Effect of family history on disclosure patterns of cystic fibrosis carrier status.код для вставкиСкачать
American Journal of Medical Genetics Part C (Semin. Med. Genet.) 119C:70 – 77 (2003) A R T I C L E Effect of Family History on Disclosure Patterns of Cystic Fibrosis Carrier Status KELLY E. ORMOND,* PATTI L. MILLS, LUCILLE A. LESTER, AND LAINIE F. ROSS As general population screening becomes more common, an increasing number of cystic fibrosis (CF) carriers will be identified who do not have a family history of CF. Whether these carriers inform their relatives of their carrier status and whether their relatives are motivated to pursue carrier screening is unknown. We surveyed CF carriers with and without a family history of CF to understand whether and how information dissemination patterns differ, why information is or is not shared, and to what extent relatives are known to undergo testing. CF carriers were identified from a general population carrier screening clinic (group B ¼ 18) or were parents of affected children followed at a CF clinic (group A ¼ 30). CF carriers with a family history told essentially 100% of their living parents, siblings, and halfsiblings, while those without a family history told 84% of living parents and 56% of siblings (P < 0.05). Despite the high rate of information dissemination in both groups, few siblings were known to have undergone carrier screening (14/74). Significantly fewer second- and third-degree relatives were informed about carrier status or were known to have undergone carrier screening. Group A was more likely to inform second- and third-degree relatives about carrier status. Our study documents that the frequency and reasons for disclosing CF carrier status differ between individuals with and without a family history of CF despite the fact that the reproductive risks for their relatives are the same. ß 2003 Wiley-Liss, Inc. KEY WORDS: cystic fibrosis; disclosure; genetic screening; prenatal screening; mass screening; family relations; obligate carrier; carrier testing; heterozygote; heterozygote detection INTRODUCTION Historically, most reproductive genetic risk counseling for cystic fibrosis (CF) was offered based on family history. In the early 1990s, CF mutation analysis became available to individuals with a family history, and since the late 1990s, Kelly E. Ormond, MS, is an Assistant in the Department of Obstetrics and Gynecology, Section of Reproductive Genetics, at Northwestern University, as well as a fellow at the MacLean Center for Medical Ethics at the University of Chicago, both in Chicago, Illinois. Patti L. Mills, MS, is a genetic counselor in the Department of Human Genetics at the University of Chicago in Chicago, Illinois. Lucille A. Lester, M.D., is a Professor in the Department of Pediatrics at Rush Presbyterian St. Lukes Medical Center in Chicago, Illinois. Lainie F. Ross, M.D., Ph.D., is an Associate Professor in the Department of Pediatrics and Assistant Director of the MacLean Center for Medical Ethics, both at the University of Chicago in Chicago, Illinois. *Correspondence to: Kelly E. Ormond, 76 N. Sr. Cair #1280, Chicago, IL 60611. E-mail: Kfirstname.lastname@example.org DOI 10.1002/ajmg.c.10008 ß 2003 Wiley-Liss, Inc. genetic carrier testing has been available to a wider population. In 1997, a National Institutes of Health (NIH) consensus statement recommended offering CF carrier screening to Caucasian individuals of reproductive age regardless of family history [National Institutes of Health, 1997]. In 2001, the American College of Obstetrics and Gynecology (ACOG) and the American College of Medical Genetics (ACMG) jointly issued a similar recommendation [American College of Obstetrics and Gynecology and American College of Medical Genetics, 2001; Grody et al., 2001]. One of the main barriers to carrier testing is lack of public awareness about CF and what CF carrier testing One of the main barriers to carrier testing is lack of public awareness about CF and what CF carrier testing can and cannot do. can and cannot do. While the consensus statements seek to detect CF carriers in the general population, first- and second-degree relatives of those already known to be CF carriers are themselves at higher risk of being carriers. In order to counsel these patients accurately about their risk of being a CF carrier, they must be aware of their relative’s CF carrier status and inform the clinician of this information. Barriers to information dissemination include lack of information about CF [Suhr et al., 1994; Fanos and Johnson, 1995a]; the perception that relatives would not be interested in carrier screening [Suhr et al., 1994; Sorenson et al., 1996; Williams and Schutte, 1997]; and gender (studies