Tobacco Documents Online

Consensus Statement Genetic Testing for Susceptibility to Adult-Onset Cancer The Process and Content of Informed Consent Gall Geller, ScD; Jeffrey R. Botkin, MD, MPH; Michael J. Green, MD, MS; Nancy Press, PhD; Barbara B. Biesecker, MS; Benjamin Wilfond, MD; Generosa Grana MD; Mary B. Daly, MD, PhD; Katherine Schneider, MPH; Mary Jo Ellis Kahn, RN, MSN Objective.---To provide guidance on informed consent to clinicians offering cancer susceptibility testing. Participant~--The Task Force on Informed Consent is part of the Cancer Ge- netics Studies Conso~um (CGSC), whose members were recipients of National Institutes of Health grants to assess the implications of cancer susceptibility test- ing. The 10 task force members represent a range of relevant backgrounds, including various medical specialties, social science, genetic counseling, and con- sumer advocacy. Evidence.--The CGSC held 3 public meetings from 1994 to 1996. At its first meeting, the task force jointly established a list of topics. The cochalrs (G.G. and J.R.B) then developed an outline and assigned each topic to an appropriate writer and reviewer. Wdters summarized the literature on their topics and drafted recom- mendations, which were then revised by the reviewers. The cochalrs compiled and edited the entire manuscript. All members were involved in writing this report. Consensus Process.--The first draft was distributed to task force members, after which a meeting was held to discuss its content and organization. Consensus was reached by voting. A subsequent draft was presented to the entire CGSC at its third meeting, and comments were incorporated. conclusions.raThe task force recommends that informed consent for cancer susceptibility testing be an ongoing process of education and counseling in which (1) providers elicit participant, family, and community values and disclose their own, (2) decision making is shared, (3) the style of information disclosure is individual- ized, and (4) specific content areas are discussed. JAM.4. 1997'~77:1467-1474 From Genetics and Public Policy Studies, Depart- merit of Pediatrics, Johns Hopkins University, Balti- more, Md (Dr Geller); Utah Center for Human Genome Research, Eccles Institute of Human Genetics, Univer- shy of Utah, Salt Lake City (Dr Bo~Jn); Department of Humanities, Hershey Medical Center, Hershey, Pa (Dr Green); Department of Psychiatry and Biobehav- ioral Sciences, School of Medicine, University of Cali- fornia at Los Angeles (Dr Press); Medical Genetics Branch, National Human Genome Research Institute, Bethesda, Md (Ms Biesecker); Department of Pediat- rics, University of Arizona, Tucson (Dr Wilfond); Cancer Center of So~t~em New Jersey, Camden, NJ (Dr Grana); Fox Chase Cancer Center, Philadelphia, Pa (Dr Daly); Dana Farber Cancer Institute, Boston, Mass (Ms Schneider); and National Breast Cancer Coalition, Richmond, Va (Ms Kahn). Reprints: Gall Geller, ScD, Johns Hopkins University School of Medicine° 550 N Broadway, Suite 511, tirnere, MD 21205. EVOLUTION OF CANCER SUSCEPTIBILITY TESTING Heredita~ and sequ~'ed genetic mu- tations play a fundamental role in the development of cancer.1VChile most can- cers result from acquired genetic alter- ations in somatic cells, some cancer pa- tients have inherited mutations that play an etiologic role in their disease. More than 50 different types of cancer dem- onstrate a familial clustering suggestive of an inherited predisposition, including tumors of the breast, ovary, colon, and prostate.=~ Genetic and epidemiologic research in families with an increased incidence of cancer has been useful for identifying risk-conferr~g mutations that are pres- ent in germline cells. The ultimate pur- pose of such research is to identify effective interventions for cancer pre- vention and treatment, and the pace of current discovery offers great promise in this respect. Unfortunately," at the present time, predictive genetic testing for many cancer-predisposing mutations, although technically possible, has lim- ited utility for prevention or early de- tection of disease.9,1° This gap between our ability to test for cancer suscepti- bility and our uncertain or limited abil- ityto reduce mortality or morbidity con- tributes to psychosocial, ethical, and policy complexities.11-14 These complexi- ties make the process of educating tential test consumers about testing par- JAMA, May 14, 1997--Vol 277, No. 18 Genetic Testing for Cancer--Geller et al 1467

ticularly important and problematic. As research about cancer-predisposing mu- tations progresses and test results are provided to consumers, the informed consent process must reflect these com- plexities. Despite current recommendations by a number of professional societies that genetic testing for cancer susceptibility be confined to research protocois,~5'z~ such testing is being introduced into clinical practice,z7 Since informed con- sent is as important in the clinical set- ting as it is in research, particular care must be taken by clinicians who do not have the benefit of institutional review board evaluations of their testing pro- tocols. In 1994, the National Institutes of Health funded a group of studies to assess the ethical and psychosocial im- plications of testing for cancer-predis- posinK mutations. These and a small number of related projects formed the Cancer Genetics Studies Consortium (CGSC), which established a task force to explore the issue of informed consent for cancer susceptibility testing in both clinical and research settings. This state- ment is the result of the work of that task force. WHY INFORMED CONSENT FOR GENETIC TESTING REQUIRES SPECIAL CONSIDERATION One of the issues the task force' dis- cussed was whether such testing is suf- ficiently different from other types of medical tests to require special consid- eration for informed consent. There is no doubt that genetic information is simi- lar in many respects to other kinds of medical information and that we can learn fi, om our past experience with simi- lax circumstances (eg, involving the hu- man immunodeficiency virusm). The po- sition we have taken is that even if genetic information is not unique, the challenges genetic testing presents to informed consent are sufficiently numer- ous to warrant spedal attention. Ge- netic information possesses all of the following characteristics at the same time: (1) it affects an entire family, rather than just a single individuall~ (2) it is laden with symbolic meaning in our cul- ture, partly because of the tendency to consider genes as uniquely determinis- tic of future health and behavior=; and (3) it presents unique challenges for medi- cal professionals because of its probabi- listic nature~ and its implications for the reclassification of "patients" from healthy to at-risl~ Specific to informa- tion about inherited susceptibility to can- cer, the level of prediction is not cer- tain,= yet no independent test is available to separate true from false positives, and the primary risks and benefits of testing at thi~ time are psychological and social rather than physical (because the effi- cacy of preventive and therapeutic strat- egies has not been proven). The simul. taneous presence of these characteristics makes genetic information more complex than other kinds of medical information. The purpose of this article is to pro- vide recommendations regarding in- formed consent for genetic testing for susceptibility to adult-onset cancers. This discussion is relevant to any test- ing context in which results are pro- vided to patient-participants (hereafter referred to as participants). Most of our examples will refer to predictive testing for breast and ovarian cancers, but our recommendations are relevant to other forms of adult-onset cancer. This article will focus only on the testing of adults because predictive testing in children involves additional layers of complex- ity.2~ After providing a general back- ground on informed consent, we will make specific recommendations for the process and content of consent for can- cer susceptibility testing, as well as rec- ommendations for future research on informed consent in this context. BACKGROUND ON INFORMED CONSENT General Theory of Informed Consent The idea of informed consent has its origins in law, ethics, and our contem- porary understanding in medicine about the nature of the physician-patient re- lationship.