Translating Advances in Cardiogenetics Into Effective Clinical Practice

Abstract

In this article we describe a qualitative research study in which we explored individuals’ subjective experiences of both genetic testing and cardiogenetic disorders. Using a grounded theory approach, we coded and analyzed interview and focus group transcripts from 50 participants. We found that just under half of the participants who received their diagnosis during the study reported difficulty understanding information about both the purpose of genetic testing and their cardiac disease. A high level of anxiety about genetic testing and cardiac symptoms exacerbated individuals’ cognitive confusion. Participants reported both positive and negative interactions with the medical community, depending on health care professionals’ knowledge of cardiogenetic disorders. Overall, participants expressed a range of attitudes—positive, negative, and ambivalent—toward genetic testing. We conclude with a discussion of the barriers to achieving effective clinical care for genetic conditions and offer suggestions for improving collaborative decision making between physicians and patients.

Whole-genome sequencing now makes it possible to test for mutations throughout the entire genome (Tucker, Marra, & Friedman, 2009). Better technology, however, does not automatically translate into better clinical care. The decision about whether to initiate specific genetic testing, and the ways in which the results of that testing will inform subsequent medical care, is the result of a complex series of interactions between medical professionals, individuals, and families. One group of disorders that can be considered the “leading edge” of genomic medicine is sudden cardiac death resulting from genetic conditions such as long-QT syndrome (LQTS), catecholaminergic polymorphic ventricular tachycardia (CPVT), and Brugada syndrome (Hofman, Tan, Alders, van Langen, & Wilde, 2010).

This group of inherited cardiac arrhythmias represents a serious public health issue and might account for as many as 200,000 deaths annually in the United States (Ackerman et al., 2011). The prevalence of inherited LQTS in the United States is estimated at about 1:7,000 persons, causing perhaps 2,000 to 3,000 sudden deaths in children and young adults each year (Schwartz et al., 2009). The prevalence of CPVT is estimated to be about 1 in 10,000 people (Liu, Ruan, & Priori, 2008). The prevalence of symptomatic Brugada is estimated to be 1 in 5,000 in Western countries, with a higher incidence in Asia. The prevalence of clinically silent type 1 Brugada is estimated to be much higher (Li & Behr, 2013).

A number of factors complicate the interactions between patients and medical personnel in the context of these cardiogenetic disorders. Because of the rarity of these diseases, many emergency medical technicians and primary care physicians (internists and pediatricians) are not familiar with the complexities of inherited cardiac arrhythmias. For example, in a study focused on screening young athletes to prevent sudden cardiac death, researchers at Stanford University showed eight abnormal and 10 normal electrocardiograms (EKGs) to 53 experienced pediatric and adult cardiologists. Thirty-two percent of teenagers with abnormal EKGs were never identified. Twenty-six percent of normal patients were warned against exercise, and 19% of patients who should have been restricted received no warning (Hill, Miyake, Grady, & Dubin, 2011). The wide variability in the presence or absence of diverse symptoms makes these diseases difficult to diagnose. Moreover, their genetic basis has only recently been discovered.

Thus, genetic testing might not be requested in response to symptoms suggestive of these conditions, or as part of an autopsy following a sudden death. Deaths from these arrhythmias are frequently relegated to the category of “death from unknown causes.” The absence of genetic testing in this context is particularly problematic because detecting these genetic conditions can identify other family members who might be at risk of sudden cardiac death (Tester & Ackerman, 2012). Misdiagnosis, in contrast, can cause additional stress to family members who are ill or grieving the illness or death of a loved one.

In addition to a lack of familiarity with these diseases among medical professionals, an understanding of genetics is lacking in the population at large. Lea, Kaphingst, Bowen, Lipkus, and Hadley (2011), in a review of the research, found that members of the public had some familiarity with genetic and genomic terms, but had difficulty understanding the underlying concepts. These authors concluded that improvements in both health and numeric literacy were needed for people to understand and apply genetic information. In response to these problems, Tesla and colleagues (2009) set up an online service to answer questions about medical genetics. Between 2003 and 2009 they responded to 4,497 questions from every U.S. state and 84 countries. Twenty percent of survey respondents (n = 247) reported that, prior to visiting this site, they were unaware of genetic services from physicians or genetic counselors. Even among users with access to the Internet, 1 in 5 people have limited or no knowledge of genetics, especially Latinos and African Americans.

In the current study, we interviewed individuals and families about their experiences after a serious cardiac event or a sudden cardiac death of a family member. These families had to make a series of difficult decisions, including whether to (a) undergo genetic testing; (b) share the results of that testing with other family members; and (c) comply with recommendations for medication, lifestyle changes, and in some cases the implantation of a cardiac defibrillator. The goal of the project was to understand more about patients’ subjective experiences so as to improve clinical care.

The Medical Context: Cardiac Arrhythmias

In long-QT syndrome (LQTS), the interval between the beginning of a heartbeat (known in medicine as the Q wave on an EKG) and the end of recovery from a heartbeat (the T wave) is too long (Goldenberg, Zereba, & Moss, 2008). This prolongation can lead to disruption of the normal heart rhythm, causing an arrhythmia in which the heart beats too fast, thereby impairing effective blood pumping function. LQTS might be caused by inherited gene mutations that alter the way cardiac channels are formed. The symptoms include fainting, irregular or rapid heartbeat, or the heart can stop beating altogether. Currently, approximately 13 loci of LQTS have been identified, each caused by a mutation in a specific gene that is susceptible to a particular triggering activity.

