Abstract
The last decade has experienced unprecedented uptake of noninvasive prenatal testing (NIPT), creating significant changes in the way prenatal clinicians provide services. Through the lens of social shaping of technology, we examine the effects of the introduction of this technology on the health care system in Ontario, Canada. Using a qualitative descriptive approach, we conducted a cross-sectional study investigating clinicians’ perspectives of NIPT in 2014, 2016, and 2018. Through in-depth interviews (n = 37), we explored their perspectives on the impact of NIPT on their referral practices, workload, coordination of testing modalities, education and counseling, and elicited their views on recent expansions of the test. Findings suggest that the introduction of NIPT has created unintended consequences with respect to clinician workload and wellness, clinician education, equity of access, and public system resources. Responsiveness from decision makers is key to ensuring the responsible use of NIPT in the health care system.
Keywords
Introduction
Technological developments have shaped the field of prenatal genetic testing for chromosomal anomalies since the 1960s (Löwy, 2014), altering conceptions of pregnancy, disability, and normality (Parens & Asch, 2000; Rothman, 1993; Taylor, 2008). The relatively recent and rapid introduction of noninvasive prenatal testing (NIPT) into the prenatal pathway has produced a “paradigm shift” (Chitty & Bianchi, 2013, p. 512) in how prenatal testing is performed and, correspondingly, in how prenatal testing is conceptualized (Ravitsky, 2017). NIPT is a screening test that uses the cell-free DNA in maternal blood to detect fetal chromosomal aneuploidies (Cernat et al., 2019). The sensitivity and specificity rates are estimated around 90% to 99% and 99% to 100%, respectively, for the detection of trisomy 13, 18, and 21, making it the most accurate screening test available for these conditions (Cernat et al., 2019; Mackie et al., 2017). Although NIPT is not diagnostic, its utility arises from its ability to screen early in pregnancy with high accuracy while avoiding the risk of pregnancy loss that can occur with invasive diagnostic tests such as amniocentesis and chorionic villus sampling (Cernat et al., 2019; Skirton et al., 2014). Clinical practice guidelines currently recommend that positive NIPT results are followed by a diagnostic test before a decision about pregnancy termination is made (Audibert et al., 2017; Cernat et al., 2019; Skirton et al., 2014; Wilson et al., 2013).
In Ontario, Canada, where this study was conducted, NIPT became clinically available in 2011 and became publicly funded for those at high risk of fetal aneuploidy in early 2014 (Vanstone et al., 2015a). Initially, the policy to provide public funding for NIPT was unwritten and unpublicized, relying on clinician word of mouth that requests for funding approval for individual patients were now being approved. Funding criteria were eventually codified and circulated more widely, but individual approvals were required until 2018. In 2017, Ontario created a provincial prenatal screening oversight organization (Prenatal Screening Ontario) and commissioned a technology assessment of NIPT which affirmed the policy of providing public funding for those at high risk of fetal aneuploidy (Health Quality Ontario, 2019). Ontario is the most populous Canadian province (Statistics Canada, n.d.), and operates a publicly funded single-payer health care system. Clinicians are reimbursed through different methods, including fee-for-service; however, no additional reimbursement is obtained for patient uptake of NIPT (Ministry of Health and Long-Term Care [MOHLTC], 2019; Vanstone et al., 2015a). Clinicians are reimbursed for offering and counseling about NIPT at the same rate as counseling for other prenatal screening technologies. Similar to many other jurisdictions (Chitty & Bianchi, 2013; Long & Goldblatt, 2014; Minear et al., 2015), the implementation of NIPT into the public system has been characterized as “rapid” and “atypical” (Vanstone et al., 2015b, p. 55), in part due to the efforts to increase uptake by industry actors (Vanstone et al., 2018). Many have argued that NIPT raises a number of ethical concerns because of its potential to increase the number of terminations performed for reasons of chromosomal anomaly (de Jong et al., 2010; Haidar et al., 2018); be used for sex selection (de Jong et al., 2010; Haidar et al., 2016); increase anxiety for pregnant women due to the “burden of choice” (Kibel & Vanstone, 2017; Vanstone et al., 2015a; Wilson et al., 2013); and promote routinization of screening, ultimately resulting in an erosion of informed choice (Haidar et al., 2016; Lewis et al., 2013). There are additional concerns around the use of NIPT due to the role of industry in its adoption and promotion (Haidar et al., 2016; Vanstone et al., 2018); its implications for the disabled community and how this community is perceived (Farrimond & Kelly, 2013; Haidar et al., 2016); its potential to encourage eugenic trends (Farrimond & Kelly, 2013; Haidar et al., 2016); equity of access issues (Allyse et al., 2015; Haidar et al., 2016); and the strain it places on already limited health care resources (Dondorp et al., 2015; Haidar et al., 2016). In addition, some contend that the implementation of NIPT may pose consequences that are “unexpected or inevitable” (Vanstone et al., 2015b, p. 55) such as its impact on the ways in which clinicians practice, or its effect on the structure of the prenatal care pathway (Vanstone et al., 2015b).
