Factors Associated with Acceptability, Consideration and Intention of Uptake of Direct-To-Consumer Genetic Testing: A Survey Study

Abstract

Background

With interest in personalised health care growing, so is interest in personal genetic testing. This is now offered direct-to-consumer, thereby referred to as direct-to-consumer genetic testing (DTC-GT). Criticisms have been expressed on whether a truly informed decision to undergo testing is made with regard to these services. In order to provide relevant information to achieve this, knowing the characteristics of the expected user population is helpful. Therefore, the aim of this study is to identify characteristics of individuals who (1) find the concept of DTC-GT acceptable and (2) consider undergoing DTC-GT in the distant or near future.

Methods

This cross-sectional study investigated factors associated with acceptability, consideration and intention in the Dutch general population. Studied variables included awareness, principles and how-to knowledge, attitude, innovativeness, and multiple demographic characteristics. Generalised linear models were applied to identify associated variables.

Results

Full data was obtained for 836 respondents. Of those, 18.3% found DTC-GT somewhat or totally acceptable, whereas 12.6% considered and 5.5% intended to undergo DTC-GT in the distant or near future. Acceptability was greater with lower principles knowledge, and consideration and intention with lower how-to knowledge. A more positive attitude and greater innovativeness were associated with an increase in all 3 outcomes.

Conclusion

Informed decision making may be hampered as individuals with lower how-to knowledge were found to be more interested in pursuing testing. The identified characteristics can be used in development and distribution of public and personalized information, in order to help consumers make a truly informed decision.

Introduction

Personalised health care, that is, health care tailored to an individual's characteristics and needs, is undergoing rapid developments. This is, at least partially, the result of increased pressure from authorities of taking responsibility for one's own health. In this light, an interesting and emerging scientific field is that of personal genetic disease-risk testing, which aims to estimate risks of developing specific common chronic diseases in the future, based on genetic profiling. In contrast to the traditional genetic testing in the medical setting, which is often of a diagnostic nature, personal genetic disease-risk testing is mostly non-diagnostic but of a preventive nature, and aimed directly at the consumer without interference of a health care professional. Therefore, it is often referred to as direct-to-consumer genetic testing (DTC-GT). Such services are mostly offered online; consumers send their DNA sample by mail and receive their results in an online report [1].

One of the criticisms of DTC-GT is the dubiousness of attaining a truly informed decision of the customer to undergo testing. Indeed, companies have now mostly offered their product as a non-medical product intended “for informational purposes only”, thereby deferring attaining a true informed consent to a mere agreement to terms of purchase, such as agreeing to use and store the consumer's samples and data [2, 3]. Yet, a majority of companies simultaneously and conflictingly state on their websites that results can be used to inform health decisions [4]. Therefore, considering the nature of the information from such genetic tests, consent should not be reduced to merely ticking a box but instead companies should strive to attain a well-informed decision [5].

An informed decision should, at least, be (1) based on sufficient relevant information and (2) consistent with the values of the individual making the decision [6]. This means that, primarily, complete and objective information provision is essential. However, although DTC-GT companies should feel morally obliged to provide this information, the majority of their websites were found to be misleading, overemphasising benefits and requiring high literacy [4, 7, 8]. Therefore, other stakeholders such as governmental authorities or the medical community should step in to ensure adequate pre-test education among others through the availability of high-quality, complete and objective information. In order to develop and distribute this information more effectively, it is helpful to know the characteristics of the expected user population. Therefore, the aim of this study is to identify characteristics of individuals who (1) find the concept of DTC-GT acceptable, (2) consider undergoing DTC-GT at some point future, or (3) intend to undergo DTC-GT in the coming year.

Methods

Design

This cross-sectional analysis is part of a study into knowledge and attitude towards DTC-GT, and determinants of intended use in the Netherlands. It is based on an online survey, which was distributed in June 2017. The survey was anonymous, confidential and voluntary. The study protocol was approved by the Medical Ethics Review Committee of Zuyderland (number 17-N-58).

Participants

Participants were recruited from an online panel (Flycatcher, Maastricht, The Netherlands) representative of the Dutch adult (18+) population, based on age, gender and education level. Panel members were approached via e-mail and were required to sign an electronic informed consent form before participating. They received a personalised link to the web questionnaire. A reminder e-mail was sent after 5 days in case of non-response.