show that men are less likely than women to inform family members of their genetic risks [Callanan et al., 1995; Fanos and Johnson, 1995a]). Even when individuals know that they are at increased risk for carrying a genetic abnormality, many choose not to undergo testing. In studies that assess individuals at risk for carrying a disease mutation or a balanced chromosome translocation based on family history, ARTICLE testing uptake has been low [Ayme et al., 1993; Suhr et al., 1994], and first-degree relatives are significantly more likely to undergo carrier testing than more distant (second- and third-degree) relatives [Suslak et al., 1985]. Studies that specifically addressed the interest in and acceptance of CF screening in individuals with a family history of CF [Callanan et al., 1995; Sorenson et al., 1997; Henneman et al., 2001] and in nonpregnant individuals in the general population [Clayton et al., 1996] demonstrate significantly lower screening uptake by individuals without a family history of CF. Awareness of and personal Studies that specifically addressed the interest in and acceptance of CF screening in individuals with a family history of CF and in nonpregnant individuals in the general population demonstrate significantly lower screening uptake by individuals without a family history of CF. experience with CF has been associated with both increased [Sorenson et al., 1997; Wake et al., 1997] and decreased [Botkin and Alemagno, 1992] uptake of carrier screening, suggesting that the perceived burden of CF may play a role in whether testing is pursued [Fanos and Johnson, 1995b]. There may also be logistical barriers to uptake of testing such as cost, accessibility to facilities offering testing, the desire to avoid phlebotomy, and family myths about who is at risk for carrying a disease mutation [Suhr et al., 1994; Fanos and Johnson, 1995b; Clayton et al., 1996; Sorenson et al., 1997]. Given the current NIH, ACMG, and ACOG recommendations, more CF carriers without a family history are likely to be detected. Our study addresses whether the presence of a AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) family history of CF affects carrier status information dissemination patterns and the uptake of carrier testing in at-risk relatives; and which factors facilitate or hinder information dissemination. MATERIALS AND METHODS Subject Recruitment Group A consisted of obligate carriers identified through a CF clinic population at Rush Presbyterian St. Lukes Medical Center (Rush/St. Lukes) in Chicago; both mothers and fathers of children 2–10 years in age with CF were recruited. Since most patients are diagnosed in the first year of life, the lower limit of 2 years was chosen to ensure that parents had sufficient opportunity to disclose CF carrier information to their relatives. The upper limit of 10 years was chosen because it is only within the last 10 years that genetic testing has been available and routinely offered to families in the context of a new diagnosis of CF in a family member. Fifty families were included in this study population, and assuming intact families, questionnaires were mailed for both parents to complete individually. These families had been counseled about the reproductive risks to themselves and their relatives by a pulmonologist (L.A.L.) or by a genetic counselor associated with the CF clinic. Group B consisted of CF carriers identified through general population screening at Northwestern Memorial Hospital (NMH) in Chicago between January 1998 and August 2000. During that time, NMH offered CF screening to all Caucasian prenatal patients seen for genetic counseling (e.g., preprocedure amniocentesis/CVS counseling); to all patients requesting CF screening; and to individuals of Ashkenazi Jewish (AJ) ancestry as part of a broader panel of screening tests. All patients seen for carrier screening received a short genetic counseling session (by K.E.O. or another genetic counselor) prior to screening and a follow-up letter with their results. Those determined to be carriers were also counseled about the reproductive risks to themselves and to relatives and received a generic letter to share with 71 family members regarding their carrier risk and the availability of screening. Between January 1998 and August 2000, approximately 350 individuals were screened for CF and 850 were screened for the AJ panel. Inclusion criteria for this study included individuals who were found to be CF carriers greater than 6 months prior to the beginning of the study period. Fiftyone individuals, including two at risk couples (neither of the women was pregnant at the time of our study) and seven individuals with a family history of CF carriers but no affected