= The core notion is that any decisions about care involving a compe- tent person are to be made in a collabo- rative manner between patient and phy- sician and that the patient's authorization for a diagnostic procedure or treatment option be 'Sntentional, substantially non- controlled, and based on substantial understanding.'~ These criteria apply whether informed consent is being ob- tained for a clinical service or for par- tidpation in research. There is some guid- ance in the literature regarding informed consent for genetic services~ and for human subjects research both gener- ally~ and with specific reference to ge- netic research.~ Obstade~ to Informed Consent Much has been written about both participant and medical professional im- pediments to achieving adequate in- formed consent.~ One such obstacle results from the imbalance of power in- herent in the relationship between par- ticipants and medical professionals. Par- ticipants often feel intimidated in the face of the authority of the professional and, ~f they are ill, feel vulnerable as a result of their iliness.~ There are other barriers that are par- ticularly relevant to consent for cancer susceptibility testing. Obstacles to en- suring participant understanding include variability in the participant's experi- ence with cancer and inherent difficul- ties in understanding probabilistic in- formation29 Obstacles to ensuring that participants are substantially noncon- trolled include the unwillingness of some participants to assume an active role in decision making,~° their often unques- tioned trust in the health care system and in their medical professionals, pres- sure by family members or medical pro- fessionals, and financial incentives or constraints due to the cost of care and the status of their insurance coverage. Obstacles to medical professionals' as- sisting participants in understanding ge- netic information include their own lack of genetics knowledge,~ their inad- equacy at communicating probabilistic information, their cynicism about the validity of the informed consent pro- cess,41.~ and their tendency to be direc- tive.~.~ Cross-Cultural/Ethnic Dlfference~ Adequate informed consent in a plu- ralistic society requires that medical professionals be sensitive to the educa- tional, linguistic, and religious differ- ences among participants and between medical professionals and participants.~ Medical professionals also must recog- nize the potential impact of sex, race, ethnicity, class, and culture on the de- cision-making process.~-~ For example, there is evidence that AshkenaZi Jewish women have different reactions than Af- rican-American women regarding the availability of breast cancer suscepti- bility testing and their familial obliga- tions if the test is positive2° Knowledge of ethnocultural values, beliefs, health practices, and communication styles is fundamental for the effective provision of genetic services.~'In recent years, the genetic counseling community has published extensively on the topic of cross-cultural counseling.~ The in- volvement of a community advisory group could assist medical professionals in the planning and conducting of ge- netic testing protocols.~ PROCESS OF INFORMED CONSENT In this section, we recommend that informed consent be an ongoing process and that medical professionals and par- ticipants become partners in decision making. This presupposes that the medi- cal professional is sufficiently informed regarding genetics or knows how to ob- tain the necessary factual genetics in- formation. We describe the counseling 1468 JAMA, May 14, 1997--Vol 277, No. 18 Genetic Testing for Cancer~etler et al

and education components of this pro- cess in greater detail. Process ot Decision Making It is a challenge in complex decision making to respect both the ideal of in- formed consent and the constraints of the clinical and research environments. Informed consent can be conceptualized in at least 2 ways: the "event model" and the "process model."~: In the event model, the participant is presented with several options from which to choose, the medical professional usually in~cates what he or she thinks is in the partici- pant's best interest, and the participant has the opportunity to agree with or reject the medical professional's advice. The event model documents informed consent through the completion of a writ- ten form summarizing the proposed medical intervention that has been pre- sented orally to the participant by the health care provider prior to obtaining consent. The consent form typically car- ties legal weight, even though it is widely acknowledged that such documents are only as good as the education that pre- cedes them. The event model is limited, however, in that it takes place at a single point in time. Even though this is not how people make complicated decisions, informed consent typically follows the event model in both the clinical and re- search contexts. An alternative way to understand in- formed consent is to adopt a process model of decision making. In contrast to the event model, the process model as- sumes an ongoing relationship in which medical professionals elicit participants' values as well as disclose their own, and decision making is shared over time. Many adult participants are accustomed to interactive decision making in which they solicit and receive their medical professional's advice. In I study of wom- en's attitudes toward informed consent for breast cancer susceptibility testing, most respondents expressed a desire for their medical professional's recommen- dations.~9 Since the decision to undergo genetic testing for cancer susceptibility is complex, the task force believes that informed consent ih this context should be approached as a multistep process, and those considering genetic test'.mg must be given every chance to rethink and confirm their final decision. We therefore recommend that, whenever possible, informed consent for cancer susceptibility testing follow a process model. Further, counseling for such test- ing should be made in the context of dialogue with families and an under- standing of relevant community values. Informed consent, according to a pro- cess model, is more than the transmis- JAMA, May 14, 1997--Vol 277, No, 18 sion of information from an active medi- cal professional to a passive participant. Integral to informed, shared decision making about genetic testing is the en- tire education and counseling process that occurs when someone faces such a choice.® Such counseling should be ex- tensiv~ and address personal aspects of the participant's decision. Participants can have powerful reactions to genetic information that do not necessarily re- flect the stability or integrity of the par- ticipant. Since every participant is unique, it is important for the medical professional to acknowledge the poten- tially life-altering nature of this infor- mation for some participants. The informed consent process, while intending to help the participant antici- pate issues or worries, may generate new concerns for the participant. Sen- sitivity to the harms as well as the ben- efits inherent in the informed consent process itself will allow for a more mu- tually productive relationship between participant and professional. Partici- pants may benefit from referral to a genetic counselor or a physician or nurse with special expertise in this domain for assistance in making a decision that is consistent with their values and needs at the time. Participants who have the opportunity to explore their own moti- vations and fears through counseling will be better able to hear and understand information that is disclosed during the educational part of the process. Process of Education: Formats of Disclosure Because adul~ have d~erent le~- ing styles, a ~az~ety of c~cios~e pro- eedures have been suggested for impro~£,~g the consent preeess. We rec- ommend that the process be adaptive to many styles of d~Josure and deci- sion making, and that medical profes- sionals consider using multiple educa- tional formats as described below. Face-to-Face Formats.