Evaluation of pre-test counselling offered for non-invasive prenatal testing (NIPT) as a primary screening tool

Abstract

The increasing popularity and expansion of non-invasive prenatal testing (NIPT) to screen for rare conditions beyond common trisomies prompts evaluation of pre-test counselling currently offered. We conducted a prospective survey to assess women’s knowledge of NIPT in those who had undergone NIPT (study group) and those who were planning to have NIPT (control group). Out of the 189 questionnaires analysed, the study group did not show a higher knowledge score compared to the control group (P = 0.097). 44% misunderstood that NIPT can identify more conditions than invasive testing, 69.8% were unaware of the recommended need for nuchal translucency measurement and 52.6% were unaware of the possibility of incidental findings. 31% even considered discussing termination of pregnancy as one of the next steps if NIPT shows high risk for Down syndrome. This study shows that current pre-test counselling is inadequate. Service providers should address these knowledge gaps and assist women to make informed choices.

Impact Statement

• What is already known on this subject? Pre-test counselling for non-invasive prenatal testing (NIPT) should be conducted to assist women in making an informed consent.

• What do the results of this study add? Our results show that a significant proportion of women are unaware of the limitations of NIPT.

• What are the implications of these findings for clinical practice and/or further research? Service providers should improve their pre-test counselling focusing on areas of knowledge deficiencies and misunderstanding on NIPT identified in this study.

Keywords:

• Knowledge
• non-invasive prenatal testing
• pre-post tests
• prenatal genetic counselling
• surveys and questionnaire

Introduction

The breakthrough discovery of cell-free foetal DNA in maternal plasma in 1997 paved the way for integration of non-invasive prenatal testing (NIPT) into clinical practice for detection of common trisomies in 2011 (Lau et al. 2012)^. Since then, the use of NIPT is gaining popularity worldwide (Minear et al. 2015, Gadsboll et al. 2020) owing to its higher sensitivity and lower false positive rate over combined first trimester screening test (American College of Obstetricians and Gynecologists’ Committee on Practice Bulletins—Obstetrics; Committee on Genetics; Society for Maternal-Fetal Medicine 2020) and is increasingly provided as a primary screening test (PW Soothill 2014, Kostenko et al. 2019).

In recent years, coverage of NIPT has rapidly expanded stepwise to include sex chromosome aneuploidies (SCA), panels of sub-microscopic copy number variants (CNV) as well as autosomal trisomies (Shaw et al. 2020). Within a decade, NIPT has ventured beyond basic trisomy screening to include rare genetic conditions with much lower prevalence and therefore lower positive predictive value (PPV) and higher false-positive rates (Petersen et al. 2017). As an example, PPV for Turner syndrome ranges from 9% to 40% (Bedei et al. 2021) and PPV for CNV ranges from 14% to 50% depending on the size of CNV (Chen et al. 2019, Pei et al. 2020, Pang et al. 2021). Therefore, international guidelines do not recommend NIPT panels to be offered beyond basic common trisomies until further supportive clinical validity and utility research (Gregg et al. 2016, Kozlowski et al. 2019). However, commercially driven expanded NIPT panels have already been performed in various countries and often in the general population (American College of Obstetricians and Gynecologists’ Committee on Practice Bulletins—Obstetrics; Committee on Genetics; Society for Maternal-Fetal Medicine 2020). Hong Kong was the first place worldwide to introduce NIPT for clinical use in August 2011 (Allyse et al. 2015). There are now multiple private laboratories offering various NIPT panels including basic trisomy screening, extended panel for screening of SCA and five to seven relatively more common microdeletion syndromes, and advance panel to screen for over 80 rare CNVs.

It is widely recognised that for any genetic test, be it screening or diagnostic, optimal pre-test and post-test counselling should be offered (Fonda Allen et al. 2016). This is particularly true for pregnant women as suboptimal pre-test counselling may lead to unnecessary screening tests, anxiety and in extreme situations, ill-informed decisions for termination of pregnancy (TOP). The procedural ease of NIPT by mere blood taking may hinder women from paying much attention to the possible ethical and medical issues involved in these expanded screening (Kater-Kuipers et al. 2020). Given the rapid evolution of NIPT, it is necessary to investigate if pre-test counselling currently offered has kept up the pace of these changes.

The objective of our study is to evaluate the adequacy of pre-test counselling by assessing women’s knowledge on NIPT, and to identify areas of knowledge deficit and common misunderstandings to help counsellors guide patients in making informed decisions.