The importance of studying these inevitable and unintended consequences (Russell & Williams, 2002) becomes evident when considered through the lens of the theory of social shaping of technology (SST; Wajcman & MacKenzie, 1988; Williams & Edge, 1996). Arising out of critiques of technological determinism, which suggests that technology develops autonomously and without influence from societal factors, SST posits that technology and society co-constitute one another (Bijker, 2009; Williams & Edge, 1996). That is, society influences the development of technology and, in turn, technology influences society. Society’s influence on technology is manifested through conscious and unconscious decisions which are the products of societal factors such as culture, economics, and belief systems (Williams & Edge, 1996). Understanding this influence allows for the consideration that the trajectory of a technology is only one of many possible trajectories, affording the opportunity to engage in a critical examination and problematization of the current trajectory (Williams & Edge, 1996). In the context of NIPT, viewing the issue through an SST lens makes it possible to understand that (a) the ways in which NIPT is implemented into a health care system necessarily affect how this technology is used and accessed, and (b) how NIPT is used and accessed necessarily affects the health care system and broader society in which it resides. Specifically, the implementation of this morally complex technology has the potential to affect the structure of a health care system, the specific ways in which clinicians practice, and the quality of care received by patients. Engaging in critical inquiry about the effect of this technology on society is essential to the responsible development and innovation of this health care technology (Dondorp et al., 2015).
To elicit and examine some potential consequences of the rapid, industry-driven implementation of NIPT, we conducted a cross-sectional qualitative descriptive study that explores clinicians’ perspectives on the implementation of NIPT in the Ontario health care system at three distinct time points: 2014, when publicly funded NIPT was introduced in Ontario; 2016; and 2018, when public funding was reassessed (Health Quality Ontario, 2019). This study was guided by two overarching research questions:
Method
This study used a qualitative description approach (Sandelowski, 2010) to elicit experiences of clinical work with NIPT from a range of front-line clinicians. Data were collected cross-sectionally, beginning in 2014 when publicly funded NIPT was implemented in Ontario, Canada. This research design allowed us to understand how the technology was experienced at each time point as well as longitudinally over time (Calman et al., 2013; Cummings, 2017). We used the theory of SST as a “sensitizing concept” (Blumer, 1954; Bowen, 2006), a “background idea” (Charmaz, 2003) used to “see, organize, and understand experience” (Blumer, 1954) as opposed to a prescriptive theoretical framework. We used SST to develop the research question, to design the study and data collection instruments, and to interpret the findings. This research received ethical approval from the Hamilton Integrated Research Ethics Board, and all participants provided informed consent.
Sampling and Recruitment
Ontario clinicians whose area of clinical practice included exposure to NIPT were eligible to participate in this study. This included family physicians, midwives, obstetricians, genetic counselors, clinical geneticists, and nurses. We used a purposive approach to sampling, which evolved as the technology diffused through the health care system and more primary care providers regularly encountered it in practice.
In 2014, we sampled participants by contacting clinics with an identified alignment to prenatal screening, asking the clinic to circulate our recruitment notice to all relevant clinicians. As participants were identified, we engaged in snowball sampling, asking initial participants to identify colleagues who would have relevant but different experiences with NIPT. For example, we asked obstetricians with a specialist maternal–fetal medicine practice to refer us to community obstetrician colleagues who were regular sources of referrals. To gather a front-line perspective, we contacted several community clinical groups through publicly available contact information and asked for one participant who could represent a typical exposure to this new technology.
In 2016 and 2018, knowledge about NIPT was more widespread, and regular use of this technology by a variety of different types of clinicians had increased. In response to this new context, we prioritized recruitment of primary care clinicians who were now the most common clinical users of the technology. We completed a small number of additional interviews with those who had a specialized clinical practice for which NIPT was relevant and who were knowledgeable about the policy landscape of this evolving technology. None of the participants were interviewed at multiple time points. Our goal was not to present perspectives from the same group over time but rather to gain an overall understanding of the implementation of the NIPT as the technological and policy landscape changed over time.
Throughout recruitment, we adopted the principle of maximum variation, seeking participants with different specialties and sizes of practice, and who worked in a variety of academic centers, community settings, and geographic locations throughout the province. Data collection was considered complete when theoretical saturation was reached, which we determined as occurring when additional interviews that year did not yield new data of analytic significance (Morse, 1995). Supplemental Table 1 provides information on clinician type and year of participation as well as years of clinical experience.
