Study characteristics
Sixteen studies were included in the analysis [46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61]. Most were quantitative (n=12), using questionnaires to assess public perceptions [46,47,48,49,50,51,52, 54, 56, 57, 59, 61]. Three studies conducted focus groups [53, 55, 60] while one study used both focus groups and a survey [58]. The US has contributed the most to this field thus far, undertaking six of the 16 studies identified in the literature search [49, 50, 52, 53, 55, 58]. This is followed by Canada (n=2) [48, 51] and Japan (n=2) [54, 59]. Each of the following countries contributed one study: Jordan [56], Korea [57], The Netherlands [61], Singapore [60], Qatar [46] and the UK [47]. Ten of the studies attempted to recruit a representative sample [46,47,48,49, 51, 53, 54, 56, 59, 61]. Higher educated participant populations (compared to the general population) were noted in four studies [48, 59,60,61]. Three studies recruited participants from specific sites [52, 55, 57]. No studies attempted to discern the views of underrepresented populations aside Mallow et al. [58] who conducted focus groups with a rural community (Table3, Supplementary File3).
Education level influenced decisions to hypothetically partake in genomic testing in different ways [49, 51, 56, 59, 61]. Three studies found that more educated individuals were more likely to be interested in testing [49, 56, 59], while two other studies found that being more educated led to more critical attitudes towards testing [51, 61]. One study found no association between education level and attitude towards testing [57]. Khadir, Al-Qerem and Jarrar [56] found that having a low perceived knowledge of genomic testings social consequences reduced the likelihood of having a reserved attitude. Abdul Rahim et al. [46] found genetic/genomic knowledge did not impact whether a participant would engage in testing.
The age of the participant was reported to influence decision making [49, 54, 56, 57, 59], with no consensus on attitudes of older versus younger adults. Lee et al. [57] found that older adults were more likely to approve of integrating personalised medicine testing into standard healthcare. Two other studies also found that older adults were slightly more interested in genomic testing [54, 56]. In contrast, Okita et al. [59] found that older adults were less willing to partake in genomic testing, while Dodson et al. [49] found no association between age and likeliness to have testing.
Abdul Rahim et al. [46] found that marital status was not significantly associated with willingness to partake in testing in Qatari adults, while Dodson et al. [49] found American participants planning to have children in the next five years had significantly increased interest in testing. Dodson et al. [49] was the only study to investigate whether ethnicity influenced decision-making, showing no association.
Okita et al. [59] assessed the influence of employment status on willingness to partake, reporting that students had significantly more positive attitudes towards testing compared to employed respondents. Bombard et al. [48] found that having an income of more than CAD$80,000 led to a 11-12% decrease in likeliness of believing parents have a responsibility to have their child tested via expanded NBS. No study assessed the impact of sex on attitude towards testing, however Lee et al. [57] found that sex did not significantly influence whether the participant had heard of personalised medicine.
Using the NASSS domains we were able to map primary source data to technology (Domain 2), value proposition (Domain 3), the adopter system (Domain 4) and the wider context (Domain 6) (Fig.2). Greenhalgh et al. [39] does not provide specific definitions for their domains, rather they frame these domains in the form of questions that need to be answered. We replicated this approach and adapted the questions to align with our study questions (Supplementary File4).
The NASSS Framework considers the influences on adoption, nonadoption, abandonment, spread, scale-up, and sustainability of healthcare technologies. Domains 2 (Technology), 3 (Value proposition), 4 (Adopter system) and 6 (Wider context) of the NASSS Framework have been addressed in this scoping review to consider how public perceptions are incorporated in the framework.
Domain 2 considers the technical aspects of the technology that will influence its implementation [39]. Questions 2B, types of data generated; 2C, knowledge needed to use the technology; and 2E, Who owns the IP generated by the technology?, are addressed in the primary sources.
This question considers the knowledge generated by the technology and how this is perceived by patients and/or caregivers. Two studies cited the accuracy of genetic information as an issue for their participants [54, 58].
Greenhalgh et al. [39] defines this as the type of knowledge needed by both healthcare providers and patients to use the technology. However, we will only focus on the views of the general public. Although patients of genomic testing do not necessarily need knowledge to undertake testing, the informed consent process is essential. To gain informed consent from patients, understanding the baseline genomic knowledge of the public is beneficial for those taking consent. Knowledge of genetics and genomics was assessed in several different ways across the included articles [46, 52,53,54, 56, 58, 60]. These included asking participants if they had heard of various genetic and/or genomic terms, how they had heard about genomic testing, how participants describe genomics (in a focus group setting) and questions on genetics knowledge.