--One-on-One D/scuss/on,s.--The most common method for disclosing information about a medi- cal intervention is via a one-on-one dis- cussion between a medical professional and a participant. There is some evidence that oral discussions of any kind are su- perior to other forms of information disclosure~ and that comprehension of information is improved when nonphy- sicians present the information.~ Re- gardiess of who does the disclosing, un- derstanding seems directly correlated with the amount of time spent with the learner.~ Since genetic counselors have special- ized training in disoussing inherited risk, they are often considered to be the "gold standard" for suchdiscussions. However, as the number of genetic tests increases, it may be unrealistic to expect all indi- viduals to have access to an in-depth dis- cussion with a genetic counselor before the test. Alternative methods of infor- mation disclosure need to be explored. Group D/scu~s/ons.--Group discus- sions are another way to disseminate information about inherited susceptibil- ity to cancer. Like one-on-one discus- sions, group discussions involve contact with a medical professional. They also have the added advantage of peer in- teraction, which may promote satisfac- tion and social acceptance. However, group encounters have limitations. Some people are embarrassed about speaking in groups, public disclosure may jeop- ardize confidentiality, and an individu- al's specific needs may not be met. Nonpersonal Formats.--Wr/tten Bro- chures and Pamphlets.--Written bro- chures or pampbletsare a less costly way to disclose substantive information about inherited cancer risk to participants. Written materials have been shown to improve knowledge about genetic is- sues~ but have some obvious limita- tions-they are not effective for indi- viduals who do not read, they need to be written in a variety of languages with sensitivity to cultural differences, they need to be concise, and they tend to be written for a general readership rather than the specific participant. Neverthe- less, such materials can be standard- ized and reviewed at the convenience of the participant and are generally help- ful as a supplemental learning tool. Videotapes.--Videotapes, by present- ing educational material visually, can provide consistent information in a fa- miliar medium and reach marginally lit- erate individuals. They also have the advantage of being repeatable and user controlled without requiring a great deal of time on the part of the medical pro- fessional.~ However, they have not been shown to be successful in all clinical set- tinge~ and are neither participant-spe- cific nor interactive. Interactive Videos and Computers.-- More innovative and sophisticated com- puter-based technologies, such as inter- active video or CD-ROM, have the additional potential of tailoring the com- munication process to the particular learning needs of the participant,ss~ Sev- eral features make computers a desir- able way to disclose health-related in- formation. They can be effective at educating participants'in a variety of set- tings,7.79 they are widely accepted by participants of varied age, socioeco- nomic status, and educational back- grounds,~°~ and they can provide con- sistent and accurate information. Computers also can provide rapid feed- Genetic Testing for Cancer---Geller et al 2063633707 1469

back about performance to reinforce learning~ and can be used to assess knowledge and recall.TM Computers can be used in a private setting at the par- ticipant's own pace, which may de- crease embarrassment and frustration. Finally, computers can reduce the time that practitioners spend in direct edu- cation.~ Despite the growing use of com- puters for patient education, they have not been widely used for genetic edu- cation. Studies are ongoing to evaluate their efficacy in providing information about inherited susceptibility to breast cancer (M,LG., unpublished data, 1997). Strategies for Improving Compre- hension.--Simplifying Written Consen~ Farms.--Medical consent documents are notorious for their complexity.~ Most are written at an 1 lth-grade level or higher despite the fact that half the US popula- tion currently reads at a ninth-grade level or lower, and 20% of adults read below a fifth-grade level.~ There are many strat- egies for improving readability~ as well as several computer programs for esti- mating readinglevel. These programs can be used to measure language variables within specific passages of writing,~ but because they do not assess comprehen- sion, the reading level of a consent form can be further adjusted after pretesting the document to measure the target au- dience's comprehension. Reinfarcing With Interac~ivity.--It is important for the consent process to pro- vide an opportunity for a participant to ask questions and receive answers. The simple act of asking a participant to re- call information can lead to significantly better retention of disclosed informa- tion than the standard technique of dis- closure without such "quizzing.'~1 This technique adds about 5 to 10 minutes to the discussion. Such interactivity is par- ticularly suited to consent involving dis- cuseion or interactive computer. Attending ~o the Timing of Infarmed Consent.mIdeally, consent for genetic testing should take place in several stages. In the first stage, there would be extensive genetic counseling about risks, benefits, alternatives, and the l~ke. When possible, counseling should be followed by a period of days or weeks before participants make their decision. Wait- ing can remove the pressure for an im- mediate decision, improve comprehen- sion (particularly if a consent form is brought home and read at the partici- pant's convenienceg~.~), and result in more meaningful consent. This waiting period should be followed by a second stage, at which time interested partici- pants would provide explicit consent to proceed with genetic testing. Assessing the Effectiveness of the Con- sent Process.--A prerequisite of valid consent is that information be adequately disclosed by medical professionals and understood by participants. Medical pro- fessionals should attempt to elicit the degree to which participants understand the information that is disclosed. This could be done with the help of a 2~part consent form,~ in which part 2 asks several questions about the information contained in part 1. Feedback should be provided for questions that are answered incorrectly.