Journal of Cancer Education

, Volume 27, Issue 1, pp 112–119

Preferences for Genetic Testing to Identify Hereditary Colorectal Cancer: Perspectives of High-Risk Patients, Community Members, and Clinicians

Authors

    • Division of General Internal Medicine, Department of MedicineUniversity of California
    • UCSF Women’s Health Clinical Research Center
  • Millie Arora
    • Division of General Internal Medicine, Department of MedicineUniversity of California
  • Christina Hosenfeld
    • Department of PharmacyUniversity of California
  • Uri Ladabaum
    • Division of GastroenterologyStanford University
  • Miriam Kuppermann
    • Departments of Obstetrics, Gynecology and Reproductive Sciences, and Epidemiology and BiostatisticsUniversity of California
  • Sara J. Knight
    • San Francisco VA Medical Center
    • Department of PsychiatryUniversity of California
    • Department of UrologyUniversity of California
Article

DOI: 10.1007/s13187-011-0286-z

Cite this article as:
Walsh, J., Arora, M., Hosenfeld, C. et al. J Canc Educ (2012) 27: 112. doi:10.1007/s13187-011-0286-z

Abstract

The aim of this study was to establish key characteristics that patients, consumers, and health professionals value regarding genetic testing (GT) and personalized medicine using the example of GT for hereditary Lynch syndrome. We conducted a series of focus groups with individuals recruited from a clinic that follows those at high risk for hereditary cancer, individuals recruited from the community, physicians, and genetic counselors. Participants were presented with clinical scenarios about Lynch syndrome testing and asked to identify characteristics that they perceived as important in making decisions about GT. Forty-two participants (19 community members, 8 high-risk and cancer patients, 3 genetic counselors, and 8 physicians) participated. Among community members and patients, the most frequently discussed considerations were the personal impact of GT and family impact, respectively. Among physicians, the most frequently discussed topic was the characteristics of genomic services (e.g., test invasiveness); among genetic counselors, the most frequently discussed topic was evidence and recommendations. A variety of test characteristics were important in decision making about GT. High-risk patients, community members, and health care providers had different priorities. Health care professionals should be aware of differences between their own considerations about GT and those that are important to patients.

Keywords

Colorectal cancer screeningPreferencesFocus groups

Background

While genetic testing has enormous potential to diagnose and prevent disease, its adoption presents many challenges with implications not only for individual persons but also for their family members. Although a person who is identified as being at high risk for a particular disease can receive intensified surveillance and prevention interventions, he or she also may face challenges to receiving health care insurance or develop high anxiety at the knowledge of being at high risk. These and many other factors demonstrate that making decisions about being tested for genetic disease is complex. Despite the complexity of genetic testing decisions, current evidence indicates that the general population has an interest in genetic testing [16], some knowledge about genetic tests for cancer risk [7, 8], and believes genetic testing could improve medical decision making [9]. While some genetic testing is already directly available to the public in the USA, the demand for genetic tests to personalized medicine will continue to affect physicians, who will play a part in ordering and interpreting genetic tests for consumers. This evolving role will require clinical knowledge as well as an understanding of patient interests regarding genetic testing. The goal of this study is to establish the key characteristics that patients, community members, and health professionals see as central to decisions about genetic testing and personalized medicine. Based on the well-established clinical guidelines for genetic testing to identify hereditary colorectal cancer (CRC), we used Lynch syndrome, a type of familial colorectal cancer, as the clinical example. We conducted a series of focus groups to address the question of what factors individuals consider in decision making about being tested for hereditary Lynch syndrome.

Methods

Study Design

In order to gain a wide perspective, we conducted separate focus groups with individuals at high risk for or with CRC, average-risk individuals, physicians, and genetic counselors.

Recruitment

Participants were recruited from UCSF and the surrounding community. High-risk and oncology patients were recruited through the UCSF familial gastrointestinal cancer clinic. Physicians and genetic counselors were recruited through UCSF health professional lists and directories and by personal contact from study investigators. Average-risk individuals were recruited by posting flyers in several UCSF primary care clinics.

Conduct of the Focus Groups

Focus groups were scheduled to last 2 hours in the evening. Each focus group was facilitated by two experienced qualitative researchers including one research psychologist (SJK) and one physician (JW).