Abdul Rahim et al. [46] found that less than a third (n=245) of survey respondents had heard of genomic testing while just over half (n=447) had heard of genetic testing. Gibson, Hohmeier and Smith [52] found that 54% (n=7) of their participants had heard the term pharmacogenomics. Hishiyama, Minari and Suganuma [54] found that more than two-thirds of their participants had heard of classic genetic terminology (e.g. DNA, gene, chromosome), whereas fewer participants had heard of newer, genomics terminology (e.g. personal genome and pharmacogenomics). Hahn et al. [53] found that the majority of their participants had not heard the term genomic medicine and personalised medicine. Ong et al. [60] found that English and Mandarin-speaking participants had heard of the term personalised medicine but not precision medicine, while Malay-speaking participants had not heard of either term.
Three studies questioned participants on how they had heard about genomics [46, 52, 53]. Abdul Rahim et al. [46] asked about both genetic and genomic testing whereas Gibson, Hohmeier and Smith [52] asked their participants where they had heard certain terms from. Abdul Rahim et al. [46] found that 30% (n=69) of participants who knew of genomic testing, heard about it through word of mouth. Gibson, Hohmeier and Smith [52] found that 54% (n=7) of participants had heard of pharmacogenomic testing, and other key terms associated with genomics, from the internet. Hahn et al. [53] used focus groups to discern participant understanding of the term genomic medicine, and found that some college students had heard of the term on the news and in biology classes.
Two studies used focus groups to discern genomic understanding [53, 58]. Mallow et al. [58] used a Community Participating Research approach. Community leaders suggested they use terms like genes and family health history rather than scientific terminology to assist discussions with the community. They found that participants were more likely to describe inheriting disease rather than inheriting health and wellness [58]. Hahn et al. [53] found that their focus group participants described genomic medicine in terms of genetics, family history, the genome project, using genetics to heal people and cloning. Ong et al. [60] also used focus groups to discuss baseline understanding of personalised medicine and precision medicine divided into the primary language spoken by the participants, allowing for discussions on terminology specific to the language.
Knowledge of genetic and/or genomic facts was directly assessed in two studies [46, 56]. Abdul Rahim et al. [46] and Khadir, Al-Qerem and Jarrar [56] both questioned respondents on their basic genetic literacy via survey questions. Abdul Rahim et al. [46] found that 56.1% of survey respondents (n=464) were able to answer at least 5 out of 8 genetic literacy questions correctly, while Khadir, Al-Qerem and Jarrar [56] found that participants were knowledgeable in hereditary genetic information but not other scientific facts. Khadir, Al-Qerem and Jarrar [56] also gave participants the opportunity to self-report their knowledge of genetics. Many participants reported having sufficient knowledge on basic medical uses of testing and potential social consequences, such as refusing testing and the rights of third parties to request genetic test results of individuals [56].
For genomic testing, we have interpreted this question to mean whether patients own their genetic information or if it belongs to the group that conducts sequencing. Four studies found that participants had concerns about the privacy of their or their childs genetic information [46, 53, 55, 57]. Hishiyama, Minari and Suganuma [54] also found that 37.1% (n=1112) of their participants were concerned about management and storage of genetic information.
Greenhalgh et al. [39] use this domain to consider the value placed on the technology by healthcare professionals and the patient. Question 3B, demand-side value (to patient), is addressed in the primary sources.
Greenhalgh et al. [39] define this question as the downstream value of the technology, including the evidence of benefit to patients and affordability. Willingness to pay for genomic testing was directly assessed in three studies [50, 52, 57]. Gibson, Hohmeier and Smith [52] found that if the entire cost of the pharmacogenomic test was covered by insurance, 89% of participants (n=24) would undertake testing [52]. Lee et al. [57] determined that age, gender, income, inconvenience of testing and prior knowledge all influenced whether participants would pay extra for personalised medical testing. Cost of testing was a concern for 44.8% of participants (n=316) [57]. Edgar et al. [50] found that most adoptees (72.4%) and non-adoptees (80.3%) were willing to pay between US$1 and US$499. Education level was a predictor for adoptee willingness to pay, while income predicted willingness to pay in non-adoptees [50]. Abdul Rahim et al. [46] did not directly assess willingness to pay, however they noted that a high income was associated with participant willingness to partake in testing.
Hahn et al. [53] and Ong et al. [60] did not directly assess willingness to pay for genomic sequencing, but participants did express concerns about the cost of testing to the individual and whether there would be equitable access to testing.