Methods

This study was conducted at a public hospital in Hong Kong, from March to June 2021. All pregnant women were recommended to attend our universal Down syndrome screening clinic between 11 and 13 weeks gestation for measurement of nuchal translucency, PAPP-A and fbhCG. Upon attendance of this clinic, those who had undergone NIPT were recruited to the study group and those who planned to have the test were recruited to the control group. Inclusion criteria included those aged 18 or above with singleton pregnancies. Exclusion criteria included those who could not communicate in Chinese or English. After recruitment, research participants would complete a self-administered questionnaire before their consultation.

The questionnaire, consisting of two parts, was designed by the study team with reference to pre-test counselling points for NIPT as recommended by Sachs (Sachs et al. 2015). The first part was to assess the participants’ knowledge of NIPT such as its screening nature, limitations, understanding on its sensitivity and positive and negative predictive values. One score would be given for each correct answer among the 20 questions, and none for a wrong or ‘don’t-know’ answer. The second part explored their views and experiences with NIPT such as factors affecting their choice of NIPT panel. At the end of the questionnaire, they were invited to leave any further comments. A pilot study was performed on 20 participants and the primary outcome on knowledge scores were analysed. Using software G*Power 3.1.9.6, an estimated sample size of 70 in each group would give a power of 80% with a significance p-value of <0.05. Allowing a 30% non-response rate, we planned to recruit 200 research participants.

Background demographics data such as maternal age and education level were collected. Clinical data such as parity and history of abnormal pregnancy were retrieved from electronic patient records. The participants’ final NIPT reports from both the control and study groups were traced from their medical records for verifying the NIPT panels used, as well as the gestation at which the tests were performed.

The total score on knowledge was compared between the two groups by Mann-Whitney U test. Continuous variables in subgroup analysis were performed by Students’ independent t-test or Mann-Whitney U test, while categorical variables were analysed by Chi-square test or Fisher’s exact test. A p-value of <0.05 was considered to be statically significant. All data were analysed using IBM SPSS Statistics for Windows, version 22.0 (IBM Corp., Armonk, NY, USA).

This study was approved by the Hong Kong East Cluster Research Ethics Committee (HKECREC-2021-004).

Results

Over the period of three months, among the eligible 217 women, 198 questionnaires were collected, giving a response rate of 91.2%. 5 from the control group and 4 from the study group were excluded because the questionnaires distributed were mixed, leaving a total of 189 for the final analysis, 116 from the study group and 73 from the control (Figure 1). The mean gestation at which NIPT was performed was 11 weeks and 5 days and the median time interval between NIPT and filling in the questionnaire was 6 days (IQR 4-10.75) for the study group and 3.5 days (IQR 1.25–7.00) for the control group. Background demographics including age, gestation, parity, marital status, conception method, ethnicity, education level, smoking and drinking status did not differ significantly between the two groups as shown in Table 1.

Figure 1. Recruitment of control and study group.

Table 1. Basic demographics.

	Control group  Pre NIPT N = 73 (%)	Study group  Post NIPT N = 116 (%)	p Value
Age  (years)	Mean 33.0	Mean 33.8	0.139
Range	24–42	19–41
<35	49 (67.1%)	66 (56.9%)
≥35	24 (32.9%)	50 (43.1%)
Parity			0.155
0	53 (72.6%)	74 (63.8%) 40 (34.5%)
1	17 (23.3%)	1 (0.9%)
2	3 (4.1%)	1 (0.9%)
3	0 (0%)
Gestation			0.151
11	11 (15.1%)	13 (11.2%)
12	59 (80.8%)	89 (76.7%)
13	3 (4.1%)	14 (12.1%)
Marital status			–
Married	73 (100%)	116 (100%)
Single	0 (0%)	0 (0%)
Divorced	0 (0%)	0 (0%)
Conception method			0.769
Natural	69 (94.5%)	107 (92.2%)
Assisted reproductive technology	4 (5.5%)	9 (7.8%)
Ethnicity			–
Chinese	73 (100%)	116 (100%)
Education level			0.673
Primary	0 (0%)	0 (0%)
Secondary	17 (23.3%)	24 (20.7%)
Tertiary	56 (76.7%)	92 (79.3%)
Smoking			0.711
Non smoker	69 (94.5%)	106 (91.4%)
Smoker	3 (4.1%)	8 (6.9%)
Ex-smoker	1 (1.4%)	2 (1.7%)
Drug abuser	0 (0%)	0 (0%)	–

Table 2 shows the 20 knowledge related questions on NIPT. There was no statistically significant difference on the median knowledge score between the two groups (13 vs 11, p = 0.097). The number of correct answers for each individual question also did not differ between the two groups except for one question indicating more participants who had received NIPT were aware of the possibility of reporting no results (p = 0.023).