Data Collection
Participants were invited to participate in semi-structured interviews in person or over the phone based on their personal preference and as permitted by their geographical location. We conducted in-person interviews in four different cities across Ontario (Toronto, London, Hamilton, and Ottawa). One group of colleagues at the same clinic requested a focus group rather than individual interviews, which we accommodated. Interviews were conducted by all authors, only one of whom (L.L.) is a clinician (genetic counselor).
The interview guide queried participants’ exposure to NIPT, asking them to identify benefits and disadvantages compared with existing technology, concerns and excitement about the potential of the technology, changes to workflow, areas of confusion, and preferences for continuing clinical education and policy-making. The interview guides were refined in 2016 and 2018 to reflect emerging analytic insights.
Data Analysis
We used a staged coding process, similar to that of grounded theory (Charmaz, 2006; Corbin, 2008), in which units of analysis are divided into their components and recombined in a way that illustrates the relationship between the components. We conducted initial coding in 2014 and 2016 to identify preliminary analytic themes, refine the interview guide, and determine data sufficiency for that year. Although all of the interview guides explored similar questions, the interview guide in 2014 focused on the introduction of publicly funded NIPT into the health care system, which was modified to investigate the suite of expanded testing conditions being discussed by stakeholders in 2016, and further modified to examine clinicians’ perspectives on the implementation of NIPT over time in the 2018 interviews. The majority of analytic work took place after the acquisition of the 2018 data. All authors contributed to analysis, including R.B. and A.C., who are Master’s trained qualitative researchers with some experience working on NIPT and ethical issues in heath technology implementation; L.L., a genetic counselor and qualitative researcher who collected the 2016 data and contributed to interim analysis; and M.V., the principal investigator of the project who collected the 2014 data, performed 2014 and 2016 interim analysis, and has policy and ethics expertise on NIPT and other prenatal screening technologies developed through a portfolio of work over the past decade. We used NVivo v.11 (QSR International, 2015) to manage the data.
In 2018, multiple analysts (R.B. and A.C.) independently conducted initial line-by-line coding on a series of transcripts to identify preliminary categories. Authors met to determine the analytic direction, and subsequent coding was carried out by R.B. and A.C. with regular meetings with M.V. to determine a focused coding schema and discuss the relationship and boundaries of themes and categories. Categories were formed based on the prevalence of information across multiple interviews and the relevance of that information to our research questions, with broader themes emerging from these preliminary categories.
Results
Of the 37 participants who participated in this study, 12 were family physicians, six were midwives, six were genetics counselors, five were maternal–fetal medicine specialists, five were obstetricians, two were resident physicians training to be obstetricians, and one was a nurse. Nine of the participants were interviewed in 2014, 19 were interviewed in 2016, and nine were interviewed in 2018.
Our analysis identified four primary themes: Referral Practices, Scope of Practice, and Workload; Coordination of NIPT with Other Prenatal Testing Modalities; Education and Counseling of NIPT; and Expansion and Evolution of the Test. In each theme, we describe the evolution of clinician perspectives over time.
Each of the identified themes interconnects. Specifically, the referral practices and scope of practice considerations presented in the first theme are useful in understanding how clinicians coordinate NIPT within the range of other prenatal tests that they offer. Moreover, some of the ways in which referral practices occur are a product of deficits in clinician education surrounding NIPT. Finally, a description of all of the themes sets the stage for considering clinicians’ perspectives about the expansion of NIPT to test for other conditions.
Referral Practices, Scope of Practice, and Workload
In general, the introduction of government funding for NIPT in 2014 was portrayed by clinicians across different professions as burdensome. Access to public funding for the test required case-by-case approval from Ontario’s public payer, the MOHLTC. The need for case-by-case approval created a significant amount of paperwork and follow-up labor for each test ordered. The administrative pathway required to obtain funding for each patient had a “major impact” on the workload of some clinicians, characterized by being intense because of the time-sensitive nature of the process.
Inefficiencies in the funding approval process were exacerbated by a lack of clarity about who could order the test, as well as discomfort among primary care providers around the process for doing so. There was little formal guidance available around referrals and ordering NIPT, and “there’s not great data out there I think to guide us, in the published literature . . .” This left clinicians to make decisions independently about how to order this testing. Those who worked in practice groups described using collaborative decision-making. As one maternal–fetal medicine specialist put it, “We use a kind of shared group wisdom.”