~ Maintaining Confidentiality To minimize the risk of stigmatization and discrimination, access to genetic in- formation should be limited. Medical pro- fessionals should.be prudent in their han- dling of the information. They should inform partidpants about the importance of confidentiality and familial implica-. tlons of genetic testing. This discussion' may alert participants to the possibility that genetic test results may be re- quested by third parties such as rela- tives, in~urers, employers, or other phy- sicians. We recommend that genetic test results only be made available to third parties with explicit consent of the par- ticipant. The consent process should in- dude a discussion of whether results will be given to anyone other than the participant and safeguards for protect- ing confidentiality. These safeguards in- dude use of linked identifiers when han- dling specimens and data, use of a coding sheet that is kept in a secure place and accessible only to the provider, and, in a research context~ use of a Certificate of Confidentiality from the US Depart- ment of Health and Human Services~ to protect against subpoenas. CONTENT OF INFORMED CONSENT Ideally, the informed consent process should involve an assessment of the par- ticipant's prior experiences, beliefs, atti- tudes, biases, motivations, and existen- ~issuesinvolvingthe meaning of testing prior to disclosing specific aspects of test- ing. The following specific content areas can be addressed within the context of the participant's life and current needs. Purpose o! Test Participants must be dearly informed of the purpose of the genetic test and, if in a research setting, the aims and de- sign of the research project. Currently there are 2 distinct but overlapping pur- poses of genetic testing for cancer sus- ceptibility mutations: the discovery of new knowledge in the context of research protocols and the identification of per- senal genetic risk for purposes of medi- cal management in the clinical setting. If the research design includes diselo- sure of results, the research team must assure that communication of test re- sults be done with the same standard of care that is maintained in the clinical setting. Practical Aspects of Test Practical aspects of the testing pro- cess include but are not limited to the following: amount of blood to be drawn, pain and bruising that may accompany blood draw, type of analysis to be per- formed, cost of the analysis, range of possible results that may be obtained, length of time for results to become avail- able, manner in which results will be com- municated, how samples will be stored in the laboratory following completion of the analysis, and name and contact information for at least 1 medical pro- fessional who can answer questions throughout the testing process. If test- ing is being performed within a re- search protocol, the participant should be apprised of any questionnaires or in- t~-views to be included during the course of the testing process. Interpretation of Results Participants must understand that, in many circumstances, genetic testing for cancer susceptibility does not involve a test with unambiguous results. Multiple genes, each with numerous mutations, may be associated with a particular type of cancer. Thus, we recommend that in- formed consent address the variety of cancer etiologies as well as the limita- tions of the technology to detect muta- tions. Helping participants anticipate re- sults and what they may mean is complex and involves abstract thinking about probabilities as well as perception of ill- ness and issues of vulnerability. Predisposing Cancer Gene Test In- dicating the Presence of a Known Mu- tation.--In families where a gene mu- tation has been previously identified in someone with cancer, identification of the same mutation in a relative implies a significantly increased risk for cancer. It also confirms a 50% chance of passing the mutation on to each child. The de- gree of increased risk may depend on the specific mutation identified; how- ever, genotype-specific risks are not yet known. Women from high-risk breast and ovarian cancer families often report an- ticipating that they will have a muta- tion.9~ They have gro~n up fearing the time when they, too, will develop can- cer. Thus, the greatest challenge in ob- taining informed consent from women in such families may be to ensure that they understand the limited usefulness of genetic test results in making pre- dictions about their own cancer risk~ Fur- ther, it may be cHfficult to help them 1470 JA,MA, May 14, 1997--Vol 277, No. 18 Genetic Testing for Cancer~eller et al

appreciate how little data exist to guide them on the most effective methods to reduce their morbidity and mortality. Men in these families may have assumed that cancer risk was a worry only for women. Helping men to understand that they can also have a mutation, could pass it on to their children, and may have slightly increased risks of specific cancers themselvesm makes the option of testing more relevant to them. In the case of genetic testing for colon cancer susceptibility, where effective methods to reduce morbidity and mortality may be better understood,9 it is still impor- tant for participants to understand that a mutation is not equivalent to having cancer nor is it a certainty that cancer will develop. Being tested may, in a sense, replace one type of uncertainty . with another. Participants should be en- couraged to consider what learning a test result may suggest about their cur- rent risk reduction and screen/ng ac- tivities. Those who test positive should be encouraged to meet with an oncolo- gist and other specialists to discuss a strategy for prevention and screening activities. Interpreting a positive test result may be more difficult when it is the first mutation detected in a family, although such detection most likely also implies an increased risk of cancer. Participants should be told the limitations of apply- ing penetrance figures from high-risk families to those individuals who have had few or no affected family members. The relevance of population statistics for individuals remains largely unknown. There will be increasing amounts of in- formation on penetrance of specific mu- tations, but prior to this, participants need to be told of the estimation in- volved in interpreting risks. Predisposing Cancer Gene Test In- dicating That a Known Mutation Is Not PresenLnIf an individual tests negative for an inherited mutation that has previously been identified as being associated with cancer in his or her famo fly, that person's risk may be reduced to that of the general population (the per- son could have other risk factors). Test- ing in such circumstances is confined to the known mutation rather than involv- ing complete sequencing of a gene or testing of several candidate genes. This is often the type of testing done in he- reditary cancer families where multiple family members arC affected. In the event of a true-negative test result, several points should be empha- sized in the pretest and ~osttest coun- seling: cancer may still occur, the mu- tation cannot be passed on to children, • and familial relationships may still be affected, especially if other family mere- bers are found to have a mutation. Can- cer risk reduction and'screening recom- inendations relevant to the general population should be discussed. Predisposing Cancer Gene Test Re- sult Which Is Not Informative.--There are at least 2 circumstances in which a test result will not help with risk as- sessment. The first occurs when no mu- tation has been identified in the family. In this case, a test result that does not identify a mutation may be of little value and, thus, indeterminate for cancer risk. Many individuals, deciding whether to be tested for a cancer gene will be the first person in the family tested. With- out confirmation that a recognizable mu- tation has caused cancer, a negative test result may not refine one's risk for de- veloping cancer in the future. Follow-up for a negative test in this circumstance needs to be anticipated and negotiated with the participant in the consent process. Depending on the sta- tus of research findings, future testing may be indicated. In the case of breast cancer susceptibility testing, for in- stance, if BRCA1 gene sequencing does not identify a mutation in an individual from a high-risk family, offering BRC.42 testing may be warranted depending on the cancer history of the family. Another reason a test result is clas- sifted as indeterminate is if a sequence variation has been identified in the fam- ily or individual, but there is insufficient research to know whether the particu- lar variation constitutes a mutation pos- ing an increased risk of cancer or a nor- real variant (polymorphism) with no associated increased risk. In such cases, provision of cancer risk information may not be possible. If available, furtherlabo- ratory tests would need to be conducted before determining the clinical signifi- cance of the results. Individuals should be apprised about which test will be performed or whether they will need to wait for further technologic advances before such testing is possible. Psychological and Social Implications Psychological Implications.