The purpose of population-based carrier screening is to identify asymptomatic carriers of autosomal and X-linked recessive conditions and give prospective parents reproductive options to prevent the birth of an affected child. Screening programs began formally in the 1970s with screening for Tay–Sachs disease carrier status in the Ashkenazi Jewish community (Kaback, Reference Kaback, Nathan, Greenwald and Kaback1997). Subsequent programs have targeted cystic fibrosis (CF) in the United States and parts of Italy (Castellani et al., Reference Castellani, Picci, Tamanini, Girardi, Rizzotti and Assael2009; Hale et al., Reference Hale, Parad and Comeau2008), thalassaemia in Mediterranean at-risk populations (Cao et al., Reference Cao, Pintus, Lecca, Olla, Cossu, Rosatelli and Galanello1984, Reference Cao, Saba, Galanello and Rosatelli1997; Modell & Mouzouras, Reference Modell and Mouzouras1982), fragile X syndrome in Israel (Berkenstadt et al., Reference Berkenstadt, Ries-Levavi, Cuckle, Peleg and Barkai2007), and most recently, spinal muscular atrophy (SMA) in the United States and Taiwan (Su et al., Reference Su, Hung, Lin, Chen, Chern, Tsai and Lee2011; Sugarman et al., Reference Sugarman, Nagan, Zhu, Akmaev, Zhou, Rohlfs and Allitto2012). Several of these programs have reported reductions in the incidence of affected infants born with the conditions tested. Developments in genetic technology mean that it is now possible to simultaneously screen for an individual's carrier status for hundreds of inherited conditions using a single sample (Levenson, Reference Levenson2010), and such panel-based testing is likely to replace testing for individual conditions in the future.
In Australia, healthcare is available through the government-funded public health system, as well as through a user-pays private health system. The availability of carrier screening varies for different conditions and in different settings. Screening for β-thalassaemia carrier status is publically funded and generally triggered by abnormal results on full blood examination (FBE), which is performed as part of routine pregnancy care (Cousens et al., Reference Cousens, Gaff, Metcalfe and Delatycki2010). There are several well-established, non-government funded, community-based programs offering carrier screening to people of Ashkenazi (Eastern European) Jewish ancestry for conditions such as Tay–Sachs disease (Ioannou et al., Reference Ioannou, Massie, Lewis, Petrou, Gason, Metcalfe and Delatycki2010). A fee-for-service CF population carrier screening program has been in existence in the state of Victoria, Australia since 2006 (Massie et al., Reference Massie, Petrou, Forbes, Curnow, Ioannou, Dusart and Delatycki2009). In its first three years of operation, the program screened 3,200 individuals, detecting 106 carriers, and 9 carrier couples. All the couples identified through the program altered their reproductive decisions, to avoid having a child with CF. Screening for other relatively common genetic conditions, such as SMA and fragile X syndrome carrier status, are less frequently offered in Australia (Metcalfe et al., Reference Metcalfe, Jacques, Archibald, Burgess, Collins, Henry and Cohen2008).
The uptake of carrier screening is generally higher when offered in pregnancy than when offered to the non-pregnant population (Harris et al., Reference Harris, Scotcher, Hartley, Wallace, Craufurd and Harris1996; Mennie et al., Reference Mennie, Gilfillan, Compton, Curtis, Liston, Pullen and Brock1992; Wald et al., Reference Wald, George, Wald and Mackenzie1993). The reasons may be that the first contact with a health professional does not occur until the woman is pregnant or that screening does not become a priority until this time. In addition, couples may not believe they need to consider screening until pregnant (Delatycki, Reference Delatycki2008). There is little known about the attitudes of Australian obstetricians toward carrier screening for inherited conditions. Knowledge in relation to these attitudes is of great importance as the frequency with which obstetricians and other pregnancy healthcare providers offer tests to patients is a major determinant of the success of population-based carrier screening programs.
The aim of this study was to gather information about the current practice and attitudes of Australian obstetricians toward carrier screening for genetic conditions as part of routine pregnancy care.
Methods
Participants
Australian Fellows of the Royal Australian and New Zealand College of Obstetricians and Gynaecologists (RANZCOG) were invited by e-mail to complete an anonymous online survey between January 27 and March 3, 2011. Only Fellows actively practicing obstetrics were requested to complete the whole survey. A reminder was sent three weeks after the initial invitation to participate, inviting those who had not completed the survey to do so.