However, for some clinicians, this lack of guidance around referrals translated into systemic inefficiencies, particularly for specialist services: I think it would be really great if obstetrical care providers, the primary providers, would be comfortable with filling out the paperwork . . . Because in my mind it’s an extra expense for OHIP [Ontario health insurance scheme] for a patient to come and see me for a consultation when in reality we are expecting the primary care obstetrical providers to counsel about screening anyway . . . we’re already expecting them to have that skill set and for them to order NIPT in cases where it’s very clearly indicated and then only refer to us if it’s positive or if there’s something else.
The increased demand for genetics services and the time-sensitive nature of the referrals led some genetic counselors to worry that the work associated with NIPT was detracting from their own group and individual wellness.
This has potential ramifications on patient care: . . . we are terrified we are going to cause some harm . . . when we get that opportunity to review together there is a consistency to how we approach the cases. So I think there is a patient-care piece to it but there is a self-care piece to it, which losing it is not small. I don’t think we can completely blame NIPT on it but it’s absolutely a major contributing factor. And some of that would be alleviated by a different process for NIPT.
By 2016, the process for ordering NIPT had been streamlined, and clinicians no longer needed to seek preapproval from the Ministry of Health before ordering the test, which helped decrease administrative burden. Some clinicians found that “small little things make a big difference” in terms of their practice efficiency. For example, one family physician described asking patients early on whether they would consider NIPT so that their care could be better planned in advance.
By 2018, community-based clinicians were becoming more familiar with the test: “there were a few obstetricians in that community who were definitely referring their clients to us in the beginning and I found that they were just doing it on their own towards the end of my term there.” Moreover, some specialist clinicians “pushed back” with respect to accommodating referrals from cases that could have been managed in primary care. Many perceived the workload associated with NIPT to be more reasonable, though one clinician qualified, “overall we have more people that are eligible, that we see, as people are becoming more aware that NIPT exists, and so that patient volume overall, I think has probably increased.”
Midwives in Ontario are independent prenatal clinicians who provide care for low-risk pregnant women. Throughout the study time period, they are unable to order NIPT and were required to refer their patients to a physician for counseling and follow-up. This “jumping through hoops” was described as “frustrating” for both types of clinicians, creating an unnecessary time and cost burden in an already time-sensitive process.
Coordination of NIPT With Other Prenatal Testing Modalities
Across the three time periods, clinicians demonstrate a variety of opinions about how to coordinate NIPT with other prenatal testing modalities. Although clinicians’ testing strategy appeared to become more cohesive over time, clinicians in 2018 still did not appear to reach an equilibrium when it came to coordinating testing modalities: “sometimes you’d say well if you’re going to do NIPT should you even bother doing [other prenatal screening tests] and that is an ongoing question that I don’t think is terribly well understood.”
Since 2014, the public funding structure for NIPT in Ontario has remained the same. That is, NIPT is funded for pregnant patients at high risk of fetal aneuploidy and available as a private-pay test to all patients. Some providers agreed with the model of funding for high-risk individuals only, whereas others felt that the test should be offered to everyone given its superior ability to test for probability of aneuploidy when compared with other screening modalities, as well as its ability to reduce “all of that uncertainty and anxiety” associated with a positive result from a less accurate screening modality. Many providers felt that the advantages of NIPT to the individual patient needed to be balanced with broader considerations of health care system resources. Most clinicians felt that the criteria are broad enough to capture all individuals who should be tested, and believed the funding criteria to be fairly clear and reasonable. The criterion of previous aneuploidy was perceived to require refinement as the only relevant trisomies are 13, 18, and 21.
Due to the funding requirements for NIPT, clinicians discussed having to decide whether to introduce the offer of private-pay NIPT to patients who do not qualify for government funding. They expressed that there are concomitant decisions to be made about how to coordinate NIPT with other prenatal screens for this group. Across the study period, some clinicians spoke of offering the test to all average-risk patients, whereas others only offered it to those they perceived had the ability to pay for the test. For example, one family physician noted, “Right or wrong . . . I don’t include it as a current piece of my initial counselling, simply because I know the costs would be prohibitive for that patient.” Others took a hybrid approach, informing patients about all available options and recommending a testing pathway based on their knowledge of the patient, clinical indications, interests, and ability to pay. As described by one midwife, “generally speaking it’s not something that we would actively encourage people to do unless there was a particular reason.” These choices were thoughtfully considered, and often linked to autonomy: “even though we think maybe it’s not the best [option for that patient], it doesn’t come down to us. It comes down to the patient’s choice.” Decisions also appeared to be based on the clinician’s individual practice strategy, the customs of close colleagues, and the recommendations put forth in clinical practice guidelines. This sometimes led to discomfort with the variation in how tests are coordinated: I still don’t always feel comfortable because I feel like there’s so much variance. For example, there are a lot of obstetrical care providers at one hospital in [City] that are still ordering IPS [Integrated Prenatal Screening], where there’s other obstetrical care providers that are telling family physicians to please order FTS [First Trimester Screening] . . . there are different preferences and it all has to do with potentially them ordering the NIPT test.