--Partici- pants should be informed of potential adverse responses in those found to be mutat/on carriers such as amdety, de- pression, anger, and feelings of vulner- ability. In addition, participants may ex- perience guilt over the possibility of having passed the mutation to offspring. Participants who find they are not mu- tation carr/ers may also experience guilt if other close family members are found to carry the mutation--so-called survi- vor guilt. Those who are not mutation carriers may experience regrets if they have made major life decisions, such as prophylactic surgery, based on their per- ception of risk prior to testing. Individu- als who decline testing or who do not receive definitive results may exper/ence persistent anxiety over their risk status. The short-term and long-term psycho- logical effects of genetic testing for sus- ceptibility to cancer and other adult-on- set diseases are not fully known. Potential positive responses include relief from tin- certainty, an increased feeling of control as well as joy, and stronger motivation to pursue additional screening or early de- tection measures. Serious adverse psy- chological effects do not appear to be common,~4~,~°° although the research is limited. Despite these reassuring pre- liminary reports, many individuals ex- perience some degree of psychological distress~°I in response to genetic testing for cancer susceptibil/ty.~°~1~ Insurance Risks.~There are sub- stantial market and regulatory differ- ences between health, life, and disabil- ity insurances. Insurance discrimination based on genetic test results has been a major concern for state and federal governments, the National Center for Human Genome Research (now the Na- tional Human Genome Research Insti- tute), the insurance industry, profes- sional and lay societies, and scholarly commentators.~3 Participants should be informed that genetic testing for can- cer susceptibility may limit their ability to obtain health, life, or disability insur- ance, may lead to limitations in coverage, or may result in higher premiums for in- surance products. At present, insurers do not themselves conduct genetic test- ing of applicants, although they are likely to be quite interested in knowing the test results of applicants who are tested else- where. Participants should be informed that insurance providers may gain access to their genetic test results through their physician's records, hospital records, or disclosure by the participant in an insur- ance application process. Participants who provide false information to insurance companies are at risk for policy cancel- lation or for legal prosecution for insur- ance fraud. Whether individuals are at risk for these adverse outcomes will depend on a number of factors. Participants who are at greatest risk of insurance dis- crimination are those who might undergo individual underwriting for a policy, and those who attempt to change coverage providers. In the private market, 85% to 90% of health policies are purchased through large groups that do not re- quire individual underwriting.I°~ In con- trast, life insurance is obtained largely through individual policies with approxi- mately 71% of policies purchased on an individual basis in 1989.~°9 JAMA, May 14, 1997--Vol 277, No. 18 Genetic Testing for Cancer--Geller et al 1471

The incidence of insurance discrimi- nation based on genetic information has been difficult to document,n~ A growing number of states have passed legisla- tion to restrict health insurance discrimi- nation based on genetic testing.1°~'° However, this type'of legislatio~t does not cover many individuals who have insurance provided by a self-insured em- ployer. Passage of the Health Insurance Portability and Accountability Act of 1996 (the Kennedy-Kassebaum bill), stating that no one can be denied insur- ance for a preexisting condition,m dem- onstrates growing attention to this prob- lem on a national level. However, its effectiveness in protecting individuals with inherited predispositions still needs to be determined. It is appropriate, therefore, to provide detailed informa- tion to all participants about the poten- tial for insurance discrimination, paying attention to individual circumstances. Employment Discrimination.mPar- ticipants should be informed that genetic testing may pose a risk to their present or future employment. The magnitude of this risk is unknown. Federal legislation provides some protection against dis- crimination in the workplace. The Reha- bilitation Act of 1973 and the Americans With Disabilities Act of 1990 (ADA) pro- lu~it employment discrimination against those with disabilities or against those who are considered to have a disability, tmless the disability precludes perfor- mance of the job.uS.Ira In 1995, the Equal Employment Opportunity Commission issued a ruling that the definition of dis- ability covers individuals at risk for fu- ture health problems based on genetic 117 abnormalities. This interpretation has not been tested in the courts, so the strength of protection provided by the ADA is uncertain. Insurance companies are not regulated under the ADA. While an employer may be prohibited from de- nying employment to the applicant, he or she may limit or exclude insurance cov- erage for the employee. Thus, an indi- vidual may be effectively barred from a job because of limits placed on insurance benefits associated with the employment. Options for Medical "Follow-up. Although some individuals will seek genetic testing primarily to reduce un- certainty and attendant anxiety, the ma- jor motivation for offering and obtaining genetic testing is the hope that medical management following a positive test re- sult can help reduce cancer morbidity and mortality. Participants must be informed of the limits of current knowledge with respect to the treatment, prevention, or early detection of disease based on ge- netic test results. In general, interven- tions to reduce morbidity and mortality attributable to cancer can be broken down into 3 categories: (1) risk reduction, including lifestyle changes and prophy- lactic surgical interventions; (2) early de- tection interventions; and (3) chemopre- vention research on the use of dietary and synthetic agents to block or inter° rupt the process of carcinogenesis. While epidemiologic research has suggested that dietary and lifestyle factors may be as- sodated with risk for some types of can- cer (for e.x~ple, dietary fat intake and colorectal cancer)r the data for specific cancers or specific interventions, even in the general population, are lacldng~ and no such data exist for genetically at-risk individuals."8 Privacy and Confidentiality Medical professionals should not only make every effort to maintain partici- pant confidentiality (see "Maintaining Confidentiality," above), but should ex- plicitly discuss these efforts prior to test- ing. They should describe the mechanisms for keeping information secure, identify people who might have access to the in- formation, and explain that breaches of confidentiality may occur as a result of who is given information by the partici- pant. Participants also will need to con- sider whether they wish to share this information with relatives and other medical professionals. Some relatives may not be aware of their family history, and notifying them of their risk may change their sense of privacy or psychosocial well-being,m Although such disclosure may be important for the relative, it may pose risks to the participant. Tissue Storage and Reuse A unique issue in genetic testing is the ability to store tissue for future test- ing that is unanticipated at the time the sample is obtained. There have been a number of consensus statements on this issue.