The focus groups were conducted with the use of a structured interview guide. After general introductions, participants were asked questions about their general understanding of personalized medicine and genetic testing. Participants were then presented with two clinical scenarios about Lynch syndrome testing (Box). After each scenario, they discussed a series of questions relating to the decision about whether or not to be tested. The purpose of this discussion was to identify and delineate the characteristics that participants perceived as important in making decisions about genetic testing. A list of attributes was generated and then attributes that had been determined a priori were added to the list. A priori attributes included the test required, follow-up, sensitivity, specificity, cost, and privacy. Participants were then asked to generate levels of each attribute: for example, for sensitivity, individuals considered whether or not they would be tested if test sensitivity was 100%, 80%, or 70%.
https://static-content.springer.com/image/art%3A10.1007%2Fs13187-011-0286-z/MediaObjects/13187_2011_286_Fig1_HTML.gif

All participants gave written informed consent. IRB approval was received from the UCSF Committee on Human Research. Non-physician participants received $50 worth of gift card for participating and physicians received $100 honorarium.

Analysis

All focus groups were taped and transcribed. Verified transcripts were analyzed using content analysis. The initial analysis included a review of the transcripts and development of a preliminary coding scheme (e.g., categories and definitions). Two trained raters followed the initial coding structure and coded each transcript. Discrepancies between raters were resolved through discussion with all three raters (SJK, JW, MA). The coding scheme was then refined based on these discussions. The raters then conducted a final coding book with individual unique coding categories, with discrepancies being resolved through discussion with all three raters. Data were summarized using the frequency of mentions for each category and the “line counts” in the text dedicated to each category.

Results

Characteristics of Study Participants

The demographics of study participants are presented in Table 1. A total of 42 individuals participated in six focus groups. There were two groups of average-risk community members (n = 19), two groups of high-risk clinic and cancer patients (n = 8), one group of genetic counselors (n = 3), and two groups of physicians (n = 12). Overall, there were slightly more women than men and the majority were Caucasian, although other ethnic groups were well-represented. Among participating physicians, several specialties were represented, including internal medicine, medical oncology, and gastroenterology.
Table 1

Demographics of study participants

 

Average risk N = 19

High risk and cancer N = 8

Genetic counselors N = 3

Physicians N = 12

Gender

 Male

7 (37%)

4 (50%)

 

6 (50%)

 Female

12 (63%)

4 (50%)

3 (100%)

6 (50%)

Marital statusa

 Never married

7 (37%)

1 (13%)

1 (33%)

 

 Divorced/separated

4 (21%)

   

 Living w/partner

4 (21%)

   

 Widowed

1 (5%)

2 (25%)

  

 Currently married

3 (16%)

5 (63%)

2 (67%)

 

Race/ethnicity

 Alaskan Indian

   

1 (8%)

 American Indian

    

 Asian/Pacific Islander

2 (11%)

2 (25%)

1 (33%)

2 (17%)

 African American/Black

4 (21%)

  

1 (8%)

 Caucasian

11 (58%)

5 (63%)

2 (67%)

7 (58%)

 Latino/Hispanic

1 (5%)

1 (13%)

 

1 (8%)

 Other

2 (11%)

1 (13%)

  

Highest level of education completed

 Grade school

    

 Some high school

    

 High school grad

1 (5)%

   

 GED

    

 Some college

6 (32%)

3 (38%)

  

 College grad

4 (21%)

3 (38%)

  

 Education post-college

4 (21%)

   

 Grad school

4 (21%)

2 (25%)

3 (100%)

12 (100%)

Type of employment

 Paid fulltime

3 (16%)

3 (38%)

2 (67%)

12 (100%)

 Paid part-time

1 (5%)

 

1 (33%)

 

 On med leave or LOA

    

 Unemployed, looking

5 (26%)

3 (38%)

  

 Unemployed, not looking

    

 Retired, not working

1 (5%)

1 (13%)

  

 Retired, paid

    

 Retired, volunteer

2 (11%)

   

 Permanent disable

4 (21%)

1 (13%)

  

 Other

3 (16%)

   

Incomeb

 $0–<$10 K

5 (26%)

1 (13%)

  

 $10–<$30 K

7 (37%)

   

 $30–75 K

3 (16%)

3 (37%)

  

 >$75K

3 (16%)

4 (50%)

3 (100%)

12 (100%)