Inefficient coordination of screening technologies also provides the potential for “duplication of health services” with some patients accessing multiple prenatal tests that identified the same conditions. A few providers described this ad hoc system of prenatal testing as a redundancy that places an undue burden on the health care system. Others did not object to the duplication of services, identifying NIPT as “just a piece of the puzzle” because it cannot detect other important prenatal issues such as placental function. Still others questioned the utility of prenatal testing in general because “you get into the issue of limited funds for health care and where you’re putting them,” with one participant suggesting, “if you were to take $500.00 and put [it] into mental health care for women, you’d have a much better outcome . . . or formula for the kid.” Overall, the future role of NIPT and other prenatal tests was uncertain even after the test had been used in the health care system for 4 years, and clinicians expressed a range of preferences for the integration of NIPT within the prenatal screening pathway.
Education and Counseling of NIPT
Many clinicians across the interview period spoke about the variability of clinicians’ knowledge of NIPT and their ability to counsel about it. This issue appeared most acutely in 2014 when the test was introduced, and appeared to improve over the study time period as more clinicians became aware of the test. Yet, due to the shift in the type of clinicians counseling about the test (from specialists to primary care providers), the information gap remained, and even in 2018, there was a perceived need for greater education and awareness among clinicians. As described by one family physician, I can tell you that a lot of my colleagues, who are family physicians practicing just like myself in [City], have not had the education about this because it’s not as easily accessible as I think it should be. And I think that that’s a huge, huge, huge problem, because I think the variation in knowledge-base with regard to family doctors across, I can only speak of [City], is very varied, as well as the knowledge-base and level of education of the patients. So, I think that the tests that are being offered and how people are being counselled is very much dependent on who happens to be your care provider and how much they know about this.
Family physicians engaged in generalist practice described NIPT counseling as “challenging” because it is a relatively small part of their practice and they have very little time to give the patient a lot of information. Some feel that the test was implemented very quickly and their “practice had to change overnight.” Many also felt that more education and informational support is needed for clinicians, including updates about the technology and what it is able to test for given the rapid pace at which it is evolving, and the requirements for OHIP funding. In Ontario, different companies and labs offer different versions of NIPT; many clinicians were unfamiliar with the differences between them.
Clinicians reported learning about NIPT through conferences, journal articles, rounds, information from the genetics department at their respective institutions, information disseminated by labs and companies offering NIPT, word of mouth from other clinicians, and prenatal “champions” within their practice group/area. Many mentioned the need for better education of family physicians, although some also spoke of the lack of information available for specialists. Some felt that “a lot of the education has been reactive.” Accordingly, clinicians identified a need for continuing education on NIPT from a trusted, nonbiased source. Requested topics included guidance around counseling for NIPT, referring to specialists, coordinating NIPT with other screening technologies, and responding to ethical concerns.
A need for more patient-facing educational resources for patients to make sense of “the confusing alphabet soup of prenatal screening tests” was also described. Most providers directed patients toward the company websites or the Society of Obstetricians and Gynecologists of Canada (SOGC) website; for the general population, this information could cause “a lot of confusion potentially, depending on their level of understanding of genetics information.” Some reported that educational resources such as a pamphlet could empower a broader patient population to become informed about the test before they face the time-sensitive decision of whether or not to undergo it themselves. Clinicians mentioned that more patients appear to be informed about the test as time progresses; however, this seems to be dependent on the socioeconomic status of the patient population.
Expansion and Evolution of Test
Since 2014, the functionality of NIPT has increased to include the detection of microdeletions, as well as sex chromosomal aneuploidies, duplications, large copy number variations, and a limited number of single-gene anomalies (Chitty et al., 2018). However, there is a lack of evidence for accuracy regarding these additional conditions, and it is generally accepted that the positive-predictive values are lower than those for trisomies 13, 18, and 21 (Chitty et al., 2018; K. Martin et al., 2018).
Clinicians in this study discussed specific concerns regarding the expanded testing options; these issues were probed primarily in the 2016 interviews but were also discussed in the 2018 interviews. Generally, the clinicians recognized that the accuracy for the detection of these other conditions is lower than that for the trisomies: “I don’t think it’s the best test right now to be able to look at microdeletions . . . it can definitely lower the sensitivity of the test.”