Table 2. Knowledge questions related to NIPT.

Knowledge on NIPT (1 score for each correct answer)		Planning for NIPT Control group (%), N = 73			Undergone NIPT Study group (%), N = 116
Do you think:		Yes	No	Don’t know	Yes	No	Don’t know	p Value
1.	NIPT can be performed from 8 weeks	15.1	72.6^a	12.3	11.2	74.1^a	14.7	0.816
2.	NIPT looks at cell-free foetal DNA in maternal circulation	74.0^a	6.8	19.2	76.7^a	6.9	16.4	0.668
3.	NIPT can detect more chromosomal abnormalities than chorionic villus sampling or amniocentesis	57.5	23.3^a	19.2	44.0	26.7^a	29.3	0.597
4.	Invasive test is required if NIPT shows abnormal result	63.0^a	8.2	28.8	71.6^a	7.8	20.7	0.220
5.	NIPT report can reveal conditions other than those listed on your panel	46.6^a	12.3	41.1	47.4^a	11.2	41.4	0.910
6.	NIPT can yield no result	37.0^a	24.7	38.4	53.4^a	19.0	26.7	0.023*
7.	Ultrasound to measure nuchal translucency should be performed before NIPT	43.8^a	37.0	19.2	30.2^a	51.7	18.1	0.056
Which of these conditions do you think NIPT can screen for?		Can	Can’t	Don’t know	Can	Can’t	Don’t know
8.	T21 Down syndrome	94.5^a	1.4	4.1	97^a	0	2.6	0.305
9.	T18 Edwards syndrome	80.8^a	4.1	15.1	88.8^a	0.9	10.3	0.127
10.	Structural abnormalities	43.8	43.8^a	12.3	26.7	53.4^a	19.8	0.198
11.	Abnormal function e.g. eyesight, hearing	13.7	63.0^a	23.3	13.8	59.5^a	26.7	0.628
12.	Low intelligence quotient (IQ)	35.6	43.8^a	20.5	22.4	57.8^a	19.8	0.062
13.	Thalassaemia	52.1	24.7^a	23.3	31.0	37.9^a	30.2	0.053
14.	Autism	12.3	64.4^a	21.9	8.6	61.2^a	30.2	0.575
		Correct	Wrong	Don’t know	Correct	Wrong	Don’t know
15.	Imagine 100 women with a low-risk NIPT result for Downs syndrome, around how many babies do you think will still have Downs syndrome?	47.9^a	32.9	19.2	57.8^a	19.8	22.4	0.188
16.	Imagine 100 women with a high-risk NIPT result for Downs syndrome, around how many babies do you think will have Downs syndrome?	4.1^a	63.0	32.9	12.1^a	44.8	43.1	0.063
When NIPT shows high risk for Down syndrome, what should be done next?		Yes	No		Yes	No
17.	Repeat NIPT 1–2 weeks later	23.3	76.7^a		21.6	78.4^a		0.780
18.	Perform combined Downs syndrome screening (Blood for biochemical markers and ultrasound for nuchal translucency)	26.0	74.0^a		29.3	70.7^a		0.625
19.	Perform chorionic villus sampling or amniocentesis	86.3^a	13.7		87.9^a	12.1		0.743
20.	Discuss termination of pregnancy with doctor	37.0	63.0^a		31.0	69.0^a		0.398
	Total score (20):
	Median	11			13			0.097
	Interquartile range	9 (25^th), 14 (75^th)			10 (25^th), 15 (75^th)
	Score range	3–18			2–19

^aCorrect answer.

Within the study group, the question with the highest correct answer came from their understanding of the main aim of NIPT to screen for common trisomies such as T21 (113/116, 97%) and T18 (103/116, 88.8%). However, there were areas of knowledge deficits identified. About 44% (51/116) of them misunderstood that NIPT can identify more conditions than invasive prenatal testing and 69.8% (81/116) were unaware of the recommended need for nuchal translucency measurement prior to NIPT. Nearly half of the participants were unaware of the possibility of incidental findings (52.6%, 61/116) and reporting no results (45.7%, 53/116) with NIPT.

Furthermore, only about 50–60% of the participants answered correctly that NIPT cannot screen for structural abnormalities (62/116, 53.4%), low intelligent quotient (67/116, 57.8%), functional abnormalities (69/116, 59.5%) and autism (71/116, 61.2%). Noticeably, this percentage was lowest for thalassaemia (37.9%, 44/116). When asked what should be done next if NIPT turned out to be positive for Down syndrome, most participants (102/116, 87.9%) answered that invasive testing is indicated over a repeat NIPT (25/116, 21.6%) or a combined first trimester screening (34/116, 29.3%). However, 31% (36/116) reported that they would directly discuss TOP with their doctor. In addition, 43.1% (50/116) of participants did not know the PPV of NIPT for Down syndrome screening.