Greenhalgh et al. [39] use this domain to consider the adoption of the technology. The adopter system includes caregivers, healthcare professionals and patients. Question 4B addresses whether patients will adopt a technology, while 4C addresses if lay caregivers are available to facilitate adoption. As we did not include patients or lay caregivers in our review, we have adapted these definitions to incorporate hypothetical patients and/or carers under the term genomic naive public. Greenhalgh et al. [39] also emphasise patient acceptance and family conflict as influencing factors on use of technology.
Several personal values were identified across the included studies [46, 48,49,50,51,52,53,54, 56, 59]. Abdul Rahim et al. [46] and Hishiyama, Minari and Suganuma [54] found that contributing to science and medical research were reasons to partake [46, 54]. Other reasons for partaking in genomic testing suggested by Qatari adults included improved health knowledge and prevention of future health conditions [46]. This was also suggested by participants in Etchegary et al. [51], Hahn et al. [53] Khadir, Al-Qerem and Jarrar [56].
Bombard et al. [48] found that most of their participants preferred using scientific evidence (82%, n=994) and receiving expert advice (74%, n=897) when making important healthcare decisions. However, only half (53%) of participants had trust in healthcare (n=639). Hahn et al. [53] also found that many participants were sceptical of genomic medicine specifically, and often associated it with genetic engineering and cloning despite these not being directly related to genomic testing. Some participants felt they did not need the information genomic testing could provide, while others who would hypothetically want testing, believed it could promote the development of new treatments and provide more information on family history [53].
Primary reasons for not willing to partake in testing, as noted by Abdul Rahim et al. [46] were lack of time, information or knowledge, and privacy concerns. Similar concerns were suggested by Hahn et al. [53] and Lee et al. [57]. Fear of the unknown was also suggested in Hahn et al. [53] and Mallow et al. [58]. Participants in Hahn et al. [53] also noted they may be uncomfortable with the results, and the results may be too deterministic.
Aside from general concerns about the nature of genomic testing, concern regarding communication of genetic information among family members was also highlighted [47, 51, 53, 56, 58, 61]. Ballard et al. [47] noted that most participants, whether asked to imagine either they or a family member had a genetic condition, believed other family members who might also be affected should be notified. Etchegary et al. [51] and Khadir, Al-Qerem and Jarrar [56] also found that most participants would share genomic test results with family members. Participants in Hahn et al. [53] generally had a positive view of learning about genetic information if it would help other family members as some had family members who had passed away without explanation. Mallow et al. [58], however, found that communicating genetic information to family members may be an issue. Participants cited several reasons for this including: upsetting children and the creation of family issues, older family members not willing to disclose information and stigmatisation by the community, particularly if the information in question regarded mental illness or substance abuse disorders [58]. Participants also suggested they would only discuss genetic risk if there was a health crisis in the family [58]. Etchegary et al. [51], although noting that many participants would want to share information, found that those with the highest education levels and income were less likely to share results with family members. Vermeulen et al. [61] also found that 17% of their participants (n=160) were worried about causing friction within their families. However, participants who believed family history assessments were worthwhile cited disease prevention as a benefit to involving family members [61].
Greenhalgh et al. [39] describe the wider context as the institutional and sociocultural contexts. Examples of the wider context include health policy, fiscal policy, statements and positions of professional and peak bodies, as well as law and regulation. Here, in order to respond to our research questions, we focus on the socio-cultural aspects of the public.
Societal concerns were noted in many studies [51, 53,54,55,56, 58, 60, 61]. Twenty-two percent of participants (n=1425) in the Hishiyama, Minari and Suganuma [54] study noted employment and insurance discrimination as a concern. This was also noted in Etchegary et al. [51] and Khadir, Al-Qerem and Jarrar [56]. Participants in Hahn et al. [53] and Mallow et al. [58] noted discrimination and segregation as key societal issues that may arise. One-third of participants (n=311) in Vermeulen et al. [61] thought that individuals may be coerced into testing if it is normalised.
Cultural context may influence participant responses. For example, Abdul-Rahim et al. found the 45.1% of their respondents (n=241) were in consanguineous relationships [46]. No other study reported on consanguinity, demonstrating that different cultures prioritise different elements when reporting. Abdul-Rahim et al. found that 70.9% population (n=584) were willing to undergo genomic testing [46], whereas Dodson et al. found that 39.5% of their US population (n=805) were somewhat interested and 19.1% (n=389) were definitely interested in genomic testing [49]. These papers demonstrates that different cultures can influence perceptions of genomic testing. However, the Caucasian US population in Gibson et al. were more willing to undergo testing at 81.0% (n=21) [52], showing that even within the same country there can be cultural differences that may lead to differences in perception.
Read more here:
How does the genomic naive public perceive whole genomic testing for health purposes? A scoping review | European Journal of Human Genetics -...
Recommendation and review posted by G. Smith