Measurement Tool
The survey was developed by a panel of clinicians and researchers with expertise in carrier screening programs and was informed by previous international studies (Morgan et al., Reference Morgan, Driscoll, Mennuti and Schulkin2004, Reference Morgan, Driscoll, Zinberg, Schulkin and Mennuti2005; Wilkins-Haug et al., Reference Wilkins-Haug, Hill, Schmidt, Holzman and Schulkin1999). The survey was reviewed by the RANZCOG Continuing Professional Development committee, and the content was modified in response to their feedback. The survey was divided into four parts: demographic information, current practice and attitudes toward screening for β-thalassaemia, current practice and attitudes toward screening for CF, and attitudes toward population-based genetic carrier screening in general. Respondents were asked to rate certain aspects of β-thalassaemia and CF screening tests on a 5-point scale, where 1 was very poor, 2 poor, 3 satisfactory, 4 good, and 5 excellent. They were also asked to rate their level of concern regarding general aspects of population-based screening for genetic conditions using a 4-point scale, where 1 was no concern, 2 minor concern, 3 moderate concern, and 4 major concern. Respondents were provided with two, free text boxes at the end of the survey and asked for any additional comments about specific barriers to screening or general comments. Responses to the questionnaire were anonymous.
Data Analysis
LimeSurvey software was used to generate the electronic version of the survey, and to store and analyze the responses. Using content analysis, open-ended responses were categorized independently by Zornitza Stark, Belinda McClaren, and Sylvia Metcalfe based on similarity and differences. Numbers of responses in categories are reported.
Ethics Committee Approval
The study was approved by the Royal Children's Hospital, Victoria, Australia Human Research Ethics Committee (HREC 30068).
Results
Demographic Details
A total of 1,206 e-mails were sent to practicing Australian Fellows of the RANZCOG who had supplied the College with an e-mail address. Eighty-five e-mails were returned as undeliverable, leaving 1,121 potential respondents. From those, 237 responses were received (response rate minimum of 21.1%) with representative proportion of responses received from each state/territory; 55 respondents identified themselves as not practicing obstetrics, and 26 did not complete the survey sufficiently for their responses to be included in the analysis. One hundred and fifty-six eligible Fellows completed the full survey. Demographic information and type of practice of the respondents is shown in Table 1.
Current Practice and Attitudes Toward Carrier Testing for β-Thalassaemia and Cystic Fibrosis
One hundred and fifty-two obstetricians (97%) supported carrier testing for β-thalassaemia in pregnancy, and 130 (83%) supported carrier screening for CF. Self-reported current practice patterns with respect to these two conditions are summarized in Table 2. The opinion of obstetricians regarding certain aspects of β-thalassaemia and CF carrier screening are presented in Table 3.
*All 12 practiced in states where there are established fee-for-service CF carrier testing programs (Victoria and New South Wales).
Current Practice and Attitudes Toward Carrier Testing for Other Inherited Conditions
One hundred and thirty obstetricians (83%) supported population-based carrier screening for at least some inherited conditions. However, only 9 (6%) felt this should take place during pregnancy, with 90 respondents (58%) stating it should ideally take place in adulthood before pregnancy, 33 (21%) at birth and 24 (15%) in high school. A very low number of obstetricians reported routinely offering carrier tests for any other conditions: 7 (5%) for Tay–Sachs disease, 8 (5%) for fragile X syndrome, and 3 (2%) for SMA.
The participants were asked to rate their level of concern regarding various aspects of population-based carrier screening and their responses are presented in Figure 1. Ninety-three (60%) of survey participants stated they would like more training in this area.
Additional Comments
Forty-six participants provided further comments in the open-ended questions. Five of these commented on survey design only and these are not reported. The responses of the remaining 41 participants were categorized based on similarity of content. Some participants’ responses covered more than one topic and their comments were coded into more than one category. Forty-nine comments addressed topics raised in Tables 2 and 3 and Figure 1. There were eight new topics raised in 25 comments that were not already covered in the survey. These were: concern about equity of access and distributive justice, from the perspective of reaching disadvantaged or multicultural populations, or limiting testing to high genetic risk populations, or targeting populations such as preconception/pregnant couples (n = 14), potential for causing harm through creating a perception of eugenics in society (n = 1), potential for stigmatization (n = 1), or raising questions regarding paternity (n = 1), impacting on life insurance (n = 2), the lack of evidence of cost-benefit (n = 3), and the need for screening to be policy driven (n = 1). Two respondents commented on their personal view that screening has eugenic undertones.