Patients with LQTS-1 tend to have cardiac events during vigorous exercise; for those with LQTS-2, events occur in response to sudden emotion or loud auditory stimuli; and events for patients with LQTS-3 can occur during sleep. Treatment can include medication, lifestyle changes, and an implantable cardiac defibrillator (ICD). Despite the constant risk of arrhythmias, LQTS patients are asymptomatic most of the time (Priori et al., 2003).

CPVT is a hereditary condition that predisposes patients to symptomatic and life-threatening tachyarrhythmias. Affected individuals are generally asymptomatic at rest, but when conditions arise that require increased sympathetic tone (e.g., exercise or emotional excitement), the increase in epinephrine/norepinephrine might trigger the arrhythmia. Onset of symptoms usually occurs in young adulthood and often progresses in severity with age. CPVT arrhythmias are generally amenable to beta-adrenergic blocker therapy; however, an ICD might be required for some (Venetucci, Denegri, Napolitano, & Priori, 2012).

Brugada syndrome is another hereditary arrhythmia condition whose genetic basis is less well understood. It includes characteristic EKG abnormalities and an increased risk for sudden death from ventricular tachyarrhythmia. Arrhythmia events often occur during sleep and might also be triggered by fever. Men are at higher risk for developing these arrhythmias; the evidence, however, does not indicate that the condition is X-linked. Although no current consensus has been reached concerning the most appropriate therapy, if an affected individual has experienced an aborted cardiac arrest or has had documented ventricular tachycardia, then an ICD is usually recommended (Li & Behr, 2013).

Methods

Medical Setting and Recruitment

Our study team was an interdisciplinary group of medical professionals at a cardiogenetics clinic in an academic medical center located in a large city in the northeastern United States. The clinical team consisted of an adult and a pediatric cardiologist, reproductive and pediatric geneticists, a pediatric cardiac nurse practitioner, a genetic counselor, a pediatric genetics fellow, and an ethicist. The research team included a senior psychologist, a postdoctoral fellow, and a psychology graduate student. An ethicist and a study coordinator completed the study team. The institutional review board (IRB) of a large urban medical school reviewed and approved the design of the study.

Recruitment involved three different samples. Participants who were being evaluated and treated at the cardiogenetics clinic comprised Sample 1 (n = 27). The nurse practitioner, using information from a referring physician, did an intake interview over the telephone and then presented this information to the team. Representatives from each specialty then met together with each patient or family to take a more extended history that included a pedigree (family tree or genogram). After gathering this information, the team met a second time and developed a more detailed plan for assessing the medical condition of the patient and the appropriateness of genetic testing. This plan was then discussed with the patient and, if appropriate, blood was drawn for testing.

Inclusion criteria included patients who had LQTS, CPVT, and Brugada syndrome and their families; or individuals whose family members had died suddenly from these disorders and who were considering genetic testing for themselves or for a child. In addition, a genetic counseling graduate student who had LQTS and was observing in the cardiogenetics clinic volunteered for the study; because so few data exist on LQTS, we decided to include her in the study. The only exclusion criterion was if no one in the family spoke English.

After patients received the results of genetic testing, a researcher contacted the patient to discuss the research, either by telephone or at a regularly scheduled appointment. The time lag between receipt of the test results and contact varied, but was never less than a month. The researcher described the study procedure and reviewed the IRB-approved consent form. This consent form explained that the patient would be asked some questions about his or her ethnicity, education, income, and marital status, and would be interviewed about experiences at the clinic. The patients were also assured that declining to participate in the research would not affect their treatment. If the patient and family members consented to participate, at their next scheduled medical appointment the research team conducted the interview or focus group.

To expand the size of our sample, we reached out to two advocacy organizations. Sample 2 (n = 14) was recruited from an organization for families in which an individual had died or had a cardiac condition that could result in sudden death. Sample 3 (n = 9) included members of an organization for families that had lost a child over the age of 1 because of unknown etiology. Liaisons at each organization informed their members about this research either through newsletters or by speaking directly to specific families. If the families agreed to participate, the liaisons provided us with their contact information. We then sent them a written version of the IRB-approved consent form and arranged for an interview or focus group participation.

Data Collection and Coding Procedure

We planned to include both focus groups and family interviews in the study, because in previous studies we found that some individuals respond to the privacy afforded by a single interviewer, whereas other participants speak more freely in groups of people with shared experiences (e.g., Dorfman, Stolerman, & Silverstein, 2011; Silverstein, Auerbach, & Levant, 2002). Moreover, we found that focus groups often stimulated participants to talk to each other about aspects of their experiences that we had not thought to ask (e.g., Schacher, Auerbach, & Silverstein, 2005). Because the data from both interviews and focus groups were textual data, our experience indicated that a grounded theory approach would be appropriate for coding data from both contexts. In the current study, no differences between the patterns in these two contexts emerged, so the data were combined.