Data Collection

Due to the presumed relative unawareness of DTC-GT among the Dutch population, the survey started with a short description of the concept, particularly focussing on disease-related and lifestyle-related DTC-GT (online suppl. Text 1; for all online suppl. material, see www.karger.com/doi/10.1159/000492960). The results of the current report focuses solely on 2 types of disease-related testing: (1) genetic testing to estimate disease risks for future development of common chronic diseases, and (2) carrier status testing, which tests whether an individual carries a single copy of a genetic mutation which will cause a genetic disorder when present in both copies. This latter type of testing is relevant for (future) offspring, as it could result in disease if both parents each pass on one mutated copy.

The items included in the questionnaire were composed based on existing models of behaviour change [9, 10], previous literature [11, 12, 13, 14, 15, 16, 17] and expert opinion in the field of genetics, DTC-GT, health education and informed decision-making.

The following quantitative items were collected through the survey:

Outcome Parameters

Acceptability of the DTC approach in genetic testing was measured as “How acceptable is it in your opinion that private companies provide results of genetic tests directly to consumer without mandatory involvement of a health professional (doctor, geneticist, counsellor)?” Respondents had to rate their answer on a 5-point scale (1 – Completely unacceptable to 5 – Completely acceptable).

Consideration referred to a general intent to undergo DTC-GT for disease-related purposes at some time in the future, reflecting a willingness to think about undergoing testing but not yet intending to do so. Consideration was measured using the following question: “Would you consider doing DTC genetic testing for disease-related purposes in the future?” This was rated on a 5-point scale (1 – Certainly no to 5 – Certainly yes).

Intention to undergo testing was operationalized as the intention to undergo testing in the next year, reflecting a more committing intent compared to consideration: “Do you intend to use a DTC genetic test for disease-related purposes in the next year?” Both were rated on a 5-point scale (1 – Certainly no to 5 – Certainly yes).

Independent Variables

Awareness of DTC-GT for disease-related purposes was asked with 1 question per purpose: “Prior to participating in this research, had you ever heard of DTC genetic tests for disease-related purposes?” This resulted in a yes/no answer.

Previous use of DTC-GT for disease-related purposes was measured by first asking whether the participant had ever undergone genetic testing to receive any type of information on their personal genetic profile. Second, if yes, whether this genetic test was obtained direct-to-consumer, through a medical professional or otherwise. Third, the purpose of this test was asked through multiple-choice options, including the answer options of disease-related purposes. This resulted in a yes/no answer.

Two types of knowledge were distinguished: principles knowledge and how-to knowledge [9]. Principles knowledge, referring to the theoretical information about the functioning principles of an innovation (i.e., DTC-GT), refers in our context to the understanding of principles of genetics, which was measured with 14 true/false questions, with the added option of “don't know” (online suppl. Text 2). Answers were rated as correct or incorrect (with “don't know” considered incorrect) and a total sum of correct answers was calculated. How-to knowledge, originally described as the practical knowledge required to use an innovation properly, refers in our context to the understanding of the DTC-GT results. Sample results based on reports from existing companies were given, reflecting 2 types of information: on carrier status and on risk of disease (online suppl. Text 3). Understanding of these results was evaluated with 6 multiple-choice or open questions, and rated as correct or incorrect to come to a total sum of correct answers (online suppl. Text 4).

Attitude towards DTC-GT for disease-related purposes was measured using 8 items of opposite meaning, which was rated on a semantic differential 5-point scale. Respondents were asked: “DTC-GT for disease-related purposes is…,” with the following answer items: beneficial to the health of the user – harmful to the health of the user, beneficial for society – harmful for society, leading to healthier lifestyle of its users – leading to unhealthier lifestyle of its users, leading to increased feelings of privacy of the user – leading to decreased feelings of privacy of the user, ethical – unethical, useful – not useful, reassuring – frightening, important – not important. These items were based on previously reported attributes towards DTC-GT [18, 19, 20, 21, 22, 23] combined with expert opinion. Negative items were reverse scored so that a higher score represents a more positive attitude. Reliability of the resulting scale was high at α = 0.85.

Innovativeness is an individual's value towards innovation (i.e., DTC-GT for disease-related purposes), reflecting the relative earliness of adopting new ideas compared to others [9]. In our study, this was operationalized based on how open the participant is in terms of using new tests or treatments in the medical domain. For this, a medical domain-specific innovativeness scale was used [24, 25]. Participants were asked to rate 3 statements (1, “If I heard of a new medical test or treatment, I would try to find out more about it”; 2, “Among my friends, I am usually one of the first to find out about a new medical test or treatment”; and 3 [reverse-scored], “In general, I am hesitant to undergo a new medical test or treatment,”) on a 5-point Likert scale from strongly disagree to strongly agree. Higher scores reflect a greater level of innovativeness. Reliability was satisfactory at α = 0.69.