Discussion
This is the first study to examine current practice and attitudes of Australian obstetricians toward population-based carrier screening for genetic conditions in routine pregnancy care. The majority of obstetricians expressed support for population-based carrier screening for at least some conditions, with 97% supporting carrier screening for β-thalassaemia, and 83% supporting carrier screening for CF.
The largely positive attitudes toward universal carrier screening among Australian obstetricians are not translated into practice, with only 20% reporting they routinely offer β-thalassaemia screening to all their patients, 8% offering carrier screening for CF, 5% for fragile X syndrome, and 2% for SMA. This contrasts with self-reported practice among obstetricians in the United States, where a similar survey found that 65.8% of respondents offered CF carrier screening to all prenatal patients (Morgan et al., Reference Morgan, Driscoll, Mennuti and Schulkin2004). The low number of Australian obstetricians reporting offering β-thalassaemia screening to all patients, when in practice the majority of pregnant women have a FBE performed, most likely reflects the indirect nature of β-thalassaemia screening, with FBE testing not being perceived by obstetricians as a screening test for β-thalassaemia. This finding is consistent with patients found to be carriers for β-thalassaemia typically reporting that they were unaware that screening had taken place (Locock & Kai, Reference Locock and Kai2008).
Most Australian obstetricians report offering carrier testing in specific circumstances, most commonly in the presence of a personal or family history of a genetic condition. However, the majority of babies with CF are born to families with no family history of CF (McClaren et al., Reference McClaren, Metcalfe, Amor, Aitken and Massie2011), and even when a family history of CF is known, only a small proportion of relatives undertake carrier testing (McClaren et al., Reference McClaren, Metcalfe, Aitken, Massie, Ukoumunne and Amor2010). This is not unique to CF, but applies to all recessively inherited conditions. Therefore, the family history-based approach (so-called ‘cascade testing’) is likely to identify only a small proportion of couples who are at risk of having a child affected by an autosomal recessive condition.
In 2001, the American College of Medical Genetics (ACMG) and the American College of Obstetricians and Gynecologists (ACOG) issued joint guidelines recommending healthcare providers to offer CF carrier screening to all couples planning a pregnancy or seeking prenatal testing (ACOG/ACMG, 2001), and a similar position statement has been issued by the Human Genetics Society of Australasia (HGSA, 2009). The RANZCOG specifically recommends β-thalassaemia screening in routine pregnancy care (RANZCOG, 2009), and with regard to other conditions suggests that ‘counselling should address availability of carrier status screening for genetic conditions of perceived high prevalence or consequence’ (RANZCOG, 2010).
A number of practical issues were raised by the obstetricians in this survey as barriers to offering universal carrier screening. Chief among these was financial cost, both to individual patients and to the health system as a whole. As a guide, one Australian laboratory charges A$220 for CF carrier screening, A$250 for fragile X screening, and A$350 for SMA screening. With the exception of the FBE and hemoglobin electrophoresis that can diagnose carrier status for β-thalassaemia, the cost of population-based genetic carrier screening tests is not currently covered by government funding or private health insurance. By contrast, several health insurance providers in the United States cover the cost of such testing. It should be noted that some of the cost of the first trimester trisomy 21 screen is similarly not covered by government funding or private health insurance in Australia. This is commonly offered as part of routine pregnancy care, and the majority of pregnant women choose to pay between A$200–300 to include maternal serum screening and a nuchal translucency measurement by ultrasound. Carrier testing has the advantage that it only needs to occur once in each individual's lifetime rather than in each pregnancy, provided partners remain unchanged. Nevertheless, cost is an important barrier to universal carrier screening. The current arrangement in Australia creates inequity in healthcare, with only those that can afford it being in a position to take up carrier screening.