Some worried about the implications of primary care providers ordering these advanced tests: . . . that’s the problem with NIPT and all these extra testing now, is that the tests are available but the ordering physicians aren’t highly educated in that field and don’t have the knowledge-base or the experience to appropriately counsel people on what the test is actually looking for and what the results mean. For example, how sensitive and specific are those tests? . . . All of those things are things that are a little bit beyond my knowledge as a family physician, so I don’t think I would feel comfortable ordering the test where the implications of the results are huge, but I don’t actually know what it means. Moreover, I think the speed at which the tests are being developed is a lot more ahead of the knowledge-base of the physicians ordering the tests, and I think that’s one of the problems.
A few clinicians felt that this type of advanced screening should be “indicated screening, not screening for the sake of screening,” and should not be received based solely on the patient’s ability to pay. Some worried about “how much information is too much information” for the patient. For instance, one genetics counselor explained: “We always worry about those people for whom the test really isn’t clinically relevant, and they might take a low-risk result as meaning more than it does.” Others worried that “it would be tough for [the child] to ever get insurance, having known that sort of information from birth.” In contrast, some did not object to the availability of this information: “if people are looking for that information and we have a way to give it to them, I’m fine with that.” There was a concern that the expanded testing created an unnecessary time burden on providers as a significant amount of time is required to counsel about these extra conditions. Moreover, the clinical utility of these findings is unknown because the same genetic anomalies may manifest with different phenotypes and varying levels of severity. Some clinicians thought that expanded testing should be delivered on the basis of the severity of the condition, and how it affects the individual’s quality of life, chances of survival, and function; others thought that testing would be relevant if the condition could be effectively treated.
In addition, some clinicians raised concerns about “where do we stop?” in terms of making decisions about whether to terminate a pregnancy based on disability. They expressed worry about the negative messages that funding NIPT may send to the public about disablement and how it sets the tone for “designer babies.”
Discussion
In this study, we describe clinicians’ perspectives on the implementation and evolution of NIPT in Ontario, Canada, in terms of both implementation considerations with practical utility, and ethical issues arising as a function of the interplay between society and technology. In general, examining this interplay becomes increasingly important as the moral complexity of a technology increases or its implementation occurs more drastically or disruptively, and NIPT is characterized by both of these factors (Bianchi et al., 2014; Vanstone et al., 2018).
The clinicians in this study described the implementation of NIPT as disruptive in a number of ways. Initially, clinicians found that NIPT had a major impact on their workload, which was exacerbated by systematic inefficiencies, including the onerous process for receiving funding, lack of appropriate education surrounding how to order the test, and provincial policy restricting midwives from ordering the test. Importantly, clinicians felt that this led to a decline in their ability to provide quality patient care, as well as a decrease in their wellness. These consequences associated with the rapid implementation of NIPT should not be underestimated, especially when considered in the context of clinical professions that experience staggeringly high rates of burnout, loss of empathy, and depression (Aiken et al., 2005; IsHak et al., 2013; Rotenstein et al., 2016). In medical ethics, the concept of beneficence captures the idea that clinicians must balance the potential benefit of a treatment, service, test, or technology against the potential harm it could cause to their patients (Beauchamp & Childress, 2001; ter Meulen, 2005). With respect to NIPT, decision makers (e.g., policymakers and regulatory bodies) are attempting to balance the potential benefits of this relatively new technology with the potential for it to cause harm based on how it is implemented. Carefully considering the potential to cause clinician or patient harm before, during, and after implementation, and anticipating and responding accordingly, is a key tenet of responsible health care decision-making (Demers-Payette et al., 2016).
By 2016 and 2018, participants had adopted more effective ways of organizing their practice to accommodate NIPT, and described a lower burden on their practice overall. Various models (e.g., Fixsen et al., 2007; May & Finch, 2009; Metz & Bartley, 2012; Rogers, 2003) describe the introduction of technologies into society as happening in identifiable stages which are iterative and not necessarily linear. The response of clinicians to the introduction of NIPT into the Ontario health care system follows Tuckman’s model of group development (Bonebright, 2010; Tuckman, 1965), originally created as a framework for team progression in the field of human resource development. This model describes several stages of group development: (a) forming, which is characterized by learning the task at hand and creating organizational standards; (b) storming, in which there is disruption and conflict; (c) norming, when relative harmonization and cohesion are achieved; and, finally, (d) performing, which occurs when the group most effectively tackles the task by way of the processes developed throughout the other stages. With respect to NIPT, the clinicians’ descriptions provide evidence of progress through the forming (i.e., figuring out how to offer NIPT) and storming (i.e., dealing with the subsequent effects on their practice) stages of Tuckman’s model. From 2014 to 2018, some norming appears to have occurred, as evidenced by the development of more efficient processes for offering NIPT. Our data do not provide evidence that clinicians have yet reached a performing state, primarily due to unresolved issues, including the inability for midwives to order NIPT, inequity of access to the technology, and inadequate education of clinicians and patients. This could, however, be a result of our specific sample; it is possible that the clinicians captured in our study may be relatively late adopters, and other clinicians who have reached a performing state may be underrepresented. Despite this, our findings do indicate a need for attention and resources to be directed to these unresolved implementation issues to enable all clinicians to reach a performing state. The consequences of not doing so are serious and include violations of key tenets of most health care systems, including the principles of universality, accessibility, and quality (Government of the United Kingdom, 2015; Health Canada, 2018; Ministry of Health Welfare and Sport, 2018).