Finally, information on the following variables was collected: age, gender, educational level, being religious (yes/no), having a partner (yes/no), having biological children (yes/no), having adopted or step children (yes/no), planning to have children (yes/maybe/don't know/no), having family members diagnosed with a genetic disease (yes/no/don't know), having a chronic disease diagnosed by a physician (yes/no), and self-rated health (1-Poor to 5-Excellent).

Data Analysis

To identify factors associated with acceptability, consideration and intention of DTC-GT we applied a generalized linear model. First, we ran univariate analyses with an α = 0.20 cut-off to identify the individual factors to be entered into the multivariate model. We checked linearity for each variable that originated from a Likert scale question. If the association appeared to be non-linear, the variable was then entered into the model as categorical variables. In the final model, a α = 0.05 cut-off was used to finally determine the set of associated variables. All analyses were done in Stata version 14 (StataCorp, 2015).

Results

Study Population

Of the 1,693 panellists who were contacted, 1,063 responded to the questionnaire (63% response rate). Of these, 9 were excluded due to low response quality (e.g., repeated patterns of answers) and 218 had to be excluded due to missing answers, leaving 836 respondents with complete data to be included in our analyses.

The main characteristics of the study sample are displayed in Table 1. The population was comparable to the Dutch population with regard to gender, age and education. The middle education level group was overrepresented, with 43.3 compared to 32.5 and 24.2% for low and high level respectively, but not different from distribution in the Dutch population. Most respondents had a partner and were non-religious. The great majority (76.6%) did not plan to have children, 62.0% had biological children and 6.8% had adopted or step children. Most respondents (51.6%) rated their health as “good.” Approximately a third of respondents had a chronic disease and 22.7% had a family member with a genetic disease. Compared to non-respondents, respondents did not differ by gender or education, but a greater proportion of non-responders was seen in the youngest cat egory and a smaller proportion in the older category (Χ²[2] = 15.78; p < 0.001).

Table 1.

Descriptive statistics of the study sample

	n	%	Dutch general populationa, b, %
Total	836
Age
18–39	250	29.9	35
40–59	299	35.8	38
60+	287	34.3	30
Gender
Male	422	50.5	49
Female	414	49.5	51
Education level
Low	272	32.5	31
Middle	362	43.3	44
High	202	24.2	25
Partner
Yes	600	71.8
No	236	28.2
Religious
Yes	357	42.7
No	479	57.3
Planning to have children
Yes	129	15.4
Maybe	29	3.5
No	640	76.6
Don't know	38	4.5
Biological children
Yes	518	62.0
No	318	38.0
Adopted or step-children
Yes	57	6.8
No	779	93.2
Self-rated health
Excellent	46	5.5
Very good	144	17.2
Good	431	51.6
Fair	187	22.4
Poor	28	3.3
Having a chronic disease
Yes	303	36.2
No	533	63.8
Genetic disease in the family
Yes	190	22.7
No	450	53.8
Don't know	196	23.4
Previous use of DTC-GT for disease-related purposes
Yes	2	0.2
No	0	99.8
Awareness of DTC-GT for disease-related purposes
Yes	238	28.5
No	598	71.5

^aGouden standard – CBS in collaboration with MOA 2016.

^bPercentages may not add up to 100% due to rounding.

Correlates of Acceptability, Consideration and Intention

Nearly a fifth of respondents (18.3%) found DTC-GT somewhat or totally acceptable. Up to 12.6% probably or definitely considered undergoing DTC-GT in the distant future, whereas 5.5% intended to undergo DTC-GT in the coming year. Details on all answer categories are displayed in Table 2.

Table 2.

Summary statistics for outcome parameters

	n	%
Acceptability
Totally unacceptable	229	27.4
Somewhat unacceptable	243	29.1
Neutral	211	25.2
Somewhat acceptable	101	12.1
Totally acceptable	52	6.2
Consideration
Definitely not	242	28.9
Probably not	214	25.6
Maybe/maybe not	275	32.9
Probably yes	81	9.7
Definitely yes	24	2.9
Intention
Definitely not	438	52.4
Probably not	210	25.1
Maybe/maybe not	142	17.0
Probably yes	30	3.6
Definitely yes	16	1.9

n, number of respondents.