relatives met the study criteria. Subjects in both groups were limited to English-speaking adults over 18 years old. Questionnaire All participants were asked to complete a written survey, which consisted of questions about demographic information and questions to ascertain the total number of first-, second-, and thirddegree relatives, their maternal/paternal relationships to the study subject, and their relatives’ gender; whether or not study subjects informed their first-, second-, and third-degree relatives about their own carrier status; the rate at which they did so (e.g., the time period that passed between the subjects learning of their own carrier status and informing their relatives); and their awareness of whether the informed relatives sought testing. In order to assess the reasons for disclosing carrier status and barriers to information dissemination, subjects were asked either to select from a list of reasons for sharing or not sharing carrier status information (Table I) or to describe their reasons in their own words. Approximately onethird to half of the participants provided unlisted reasons, which were coded to correspond to listed reasons; new categories were created for unlisted reasons. The questionnaire was mailed in February 2001 and remailed to nonresponders 1 month later. A third mailing was sent to families in group A if neither parent had responded by June 2001. Institutional review boards at all 72 AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) ARTICLE TABLE I. Possible Factors Involved in Disclosing CF Carrier Status to Relatives The age of the relative The relative’s level of maturity or ability to understand the information and its implications The relative’s gender Whether or not the relative lives nearby or if you talk frequently Your social relationship with that relative (whether or not you are comfortable having personal conversations with the relative) Concerns about privacy Concerns about financial discrimination (related to insurability, employability) Your own feelings about being a carrier Your feelings about the severity of CF Whether or not you thought it would be helpful for you to talk about your CF carrier status (i.e., to help you cope or sort out your own feelings about being a carrier) Concern about what the relative might think of you Concern about causing a relative anxiety or stress about his/her own health or carrier status Concern about causing a relative anxiety or stress about his/her own chance of having an affected child Concern that the relative or his/her child(ren) could have an affected baby Your beliefs about prenatal testing and/or pregnancy termination Your relatives’ beliefs about prenatal testing and/or pregnancy termination Recommendations from your/your child’s doctor or genetic counselor Recommendations from your social support network (e.g., close friends or family members, spiritual/religious advisors) Medical issues related to a child affected with CFa Forgot or did not think of ita Someone else (e.g., parent) disseminated relevant informationa a Categories created to incorporate responses subjects provided in their own words. three participating institutions approved the study design as described. Analysis SPSS 10 was used to perform descriptive statistics, independent t-tests, one-way ANOVA, and Pearson chi-square values as appropriate; values greater than P ¼ 0.05 were considered significant. Because disclosure to relatives did not vary significantly by the subjects’ or relatives’ gender, the categories of familial relationships were combined (e.g., sisters and brothers were recorded as siblings) to provide more power to the analyses. Carrier information dissemination patterns between groups A and B were compared. Based on previous literature, age, gender, experience with CF, and perceived severity of CF were considered as potentially confounding factors in the analysis. RESULTS Thirty-five surveys were returned from group A (parents of children with CF), with five responses excluded because either a parent (n ¼ 2) or the affected child (n ¼ 2) was adopted, or because the survey was not adequately completed (n ¼ 1). The total response rate was 30/95 (33%). For group B, 51 surveys were sent, but 3 surveys were returned as undeliverable. Of the remaining 48, 18 were returned (38%). Therefore, a total of 48 valid surveys make up the study data: 30 from group A (family history of an individual affected with CF) and 18 from group B (no family history of an individual affected with CF). Subject demographics are reported in Table II. There were no significant differences between the two groups with regards to age, gender, or percent in a married or committed relationship. There were, however, significant