~m The medical professional should assure participants that no ad- ditional testing on identifiable samples will be done without separate informed consent. Specific consent should be ob- tained from the participants if samples are to be made anonymous and retained for future research. The length of stor- age should also be discussed. Altematives An essential element of the informed consent process is the description of al- ternatives that may be relevant to the participant's decision to pargcipate. One alternative is to forgo testing altogether, since risk assessments can be made on the basis of clinical history without test- ing. Raising this option acknowledges the experimental and discretionary na- ture of testing. Options for obtaining testing through other research proto- cols or clinical programs and the pros and cons of other options should be out- lined. Conversely, medical profession- als should inform individuals if relevant research protocols are available. RECOMMENDATIONS FOR FUTURE RESEARCH ON PRETEST INFORMED CONSENT Genetic research offers substantial promise for the prevention and treat- ment of cancer. At the present time, the emerging ability to test for cancer sus- ceptibility requires that longer-term pro- spective research be conducted to bet- ter define the utility and applicability of genetic testing in higher and lower risk individuals in the population. As a com- ponent of this work, the efficacy of cur- rent preventive and early detection strategies must be assessed in men and women at increased risk. Further, ad- ditional epidemiologic, behavioral, and ethnographic research is needed to bet- ter inform the content and process of pretest education and counseling. We need more answers to questions regard- ing the nature and number of mutations involved (heterogeneity), the cancer risks associated with mutations (pen- etrance), and the types and severity of associated cancers (expressivity). Per- sons with mutations or at high risk should be informed regarding the avail- ability of clinical trials, if appropriate. Research should also pay more atten- tion to understanding psychological re- sponses to testing in terms of their na- ture and timing, as well as how to identify those individuals at greatest risk of se- rions adverse effects. In addition, ef- forts should be made to assess the im- pact of cancer susceptibility testing on family relationships; extent and impact of insurance and employment discrimi- nation; impact of testing children and adolescents, as well as how and whether the informed consent process should dif- fer for children and adolescents; and im- pact of culture-based testing (eg, among Ashkenazi Jews) on perceived social ob- ligation to participate in research pro- tocols, as well as social stigma. We must also evaluate various formats for infor- mation disclosure and counseling includ- ing their provision by different health care professionals. Finally, we must learn whether pretest education and counsel- ing are effective at helping participants understand the meaning of positive and negative test results, whether more ef- fective for some participants than oth- ers, whether and when participants want their providers to make recommenda- tions about testing, and what accounts for differences in efficacy of counseling and style of decision making. 1472 JAMA, May 14, 1997--Vol 277, No. 18 Genetic Testing for Cancer~eller et al

The work of the CGSC was supported by the National Human Genome Research Institute, the National Cancer Institute, the National Institute for Nursing Research, and the National institute for Mental Health. The authors gratefully acknowledge the organi- zation of the CGSC by Elizabeth Thompson, MS, RN; the helpful comments of the other Consortium mem- bers including Deborah Bowen, PhD, Wylie Burke, MD, PhD, Bonnie J. Flick, MD, Judy Garber, MD, MPH, Karen Glanz, PhD, MPH, Ellen R. Gritz, PhD, Don Hadley, MS, Caryn Lerman, PhD, Kenneth Of- fit, MD, MPH, Gloria M. Petersen, PhD, and Kethryn M. Taylor, PhD; and the suggestions of Consor- tium consultants including Ellen Wright Clayton, MD, JD, Nell A. Holtzman, MD, MPH, and Karen Rothenberg, JD. References L Levine AJ. The genetic oMgins of neoplasia JAMA. 1995;273:592. 2. Rowley JD, Aster JC, Sklar J. The clinical ap- plications of new DNA diagnostic technology on the management of cancer patients. JAMA. 1993;270: 2331-2337. 3. Liu B, Parsons RE, Hamilton SR, et al. hMSH2 mutations in hereditary nonpolyposis colorectal can- cer kindreds. Cancer Res. 1994;54:4590-4594. 4. Papadopoulas N, Nicolaldes NC, Wei Y-F, etal. Mutation of a mutL homolog in hereditary colon cancer. Selene. 1994;263:1625-1629. 5. Nicolaides NC, Papadopoulas N, Liu B, et al. Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature. 1994;371:75-80. 6. Mild Y, Swensen J, Shattuck-Eidens D, etal. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science. 1994";266.~)-71. 7. Wooster R, Neuhansen SL, Mangion J, et al. Localization of a breast cancer susceptibility gene, BRC.4~, to chromosome 13q12-13. Scisnce. 1994; 265;2~2090. 8. Malldn D, Li FP, Strong LC, et al. Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas and other neoplasms. Science. 1990;250: 1233-1238. 9. Burke W, Petorsen G, Lynch P, etal. Recom- mendations for follow-up care of individuals with an inherited predisposition to cancer, I: hereditary non- polypesis colon cancer. JAMA. 1997;277:915-919. 10. Burke W, Daly M, Garber J, etal. Recommen- dations for follow-up care of individuals with an inherited predisposition to cancer, If: BRCAJ and BRC.4.~. JAMA. 199"/;277:99"7-10{)3. II. National Advisory Council for Human Genome Research. Statement on use of DNA testing for presymptomaticidentification ofcsncer risk. JAM.d. 1994;271.'785. 12. Garber JE, Schrag D. Testing for cancer sus- ceptibdity. J.4M~ 1996;275:1928-1929. 13. Li FP, Gather JE, Friend SH, et al. Recom- mendations on predictive testing for germ line mutations among cancer-prone individuala J Cancer I~st. 1992;84:1156-1160. 14. Lerrfmn C, Narod S, Schulman K, et al. BRCAI testing in families with hereditery breast-ovarian cancer, a prospective study of patient decision mak- ing and outcomes. J.4J~.4. 1996;275:1885-1892. 15. Statement of the American Society of Human Genetics on genetic testing for breast and ovarian cancer predisposition. Am J Hur~ ~ 1994;55~-iVo 16. Presymptomatic genetic testing for heritable breast cancer risk. Press release of the National Breast Cancer Coalition. Washington, DC, Septem- ber 28, 1995. 17. Kahn P. Coming to grips with genes and risk. Science. 1996;274:496-498. 18. Geller G, Kass NE. Informed consent in the conte~ of prenatal HIV screening. In: Faden RR, Geller G, Powers GM, eds. ~k[DS, Women a~r~ the ~Ve~ Gestation: To~zrc~ a Mo~zll~ Accep~zble Public Po/ic~/ on HI, Screening of Pre~uz~zt Wom- en and Ne~,bor~s. New York, NY: Oxford Univer~ sity Press; 1991. 19. Garber JE, Patenaude AF. Ethical, social and counselling isanns in hereditary cancer susceptibil- ity. Cancer S,,rv. 1995;25:381-397. 29. Lippman A. Prenatal genetic testing and screen- ing: constructing needs and reinforcing inequities. Am $ Law Med. 1991;17:15-50. 