SEARCH

SEARCH BY CITATION

Keywords:

  • direct-to-consumer;
  • disclosure of genetic results;
  • genetic counseling;
  • genomic education;
  • genomics;
  • outcomes;
  • personal genomic testing;
  • personalized medicine;
  • utilization

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

To describe consumers' perceptions of genetic counseling services in the context of direct-to-consumer personal genomic testing is the purpose of this research. Utilizing data from the Scripps Genomic Health Initiative, we assessed direct-to-consumer genomic test consumers' utilization and perceptions of genetic counseling services. At long-term follow-up, approximately 14 months post-testing, participants were asked to respond to several items gauging their interactions, if any, with a Navigenics genetic counselor, and their perceptions of those interactions. Out of 1325 individuals who completed long-term follow-up, 187 (14.1%) indicated that they had spoken with a genetic counselor. The most commonly given reason for not utilizing the counseling service was a lack of need due to the perception of already understanding one's results (55.6%). The most common reasons for utilizing the service included wanting to take advantage of a free service (43.9%) and wanting more information on risk calculations (42.2%). Among those who utilized the service, a large fraction reported that counseling improved their understanding of their results (54.5%) and genetics in general (43.9%). A relatively small proportion of participants utilized genetic counseling after direct-to-consumer personal genomic testing. Among those individuals who did utilize the service, however, a large fraction perceived it to be informative, and thus presumably beneficial.

Direct-to-consumer (DTC) personal genomic testing has garnered controversy, in part due to the availability of testing without a health care intermediary. For instance, one of the major concerns expressed with respect to DTC genomic testing is that test consumers may lack the requisite knowledge of genomics to fully understand their risk results [1]. As such, there has been much debate about the DTC delivery model and whether personal genomic testing should require the involvement of a healthcare provider, such as a genetic counselor (GC) [2-4]. Unfortunately, however, there is little empirical data to inform this debate. For instance, studies to better understand consumers' level of interest in utilizing genetic counseling as an adjunct to testing, or their perceptions of the value or benefits of such services following utilization, could be useful in formulating regulatory and other policy guidelines with respect to DTC genomic testing. As such, the purpose of the current investigation was to describe consumers' utilization and perceptions of genetic counseling services following DTC personal genomic testing for common disease risk.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We performed a secondary analysis of data from the Scripps Genomic Health Initiative (SGHI), a longitudinal cohort study of adults originally designed to assess the psychological and behavioral impacts of DTC genomic testing for common disease. Participant enrollment and return of genomic results took place from October 2008 to December 2009. Participants purchased the Navigenics Health Compass at a discounted rate, and were offered genetic counseling at no additional charge. Details describing the SGHI, the Navigenics test, and the conditions included in the test have been previously published [5-8]. Analyses presented are based on web-based assessments performed at baseline and long-term follow-up. Specifically, the demographic data used in the present study was taken from the baseline assessment and compared between consumers who spoke with a GC and those who did not speak with a GC. Data on utilization of GCs, as well as perceptions of the genetic counseling service among utilizers, was taken from the long-term follow-up assessment, which was completed an average of 14 months (sd = 1.3) after receiving genomic test results. Data collection for the long-term follow-up assessment was complete as of March 2011.

At long-term follow-up, participants were asked to respond to several items gauging their interactions, if any, with a Navigenics GC. Those who did utilize a GC were asked about their perceptions of those interactions and the GC service. Participants were further asked about their reasons for speaking with a GC. Specifically, they were asked to select as many items that applied from the following list: ‘A genetic counselor contacted me’; ‘I did not understand my genetic test results’; ‘I wanted more information on how genetic risk was calculated/determined’; ‘I wanted to discuss my family history’; ‘It was a free service I thought I would take advantage of’; and ‘I wanted guidance in terms of steps I could take to follow-up my risk results.’ Similarly, those who did not speak with a GC were also asked about their reasons for not doing so and were asked to select as many items that applied from the following list: ‘I already understood my results’; ‘I did not know the service was available’; ‘I wanted to but was too busy/never found the time’; and ‘Other’.