It is clear from our data that the implementation of NIPT has had a significant impact on the way in which prenatal care providers in Ontario practice. The coordination of NIPT with other available screening technologies appears to be driven by patient characteristics, SOGC guidelines (Audibert et al., 2017), and the ways in which the practitioner’s local colleagues practice. Many of the clinicians described variability in how prenatal testing is coordinated for a particular patient and questioned the role that clinicians play as “gatekeepers” to the test. This gatekeeping role manifested itself through both the clinician’s knowledge of the test and whom they decided to offer it to. Some commented on their decision not to offer it to patients whom they perceived could not afford the test; such a practice could be considered paternalistic and are likely to disadvantage women of perceived low-income status. Moreover, when this is combined with the lack of patient-facing educational resources on NIPT, these individuals may be facing a double-bind in which they are disempowered by both the inaccessibility of information and by prior predictions about their decision to test. Clinicians also admitted that they had little understanding of the differences between different brands of the test and would often direct patients toward the company websites for more information. As these companies are inherently biased because they are providing information as well as selling their own product, one clear implication of this research is that the ethical implementation of NIPT would be well supported through the creation of both provider and patient-facing resources by an organization which is not invested in selling the test (Vanstone et al., 2015b). These resources should outline the risks and benefits of the test, and compare different brands to aid in decision-making about whether NIPT is a desired test. Comprehensive information from a neutral party is essential to facilitating a shared decision between clinician and patient about what type and brand of prenatal test is preferred by the patient (Vanstone et al., 2012, 2018).
Moreover, while some clinicians in our study heralded NIPT as superior to other screening technologies, others questioned the utility of prenatal genetic testing in general. The theory of SST refers to the concept of entrenchment to describe how certain ideas become “entrenched” in a society and fail to be reexamined (Collingridge, 1980; Williams, 1997). In this way, chromosomal anomalies have been defined as a “problem,” rendering NIPT the “solution” (see Bacchi, 2012). As described by one participant, focusing on other identified problems in the prenatal pathway, such as maternal and child nutrition, may reveal other solutions to which health care funds could be directed. Moreover, Lehoux and colleagues (2016) argue that rather than replacing existing technologies, new technologies are often added to the health care system. Silva et al. (2018) further alert us to the “steadily growing number of new technologies being introduced in health systems” (p. 1) and the issues this creates for sustainability and equity. This was reflected in one participant’s considerations of the cost to the health care system associated with ordering numerous screening tests for one individual patient.
Overall, our findings reveal that some of the specific ways in which NIPT is being funded, introduced, and used has important consequences for the health care system. The expansion of testing options has the potential to not only exacerbate existing ethical issues but also to generate new ones. For example, many clinicians expressed worries with respect to the knowledge-base of primary care providers when it came to considering the expanded set of conditions. Some argued that primary care providers are ill-prepared to adequately counsel patients on this set of expanded conditions due to a lack of specialized training in genetics. This was particularly concerning when considering the potential to cause harm through the provision of results without the proper tools to understand their implications. Moreover, providers expressed concerns about offering this testing when the clinical relevance of such findings is unknown. This consideration is not unlike the concerns about direct-to-consumer genetic testing, including the potential to create harm for the participant through the provision of false positive or negative results and their associated implications, with potentially unknown or little benefit (Wasson et al., 2006). Again, this discussion could benefit from the core tenets of medical ethics (Beauchamp & Childress, 2001): patients should be adequately counseled on the potential for additional information to be gained from the test and its possible impact on their mental health and well-being (Dondorp et al., 2015). Failure to do so affords the opportunity for this technology to create more harm than good. Accordingly, some of the clinicians in this study argued that the expanded screening for microdeletions should only be done if indicated, if the severity or clinical significance of the condition could be reasonably ascertained, and/or if the condition was treatable. It is likely that the most appropriate and responsible manner for offering NIPT will need to be renegotiated in the context of the suite of expanded conditions. This has the potential to push clinicians back to Tuckman’s “storming” phase (Bonebright, 2010; Tuckman, 1965), in which the conditions of prenatal practice are once again disrupted and reorganized.