Because families in all three samples came from widely separated geographic areas, we were able to arrange only one focus group each for Samples 1 and 2. Participants from Sample 3, in contrast, were from New York and New Jersey, and were seen in three focus groups. Each focus group included individuals from two or three families. All of the interviews from Sample 1 included the patient and one or more members from the same family.

The first author conducted all interviews and facilitated all focus groups in combination with the second and third authors, with the exception of two focus groups with Sample 3 participants that were facilitated by the second author alone. Before each interview or focus group began, the researcher reviewed the consent form and reminded participants that they could withdraw from the study at any time. Participants then completed a demographic questionnaire with questions about race, ethnicity, educational background, income, and marital status. For Sample 2, the demographic questionnaires were mailed to the participants and returned by mail to the research team. The interviews and focus groups ranged from 1.5 to 2 hours in duration and were conducted in person for Samples1 and 3, and via telephone for Sample 2 because this group included individuals from many different U.S. states. In all three samples, each family member was given a $25 gift card after completing the interview or focus group. For Samples 1 and 3, the participants’ travel expenses were reimbursed.

We began with a general, open-ended question: “Tell me the story of your [or your family member’s] experience of the cardiac event [or sudden death].” Beginning with a general question generates more “bottom-up” data (i.e., information organized by the participants) rather than “top-down” data (i.e., responses to researcher-constructed questions). After the participants had responded fully to this opening question and follow-up probes, we then asked more specific questions about attitudes toward genetic testing and toward sharing genetic information with other family members.

The research team audiorecorded the interviews, which were then typed by a professional transcription agency. The transcripts were stored on the study coordinator’s password-protected computer located in Siobhan Dolan’s office. All identifying information was removed from the transcripts, and each patient was assigned a number from 1 to 50. The research team coded each transcript separately, based on a grounded theory approach developed by Auerbach and Silverstein (2003). This data coding technique began by identifying relevant text (i.e., text pertinent to our research question). These direct quotations were then organized into repeating ideas, or words and phrases used by multiple participants. Finally these repeating ideas were grouped into a second, more abstract level of themes, or concepts implicit in the repeating ideas.

At each stage of data analysis, we discussed our individual coding to reach consensus. After these discussions yielded a single list of repeating ideas and themes, the data were presented to the entire interdisciplinary team for feedback, and ultimately to the project’s advisory board, which was composed of experts in genetics and cardiovascular disease. These multiple levels of feedback ensured that the results were transparent (other readers could check our coding), communicable (the categories made sense to other readers), and coherent (categories were internally consistent and reflected individual differences).

We then sent the list of themes to all participants and invited them to attend four scheduled feedback sessions. These kinds of participant feedback sessions were designed to check the accuracy of the researchers’ interpretation of participants’ responses, and to give participants an opportunity to debrief and to reflect on the experience of sharing their stories. The Sample 1 feedback sessions were scheduled to meet at the clinic. The Sample 2 and 3 sessions were scheduled on the phone. None of the Sample 1 participants agreed to attend a feedback session. Only two Sample 2 participants and one Sample 3 participant attended these sessions. This low level of participation in feedback sessions is typical of our experience of multiple studies in qualitative research in the New York City area. The 3 participants who did attend were not able to provide any feedback about their experiences participating in the research; rather, they continued to talk about the same experiences they had reported in their original interview or focus group. Thus, the transcripts of the feedback sessions did not yield any new data.

In Table 1 we present the demographic characteristics of the samples. Three of the Sample 1 families who chose to have genetic testing declined to participate in the research project. One Latino family cited language as a barrier. Two White families declined based on parent preferences not to do genetic testing of their children. Six other Sample 1 patients initially agreed to participate in focus groups, but did not attend. These families did not respond to follow-up phone calls and did not recontact the clinic. It is unclear how many Sample 2 and 3 families refused to participate because staff of the two organizations conducted the initial recruitment efforts.

Table 1.

Individual and Family Characteristics of the Sample.

Characteristic	Sample 1 (n = 27a)	Sample 2 (n = 14)	Sample 3 (n = 9)	Total (n = 50)	Percent
Sex
Male	8	1	2	11	22%
Female	19	13	7	39	78%
Race/Ethnicity
White	12	12	8	32	64%
African American	4	1	1	6	12%
African Caribbean	1	0	0	1	2%
Latino/Hispanic	10	0	0	10	20%
Biracial (White Asian)	0	1	0	1	2%
Education
Less than 9th grade	1	0	0	1	2%
Currently in high school	2	0	0	2	4%
General equivalency diploma	1	1	0	2	4%
High school	4	1	0	5	10%
Some college	5	5	1	11	24%
College degree	3	5	4	12	24%
Graduate degree	9	3	3	15	30%
Unknown	2	0	0	2	4%
Annual Household Income
< $25,000	3	2	1	6	12%
$26,000–$50,000	4	2	0	6	12%
$51,000–$80,000	7	3	0	10	20%
> $80,000	7	7	8	22	44%
Refused to state	2	0	0	2	4%
Unknown	4	0	0	4	4%
Age
< 20 years	3	0	0	3	6%
20–30 years	8	2	0	10	20%
31–50 years	4	7	8	19	38%
51–60 years	8	3	0	11	22%
> 60 years	2	2	1	5	10%
Unknown	2	0	0	2	4%
Marital Status
Married	10	11	7	28	56%
Cohabiting	2	0	0	2	4%
Separated	1	1	0	2	4%
Divorced	0	0	1	1	2%
Single	13	2	1	16	32%
Widowed	1	0	0	1	2%
Participants with sudden death in immediate family	13	9	9	31	62%
Families with adolescent or child as patient	6	9	7	22	44%
Families in which at least one immediate family member had genetic testing	14	11	2	27	54%
Families reporting refusal of other family members to have testing	6	8	0	14	28%
Participants who had had or whose children had had implantable cardioverter defibrillators or pacemakers	5	9	0	14	28%

^aThe graduate student participant is included with Sample 1 participants.