Among the 153 NIPT reports traced (81.0%), SafeT21 Express was the most frequently used platform (117/153, 76.4%), followed by NIFTYpro (33/153, 21.6%). Regarding the extent of NIPT, extended panel was performed in 51.6% (79/153) of participants while 47.7% (73/153) received the advance panel. About half (56.4%, 62/116) of the participants were only offered one NIPT panel option. Comparing their perceived NIPT panel performed and the actual one, only 46.4% (45/97) were concordant.

Figure 2 shows that 81% (94/116) of participants in the study group received pre-test counselling, most commonly by means of face-to-face counselling (67%, 63/94) conducted primarily by doctors (93.7%, 59/63). Some also received information from pre-printed pamphlets (61.7%, 58/94) and digital media (35.1%, 33/94). Reported time on pre-test counselling was within 5 minutes in 50.9% (27/53) and 5 to 10 minutes in 43.4% (23/53). Overall, 70% (81/116) of participants considered that information received was sufficient.

Figure 2. Pre-test counselling in study group.

The contents covered during pre-test counselling includes: the voluntary nature of the test (91.4%, 106/116), sensitivity (77.6%, 90/116), reporting time (68.1%, 79/116), need for invasive test for confirmation (59.5%, 69/116) and possible false positive results (54.3%, 63/116). However, information on it being a screening test (44%, 51/116), background technology (29.3%, 34/116), possibility of incidental findings (21.6%, 25/116) and clinical features of screened conditions (14.7%, 17/116) were not often discussed.

Most, but not all, participants (88.7%,157/177) preferred being counselled on the various NIPT options. Overall, 84.9% (95/112) of participants were satisfied with their NIPT experience and almost all of them would recommend it to their friends and perform the test again in their future pregnancies.

Discussion

It was hypothesised that those who have undergone NIPT would have received pre-test counselling and scored higher marks than those who haven’t. Interestingly, this study showed that the two groups did not differ in their knowledge scores. Although this might be related to the control group already actively searching for related information before their scheduled NIPT, knowledge deficit in certain core principles of NIPT in the study group illustrated deficiencies in pre-test counselling.

About two-third of our participants were not aware of the recommendation to have NT measurement prior to NIPT. Increased NT is known to be associated with chromosomal abnormalities and structural malformations. About one-third of foetal congenital abnormalities would have been missed if only NIPT was performed as screening test (Bardi et al. 2020). There is a consensus that NIPT should be offered in combination with ultrasound to detect markedly increased NT and structural foetal anomalies to avoid missed diagnosis or delaying diagnosis to the second trimester (Kozlowski et al. 2019, Bedei et al. 2021). Although markedly increased NT and foetal abnormalities are considered exclusion criteria for NIPT, some women might have NIPT at the earliest opportunity of 10 weeks, when nuchal translucency could only be performed after 11 weeks. This is the reason why some would suggest postponing NIPT until 13–14 weeks when an early anomaly scan can be performed (Zalel 2015).

Notably, a third of participants reported that they would directly discuss TOP as the next step of management if NIPT showed high risk for Down syndrome. This could be related to the misinterpretation of ‘high accuracy’ of NIPT often promoted by commercial companies. They highlight the high sensitivity and low false-positive rate of NIPT but the PPV is often not mentioned. Even when mentioned, the figure quoted is usually generated from a high-risk rather than a general population (Zhang et al. 2015). Although PPV might be more difficult to comprehend, it is an important parameter in pre-test counselling to avoid undue anxiety and unfounded decisions for direct TOP (Oepkes et al. 2014).

In the initial stage when NIPT was only performed for screening of common trisomies, conducting pre-test counselling based on well-established data was straightforward. Now with the cost for basic or expanded panels more or less the same and most laboratories only providing expanded panels (Yang and Tan 2020), the trend to also screen for SCA and CNV is climbing. This poses more challenges on health care providers to deliver an adequate pre-test counselling aiming to promote reproductive autonomy and guiding women and their partners in making their informed choices.