21. Hofman KA, Tambor ES. Chase GA, Geiler G, Faden RR, Holtcman NA. Physicians' knowledge of genetics and genetic tests. Acad Med. 1993;68: 22. Lee JM. Screening and informed consent. N gngl J Med. 1993;328:438-440. 23. Holtzman NA. Are we ready to screen for in- herited susceptibility to cancer? O~co/ogy. 1996; 1(P.57-64. 24. ASHG/ACMG Report. Points to consider, ethi- cal, legal and psychosocial implications of genetic testing in children and adolescents. Am ? Hum Cr~aet. 1995;57:1233-1241. 25. Appelbaum PS, Lidz CW, Meisel A. lnformed Co~sen~. Legal Theurtl and Clinical Practice. New York, NY: Oxford University Press; 1987. 26. Faden RR, Beauchamp TL, King NP. A H/s- tor~ and Theory of Informed Con.sent. New York, NY: Oxford University Press, 1986:86. 27. Elias S, Annas GJ. Generic consent for genetic screen£ng. N Engl J Med. 1994;330:1611-1613. 28. Andrews LB, Fullarton JE, Holtzman NA, Motulsky AG, eds. Assessing Genetic Ri~k~: plicatia~s for Health and Social Policy. Washing° ton, DC: National Academy Press; 1994-~259-282. 29. Press N, Browner CH. Risk, autonomy, and respons~ility:, informed consent for prenatal test- ing. Ha~tings Cent Rep. 1995;25:$9-$12. 30. Marteau TM. Towards i.nformed derisions about prenatal testing: a review./b'enat Diag,. 1996;14: 1215-1226. 31. Holtzman NA, Faden RR, Chwalow JA, Horn SD. Effect of informed consent on mothers' knowl- edge of newborn screening. Pediatrics. 1983;72:807- 812. 32. Sharp NF. Informed consent and Huntington disease: a model for communication. Am J Med Ge~*et. 1994;50"239-246. 33. President's Commission for the Protection of Human Subjects of Biomedical and Behavioral Re- search. The Belmwnt Report: Ethical Principles and Guidelines for the Protectia~ of Hu~nan Sub- jeers of Research. Washington, DC: Public Health Service; 1979. 34. President's Commission for the Study of Ethi- cal Problems in Medicine and Biomedical and Be- havioral Research. The Ethical and Legal Impli- cati~ts of Info~-med Cocsent in the Patit~t- Practitioner Relationship. Washington, DC: US Government Printing Office; 1992;1:74-76. 35. Andrews LB. Regulation ofgenntic research in i~um~ns. In: Medical Genetics: A Legal Fruntier. Chicago, Ill: American Bar Foundation; 1987:38-41. 36. Office of Protection From Research Risim. Pro- ~scting Human Rescazct~ Subjects: lcstitutional Re~/ew Board Gu/debook. Bethesda, Md: National Institutes of Health; 1993:42-63. 37. Kent G. Shared understandings for informed consent~ the relevance of psycbological research on the provision of information. Soc Sci Me& 1996;43: 1517-1523. 38. Katz J. The Silent World of Doctor and Pat~nt. New York, NY: Free Press; 1984. 39. Mazur DJ. Why the goals of informed consent are not realized: treatise on informed consent for the primary care physician. J Gen I~ztern Med. 1988; 3",370~379. 40. Lidz CW, Meisel A, Osterweis M, Holden JL, .'darx JH, Mnnetz MR. Barriers to informed con° sent. Ann Intern Meal. 1983;99:539-543. 41. Ketz J. Why doctors don't disclose uncertainty. Hctstings Cent Rep. 1984;38:35-44. 42. Taylor KM, Keiner M. Informed consent: the phy- sichn's perspective Soc Sc/,kred. 1~87;24:135-143. 43. Carter MA. Patiant-provider relatiouship in the context of genetic testing for hereditary cancers. J ,Vat/Cancer [nst Mo~zog~r. 1995;17:119-121. 44. Geller G, Tambor ES, Chase GA, Hofman ICT, Faden RR, Holtzman NA. Incorporation of genet- ics into primary care practice: will physicians do the counseling and will they be directive? Arcl~ Faro ,'d'ed. 1993;2:1119-1125. 45. Lidz CW, Applebaum P, Meisel A. Two models ofimpiementinginformed consent. Arch Inter~ Med. 1988;143:1385-1389. 46. Asch A, Geller G. Feminism. bioethics and ge- netics. In: Wolf S, ed. Feminism a~td Bioethics: Beyond Rep~ductian. New York, NY: Oxford Uni- versity Press; 1996. 47. Betancourt H, Lopez SR. The study of culture, ethnicity, and race in American psychology. Am PsychoL 1993;43:629-637. 48. Gest£n LO. Informed consent, cultural sensi- tivity, and respect for persons. JAMA. ~995;274: 844-845. 49. Carrese J, Rhodes LA. Western bioethics on the Navajo reservation: benefit or harm? JAMA. 1995~74-~25-~9. 50. Geller G, Bernhardt BA, He[zlsouer KA, Holtz- man NA, Stefanek MM, Wilcox PM. Informed con- sent and ~RC~k/testing. Nat C-enet. 1995;11:364. 51. Wang V, Marsh FH. Ethical principles and cul- tural integrity in health care delivery: Asian eth- nocultural perspectives in genetic services. J Genet Co~. 1992;1"~1-92. 22. Wang WO. Cultural competency in genetic couno sellng. J C-e~t Coup. 1994;3".')67o277. 53. Punales-Morejon D, Rapp R. Ethnocultural di- versity and genetic counseling training: the chal- lenge for a twenty-first century. J Gene~ Co~s. 1993;2:155-158. 54. Well J, Mittman L A teaching framework for cross-cultural genetic counseling. J Geezer Co~n~. 1993;2:159-169. 55. Dixeon B, Dang V, Cleveland JO, Peterson RM. An education program to overcome language and cultural barriers to genetic services. J Genet Cou~.s. 1992;1.')67-274. 56. Botkin J, Croyle RT, Smith KR, etal. A mode! protocol for evaluating the behavioral and psycho- social effects of BRC.4/ testing. J" Nat~ Cancer 57. Resoff AJ. [~formed Co~e~t: A ~u~e for Health Care ~. Rockvllle. Md: Aspen Pub- licatinns; 1981. 58. Bernhardt BA, Geller G, Strauss M, et al. To- wards a model informed consent process: a quali- tative assessment of women's attitudes about ge- netic testing for breast cancer risk. J Ge~et Coups. In press. 59. Geller G, Strauss M, Bernhardt BA, Holtzman NA. 'Decoding' informed consent: insighte from women regarding genetic testing for breast cancer susceptibility. Hasti~@s Cent Rep. 1997;27"2~33. 60. Singer GGS. Clarifying the duties and goals of genetic counselors: implications for nondirective- hess. [m Gert B, Berger EM, Cahill GF, etal, eds. d~m~s ami Hea/th Care Pro~/~r~. Sudbury, Mass: Jones & Bartlett Publishors [n~ 1996. 61. Allen KD, Hedges ED, Knudsen SK. Compar- ing four methods to inform parents about child behavior m~gement: how to inform for consent. Pediatr Dent. 1995;17:180-186. 62. Muss HB, W~te DR, Michieintte R, etal. Writ- ten informed consent in patients with breast can- cer. Cancer. 19q9;43:1549-155~ 63. Dodd M J, Mood DW. Chemotherapy: helping patients to know the drugs they are receiving and their possible side effects. Cancer ~urs. 1981;4: 311-318. 64. Silva MC, Sorrell JM. Enhancing comprehen- sion of information for informed consent: a review of emph-dcal research. [RB. 1988;10:1-5. 65. Faden RR, Chwalow AJ, Orel-Crosby E, Holtz- man NA, Chase GA, Leonard CO. What partici- pants understand about a maternal serum alpha fetoprotein screening progra~ ~ J Pub/ic Health. 1938;75:1381-1384. 66. Barbeur GL, Blumenkrants .V_J. Videotape aids hu'ormed consent decision. JAMA 1978;24~741-2742. 67. Westreich L, Levine S, Ginsburg P, Wilets L Patient knowledge about eleotroconvulsive therapy:. effect of an informational video. Co~z~ls T~zer. 1995; I1"~32-37. JAMA, May 14, 1997--Vol 277, No. 18 Genetic Testing for Cancer--Geller et al 1473