A total of 50 individuals representing 32 families participated in the study: 27 Sample 1 individuals being treated at the cardiogenetics clinic from 12 families, plus one genetic counseling graduate student who was attending the cardiogenetics clinic as part of her training and volunteered to be interviewed because she has LQTS; 14 Sample 2 individuals from 11 families; and 9 Sample 3 individuals from 7 families. The total sample was primarily female, White, middle class, and married, although 16 single individuals and 3 adolescents did participate. About half of Sample 1 was of Latino or African descent, whereas the other two samples were almost exclusively White.

Because the study of genetic testing and cardiogenetic conditions is in the earliest phases, it is important to include the widest possible variation in participants; thus we combined the data from all three subsamples. Each sample provided information-rich data, and the between-subsample differences yielded enough variation to enhance the probability that representative patterns would emerge. All three subsamples included individuals or families of individuals who had experienced cardiogenetic conditions. The samples were different in terms of number of deaths, number of participants who had undergone genetic testing, and membership in advocacy groups. We examined the responses of the genetic counseling graduate student and found that her responses reflected repeating ideas similar to those of participants from Sample 1; thus we included her in that group.

Differences in subsamples are not unusual in qualitative research. In the early stages of studying a phenomenon, samples might be chosen to provide the maximum variation of cases to allow multiple patterns to emerge. In later stages of study, theoretical sampling would be used to confirm or disconfirm emergent patterns (see Coyne, 1997, for an extensive analysis of sampling procedures in qualitative research).

Results

Data saturation was reached at 49 participants. The themes and repeating ideas relevant to health literacy and collaborative decision making in cardiogenetics are displayed in Table 2. We present selected segments of relevant text—that is, direct quotes from participants—below the appropriate repeating idea in the following section.

Table 2.

Percentage of Participants Expressing Themes and Repeating Ideas.

Selected Themes and Repeating Ideas	Sample 1 (n = 27a)	Sample 2 (n = 14)	Sample 3 (n = 9)	Total (n = 50)	%
Decision-making context
Emotional upset and anxiety	27	14	9	50	100%
Cognitive confusion	11	1	0	12	24%
Negative interactions with professionals	13	11	7	31	62%
Positive interactions with professionalsb	12	9	5	26	52%
Making the decisionc
Positive attitudes toward testing	15	13	1	29	58%
Ambivalent attitudes toward testing	8	0	4	12	24%
Negative attitudes toward testing	4	0	3	7	14%

^aThe graduate student participant is included with Sample 1 participants.

^bMany participants had both positive and negative experiences with health care professionals.

^cOne participant in both the Sample 2 and Sample 3 groups expressed no feelings about genetic testing.

The Decision-Making Context

Emotional upset and anxiety

All of the participants reported being upset and anxious during their initial meetings with clinicians. Some were mourning a family member who had died. Others were distressed because of their own or a family member’s cardiac symptoms. One member of the Sample 2 families described the ordeal:

We found our son dead in bed. … He ended up in a coma, and they did cooling therapy on him. … He was the first kid [child] to have it done. Everybody told us he was not gonna make it. Three days later he woke up and said his first words.

A man in Sample 1 expressed many of the participants’ anxiety about the seriousness of the disease and its sudden onset: “When you hear something about your heart, it’s scary because the heart and the brain are two things you can’t live without. It’s like a ticking time bomb.” This worry and emotional tension affected entire families, not just the person with the disease. One family member in Sample 1 described her attempt to monitor her sister continuously, afraid that her ICD might go off: “She gets scared that her device is gonna go off, so I’ll go over there and sleep with her, but I’m scared to sleep because I’m always making sure she’s okay.”

Some parents felt guilty about having passed the mutation on to their children, and younger patients worried about having children in the future. One mother in Sample 2 explained: “As a mother, I feel bad that something you had, you passed to your child. … Not that I ever meant to do it, but you still feel bad.”

Cognitive confusion

Individuals in the three subsamples described different levels of cognitive confusion. Because most of the Sample 3 families had experienced the death of a child many years prior to the study, when genetic testing was relatively uncommon, few of them had been told about genetic testing at that time; thus they reported little cognitive confusion. Sample 2 participants, in contrast, were members of an advocacy organization and had aggressively sought answers as to their child’s illness or death in terms of obtaining medical information online and seeking out consultations from multiple physicians. Given this relatively sophisticated knowledge base, only 1 person in this group expressed confusion about the diagnosis.