The theory of the SST puts forth that there are both conscious and unconscious decisions, influenced by societal forces, that affect the trajectory of a technology (Williams & Edge, 1996). Derived from SST theory, the concept of responsible innovation in health (Demers-Payette et al., 2016; Silva et al., 2018) emphasizes the need for health care systems to be anticipatory, reflexive, inclusive, and responsive to the effects of medical innovation on the functioning of the health care system. In this article, we are calling for the health care system to be responsive: to critically examine the effects of the implementation of NIPT, and to respond to these effects in a way that best manages patient outcomes, clinician wellness, and equity of access. Overall, the quick implementation of NIPT without system, clinician, or patient supports has resulted in inequitable distribution in the health care system. This could be ameliorated with appropriate education and guidance for both clinicians and patients, as well as funding policies which focus on reducing disparities in access and addressing other effects on clinician wellness and patient care (Haidar et al., 2020; Vanstone et al., 2014, 2018). Accountability to the effects of a medical innovation steers us away from a technological deterministic perspective in which technology is believed to develop autonomously and without impact from societal factors (Williams & Edge, 1996). Alignment with an SST approach affords us an understanding of the role we have in steering technology rather than passively allowing technology to steer us (MacKenzie & Wajcman, 1999). We further implore health care system decision makers to examine the well-marked path and to reflexively consider the implicit assumptions that serve to carve this path.
The limitations of this study include that it was conducted in a particular context (the health care system in Ontario, Canada), which has a unique set of administrative structures, policy history, and funding constraints. As such, it is unknown how generalizable these results are to health care systems in other contexts. As in any qualitative study, the findings may also reflect undetected forms of sampling or social desirability bias. For instance, although clinicians in our study still appear to be figuring out the best ways to offer NIPT to their patients within the milieu of other screening technologies, it is possible that we have not collected data from the small pockets of early adopters (Rogers, 2003) who have reached a state in which NIPT is effectively integrated into their practice. Despite these limitations, this study offers useful insight into the diffusion of NIPT into a health care system and its associated impact on the ways in which clinicians organize and enact their practice.
Conclusion
Throughout the study, we were able to observe the relatively disruptive introduction of a novel technology into the Ontario health care system. The disruption associated with the implementation of NIPT appears to decline over time as clinicians establish more effective ways of accessing the technology, often through collaboration with their practice group and other clinicians. Our data do not demonstrate that we have yet reached a state of stability with respect to NIPT. However, the constant and rapid development of this technology makes it difficult to imagine a state of sustained performing being possible, illustrating the importance of governmental supports on the appropriate use of this technology.
The characteristics of this morally complex technology mean that we must be constantly vigilant in monitoring the effects it has on our health care systems and societies as a whole. To do this entails that we are critical of the assumptions and values that underlie the use of this technology and are accountable to the unintended consequences it produces.
Supplemental Material
sj-pdf-1-qhr-10.1177_1049732320945303 – Supplemental material for Implementing Publicly Funded Noninvasive Prenatal Testing for Fetal Aneuploidy in Ontario, Canada: Clinician Experiences With a Disruptive Technology
Supplemental material, sj-pdf-1-qhr-10.1177_1049732320945303 for Implementing Publicly Funded Noninvasive Prenatal Testing for Fetal Aneuploidy in Ontario, Canada: Clinician Experiences With a Disruptive Technology by Raquel Burgess, Alexandra Cernat, Leichelle Little and Meredith Vanstone in Qualitative Health Research
Footnotes
Acknowledgements
We gratefully acknowledge the contributions of Lisa Schwartz, Mita Giacomini, and Jeff Nisker to the acquisition of funding; Barbra de Vrijer and Sarah McDonald to the recruitment of participants; and Marina Sadik to data analysis.
Author Contributions
R.B. contributed to data curation, investigation, formal analysis, project administration, writing original draft, writing review and editing. A.C contributed to data curation, investigation, formal analysis, writing review and editing. L.L. contributed to data curation, investigation, project administration, writing review and editing. M.V. contributed to funding acquisition, conceptualization, methodology, data curation, investigation, formal analysis, project administration, supervision, writing original draft, writing review and editing.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: this research was funded by the Canadian Institutes of Health Research, FRN 136734 (Principal Investigator: M.V.). A.C. is generously supported by the Hospital for Sick Children Restracomp Master’s Scholarship, as well as by a Canada Graduate Scholarships Master’s Award awarded by the Canadian Institutes of Health Research (CIHR).
Ethical Approval
This study was reviewed and approved by the Hamilton Integrated Research Ethics Board in Hamilton, ON, Canada.
Supplemental Material
Author Biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