Insurance typeb

 Private insurance

6 (32%)

6 (75%)

3 (100%)

12 (100%)

 Medicare

3 (16%)

   

 Medicaid, Medi-Cal

2 (11%)

1(13%)

  

 No payment for insurance

4 (21%)

   

 Other

3 (16%)

1(13%)

  

aMarital status not available for physicians

bNumbers do not total to 42 due to missing values

Coding Categories

A total of 10 unique coding categories were developed. They included personal impact, societal impact, family impact, characteristics of genomic service, test quality, financial issues, privacy, evidence and recommendations, discussion of risk, and experiences and emotions. The definitions for each of these categories are described in Table 2.
Table 2

Coding categories

Category

Definition

Personal impact (PER)

This category refers to discussion of benefits to the person who is deciding whether or not to go through genetic testing and/or counseling. This person could be the patient with the condition (e.g., cancer) or could be a family member of the person with the condition who is deciding to go through some testing. It includes discussion about early detection of cancer, treatment that can be done based on the outcome of genetic testing, lifestyle changes of the patient, and any knowledge that the patient may gain from genetic testing. This code will often be double-coded with the “Family Impact” code.

Societal impact (SOC)

This category refers to the discussion of any benefits or costs to society. This code can be used in discussion of knowledge for the greater good, a patient getting tested to contribute data to research and/or any knowledge that can be offered to various groups or organizations. This code should not be used for any discussion of benefits to family, finances, privacy, and experiences.

Family impact (FAM)

This category refers to any discussion of family members including parents, siblings, children, any other blood-related relatives, spouses, and partners. This code can include any benefits to the family including discussion about early detection of cancer in family members, treatment that family members can go through based on the outcome of genetic testing, lifestyle changes of family members, and any knowledge gained by the family members. It also includes any discussion about family dynamic, concern for family, and reactions by family members. This code includes discussion of impact on family. This code does not include discussion of a family member making a personal decision to undergo testing and/or counseling—this would fall under the Personal Impact code. This code does not apply to discussion mainly about financial issues, privacy, and emotional experiences.(Ex. “For me and my family this would cost too much” would be coded as Financial Issues.) This code will often be double-coded with the “Personal Impact” code.

Characteristics of genomic service (CHAR)

This category refers to discussion of features of different types of genetic testing, including tests currently available and possible tests available in the future. This code includes discussion of level of discomfort, invasiveness, pain, and amount of preparation before the test. This code also includes discussion of family history taking and documentation, genomic counseling, and seeing a geneticist. This code does not include any discussion of accuracy of the test. This code can refer to a description of the pathway that the patient will/does go through during genetic testing.

Test quality (QUAL)

This category refers to any discussion about the accuracy, precision, sensitivity, and specificity of any genetic test. This also includes discussion about the quality of the test and false positives and negatives.

Financial issues (FIN)

This category refers to any discussion of cost. It can be used to code discussions of who is paying for the test, cost to insurance companies, co-payments, out-of-pocket payments, any discussion of insurance coverage of the cost, and any mention of dollar amounts. It should also include discussion of any other expenses a person may have (e.g., groceries, rent). In discussion about family and cost, refer to the context of the discussion to decide which code to use.

Privacy (PRIV)

This category refers to any discussion of privacy concerns. This includes who receives test results (e.g., yourself, doctor, insurance company, employer), discussion of stigma and discrimination, and insurability. Context of the discussion should be noted for this code. For example, a person may be concerned about family obtaining results. In this case, context of the discussion should be noted in order to decide which code to use.

Evidence and recommendations (ENR)

This category refers to discussion of recommendations that doctors and patients use in making decisions about genetic testing and other genomic health services. It includes citing guidelines, recommendations that doctors look at, and patients being interested or not interested in established guidelines (e.g., patient wants a test regardless of guidelines). This code also includes how the genomic testing is presented (i.e., how one learns about the test) and discussion about the facts of genomic testing. This code encompasses the basis for the testing and how the practitioner applies evidence for genetic testing.

Discussion of risk (RISK)

This category refers to discussion of risk and perceived risk. This code can include pre-test probability of developing condition and post-test risk (i.e., risk of developing other diseases based on treatment after results).