Sample 1 participants, in contrast, were in the process of obtaining a diagnosis. They were in the early stages of trying to understand the implications of their diagnosis and the advantages and disadvantages of genetic testing; thus many of these participants expressed confusion about some aspect of cardiogenetic disease and/or genetic testing. Those with a lower level of education were the most confused; for example, a working-class Latina mother who had two daughters with LQTS expressed her frustration: “I don’t know what LQTS is. You gotta explain to me, talk to me in plain English. … They said it was generic. … Where is it coming from?”

The emotional upset that most Sample 1 participants experienced exacerbated their difficulty absorbing information that was essential to informed decision making. Their confusion related to the definition of the disease, the inherited nature of the disease, the advantages and limitations of testing, and the treatment options available. A mother articulated the effect of anxiety on her ability to understand the information that the doctors were providing: “Every time I go into the hospital, I’m so nervous. They might be telling me exactly what it is, but it’s just not going in there.”

Because of the anxiety surrounding sudden and potentially fatal cardiac events, even highly educated people had difficulty understanding the complex nature of the disease. One woman with a graduate degree was initially unwilling to consider that her son’s death might have had a genetic basis. Her first response was confusion: “The QT is what? The quiet time, the resting period?” After several meetings with team members, she was able to understand the situation and agreed to be tested. These kinds of experiences helped us realize that many individuals need repeated explanations over time to absorb the medical information.

Negative interactions with medical professionals

Almost two thirds (n = 31) of the total sample reported negative interactions with medical professionals. Only two Sample 1 participants criticized our clinical team directly. One family with twin daughters complained about the delay in receiving the results of genetic testing. After further delays because of inclement weather, they requested and were given the negative (no mutation) test results by telephone. In another family, the father had a cardiac arrest and testing results showed a deleterious mutation associated with LQTS; the parents did not want their 18-year-old son tested, or even to know the name of the disease. Given this context, the mother was angry when, in response to the son’s request for online information, a team member gave him the name of a Web site, where he learned about the disease. The small number of complaints in Sample 1 about their current care might have been influenced by a concern about compromising their care, and thus must be viewed with some skepticism.

Some of the Sample 1 families did have negative stories about doctors they had seen prior to coming to our clinic, reporting that many were totally unaware of cardiogenetic disease: “I had taken [daughter] to her pediatrician, but they didn’t think anything was wrong. The doctor told me her QT interval was prolonged. … It didn’t raise any red flags for him.” In one dramatic example, a husband had come home to find his wife dead on the kitchen floor. Ultimately, the autopsy showed that she had died because of falling and hitting her head; genetic testing was not done. When their daughter began to have seizures at age 9, she was sent back and forth between doctors for several years. One doctor completely dismissed her symptoms: “The pediatrician said it was psychosomatic.” At age 25, clinic doctors did genetic testing, and the results were positive for catecholaminergic polymorphic ventricular tachycardia (CPVT), an inherited cardiac arrhythmia. This diagnosis likely explained the mother’s death and the daughter’s symptoms throughout adolescence.

Participants in the other two samples had many more negative comments about doctors, hospital procedures, and medical examiners. One Sample 2 mother reported, “When I went back to tell her [pediatrician] what my daughter had died of … she had never heard of it.” Another Sample 2 mother discovered that during her son’s routine screening prior to participating in sports, he had indicated several symptoms consistent with LQTS syndrome. His mother reported, “Prior to his death, my son had checked ‘yes’ to dizziness during and after activity; ‘yes’ to chest pain; ‘yes’ to trouble breathing.” However, his pediatrician had not acted on that information. When this mother learned that her son’s reported symptoms could have been treated, she described her devastation: “That put me on my knees.”

Another mother from Sample 2 was embittered because a cardiologist had failed to collaborate with a neurologist in diagnosing her son’s condition. The cardiologist had acted based on EKG results and ignored her son’s grand mal seizure. He recommended immediate implantation of an ICD, without waiting for genetic testing. This young man’s mother pointed out, “It was a misdiagnosis. [My son] has never had any activity on the ICD, but has had two more seizures.” After many years, the family underwent genetic testing, with negative results (no mutation present). Because of this misdiagnosis, the son lost both his swimming scholarship to college and his social network that had been built around the swim team. Moreover, he was needlessly worried for 3 years about having a cardiac episode or sudden death, and his seizure disorder went untreated.

Some of the Sample 2 and Sample 3 families described the cold, unfeeling manner of medical examiners when the family tried to get the results of their child’s autopsy: “The ME [medical examiner] didn’t really grasp that she was dealing with the mother of a dead child. She was blunt, to the point, and off the phone.” These negative experiences stemmed from the professionals’ absence of empathy in the face of the family’s loss of a child, lack of familiarity with the disease, and ignorance about the possible implications of the disease for surviving family members.

Positive interactions with medical professionals

Half of the participants (n = 26) reported positive interactions with the medical community: “[Doctors at Sample 1 clinic] did a really great job explaining to me what they thought could possibly be the cause of the sudden deaths. They gave me written information. … It was very understandable, very simple.” Similarly, a mother from Sample 3 stated, “My pediatrician and my OB [obstetrician] were definitely the biggest godsends in my life because they navigated everything. … They worked with the genetics team. They had a whole conference about me.” A few participants discussed positive interactions with medical examiners. In the face of their primary care physician’s failure to suggest genetic testing, one woman from Sample 2 reported that the medical examiner urged the family to be checked for LQTS after hearing about the family history of sudden death: “The coroner wasn’t satisfied. He said, ‘This sounds like a syndrome called long QT. I want you to go get checked.’ Thank God for that coroner, because we never would have known.”