Acceptability

Respondents were more acceptable of DTC-GT when they had lower levels of principles knowledge. Also, a more positive attitude and greater innovativeness were associated with greater acceptability. In addition, younger and non-religious people, as well as those with poor or excellent self-rated health tended to find DTC-GT more acceptable (Table 3 and online suppl. Table 1).

Table 3.

Multivariate analyses for acceptability, consideration and intention

	b	SE b	p value
Acceptability
Age	Per 1 year increase	−0.015	0.003	0.000*
Education	Low	Ref.
	Medium	0.111	0.092	0.227
	High	0.040	0.110	0.719
Being religious	Yes	−0.160	0.076	0.035*
	No	Ref.
Planning to have children	Yes	−0.269	0.142	0.057
	Maybe	−0.013	0.219	0.951
	Don't know	0.069	0.198	0.727
	No	Ref.
Having biological children	Yes	0.024	0.094	0.800
	No	Ref.
Self-rated health	1	0.340	0.169	0.043*
	2	0.066	0.107	0.539
	3	Ref.
	4	0.100	0.097	0.303
	5	0.627	0.211	0.003*
Principles knowledge	Per 1 increase in score	−0.048	0.014	0.001*
Attitude	Per 1 increase in score	0.580	0.061	0.000*
Innovativeness	Per 1 increase in score	0.175	0.044	0.000*
Consideration
Age	Per 1 year increase	−0.015	0.003	0.000*
Education	Low	Ref.
	Medium	0.021	0.071	0.765
	High	0.049	0.084	0.559
Planning to have children	Yes	−0.189	0.110	0.086
	Maybe	−0.146	0.169	0.390
	Don't know	−0.030	0.153	0.846
	No	Ref.
Having biological children	Yes	−0.045	0.075	0.529
	No	Ref.
Having a genetic disease in the family	Yes	0.231	0.075	0.002*
	Don't know	0.037	0.073	0.613
	No	Ref.
Having a chronic disease	Yes	−0.051	0.071	0.470
	No	Ref.
Self-rated health	Per 1 increase in score	0.081	0.037	0.041*
How-to knowledge	Per 1 increase in score	−0.052	0.020	0.010*
Attitude	Per 1 increase in score	0.651	0.047	0.000*
Innovativeness	Per 1 increase in score	0.410	0.034	0.000*
Intention
Age	Per 1 year increase	−0.009	0.002	0.000*
Education	Low	Ref.
	Medium	0.040	0.074	0.584
	High	−0.018	0.087	0.837
Having biological children	Yes	−0.010	0.073	0.886
	No	Ref.
Having a genetic disease in the family	Yes	0.090	0.077	0.243
	Don't know	−0.017	0.076	0.827
	No	Ref.
Self-rated health	Per 1 increase in score	0.088	0.037	0.018*
How-to knowledge	Per 1 increase in score	−0.085	0.021	0.000*
Attitude	Per 1 increase in score	0.284	0.049	0.000*
Innovativeness	Per 1 increase in score	0.328	0.036	0.000*

^*α < 0.05.

Consideration and Intention

Correlates of consideration and intention were largely overlapping. First, respondents with lower levels of how-to knowledge were more likely to have greater consideration and intention to undergo DTC-GT. In addition, a more positive attitude and greater innovativeness were also associated with greater consideration and intention. Finally, respondents were more likely to consider or intend to undergoing DTC-GT when they were younger, had a better rating of health. In contrast, having a genetic disease in the family was associated with greater consideration but not intention (Table 3 and online suppl. Table 1).

Discussion

The aim of this paper was to identify characteristics related to acceptability, consideration and intention of DTC-GT.

Innovativeness

Greater innovativeness lies at the basis of early adoption of an innovation [9]. As expected, with DTC-GT being relatively new, our results show that greater innovativeness (as measured within the medical domain) was associated with greater acceptability, consideration and intention. One previous study on clinical genetic testing for BRCA1/2 also found that greater innovativeness was associated with awareness and uptake of testing [24]. To date, this is the first study to examine innovativeness with regard to DTC-GT and results indicate it as an important factor in predicting interest in DTC-GT.

Knowledge and Attitudes

Clearly, a first requirement for adoption to take place is awareness of the innovation. Nevertheless, our results showed no association between prior awareness of DTC-GT and acceptability, consideration or intention. On the other hand, our results showed that acceptability was greater with lower knowledge of principles of genetics (i.e., principles knowledge). Similarly, consideration and intention were greater with lower understanding of genetic test results (i.e., how-to knowledge). This is in line with a previous study among health-care professionals, which found that non-geneticist HCPs, who may be assumed to have lower knowledge of genetics and genetic testing, were more positive about the benefits of genetic testing [26]. Somewhat related, another study found that participants who had been exposed to information on risks of DTC BRCA testing had lower intentions of undergoing testing [27]. Our findings can be interpreted as worrisome, as they may indicate that those individuals with the least understanding of genetic test results would be most interested in pursuing testing. This could also imply that the first prerequisite for informed decision-making, namely, that it is based on sufficient relevant information and may often be violated. In addition, it may imply that with improved information provision to the general public, consideration might be lower.