Importantly, Navigenics did some genetic counseling outreach, which varied over time in terms of how it was conducted. From the outset of the study to June 2009, no outreach was conducted, but starting in July 2009, they began calling or emailing participants found to be at higher risk for certain diseases based on their genomic results with instructions on how to set up an appointment. In this initial stage, the specific disease risk criteria that triggered this outreach included any one of the following: (i) homozygous risk for Alzheimer's disease (i.e. two copies of the APOE-ϵ4 allele); (ii) multiple sclerosis estimated lifetime risk of 1.5% or greater; (iii) more than two cancers that were color coded orange (orange color code denoted high risk); (iv) more than eight conditions (overall) that were color coded orange; and (v) any condition for which the individual had greater than 60% estimated lifetime risk. In January of 2010, Navigenics began doing outreach with all participants, irrespective of risk results. Further details on Navigenics' outreach methods can be found elsewhere [9].

All statistical analyses were completed using the software package, SPSS and included the use of t-tests, χ2-squared tests, and Mann–Whitney U-tests.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Participants

Out of 3639 individuals who completed a baseline survey, a total of 2037 completed the short-term follow-up assessment and 1325 completed the long-term follow-up assessment. When considered alone, the response rate for the long-term follow-up assessment is 36.4%. However, the number of individuals who completed either or both follow-ups is consistent with an overall study response rate of 61.6%. Among those who completed long-term follow-up, 187 (14.1%) indicated that they had spoken with a GC.

Baseline demographic characteristics and outreach

Baseline demographics are depicted in Table 1. Demographic comparisons between those who utilized a GC and those who did not were performed. Individuals who spoke with a GC were found to be older in age, have a higher education and income, and to be less likely to have a health-related occupation.

Table 1. Baseline demographic characteristics of participants who utilized GC versus those who did not
DemographicUtilized GCDid not utilizep-value
  1. a

    Pearson χ2-Square.

  2. b

    Mann–Whitney U-Test.

n1871138n/a
Gender (% female)55.660.90.17a
Age (median, sd, range)51.0 (11.1), 23–7548.0 (12.3), 19–840.03b
Ethnicity (% self-reported Caucasian)85.684.80.79a
Education (modal category, %)Master's degree, 25.1Four-year degree, 24.40.009b
Income (median category, %)150,000–199,999, 15.0100,000–149,999, 24.9<0.0005b
Health-related occupation (% yes)19.828.40.01a
Marital status (modal category, %)Married, 69.5Married, 66.10.53a
Have children (% yes)56.361.00.27a

In terms of the relationship between Navigenics' outreach and SGHI participants' GC utilization, we found that 140 out of 1153 total participants in the no outreach period (12.1%) utilized genetic counseling, while 47 out of 172 participants in the proactive outreach period (27.3%) utilized counseling. This difference is significant (p < 0.0005) suggesting that proactive outreach was associated with increased GC utilization rates.

GC utilization, perceptions, and outcomes

Reasons given for utilizing and not utilizing genetic counseling services are represented in Fig. 1. The most commonly reported reason for not utilizing a GC was a lack of need to due to the perception of already understanding one's results (55.6%) and the least commonly reported reason was ‘other’ (6.2%). Those who reported ‘other’ (n = 71) were asked to provide a specific reason, which resulted in a wide variety of responses, such as not feeling the need to pursue further explanations (39.4%), already consulting with their physician about their results (14.1%), or still planning to speak with a GC (14.1%).

image

Figure 1. Reasons selected by participants for either utilizing or not utilizing GC. Participants could select more than one response.

Download figure to PowerPoint

Of those who did speak with a GC, the most commonly reported reason for utilizing the service was wanting to take advantage of a free service (43.9%), while the least commonly reported reason was a perceived lack of understanding of one's results (8.0%) (Fig. 1). The outcomes of speaking with a GC are depicted in Fig. 2. Among those who utilized the service, a large portion reported that counseling helped them to feel either more or somewhat more educated about genetics in general (75.5%) and that it either improved or somewhat improved their understanding of their own results (85.0%). A total of 30.4% reported that speaking with a GC made them either more or somewhat more likely to discuss their results with a physician.

image

Figure 2. Outcomes of speaking with a GC about DTC genomic test results (n = 187).

Download figure to PowerPoint

Given that individuals in a health-related occupation were less likely to utilize genetic counseling relative to those not in a health-related occupation, we compared these two groups on our outcomes of interest, but found few significant differences. Among those who utilized a GC, we found that individuals in a non-health-related occupation were more likely to report being contacted by a GC as their reason for utilizing the service (p = 0.04). Among those who did not utilize a GC, we found that individuals in a non-health-related occupation were more likely to report, as a reason for not utilizing the service, that they ‘wanted to [speak with a counselor], but were too busy’ (p = 0.03).