Making the Decision

Positive attitudes toward testing

Participants in the three subsamples reported contrasting attitudes about testing. Most of the Sample 3 participants had experienced the death of a child many years previously. At that time, many in the medical community had not been aware of the possibility that a child’s unexpected death could be the result of an inherited cardiac disease; thus, most Sample 3 participants had not had genetic testing. All except one person in this group were either negative or ambivalent about seeking genetic testing at the time of the study. They did not want to reopen the pain of wondering why their child had died.

In Samples 1 and 2, although some family members had negative feelings about testing, at least one person in each of the families in both samples was eager to know why a beloved child had died, or the probability of other family members inheriting the disease. One woman in Sample 1 reported experiencing a sense of closure after the testing: “[Genetic testing] makes me feel powerful. It connects the dots.” Another woman in Sample 1 wanted more information about heritability: “I’ve always wanted to know the chances of passing it on to my children.” Some families wanted to know what their triggers were so they could be careful and (it was hoped) avoid another cardiac event: “I’d rather know, because that way I can prevent it from happening.”

Ambivalent attitudes toward testing

Many of the Sample 1 participants expressed ambivalent feelings; wanting information, but dreading it as well: “You want to know and you don’t want to know, because when you know, it’s scary. And when you don’t know, it’s still scary.” Similarly, a mother in Sample 3 expressed her ambivalence about testing during the autopsy: “Part of you wants the answer so desperately, and the other part of you says, ‘What good is it if I get an answer?’”

Two families initially had negative attitudes toward testing, but ultimately agreed. A Latino father in Sample 1 was sure that the disease had not been transmitted through his family: “I don’t see no history in my family.” He then proceeded to describe symptoms in many of his siblings that strongly resembled cardiac arrhythmias, such as chest pain, syncope (fainting), and sudden death from unknown causes. One sibling even had an ICD implanted. Ultimately, his daughters convinced him to have testing, and his results were positive (a mutation was present). Another family in Sample 1, whose 28-year-old daughter had a cardiac arrest shortly after bariatric surgery, was opposed to having their 17-year-old son tested. The family met with the team, and during the discussion, the son revealed that he had already gotten information about the disease on the Internet. The parents then agreed that he could be tested.

Negative attitudes toward testing

In Samples 1 and 2, attitudes toward genetic testing were generally positive. The only patient in Sample 1 who refused testing was the adolescent whose mother was angry when a member of the team gave him an Internet source where he discovered the name of his father’s condition. Some of the Sample 1 and 2 participants (n = 14) reported that members of their family (both immediate and extended) with whom they had shared their positive genetic test results had decided against testing for themselves and often for their children as well. Although we did not have an opportunity to interview these other family members directly, these statements occurred frequently enough that we have chosen to report these data. A man in Sample 1 stated, “I have about ten cousins who have not gotten tested. Some are extremely stubborn; some are scared.” Similarly, a woman in Sample 1 reported, “My sister hasn’t gotten herself or her kids tested. I think she’s in denial.”

Our finding replicates that of Stol, Menko, and Westerman (2010), indicating that when patients, rather than physicians, inform relatives about a hereditary disease, few requests for genetic testing follow. It is possible that some relatives in the current study might not have actually been informed. However, Laurie (1999) reported that some individuals who were informed simply did not want to be tested. Future researchers will need to explore this phenomenon.

In summary, the participants in our study experienced a great deal of emotional upset and anxiety surrounding their experience of losing a loved one and/or discovering the presence of a potentially fatal genetic disease in themselves or in a family member. They faced a difficult decision-making process in which they had to make sense of complicated information about cardiac disease and genetic testing. Their emotional upset interfered with their ability to absorb this medical information. In this difficult context, participants reported both positive and negative experiences with physicians. About two thirds reported negative interactions with doctors and medical examiners, whereas about half of the participants had positive experiences with medical professionals.

Some participants welcomed genetic testing because it provided them with a strategy for managing a serious illness or explained a previously unexplained death of a family member. Others reported feeling ambivalent about testing, wanting to know and not wanting to know at the same time. Some of the patients had negative feelings about testing. Many participants who informed their family about the presence of a cardiogenetic condition in the family reported that family members refused to be tested.

Discussion

Advances in genetics and genomics hold great promise in terms of improving health outcomes in virtually every area of medicine. However, multiple barriers to collaborative decision making between patients and physicians need to be overcome before these scientific advances can be translated into quality health care. In Figure 1 we present a model that identifies both general public health issues that need improvement and specific problems that emerge in the interactions between patients and physicians in the context of genetic testing. The model provides suggestions for addressing these barriers, and we discuss it in detail in the next section.

Figure 1.

Barriers and solutions to improving clinical care in genetics.