differences between the groups with regards to educational level and religious affiliation. With regards to education, 85% of the overall population had at least a college education, with fewer in group A reporting college education or higher (t-test, P ¼ 0.001). Religious affiliation was notable for a high percentage of group B reporting Jewish affiliation (65%), while group A was almost exclusively of Christian affiliation, with no respondents of Jewish descent. Consistent with our study design, 100% of group A reported knowing someone affected with CF; somewhat surprisingly, no individual in the general population screening group knew anyone with CF (chi-square, P < 0.001). On a Likert scale of 1 (not severe) to 4 (severe), 48% of respondents stated that CF was severe, 30% moderate, 18% mild, and 2% not severe. There were significant differences between groups such that all individuals in group B (who did not know an individual with CF) ranked CF as moderate to severe (mean, 3.71), while individuals in group A, with a family history of CF, reported a broader range of perceived severity (mean, 3.00; ANOVA, P ¼ 0.004). Because of the direct association between group affiliation and knowledge of a person affected with CF, it was impossible to determine whether the family history or overall knowledge of ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) 73 TABLE II. Subject Characteristics Age Gender (% female) Marital status, (% married/in a committed relationship) Educational level (% with college or postgraduate education)a % who knew someone with CFa Average perceived severity CF (1, not severe; 4, severe)a % who would consider prenatal diagnosis for CF a Group A, family history (n ¼ 30) Group B, no family history (n ¼ 18) Total population (n ¼ 48) 38.27 63% 94% 74% 100% 3.0 50% 35.5 66% 100% 100% 0% 3.71 100% 37.42 65% 92% 85% 60% 3.27 70% P < 0.05. CF was the major factor underlying the difference in perceived sev erity of CF. Seventy percent of respondents stated they would consider prenatal diagnosis for CF; 6% stated they were against it; 14% stated that testing should be available, although they would not undergo testing themselves; 8% were undecided about prenatal diagnosis for CF; and 2% had not thought about it. There were significant differences between groups, with 100% of the general population screening group stating that they would consider prenatal diagnosis for CF, versus approximately 50% of the obligate carrier group (P ¼ 0.004). Frequency and Rate of Dissemination of CF Carrier Status Information to Relatives and Awareness of Relatives’ Decisions to Undergo Testing Living relatives of study participants included a total of 287 first-degree relatives (71 parents, 110 siblings, 12 half-siblings, 94 children), 186 second-degree relatives, and 529 third-degree relatives. Table III lists the percentage of relatives in each category who were informed about carrier status. Across all subject groups and categories of relatives, the gender of the CF carrier or the relatives did not impact information dissemination patterns or uptake of carrier screening in relatives, and therefore several of the gender-related categories were combined to add power to statistical analyses. CF carriers with a family history told virtually 100% of their living parents, siblings, and half-siblings, while carriers with no family history told 84% of living parents and 56% of siblings. Fathers CF carriers with a family history told 100% of their living parents, siblings, and half-siblings, while carriers with no family history told 84% of living parents and 56% of siblings. (t-test, P ¼ 0.05), full sisters (P ¼ 0.045), and full brothers (P ¼ 0.004) were informed more quickly by carriers who had an affected child. Regardless of the high frequency of information dissemination in both groups, few siblings were known to have undergone carrier screening (14/122; 11.5%). One-third of children at risk for being carriers reportedly had been informed of their parent’s carrier status. For all children not informed of their CF carrier risk, the primary reason was age; presumably their parents felt they were too young to understand the significance of the information. TABLE III. Disclosure of Carrier Status to Living Relatives % informed Relative a Parents Full siblingsa Half-siblings Maternal aunts/uncleb Paternal aunts/unclesc Total aunts/unclesc Maternal cousinsc Paternal cousinsc Total cousinsc a Group A, family history d 45/46 (98%) 78/78 (100%) 11/12 (92%) 55/71 (78%) 43/73 (58%) 98/144 (68%) 110/200 (55%) 99/218 (45%) 209/418 (50%) Group B, no family history 21/25 (84%) 18/32 (56%) N/A 6/19 (32%) 3/23 (13%) 9/42 (21%) 1/37 (3%)e 2/74 (3%)e 3/111 (3%)e P < 0.05. P < 0.01. c P < 