Genetic risk comparisons

We compared the personal genetic risk estimates between those who utilized a GC and those who did not. We found only one risk result out of 28 total conditions compared to be significantly different between groups. Specifically, a higher number of those who utilized a GC had an increased genetic risk for heart attack risk (p = 0.03), which does not remain significant after correction for multiple comparisons.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Our results suggest that a large fraction of those who utilized genetic counseling after receipt of DTC genomic testing for common disease risk perceived it to be informative, and thus presumably beneficial. Specifically, many participants indicated that genetic counseling improved their understanding of genetics in general, as well as their understanding of their own test results specifically. This suggests that genetic counseling may have educational value for consumers of DTC genomic testing. Ultimately, however, only a small proportion of participants utilized the free counseling service offered to them despite some degree of outreach from the genomic test provider. In general, this is consistent with another study in the literature that also showed relatively low genetic counseling utilization rates [9].

The primary reason given for not utilizing genetic counseling services was lack of need due to the perception of already understanding one's results. Consistent with this was that the least commonly reported reason for utilizing a GC was having difficulty understanding one's results. From this, we can deduce that among this selected sample, most genomic test consumers seemed comfortable with their understanding of their own results. Furthermore, there were no significant differences in the personal genetic risk estimates disclosed between utilizers and nonutilizers, suggesting that genomic disease risk itself did not drive utilization.

Limitations of the SGHI have been previously described in detail [5-7]. Briefly, our cohort is made up of individuals in good health per their self-report. Additionally, the sample is predominantly Caucasian with a high level of education and income. As such, this sample is not representative of the general public, although it is believed to be representative of the population of DTC genomic test consumers at the time of the study. In addition, there are some limitations of this analysis, specifically. First, we do not have complete data on the full list of participants who were contacted by Navigenics in the context of their outreach protocol and thus cannot explicitly control for this as a variable in our analysis or analyze this group separately. Second, this particular analysis did not include an objective assessment of how well consumers understood their data or whether counseling actually improved their genetic knowledge. Lastly, the survey used to assess perceptions of genetic counseling services was not a validated instrument, and as such, it is possible that some reasons for uptake or decline of GC services were not captured.

There are other considerations when assessing the feasibility of genetic counseling for consumers who undergo DTC genomic testing. These include the rather limited number of certified GCs in the United States and Canada (3,026 or approximately 1 for every 135,000 people) [10]. Thus, if genetic testing were to become more widespread and routine, implementing a delivery model that requires the involvement of a GC may not be sustainable. Further, in the past genetic counseling training program curricula have not focused on genomic testing for risk assessment of common complex diseases. This, taken together with the fact that DTC genomic testing is still relatively new and limited to a small population of early adopters, raises the possibility that GCs, as a group, may be less well-versed in interpretation and provision of clinical services with respect to this type of testing [11]. There is currently a strong need for a larger GC workforce to accompany the advances in genetics and genomics, as well as updated genetics training curricula focused on next generation technologies and genetic susceptibility testing.

Overall, we found relatively low utilization of GC services in a large cohort of DTC genomic test consumers. This was the case despite the fact that the service was offered free of charge. Of note is that most participants who did not utilize a GC indicated that they felt they already understood their results. In terms of the appropriate delivery model for DTC genomic testing, an empirically driven model to consider may be to provide consumers with the option of GC services to be used at their own discretion. In this situation, consumers would have continued direct access to their personal genomic information while this valuable resource is allocated to those individuals who may benefit most from its utilization.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This work was supported in part by a NIH/NHGRI R21 grant (1R21HG005747-01; C. S. B.), a NIH/NCRR flagship Clinical and Translational Science Award grant (1UL1RR025774-01; E. J. T.), and Scripps Genomic Medicine Division of Scripps Health. We acknowledge the support Laura Ornowski, MS of Scripps who assisted with data collection, as well as Vance Vanier, MD, Michele Cargill, PhD, Elana Silver, MS, and Elissa Levin, MS, CGC of Navigenics, along with their staff of genetic counselors and other personnel who helped support the project.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References