Experiences and emotions (EE)

This category includes any discussion of emotional factors and experiences of the person who is deciding to undergo testing and/or counseling. This code should be used for discussion of peace of mind, not wanting to know, cancer worry, taking or taking away responsibility, guilt, stress, anxiety, attitude, feelings, behavior, experience with cancer, and/or other diseases. This code should not be used for discussion of experiences and emotions with family.

Percentage of Discussion Devoted to Each Coded Theme

The percentage of discussion devoted to each theme is described in Table 3 Themes that were frequently discussed included personal impact, family impact, financial impact, and characteristics of genomic services. The areas of greatest focus varied among the groups. Among average-risk participants and high-risk/oncology participants, the area of most discussion was the personal impact of genetic testing. In both groups, the second most discussed area was the impact on family and the third most frequently discussed area was financial issues.
Table 3

Percentage of discussion (noted by line counts) devoted to discussing each coded theme by group

 

Average-risk patients

High-risk/oncology patients

Genetic counselors

Physicians

Personal impact

254 (23%)

274 (32%)

47 (11%)

59 (6%)

Societal impact

86 (8%)

66 (8%)

0 (0%)

0 (0%)

Family impact

190 (18%)

194 (23%)

24 (6%)

133 (13%)

Characteristics of genomic service

75 (7%)

64 (7%)

24 (6%)

273 (28%)

Test quality

107 (10%)

44 (5%)

29 (7%)

24 (2%)

Financial issues

163 (15%)

77 (9%)

10 (2%)

135 (14%)

Privacy

90(8%)

39 (4%)

18 (4%)

108 (11%)

Evidence and recommendations

0 (0%)

6 (1%)

123 (29%)

85 (9%)

Discussion of risk

50 (5%)

32 (4%)

63 (15%)

58 (6%)

Experiences and emotions

69 (6%)

58 (7%)

85 (20%)

11 (11%)

Among physicians, the most frequently discussed topic was the characteristics of genomic service, followed by financial issues and family impact. The most commonly discussed topic among genetic counselors was evidence and recommendations, while the second most frequently discussed topic was experience and emotions and the third most discussed topic was discussion of risk.

Exploration of the Coding Categories

We conducted a more detailed analysis of some of the commonly reported coding categories. For personal impact, themes included wanting to gain knowledge, regardless of whether something could be done, “needing” to know, feeling a sense of responsibility for family, wanting to make lifestyle changes and for high-risk patients, giving an explanation for things that were not able to be explained before testing. Examples included:

“I’d like to know. I’d like the knowledge so that I could prevent it, take early action, rather than just taking a chance. If there’s some way to find out if you have a specific gene, and you’re predisposed to a certain type of cancer, I think that knowledge is valuable.” — (High–risk patient)

“I think that genetics testing – you would know the results of your blood tests, and that you are predisposed to cancer, that you’d want to make a lifestyle change. Like, if you had lung cancer in your family, and you were a smoker, you would want to quit smoking. … You could probably eat a better – have a better diet, eat more vegetables. You know, maybe not – cut down on meat. Just lifestyle changes. To lower the chances of getting that particular cancer.” — (Average-risk patient)

“Yes. So I could help the other members of my family, and let them know that they need to get tested also.” — (Average-risk patient)

For family impact, a major focus was on the potential impact of genetic testing on an individual’s children. Participants often expressed a strong desire to know “for their families.” Participants expressed concern about wanting to allow family members to make their own decisions and understanding trends in one’s own family. Examples included:

“If I knew I had it, and I would want my closest family members to get checked out, so that way we could be safer for even their kids, when they have ‘em, that they won’t contract it.” — (Average-risk patient)

“Personally, I would get tested. I think one of the key things in here is that it’s up to your relative to decide. So I would. And that leaves them to decide if they want to. So I’m not making the decision for anybody, I’m just making a decision for myself.” — (High-risk patient)

“Obviously, I’m interested for my kids. That’s the biggest thing I want to get out of it.” — (High-risk patient)

For financial issues, cost, particularly whether or not a genetic test was covered by insurance or how much had to be paid out of pocket, was frequently cited as a factor in individuals’ deciding whether or not to be tested. Community members expressed concern about any out-of-pocket cost as illustrated by the following comments.