Public Health Issues

The general population

The findings of the current study point to the need for a number of improvements in health care education. The first change must take place at the level of public policy. Previously, researchers have suggested that Latino patients tend to struggle most with understanding medical information (Britigan, Murnan, & Rohas-Guyler, 2009); our data confirmed this finding. However, in our study, some White, well-educated (college and higher) patients also had difficulty absorbing some of the information about their disease and about genetic testing. These results reflect the low level of knowledge about genetic conditions in the population at large, especially in contrast to other diseases, such as diabetes or asthma (Condit, 2010). Bowling and colleagues (2008) illustrated how difficult it is to increase health literacy about genetics. They found that even college students who had taken a course in genetics did not significantly increase their scores on a test of genetic literacy.

Some efforts are in place to remedy this low level of literacy about genetics. The Web site Ask a Geneticist (http://genetics.thetech.org/ask-a-geneticist) provides genetics information “on demand.” Online access to genetics information will become an important educational resource; however, the increasing importance of online education raises concerns about potential disparities in health literacy based on unequal access to the Internet.

Physicians and medical examiners

We were surprised to discover that many physicians and medical examiners the participants had consulted were also unfamiliar with the genetic basis of some cardiac arrhythmias. Participants in all three samples reported examples of physicians—even cardiologists—who had not recognized that their symptoms could have been caused by an inherited cardiac channelopathy. This lack of information might result from the fact that genetics is both a clinical field unto itself and also crosses many different traditional medical specialties. In the United States, clinical genetics residency training is usually completed after or in conjunction with a residency in pediatrics, internal medicine, or obstetrics and gynecology. Despite this cross training, many providers view genetics as peripheral to everyday clinical concerns (Guttmacher, Porteus, & McInerney, 2007). This lack of knowledge might be remedied by the recent emergence of a number of online communities for health care professionals, such as the Human Genome Project (www.ornl.gov/sci/techresources/Human_Genome/education) and the Genetic Testing Registry (http://www.ncbi.nlm.nih.gov/gtr/). Solutions for these public health issues reside in the public education system in high schools, colleges, and medical schools, and must be solved at a policy level.

In the Consultation Room: A Tale of Two Cultures

Solution-focused culture

We turn now to a discussion of direct interactions between patients and doctors in the consultation room. The term translational is particularly apt in the context of health care because physicians and patients come from different cultures with regard to understanding and treating disease. Physicians are solution-focused, motivated to solve problems thoroughly and quickly. Physicians use a technical language that focuses on complex chemical and electrophysiological processes, and is sprinkled with acronyms. In the context of the current research project, even the psychologists on the team, another group of highly educated professionals, often needed translations of the medical terminology.

Because physicians have spent such a long time training and practicing alongside other doctors, many are not aware of how specialized their language has become; they often assume that their patients understand more than they do. The clinic physicians who treated Sample 1 participants always spent a great deal of time explaining cardiogenetic disease and the issues surrounding genetic testing. At the end of these lengthy explanations, they would ask the patients if they had any questions. Too often patients would tell the doctors they had “no questions,” and later during the research interview or focus group indicate to the researchers that they had not understood either the details of the disease or the reason for genetic testing

One technique with the potential to improve communication is active listening. Active listening includes at least three steps (Bodie, 2011): listening carefully and attending to implicit as well as explicit content and feelings, summarizing the other’s statements to insure that they feel understood, and using verbal and nonverbal communication to convey empathy. This technique allows a speaker to find out whether the listener has actually understood, and if not, to continue to explain until understanding is reached. This approach engages the patient and communicates information.

Researchers have shown that training in these kinds of communication skills can be effective; for example, Fallowfield and colleagues (2002) found that physicians who attended a communication course did show objective improvements in key communication skills. Similarly, in a meta-analysis of 106 correlational and 21 experimental studies, Zolnierek and DiMatteo (2009) found that with communication training, the odds of patient adherence were 1.62 times greater than when a physician had received no training.

Emotion-focused culture

In contrast to the physician’s focus on solving a medical problem, patients enter the consultation room mourning the loss of a loved one and/ or experiencing anxiety about their own or a family member’s health. Their anxiety makes it more difficult for them to absorb complicated medical information. To process this information and make an informed decision, patients need several things: an explanation given in simple language and in a culturally consistent format, an expression of empathy for their emotional upset, and time to process technical facts and the implications of those facts for their well-being. In our current medical context, physicians rarely have the training or time to provide this kind of clinical care.

Even in the positive ecological climate of our clinic, where the team had excellent rapport with patients, two examples typical of the cultural collisions between patients and doctors did occur: one in the family whose adult daughter had died after bariatric surgery, and another in which the father had survived a cardiac arrest. Neither family wanted their adolescent sons to know anything about the disease. This emotion-focused perspective reflected the parents’ anxiety that, if their sons tested positive for the mutation, the young men would be filled with anxiety and unable to enjoy life. Rather than educating their sons about the importance of taking medication and avoiding certain high-risk activities (e.g., experimenting with drugs), they chose to deny the medical danger inherent in keeping this information from their sons.

The physicians, in contrast, strongly objected to this stance. They wanted to resolve the genetic question so that the young men could take preventive measures if they did have the mutation. This solution-oriented approach stood in stark contrast to the parents’ emotional need to deny the risk to their sons. Neither side could understand the other’s perspective. In the end, the first family realized that their son had already learned a great deal about the disease, and so agreed to testing. The second family, however, expressed anger at the team for giving their son an online source of information about the disease. They did not opt for testing or for informing the son about the father’s cardiac disease.