Concerning the use of NIPT to screen for sex chromosome aneuploidies, there are relatively more clinical data on sensitivity (89% for monosomy X, 82–90% for XYY, XXY, XXX), false-positive rates (∼1%) and PPV (26% for monosomy X, 86% for XXY, 40% for XXX) (Petersen et al. 2017, Suciu et al. 2019) to allow for meaningful counselling. Prenatal diagnosis facilitates paediatric monitoring, timely intervention and allows parents to prepare for the birth of a child with additional needs; these are all reasons supporting the incorporation of SCA screening into NIPT. However, the majority of parents had no prior knowledge of SCA and were not even aware that they were screened for until they received the results. These unexpected false- or true-positive findings will inevitably cause anxiety, depressed feelings and negativity for prospective parents (Riggan et al. 2021). Therefore SCA screening should only be offered after appropriate pre-test counselling (Bedei et al. 2021) and information to support informed decision-making about testing for SCA should be included (Johnston et al. 2023). An additional concern for SCA screening is the ethical dilemma on sex selection, which is one of the reasons why some countries such as Netherlands and India omit this from their publicly-funded NIPT panels (van der Meij et al. 2019, Bowman-Smart et al. 2020).

On the other hand, counselling on screening for CNV is more complex as its prevalence is extremely low, resulting in a lack of data on its clinical utility. With limited time available for pre-test counselling, it is not feasible to go through each and every CNV and provide discussion on their manifestations and prognosis, which is often unpredictable and can vary substantially due to variable penetrance and expressivity. When NIPT reports presence of CNV, it is not uncommon for the pregnant women to decide for invasive tests, which carries risk of miscarriage. On the other hand, even with genome-wide array, CNVs <3Mb will still be missed, leading to false reassurance (Chitty et al. 2018). Therefore, international guidelines do not recommend NIPT to screen for genome-wide CNVs (Gregg et al. 2016, Kozlowski et al. 2019). A reasonable compromise is to offer NIPT for selected CNVs such as DiGeorge syndrome, where the prevalence is relatively higher (1 in 1000) (Kruszka et al. 2017) with more clinical data, well-established disease profile and when antenatal diagnosis can facilitate subsequent management.

In response to the increasing volume of technical information which can be overwhelming to women, a new approach to pre-test counselling has been suggested. Apart from providing the essential information on NIPT, women’s values on whether they wish to know about genetic disorders of their unborn child should be explored. Encouraging them to be aware of what prenatal screening means to them might be more useful than going through medical jargon on various rare conditions (Kater-Kuipers et al. 2020).

Allowing more time for pre-test counselling should improve informed choice. A survey conducted in Japan showed that women’s satisfaction on NIPT counselling by a certified genetic counsellor was proportionate to the amount of time spent counselling, and remained more or less constant beyond 20 minutes (Yotsumoto et al. 2016). In this study, total time spent on pre-test counselling was less than five minutes in more than half of the cases even when most women received extended or advance NIPT panels. Some women felt that their counselling session was too short for the volume of information presented to them (Cernat et al. 2019). Furthermore, private obstetricians lack specific training in knowledge and skills in genetics counselling (Liehr 2021). A study in the United Kingdom demonstrated that following a 40-min training session for health care workers, there was a significant increase in their knowledge about NIPT and confidence in discussing NIPT with patients (Oxenford et al. 2017). Construction of non-biased training resources in a variety of formats by professional societies and independent laboratories rather than commercial companies would be ideal in prespsonse to the growing demand for NIPT.

A limitation of this study was that prior exposure of information on NIPT in the control group could not be assessed. An alternative is to perform the questionnaire before and after pre-test counselling on the same women but this would introduce bias. Most of our questions focussed on the general understanding of NIPT, which might limit the capacity to extend our findings of inadequate pre-test counselling on screening for SCAs and CNVs. Nevertheless, the lack of understanding on Downs syndrome, being the primary disease in consideration when performing NIPT, signifies that considerable areas of improvement in pre-test counselling in expanded NIPT. Higher knowledge scores might be obtained in those who had NIPT in previous pregnancies, but as the parity did not differ between the two groups, this effect should likely be comparable. As NIPT were performed in different clinics, this study could not evaluate performance of individual counsellors and the results could only allow a general representation of quality of pre-test counselling. Lastly, the experience on pre-test counselling was evaluated through receivers’ perspective and not validated by the service providers. To obtain a more comprehensive picture, both parties can be investigated in a matched manner to further identify loopholes in current practice and to improve the quality of pre-test counselling.

Conclusion

This study showed that knowledge of women who had undergone NIPT was not superior to those that had not yet had the test. Areas of knowledge deficiencies and misunderstanding were identified in the current era of increasing complexity of NIPT. This can hopefully aid NIPT providers in improving their pre-test counselling skills, with the goal of allowing parents to make more informed decisions about their pregnancy.

Acknowledgements

The authors acknowledge colleagues in the prenatal diagnostic clinic, Pamela Youde Nethersole Eastern Hospital for assistance in preparing and conducting this research.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.