0.001. d One individual was adopted out of the family and has had little contact with his/her biological father. e Several respondents did not list a number of cousins, but simply stated ‘‘too many’’; therefore, these numbers are possibly an overestimation of the percentage of informed cousins. b 74 AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) There were significant differences between groups A and B in the frequency of informing second-degree relatives (68% vs. 21%, respectively; chi-square, P < 0.001) and third-degree relatives (50% vs. 3%; P < 0.001) about carrier status. When potential confounding variables were considered, including education, gender, perceived severity of CF, and interest in prenatal diagnosis, only education was a significant factor in dissemination of carrier status to both second- (ANOVA, P ¼ 0.001) and third-degree relatives (P < 0.001). For second-degree relatives, slightly more maternal relatives than paternal relatives were informed (58% vs. 47%); no such differences were noted for third-degree relatives. In general, carriers told either everyone or no one in a family segment (e.g., all maternal aunts/ uncles/cousins were told, but no paternal relatives were told). Information was provided more slowly to more genetically distant relatives such as thirddegree relatives (e.g., an increase in the time between diagnosis and disclosure; Table IV). For both groups, few second-degree (6/186, or 3.2%) and third-degree relatives (13/539, or 2.4%) ARTICLE were known to have undergone carrier screening. Reasons for Disclosing CF Carrier Status Participants did or did not disclose CF carrier status for a number of reasons. Individuals provided between 0–7 reasons for disclosure and 0–2 responses for nondisclosure. The most common reason for disclosing information was a close social relationship with the relative, as well as the need for social support in a time of crisis (whether in the context of the diagnosis of an affected child or the stressful situation of dealing with newly discovered carrier status in the midst of pregnancy). The most common reason for disclosing information was a close social relationship with the relative, as well as the need for social support in a time of crisis. TABLE IV. Time to Disseminate Carrier Status Disclosure to whom (n ¼ total no. informed)a Mother (n ¼ 37) Father (n ¼ 29) Full sisters (n ¼ 32) Full brothers (n ¼ 25) Half-siblings (n ¼ 11) Maternal aunts (n ¼ 32) Maternal uncles (n ¼ 24) Paternal aunts (n ¼ 25) Paternal uncles (n ¼ 17) Maternal female cousins (n ¼ 43) Maternal male cousins (n ¼ 56) Paternal female cousins (n ¼ 45) Paternal male cousins (n ¼ 59) a Time between diagnosis and disclosure As soon as possible % <6 months 97% 94% 97% 92% 100% 69% 66% 69% 50% 70% 62% 67% 71% 3% 3% 3% 8% 31% 33% 25% 40% 24% 31% 20% 21% % >6 months 3% 6% 10% 6% 6% 13% 7% Numbers may slightly underestimate numbers of individuals informed during each time segment, as some respondents did not specify the number of relatives informed during the time period. Responses such as ‘‘I am very close to all four of my siblings. I told them for many reasons. One being that I needed their support’’; ‘‘We are a really close family; everyone was told’’; and ‘‘We were close—I did not hesitate’’ suggest that social closeness with relatives was a common reason for disclosing results across all relative categories. Rarely, social and/or geographical distance was cited as a reason for not disclosing carrier status: ‘‘It’s just that I haven’t spoken to some of them recently; however, I’ll probably share it with them when I do talk with them.’’ Slightly fewer relatives articulated their relatives’ reproductive risks as a reason for disclosing CF carrier status: ‘‘They should know for their future children’s health’’ and ‘‘If they were going to have children please get tested’’ (from an obligate carrier). In other instances, carriers used relatives’ unmarried status or completion of childbearing as a reason for nondisclosure: ‘‘My brother is not married and does not have a significant girlfriend. . . . Perhaps I should tell him; he too could be a carrier’’ and ‘‘[My relatives are] done having children.’’ Another major reason for disclosure of CF carrier status to relatives was secondary to the disclosure of the child’s ongoing health problems related to CF: ‘‘At the time we told our relatives, we really were not thinking of anything other than our newborn’s health’’ and ‘‘It was discussed when they were told my son has CF.’’ In general, reasons for nondisclosure were less frequently given and were related to lack of social closeness with the relatives. No individuals cited negative feelings about being a CF carrier (e.g., guilt, shame) or fear of causing anxiety in relatives as reasons for nondisclosure. DISCUSSION There are numerous reasons why some carriers of genetic abnormalities do not share carrier information: altered self-concept [McConkie Rosell and DeVellis, 2000], misunderstanding of genetic risk, the perception that their relatives would not be interested in undergoing genetic testing [Surh et al., ARTICLE 1994; Sorenseon et al., 1996], or because of their own personal feelings about disease severity and/or about the use of testing and prenatal diagnosis. Studies There are numerous reasons why some carriers of genetic abnormalities do not share carrier information: altered self-concept, misunderstanding of genetic risk, the perception that their relatives would not be interested in the information, or because of their own personal feelings about disease severity and/or about the use of testing and prenatal diagnosis. that address the psychosocial aspects of carrier status focus primarily on selfconcept and do not necessarily measure how disclosing carrier status impacts family dynamics. It is likely that such disclosures alter family relationships in both positive and negative ways, which also may vary depending on the mode of inheritance (e.g., autosomal recessive vs. sex-linked vs. translocations). Our study indicates that in both groups CF carriers are more likely to share information about their carrier status with those relatives who are more closely related genetically and with those relatives with whom there is a close social bond; CF carriers favor telling one side of the family (more often the maternal relatives than paternal ones) of their status, suggesting personal or social factors play a greater role in motivating carriers to share their status than genetic factors; and disclosure of carrier status usually occurs within 6 months of receiving the diagnosis of being a carrier, often in the context of explaining an affected child’s condition or as a means of soliciting social support. Individuals with a family history of CF (e.g., an affected child) are more likely to share AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) carrier status with relatives of all degrees, primarily because they are sharing other health information about their child. Even if one assumes some selection bias, i.e., those individuals returning questionnaires were more comfortable with sharing their carrier status, it is clear from our study that the presence of an affected family member significantly influences the disclosure pattern of CF carrier status within the family. Yet the method of identifying CF carrier status does not impact the risk assessment; relatives of CF carriers, whether identified via general population screening or due to a positive family history, have the same risks (e.g., a third-degree relative to a CF carrier has a one-in-sight chance of also being a carrier, regardless of whether there is an affected individual in the family). While this may become less significant as CF carrier screening becomes routine in the prenatal setting, it does prevent fully informed decisionmaking for those who do not seek counseling or testing prior to conception. Thus, without knowledge of a family history, individuals may be less likely to undergo counseling or carrier testing themselves because they do not view themselves at increased risk when in fact they are. One possible explanation for the difference in disclosure frequencies between those with and those without family history of an affected family member is the impact of personal experience with CF. The literature of One possible explanation for the difference in disclosure frequencies between those with and those without family history of an affected family member is the impact of personal experience with CF. disability studies supports the concept that personal experience with a disorder (e.g., CF) is related to more positive attitudes toward disability [Yuker, 1994], 75 which is consistent with our study data regarding perceived severity of CF and the influence of personal experience. And yet, although carriers without exposure to CF report higher perceived severity rankings, such that one might expect that they would be more motivated to share information about reproductive risks, they were less likely to disclose. An alternate hypothesis to explain this lower disclosure rate is that nondisclosers perceive a decreased sense of the reality of risk. Many individuals who have never experienced the occurrence of an affected child in the family may perceive it to be unlikely that the event could ever happen, regardless of how they perceive severity, while individuals with personal experience may feel an increased awareness of vulnerability [Livneh, 1982]. An alternate hypothesis to explain this lower disclosure rate is that nondisclosers perceive a decreased sense of the reality of risk. Many individuals who have never experienced the occurrence of an affected child in the family may perceive it to be unlikely that the event could ever happen, regardless of how they perceive severity, while individuals with personal experience may feel an increased awareness of vulnerability. A second possible explanation for the difference in disclosure frequencies may be that many obligate carriers did most of their telling in the context of explaining their child’s diagnosis, making carrier status an inadvertent disclosure, rather than the primary one. This mode of disclosure may be useful to some relatives, but it assumes an under- 76 AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) standing of the complexities of recessive inheritance, that relatives can determine that they are at risk, and that relatives are aware of genetic screening availability. While prenatal population screening may identify these people, it would be preferable for health care providers to encourage their patients to share this information with relatives even when they are not pregnant to give their relatives greater time to adjust to and appreciate the significance of the information and to make informed decisions about what to do with this information (e.g., whether or not to undergo carrier testing and how such information would influence reproductive plans). Finally, very few of our respondents in either group reported that their relatives had pursued carrier testing. This may suggest that very few relatives sought testing, as has been observed in other studies (e.g., Suhr et al. ), or that relatives did not tell the study subjects whether or not they had testing. In part, there was no informational utility in reporting testing decisions back to study subjects who already knew their carrier status. Relatives who underwent testing may have chosen not to reveal their status for reasons of privacy or to avoid creating a ‘‘have/have not’’ dichotomy within the family. Additionally, relatives who did undergo testing and learned that they were carriers may have sought social support from someone other than study participants in an effort to avoid increasing the perceived burden on study participants. Limitations of this study include the small population size, a relatively low response rate in both groups, and differences in education and religious affiliations between groups A and B. The relatively low response rate may be due in part to the complexity and the amount of information the surveys were intended to solicit. The differences in education and religious affiliations between the groups may represent the geographic regions from which the participants were recruited. It also reflects the participation of Northwestern Memorial Hospital in an outreach program aimed at educating AJ indivi- duals about their reproductive risks for a number of genetic conditions for which carrier screening is available. Although we hypothesized that the existence of an affected proband is the most significant factor to explain disclosure patterns, our data may not be generalizable given these confounding demographics. Another limitation of this study is that the data regarding the subjects’ disclosure to relatives and genetic testing by relatives were not independently verified in that we did not contact the relatives. It would be very informative to know what relatives understand about their own risks and their reasons for choosing to undergo or forego genetic testing. Several studies have addressed how to contact the relatives without intruding on the privacy of the proband and his or her relatives [Winter et al., 1996; Wilcke et al., 1999]. Finally, our study only assessed the dissemination of information from known carriers. We did not assess differences in the genetic counseling provided in the CF clinic versus that provided in the prenatal genetic counseling clinics, and what role these differences may have had on disclosure patterns. Nor did we question individuals in group B regarding the utility of sample letters for disclosure. Clearly, further research needs to be done to understand familial obstacles to information disclosure. Our study documents that the frequency of and reasons for disclosing CF carrier status to relatives differ between individuals with and without an affected family member despite the fact that the reproductive risks for their relatives are the same. This has significant implications for health care professionals who counsel women and couples regarding reproductive decision-making, because accurate risk assessment is not possible