Finally, attitude towards DTC-GT was positively associated with acceptability, consideration and intention, a finding that is consistent with other research [28]. Coming back to informed decision-making, this could indicate that the decision to consider undergoing testing would meet the second prerequisite, namely, that it is in line with one's personal values.

Demographic Factors

In addition, several demographic factors were found to be associated with acceptability, consideration and intention. First, younger age was associated with all 3 outcomes, which is in line with other research [29, 30]. This may be explained by the greater impact testing may still have on future life, and younger individuals may have a greater general interest in innovation, for example, in the embrace of new technology. Second, self-rated health was associated with acceptability, consideration and intention. For acceptability the association was U-shaped, with increased acceptability for both poor and excellent health. Consideration and intention both increased with more positive self-rated health. In addition, having a genetic disease in the family increased consideration. These findings may both be explained through different perceived utility. Namely, it was found that actual DTC-GT users with poorer self-reported health [8] and members of the general population with serious disease in the family [20] perceived results as having lower utility. Third, identifying with a religion resulted in lower acceptability of DTC-GT. This is consistent with other research on genetic testing, which generally showed lower interest in genetic testing among people who identified with a religion [20]. In contrast to previous research, we found no association with education [11, 31] or gender [8, 11, 30].

Methodological Strengths and Limitations

Our study has identified both new and previously identified factors associated with acceptability of and interest in undergoing DTC-GT. This was done in a large and representative sample of the Dutch population, thereby allowing generalisation of the results. Nevertheless, as with all research, findings should be critically evaluated. First, where possible, validated or previously used questionnaires were used. However, due to the relative novelty of the topic, the questionnaire included several self-developed items. To improve validity as much as possible, existing questionnaires were modified to fit our research and experts were consulted. In addition, the questionnaire was successfully pretested in a small sample. Second, many of our respondents had not yet heard of DTC-GT prior to the survey. At the start of the survey, we offered an introduction to the topic (online suppl. Text 1), which we aimed to be as objective and informative as possible. However, they had to base their opinion on the topic on this relatively brief description and do so in the relatively short amount of time they took to read it. Although there was no time limit to read and reflect on this introductory text, it is likely individuals read this without thorough pondering. As a result, decisions or answers are likely to reflect rather intuitive responses. Therefore, it may be relevant for a future study to attempt to replicate our results after informing respondents more elaborately, for example, by offering a detailed description of obtained information and risks and benefits of testing, in order to obtain results from a well-informed decision. Finally, in behavioural research a well-known phenomenon is that of the intention-behaviour gap, which refers to the discrepancy between the intentions people make and the actual behaviour they perform. As a result, the factors associated with consideration and intention that we identified may not necessarily reflect factors of actual uptake of DTC-GT. We therefore recommend that future studies focus on identifying factors associated with actual uptake.

Conclusion

Our research has identified a number of correlates of DTC-GT acceptability and the consideration to take up DTC-GT in the distant and near future. Of particular interest are the findings that lower knowledge of genetic principles was associated with greater acceptability, and that lower understanding of DTC-GT results was associated with greater consideration and intention to undergo genetic testing. These findings plea for investment in public and personalized information on basic genetic principles and the content of disease-related genetic testing results. The characteristics identified in this study can be used in the development and distribution of public and personalized information and decision support with regard to DTC-GT, in order to help consumers make an informed decision on whether or not to undergo testing. In addition, the FDA already requires positive comprehension tests for DTC-GT to receive market approval [32]. This should be expected of all DTC-GT companies, including those not under FDA jurisdiction.

Disclosure Statement

In addition to his appointment as Professor of Complex Genetics and Epidemiology at Maastricht University, M.Z. directs the University spin-off Personalised Genetics Company called MyBasePair. K.S., D.K., A.W., A.S., H.V., and L.O. have no conflicting interests to declare.

Funding Source

This project was funded by the Maastricht University Interfaculty Program “Eatwell.” The funder had no influence on the design or execution of the research.

Supplementary Material

Supplementary data

Supplementary data