“If it was free, or if it was something your insurance covered, or something you can pay out of pocket for, a big consideration for me.” — (Average-risk patient)

“What are the cost factors involved in this? And who’s payin’ for it, and is it covered – like me, I have Medicare and Medi-Cal. Would this be something that’s going to be covered under Medicare and Medi-Cal, or is this an out-of-pocket expense?” — (Average-risk patient)

“I would do it, yeah. I would do whatever I could do to try to contribute in a positive way to the situation. Like I said, I would inform as many of my family members as possible, hey, you guys might want to look into this testing. But it’s – even for me, I just have to look at if it’s possible for me to – if I can afford the test, then I’d do it.” — (Average-risk patient)

Discussion

In this qualitative study, which included the perspectives of various stakeholders, we found that a variety of issues are important in decision making about genetic testing, including personal and family impact, financial impact, and the characteristics of genomic services. Decision making about genetic testing is complex, as evidenced by the 10 different categories of topics discussed. For high-risk patients and average-risk consumers, personal impact was discussed most, followed by family impact and then followed by financial issues. For health professionals, characteristics of genomic services and evidence and recommendations ranked highest, although physicians also ranked family impact and financial issues highly.

Early findings regarding patients’ attitudes toward genetic testing demonstrated variation of specific concerns. Consumers worried that tests may pose a potential loss of privacy, and have expressed fears about the potential for discrimination, anxiety about waiting for test results, and worry about the impact of genetic information on family members [1012]. However, surveys also have shown that people are very interested in learning about a personal genetic predisposition to cancer [1]. In one study, 87% of people responding wanted to know if they had a genetic predisposition to cancer [13]. Factors that have been associated with interest in genetic testing include perceived risk for cancer and perceived benefits and risks of testing [14]. In our study, the impact on family was also a major factor influencing decision making about genetic testing for Lynch syndrome.

While there is high interest in genetic testing and the factors that affect its use [1518], little is known about what patients value, how patients value, and what factors determine genetic testing utilization [19]. Studies of specific diseases (e.g., Huntington’s disease, breast cancer) have identified a range of attitudes and other factors that may influence uptake of genetic testing but little is known about what attitudes and factors are more important than others from the consumer perspective. The current study shows that many factors influence decision making about genetics and that the relative importance of these factors varies among stakeholder groups.

Willingness to have genetic testing has previously been primarily studied in families at risk for hereditary cancers [2024]. In a prior study of first-degree relatives of individuals with hereditary nonpolyposis colorectal cancer, the majority indicated that learning about their children’s risks would be the most important reason to consider testing [25]. One prior study did compare knowledge of and interest in genetic testing for hereditary CRC syndromes in CRC patients, their families, and average-risk controls. All three groups were interested in testing, but had inadequate knowledge about testing and its implications, and that reasons for being tested or not tested varied among the groups [2]. The current study makes a unique contribution in that it provides information on the actual characteristics of genetic testing that are valued not only by high-risk individuals but also by average-risk individuals, physicians, and genetic counselors. The impact on family was important to all participants; although the relative importance of family varied, all felt that it was important, and certainly, it is a factor that physicians should consider when counseling patients about genetic testing.

In this era of increasing focus on genomics and genomic testing, physicians and other health professionals will increasingly consider genetic testing for high-risk individuals. Although this is an exploratory study, health professionals counseling high-risk individuals about genetic testing should certainly discuss family and personal impact and should probably discuss financial issues and characteristics of genomic services as well.

Our study had several limitations. First, it was a qualitative study, and although it suggests that different factors may have more or less importance to different groups, these results should be interpreted as hypothesis generating. Second, there were only three genetic counseling professionals, which limit the conclusions that can be drawn about them.

Despite these limitations, this study provides a unique look at the variety of characteristics that are important to different types of stakeholders in decision making about genetic testing. As the availability of genetic testing increases and as the public interest in genetic testing continues to increase when counseling individuals about genetic testing, physicians should be aware of the multitude of factors that influence decision making. Recent studies have shown that individuals want primary care physicians involved in genetic testing decisions and disclosures [26]. Thus, primary care physicians should be aware of the variety of characteristics that individuals value when making decisions about genetic testing.

Conclusion

A variety of characteristics are important to individuals in decision making about genetic testing. High-risk individuals, community members and physicians had different priorities. Primary care physicians should be aware of the variety of characteristics that individuals value when making decisions about genetic testing.

Copyright information

© Springer Science+Business Media, LLC 2011