Patient education is an integral part of informed consent. Time constraints, language and cultural barriers, uncertainty in interpreting results, and lack of supporting services were all identified as important barriers to offering screening. With the number of available screening tests set to increase, it may be that detailed counseling will need to be reserved for those couples found to be at increased risk of specific conditions. Of note, the majority of obstetricians offering CF carrier screening routinely to all patients practiced in the states where there are existing fee-for-service carrier screening programs. Having a dedicated program facilitates screening through the provision of practitioner education, a clear pathway for testing, and support with interpretation and follow-up of results.
Creating psychological harm was the most significant concern that Australian obstetricians had with regards to offering population-based carrier screening programs. In addition, some survey participants commented that those found to be carriers may be stigmatized and subject to insurance restrictions. Carriers for recessive genetic conditions are generally asymptomatic, and each person is estimated to be a carrier for several recessive conditions. The evaluation of existing carrier screening programs has shown that carriers are often initially anxious about their positive test results (Ioannou et al., Reference Ioannou, Massie, Lewis, Petrou, Gason, Metcalfe and Delatycki2010; Scriver et al., Reference Scriver, Bardanis, Cartier, Clow, Lancaster and Ostrowsky1984). However, this anxiety subsides, and the long-term follow-up of individuals who have taken part in carrier screening programs has shown that the majority have enduring positive feelings about the experience of being screened (Locock & Kai, Reference Locock and Kai2008; Zeesman et al., Reference Zeesman, Clow, Cartier and Scriver1984). Although commonly cited in professional circles and in the mass media, the concern regarding life insurance implications for those found to be carriers is unfounded (Delatycki et al., Reference Delatycki, Allen and Williamson2002).
Australian obstetricians expressed only moderate levels of concern regarding liability arising from not offering carrier screening for genetic conditions in pregnancy. This contrasts with studies of American obstetricians, who cited liability from not offering screening as their most significant concern (Morgan et al., Reference Morgan, Driscoll, Mennuti and Schulkin2004) and there are reports of ‘wrongful birth’ legal action being taken in the United States over failure to provide CF carrier screening (Hausen, Reference Hausen2012). We are not aware of successful legal action being taken for failure to offer such screening in Australia, but if this were to occur, it is likely that the level of concern would increase considerably.
One of the most notable findings of this survey was that only 6% of surveyed obstetricians felt that pregnancy is the ideal time to offer carrier screening, with most favoring preconceptual screening in adulthood. The ethical considerations in choosing a model for universal carrier screening have recently been reviewed, with CF as an example (Modra et al., Reference Modra, Massie and Delatycki2010). It has been argued that preconceptual carrier screening done outside of the medical context (e.g., in schools or workplaces) is ethically superior as it promotes greater autonomy and maximizes the number of reproductive options open to people identified to be carriers. This model works well for conditions that are limited to certain ethic groups, with attendant high degree of community support and education (e.g., Tay–Sachs disease screening programs in Ashkenazi Jews). However, whether it can be translated to the wider community remains to be seen, and in the absence of such programs, offering carrier testing in pregnancy remains important.
This electronic survey elicited responses from only 21% of those successfully e-mailed. It is likely that the distribution list included many Fellows who do not practice obstetrics, and therefore the true response rate of practicing obstetricians is considerably higher. Nevertheless, a relatively low response rate may be indicative of this issue not being perceived as relevant by Australian obstetricians, which in itself would constitute a major barrier to the development of screening programs. Fertility specialists comprise another group of RANZCOG Fellows who are well placed to perform preconceptional carrier screening and ascertaining their views, as well as the views of general practitioners who deliver a substantial part of pregnancy care, will be equally important.
The field of carrier screening for genetic conditions evokes unique ethical, legal, psychosocial, and privacy concerns. Advances in genetic technology mean that the ability to simultaneously screen for an individual's carrier status for hundreds of inherited conditions using a single sample is already a reality (Levenson, Reference Levenson2010) and the cost of such screening will continue to decline. We have identified specific practical barriers and ethical concerns among Australian obstetricians regarding the implementation of population-based carrier screening programs. Addressing some of these concerns may increase support for screening and the findings of this survey have important implications for the future planning of screening programs and genetic services in Australia and internationally.
Acknowledgments
This project was supported by the Victorian Government's Operational Infrastructure Support Program. The authors would like to thank Professor Michael Permazel for his input into the survey and Mr Lucien Stals from the Education Resource Centre at RCH for his assistance with the development of the electronic version of the survey. Martin B. Delatycki is a National Health and Medical Research Council Practitioner Fellow.