for individuals who have incomplete family history information because they are uninformed or misinformed. While encouraging health care professionals to counsel all Caucasian individuals of reproductive age about CF carrier screening regardless of family history (as recommended by ACOG, ACMG, and NIH) may make ARTICLE this less relevant, it remains unclear how widely accepted community-based CF carrier screening will become. Additional research is needed to assess ways that health care providers can promote and encourage patients to communicate with all of their relatives in order to increase informed reproductive decision making. ACKNOWLEDGMENTS The authors thank M. Justin Coffey, Heather McManus, and Erica Wood for assistance with mailing of surveys and data entry, as well as Carol Stocking for wisdom about statistical analyses. L.F.R.’s participation in this project was sponsored by a Harris Foundation grant. This research was completed as part of K.E.O.’s fellowship in clinical medical ethics at the MacLean Center for Clinical Medical Ethics, the University of Chicago. REFERENCES American College of Obstetrics and Gynecology and American College of Medical Genetics. 2001. Preconception and prenatal carrier screening for cystic fibrosis: clinical and laboratory guidelines. Washington DC: American College of Obstetrics and Gynecology. Ayme S, Macquart-Moulin G, Julian-Reynier C, Chabal F, Giraud F. 1993. Diffusion of information about genetic risk within families. Neuromusc Disord 3:571–574. Botkin JR, Alegmagno S. 1992. Carrier screening for cystic fibrosis: a pilot study of the attitudes of pregnant women. Am J Pub Health 82:723–725. Callanan NP, Bloom D, Sorenson J, DeVellis B, Cheuvront B. 1995. CF carrier testing: experience of relatives. J Genet Couns 4: 83–95. Clayton E, Hannig V, Pfotenhauer J, Parker R, Campbell P III, Phillips J III. 1996. Lack of interest by nonpregnant couples in population-based cystic fibrosis carrier screening. Am J Hum Genet 58:617–627. Fanos J, Johnson J. 1995a. Barriers to carrier testing for adult cystic fibrosis sibs: the importance of not knowing. Am J Med Genet 59:85–91. Fanos J, Johnson J. 1995b. Perception of carrier status by cystic fibrosis siblings. Am J Hum Genet 57:431–438. Grody WW, Cutting GR, Klinger KW, Richards CS, Watson MS, Desnick RJ. 2001. Laboratory standards and guidelines for population-based cystic fibrosis carrier screening. Genet Med 3:149–154. Henneman L, Bramsen I, Van Os T, Reuling I, Heyerman H, van der Laaag J, van der Ploeg H, ten Kate L. 2001. Attitudes towards reproductive issues and carrier testing among ARTICLE adult patients and parents of children with cystic fibrosis (CF). Prenat Diagn 21:1–9. Livneh H. 1982. On the origins of negative attitudes toward persons with disabilities. Rehabilitation Lit 43:338–347. McConkie-Rosell A, DeVellis B. 2000. Threat to parental role: a possible mechanism of altered self-concept related to carrier knowledge. J Genet Couns 9:285–302. National Institutes of Health. 1997. NIH consensus statement. Genet test Cystic Fibrosis 5:1–37. Sorenson J, Cheuvront B, Bruning A, Talton S, DeVellis B, Koch G, Callanan N, Fernald G. 1996. Proband and parent assistance in identifying relatives for cystic fibrosis carrier testing. Am J Med Genet 63:419–425. AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) Sorenson J, Cheuvront B, DeVellis B, Callanan N, Silverman L, Koch G, Sharp T, Fernald G. 1997. Acceptance of home and clinic-based cystic fibrosis carrier education and testing by first, second, and third degree relatives of cystic fibrosis patients. Am J Med Genet 70:121–129. Suhr L, Cappelli M, MacDonald N, Mettler G, Dales R. 1994. Cystic fibrosis carrier screening in a high-risk population. Arch Pediatr Adolesc Med 148:632– 637. Suslak L, Price D, Desposito F. 1985. Transmitting balanced translocation carrier information within families: a follow-up study. Am J Med Genet 20:227–232. Wake SA, Rogers CJ, Colley PW, Hieatt EA, Jenner CF, Turner GM. 1997. Cystic fibrosis 77 carrier screening in two New South Wales country towns. Med J Aust 164:471–474. Wilcke JTR, Seersholm N, Kok-Jensen A, Dirksen A. 1999. Transmitting genetic risk information in families: attitudes about disclosing identities of relatives. Am J Hum Genet 65:902–909. Williams JK, Schutte DL. 1997. Benefits and burdens of genetic carrier identification. W J Nursing Res 19:71–81. Winter PR, Wisner GL, Finnegan J, Bartels D, LeRoy B, Chen PL, Sellers TA. 1996. Notification of a family history of breast cancer: issues of privacy and confidentiality. Am J Med Genet 66:1–6. Yuker HE. 1994. Variables that influence attitude toward persons with disabilities: conclusions from the data. J Soc Behav Pers 9:3–22.