We do not yet know whether the therapeutic alliance with the latter family has been disrupted or simply stressed. If the son does begin to have symptoms, will the family return to the clinic for treatment? From the perspective of effective communication and empathy, if the doctors had spent several more sessions listening to the parents’ concerns, would the parents have gained more trust, over time, that the doctors would help them through the crisis if their son did indeed have the disease? Similarly, would the doctors have become more respectful, if not accepting, of the parents’ perspective? Chen (2011) wrote about the challenge of empathizing with patients and respecting their treatment choices when the physician believes the patient’s well-being is threatened. These kinds of clashes between doctor and patient problem-solving approaches are not unusual, and might increase as more genetic information becomes a part of routine medical care.

Culturally Consonant Care

Latinos are a population at particularly high risk for poor communication with medical professionals, which proved to be the case for many of the Latino patients in our study. Kaphingst, Lachance, Gepp, D’Anna, and Rios-Ellis (2009) demonstrated the utility of presenting health information in formats that are culturally consonant for patients. In the first phase of a study, 27 community members were interviewed and reported that a focus on family history was a more meaningful educational approach than a presentation of genetic facts. In the second phase, using a sample of 474 Spanish-speaking Latino participants, the project team followed this advice and used small group sessions led by lay health advisors to create family trees showing how diseases ran in families. This format increased the likelihood that participants would discuss genetic information with both their physician and other family members. This approach of designing culturally sensitive educational approaches and using lay health advisors shows promise in terms of providing translational mechanisms for culturally underserved groups.

Conclusion

Findings from the current study suggest a number of changes that would improve the medical community’s ability to translate recent advances in genetics into more effective clinical care. The low level of both general and professional understanding of genetics indicates the need for improved public and professional education on this subject. The use of active listening, culturally preferred educational approaches, and lay health advisors has the potential to decrease the cultural clash between physicians and patients from diverse communities. Finally, emphasizing the importance of empathy and cultural competence in medical training is likely to enhance doctor–patient interactions.

Limitations

The study has several limitations. The sample size was relatively large for a qualitative study, yet it was still limited. In addition, all of the participants chose to participate in the research. These participants might have held views that are different from individuals who would decline to participate in research and/or would not want genetic testing.

Although the sample did include a relatively large minority of families of color, African Americans in particular were underrepresented. Moreover, fewer than a quarter of the participants were men, despite the fact that the majority of families included men. In Sample 2, several fathers were present whereas their wives participated in a telephone interview, but declined to speak to the researchers about their child’s illness or death. This low level of participation might reflect the fact that, in general, men typically do not feel comfortable talking about their feelings, especially feelings of vulnerability (Levant, 2011). Special efforts to recruit men in subsequent research would be important so as to tailor support services to their needs.

Finally, patients in Sample 1 were being treated during the course of the study. Although they had been assured that their refusal to participate would not negatively affect their treatment, and that their interview would be kept confidential, their responses might reflect some level of distrust of the medical establishment. This distrust might have been exacerbated for families of color because neither the medical team nor the research team included a professional of color. Thus, the responses of Sample 1 participants might reflect a positive bias.

Biographies

Louise Bordeaux Silverstein, PhD, is a professor in the Department of School-Clinical Child Psychology at the Ferkauf Graduate School of Psychology, Yeshiva University, Bronx, New York, USA.

Marina Stolerman, PsyD, is in private practice in New York, New York, USA.

Nadia Hidayatallah, PsyD, is a postdoctoral fellow at the Child & Family Institute of St. Luke’s–Roosevelt Hospital Center, New York, New York, USA.

Thomas McDonald, MD, is a professor in the Department of Medicine at Albert Einstein College of Medicine/Montefiore Medical Center, Yeshiva University, Bronx, New York, USA.

Christine A. Walsh, MD, is a professor in the Department of Clinical Pediatrics and chief of the Section of Cardiology in the Department of Pediatric Cardiology at Albert Einstein College of Medicine/Montefiore Medical Center, Yeshiva University, Bronx, New York, USA.

Esma Paljevic, EdD, is an assistant professor in the Lienhard School of Nursing, Pace University, in New York, New York, USA.

Lilian L. Cohen, MD, is an assistant professor in the Department of Pediatrics at Weill Cornell Medical College/New York-Presbyterian Hospital in New York, New York, USA.

Robert W. Marion, MD, is a professor in the Department of Pediatrics, and the Ruth L. Gottesman Chair in Developmental Pediatrics in the Department of Pediatrics at Albert Einstein College of Medicine/Montefiore Medical Center, Yeshiva University, Bronx, New York, USA.

David Wasserman, JD, is a consultant to the Yeshiva University Center for Ethics, New York, New York, USA.

Siobhan M. Dolan, MD, MPH, is a professor of clinical obstetrics and gynecology and women’s health in the Department of Obstetrics & Gynecology and Women’s Health at Albert Einstein College of Medicine/Montefiore Medical Center, Yeshiva University, Bronx, New York, USA.