`health interventions in the AIDS epidemic. Calif Law Rev
`82(111): 132-133
`EEOC (Equal Employment Opportunity Commission) (1995)
`Compliance manual transmission number 915.002
`Fehrs LJ, Fleming D, Foster LR, McAllister RD, Fox V, Mode-
`sitt S. Conrad R (1988) Trial of anonymous versus confi-
`dential human immunodeficieny virus
`testing.
`Lancet
`2(8607): 379-382
`42 USC §300dd-31 (1991) Suppl III
`42 USCA §300dd-32(a) (1994) West Suppl
`(b) and (c) (1994) West Suppl
`Gostin LO, Turek-Brezina J, Powers M, Kozloff R, Faden R,
`
`397
`
`Stenauer DD (1993) Privacy and security of personal infor-
`mation in a new health care system. JAMA 270:2487-2493
`HR 435, 104th Cong, 1st Sess (1995) (Fair Health Information
`Practices Act)
`HR 4077, 103d Cong, 2d Sess (1994) (Fair Health Information
`Practices Act)
`Jolly DH, Blankenship E, Admimora A, Petz W, Own-O'Dowd
`J, Randall-David E, Meriwether R (1990) HIV reportability
`in a system with anonymous testing. Int Conf AIDS 6(1):
`332
`Jones L (1994) Potentially dangerous information: cancer ge-
`netics advances create ethical, treatment issues. Am Med
`News (May 16) p 3
`Kassler WJ, Meriwether R, Blankenship E, Harris G, Lee P,
`Bobbit W, Peterman T (1992) Anonymous vs. confidential
`HIV testing in North Carolina. Int Conf AIDS 8(2): C379
`Kegeles SM, Catania JA, Coates TJ, Pollack LM, Lo B (1990)
`Many people who seek anonymous antibody testing would
`avoid it under other circumstances. AIDS 4:585-588
`Lebacqz K (1994) Genetic privacy: no deal for the poor. Dialog
`33(1): 39-48
`Macklin R (1992) Privacy and control of genetic information.
`In: Annas G. Elias S (eds) Gene mapping: using law and
`ethics as guides. Oxford University Press, New York, pp
`157-172
`Roses A (1995) Apolipoprotein E genotyping in the differential
`diagnosis, not prediction, of Alzheimer's disease. Ann Neu-
`rol 38(1): 6-14
`Meyer PA, Jones JL, Garrison CZ, Dowda H (1994) Compari-
`son of individuals receiving anonymous and confidential
`testing for HIV. S Med J
`Paringer L, Philips K, Ho T (1991) The impact of risk, knowl-
`edge, and state regulatory policy on testing decisions. In-
`quiry 28:226-235
`Rothfelder J (1992) Privacy for sale: how computerization has
`made everyone's private life an open secret. Simon & Schus-
`ter, New York
`Siegler M (1982) Confidentiality in medicine: a decrepit con-
`cept. N Engl J Med 307:1518-1521
`Suter SM (1993) Whose genes are these anyway? familial con-
`flicts over access to genetic information. Mich Law Rev
`91:1884-1887
`Wall Street Journal (1995) May 12, p Al
`Wertz DC, Fanos JH, Reilly PR (1994) Genetic testing for
`children and adolescents: who decides? JAMA 272:875-
`881
`
`Snap's Exhibit No. 1063
`Page 5
`
`