68. Kasper JF, Mulley AG, Weunberg JE. Deveh oping shared decision-making prograr~ to improve the quality of health care. QRB Qual Rev Bull. 1992:18:183-190. 69. Hopper KD, Zajdel M, Hulse SF, et al. Inter- active method of informing patients of the risks of intravenous contrast media. Radiology. 1994;192: 67-71. 70. Wetetene SL, Sheehan TJ, Votaw RS, Peter- son biG, Rothfleld N. Evaluation of a computer be~-*d education leseon for patients with rheuma- toid arthriti~. Y Rheumatol. 1925;12~907-912. 71. Smith Jbi, Pike RA, Jalmbowski B, Pardoe JR. Survey of computer programs for diabetes man- agement and education. D/abetes Edac. 1988;14: 412-415. 72. Kumar NB, Bostow DE, Schapira DV, Kritch KM. Efficacy ofintersctive, automated programmed instruction in nutrition education for cancer pre- vention. J Cancer Edue. 1993;8"~03-211. 73. Gustafsen DH, Bosworth K, Hawkins RP, Boberg EW, Bricker E. CHESS: a computer-based system for providing information, referrals, deci- sion support and social support to people facing medical and other health related crises. In: Firose ME, ed. Sizteenth Annual Symposium on Com- puter Application~ in Medical Care. New York, NY: McGraw-Hill; 1993:161-165. 74. Slack WV, Hicks GP, Reed CE, Van Curs LJ. A computer-bnsed med/cal history system. N Engl J Med. 1966;274:194-198. 75. Slack WV, Slack CW. Pationt-compater dia- logue. N Engl J Med. 1972;286:1304-1309. 76. Tibbles L, Lewis C, Reisine S, Rippoy R, Don- aid M. Computer-assisted instruction for preopera- tive and postoperative patient education in joint replacement surgery. Camput Nuts. 1992;10"~08- 212. 77. Leirer VO, Morrow DG, Pariante GM, Sheikh JL Elders' nonadherence, its assessment, and corn- pater-assisted instruction for medication recall train- ing. J Am Gerta~r Soc. 1988;36:877-884. 78. Slack W. Patient counseling by computer. In: Zoog S, Yarnall S, eds. The Changing Health Care Teaz~" Improving Effsc~ivcn~e in Patient Care. Seattle, Wash: biedical Commun/cations and Ser- vices Assoc; 1976:108-111. 79. Ghosh A, Greist JH. Computer treatment in p~ychlatry. P~ychta~r Ann. 1988;1"8:246-250. 80. Kahn G. Computer-based patient education: a progress report. MD Comput. 1993;10:93-99. 81. Gillispie MA, Ellis LBM. Computer-based pa- tient education revisited. JMed Sy~t 1993;17:119- 125. 82. Kinzie biB, Schorling JB, Siegel bi. Prenatal alcohol education for low-income women with in- teractive multimedi~ Patient Educ Count. 1993; 21:51-60. 83. Kulik CC, Kulik JA. Effectiveness ofcomputer- based instruction: an updated analysis. Comp~t Hum Behav. 1991;7ff5-94. 84. Frice ttV. Computer-Aided Ins~" A Guid~ for Authors. Pacific Grove, Calif. Brooks/Cole Pub- li~hing Co; 1991. 8~. Blackwell bi, Niemiec R, Walberg H. CAI can be doubly effective. Phi Delta Kappa. 1986,~7:754N 751. 86. Philipson SJ, Doyle MA, Gabcam SGA, Night- ingale C, Phil/pson EH. Informed consent for re- search: a study to evaluate readability and pro- cessability to effect change. J.lnvest Med. 1995;43: 459-467. 87. Hopper KD, TenHave TR, Hartzel J. Informed consent forms for clinical and research imaging pro- cedure~ how much do patients understand? A JR Am J Roent#enoL 1995;164:493-496. 88. Grondnsr TM. On the readability of surgical consent form~. N E~gl J Med. 1980;302:9~0-902. 8~. MeadeCD, Howser DM.Consent forms: how to determine and improve their readability. Oncel Nuts Forum. 1992;19:1523-1528. 90. Doak CC, Doak LG, Root JH. Teaching Pa- tients With Low Literacy Skills. Philadelphia, Pa: JB Lippincott; 1996. 91. White CS, Mason AC, Feehan M, Templeton PA. Informed consent for percutaneons lung bi- opsy: comparison of two consent protocols based on patient recali after the procedure. A JR Am J Roent- genoL 1995;165:1139-1142. 92. Morrow G, Gootnick J, Schmale A. A simple technique for increasing cancer patients' knowl- edge of informed consent to treatment. Cancer. 1978;42.q93-799. 93. Bernhardt BA, Chase GA, Faden RR, et al. Educating patients about cystic fibrosis carrier screening in a primary care setting. Arch Faro Med. 1996;5:336-340. 94. Miller R, Willner HS. The two-part consent form: a suggestion for promoting free and informed consent. N Engl J Med. 1974;290:964-965. 95. Taub HA, Baker MT. A reevaluation ofinformed consent in the elderly, a method for improving com- prehension through direct testing. Clin Res. 1984; 32:17-21. 96. Early CL, Strong LC. Certificate of confiden- tiality: a valuable tool for protecting genetic data. Am J Hum Genet. 1995;57:727-731. 97. Biesecker BB, Boehnke M, Calzone K, et al. Genetic counseling for families with inherited sus- cepta~ility to breast and ovarian cancer. JAMA. 1993;269:.1970-1974. 98. Smith JR, Freije D, Carpten JD, et al. Major susceptibility locus for prostate cancer on chromo- some 1 suggested by a genome-wide search. ence. 1996;274:1371-1374. 99. Croyle R, Smith K, Botkin J, Baty B, Nash J. Psychological responses to BRCA1 mutation test- ing:. preliminary findings. Health Psyclml. 1997;16: 63-72. 100, Wiggins S, Whyte P, Huggins M, et al. The l~ychological consequences of predictive testing for Huntington's disease. N Engl J Med. 1992;327: 1401-1405. 101. Huggins M, Bloch M, Wiggins S, et al. Pre- dictive testing for Huntington disease in Canada" adverse effects and unexpected results in those receiving a decreased risk. Am JMed Genet. 1992; 43:5~8-515. 102. Lynch HT, Watson P, Conway TA, et al. DNA ~reaning for breast/ovarian cancer susceptibility on linked markers: a family study. Arc/t Intern Meal. 1993;153:1979-1987. 103. Lerman C, Daly M, Masny A, Balshem A. At- titudes about genetic testing for brcast-~varian can- cer susceptibility. J Clin OncoL 1994;1"2:843-850. 104. US Congress, Office of Technology Assess- ment~ Genetic Tests and Health [n.~urance: Results of a Survey: Background Paper. Washington, DC: US Government Printing Office; October 1992. Pub- lication OTA-BP-BA-93. 106. NIH-DOE Working Group on Ethical, Legal, and Social Implications of Human Genome Research. Cr~etic [~formation and Health I~umnce: Report of the Task Force on Genetic Information and Health In~ra~,ce. Washington, DC: US Dept of Health and Human Services: May 19~. NIH publication 93-368~ 106. Brown RS, Marshall K, eds. Adt~ances in netic I~for~ation." A Guide for State Policy Mak- ere. 2nd ed. Lexington, Ky: Council of State Gov- ernments; 1993. 107. American Council of Life Insurance and The Health Insurance Assoc/ation of America. Report of the ACLI.HIAA Ta~k Force on Genetic Testing. Washington. DC: American Council of Life Insar- ance; 1991. 108. Hudson KL, Rothernborg KH. Andrews LB, Kahn M J, Collins F. Genetic discrimination and health insurance: an urgent need for reform. ence. 1995;270:391-393. 109. 0strer H, Allen W, Crandall LA, et al. Insur- ance and genetic testing', where are we now? Am J Hum Genet. 1993;52:565-577. 110. Rothstein M& Genetics insurance, and the ethics of genetic counseling. In: Molecular Genetic Medicine. Vol 3. Orlando, Fla: Academic Press Inc; 1993:159-177. III. Health Insurance Portability and Accountabil- ity Act of 1996, Pub L No. 104-191, 110 Star 1936. 112. Hereditary Susceptibility Working Group of the National Action Plan on Breast Cancer. Com- mentary on ASCO (American Society of Clinical 0neology) statement on genetic testing for cancer susceptibility. J Clin OncoL 1996;14:1738-1740. 113. American Society of Clinical 0ncology. netic testing for cancer susceptibility. J Clin OncoL 1996;14:1730-1736. 114. Lapham EV, Kozma C, Weiss JO. Genetic dis- crimination: perspectives of consumer~. Sc/encs. 1996;274:621-624. 115. McEweh JE, Reilly PR. State legislative ef- forts to regulate use and potential misuse of genetic information. Am J Hum C~net. 1992;51:687-647. 116. Natowitz MR, Alper JIL Alper JS. Genetic discrim/nation and the Americans With Disabilities Act. Am J Hum Genet. 1992;51:895-897. 117. Equal Employment Opport unity Commission. Campliance Manual, vol 1. Washington, DC: Equal Employment Oppermnity Commission; biaroh 14, 1995. EEOC order 915.002, section 902. 118. Nixon DW. Status of cancer prevention clini- caltrials. De Vita VT Jr, Hellman S, R osenberg SA, eds. Cancer Prez, entio~ Update. Philadelphia, P~" JB IAppincott; 1991:1-9. 119. Winter PR, Wiesner GL, Finnegan J, et al. Notification of a family history of breast cancer:. issues of privacy and confidentiality. Am J Med C, enet. 1996;66:1-6. 120. ASHG Board of Directors Report. Statement on informed consent for genetic research. Am J Hum Genet. 1996;59:.471-474. 121. Clayton EW, Steinberg KK, Khoury M J, et al. Informed consent for genetic researoh on stored tissue samples. JAMA. 1995;274:1786-1792. 122. ACMG (American College of Medical Genet- ics) Storage of Genetic Materials Committee. State- ment on storage and use of genetic materials. Am J Hum Gen~t. 1995;57:1499-1500. 0 1474 JAMA, May 14, 1997--Vol 277, No. 18 Genetic Testing for Cancer~eller et al