- Email: [email protected]
Presenting and discussing nuchal translucency screening for fetal abnormality in the UK
ARTICLE IN PRESS Midwifery (2004) 20, 82–93
www.elsevier.com/locate/midw
Presenting and discussing nuchal translucency screening for fetal abnormality in the UK Alison M. Pilnick, BPharm, MRPharmS, MA, PhD (Senior Lecturer in Sociology)a,*, Diane M. Fraser, BEd, PhD, MTD, RM (Professor of Midwifery/Head)b, David K. James, MA, MD, FRCOG, DCH (Professor of Fetomaternal Medicine)c a
School of Sociology and Social Policy, University of Nottingham, University Park, Nottingham NG7 2RD, UK Academic Division of Midwifery, School of Human Development, University of Nottingham, UK c Division of Obstetrics and Gynaecology, School of Human Development, University of Nottingham, UK b
Received 19 February 2003; received in revised form 17 June 2003; accepted 4 July 2003
Summary Objective: to investigate the relationship between information giving by midwives and decision-making by women offered nuchal translucency (NT) screening. To establish how risk figures are discussed in practice, with the intention of relating this to the existing, and often critical, literature on women’s accounts of antenatal screening. Design: a qualitative study following women through the process of being offered and deciding to undergo NT screening. Tape recording of consultations, analysed in their entirety, was combined with post-screening interviews. Setting: a large teaching hospital in the UK. Participants: fourteen pregnant women eligible for NT screening at the time of recruitment. Data collection: (i) tape recordings of consultations between community midwives and pregnant women where nuchal translucency screening was offered; (ii) tape recordings of consultations between hospital midwives and pregnant women immediately post-screening; (iii) individual face-to-face interviews with pregnant women between two and six weeks after the screening, carried out by the first author. Findings: NT screening was in general well received, particularly by those women who had undergone serum screening with previous pregnancies. However, communicating the nature of a risk figure is an interactionally complex process. A large amount of interactional work is required by midwives both before and after screening to ensure that women comprehend this information. Despite the emphasis placed in these consultations on understanding the purpose of NT screening and the status of the results, women often framed their decision to undergo NT screening in terms of it being a formality, or of presuming that all was well. This sometimes created practical and personal difficulties in terms of decision-making.
*Corresponding author. E-mail address: [email protected] (A.M. Pilnick). 0266-6138/$ - see front matter & 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0266-6138(03)00056-1
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
83
Key conclusions and implications for practice: previous sociological and psychological research has tended to be critical of midwives in terms of ensuring informed choice in screening, but this research is often based on post hoc accounts. Examining actual consultations with these accounts helps to illustrate the other factors that affect women’s perceptions of testing, and the way in which risk, choice and decision-making are introduced and discussed in practice. Encouraging women to consider what action they might take on the basis of a personally unfavourable NT result in advance of undergoing the scan may help them to decide whether the information gained will be useful to them. Recognising the complex interactional work required in making sure that women understand the nature of the results that will be obtained is an important issue for the education and training of midwives. & 2003 Elsevier Ltd. All rights reserved.
Introduction At present, there are no specific national guidelines for antenatal screening procedures in the UK, beyond the report of a Royal College of Obstetricians and Gynaecologists (RCOG) (1997) working party on fetal abnormalities, and the RCOG recommendations that all antenatal units should have a written statement of their policy (Lane et al., 2001). National Screening Committee (NSC) recommendations are that there should be a policy of second trimester screening for Down’s syndrome using double analyses as a basic provision, and the National Institute for Clinical Excellence (NICE) is currently formulating evidence-based guidelines with the objective being a national co-ordinated approach to antenatal screening services. What this means in practice is that in the UK current policies vary in specific timing and in tests that are routinely offered on a local and regional basis. Individual hospitals also vary in their policies when women specifically ask for particular tests that are not routinely offered. In most hospitals in the UK, however, serum screening for one or more markers of fetal abnormality is a routinely offered procedure. Serum screening is carried out through a blood test at around 15–18 weeks (ideally at around week 16). Since it is not a diagnostic test, it will not in itself confirm whether or not a fetus is affected, and can give both false-positive and false-negative results. Some centres in the UK now offer an alternative form of screening, nuchal translucency (NT) screening, which can be carried out during an extended ultrasound scan at between 10 and 14 weeks. The nuchal fold is the thickness of skin overlying the cervical spine in the fetus. Nuchal translucency is a measurement of the fluid between the neck skin and the cervical spine. A number of research studies have indicated that a fetal nuchal translucency of X 3 mm is a useful first trimester marker for fetal chromosomal abnormalities, and particularly Trisomy 21 (Down’s syndrome) (e.g. Nicolaides et al., 1992; Pandya et al., 1995), although some
researchers have suggested its feasibility and promise needs to be backed up by further data obtained from low-risk populations (Bewley et al., 1995). The nuchal fold measurement, then, combined with maternal age (which is also a predictor of likelihood of Down’s) is used to produce a risk figure for fetal abnormality, expressed as a probability.
Studies of the implementation of serum screening There is a growing sociological and psychological literature on the practical implementation of serum screening programmes, which is largely critical of the way in which this service is delivered (e.g. Marteau et al., 1992, 1995; Searle, 1997). In particular, the extent to which women recognise that there is a real choice to be made here has been questioned, with research in both UK and US contexts tending to suggest that serum screening is seen by pregnant women as just another part of routine antenatal care (e.g. Rothman, 1988; Green and Statham, 1996; Press and Browner, 1997). Researchers have also explored the information that practitioners give to pregnant women about the conditions that are screened for, e.g. Williams et al.’s (2002) study of practitioners’ portrayals of Down’s syndrome, and have concluded that this can often be focused primarily on potential problems. However, since NT screening is a more recent phenomenon, there is a lack of research looking at its actual implementation in the UK.
Talking about risk The fact that antenatal screening procedures do not produce definitive results, but instead give probabilities, has been another ground on which they have been criticised. Some critical commentators have gone so far as to argue that women have no means of making sense of the information that they have been given, and that they may not
ARTICLE IN PRESS 84
understand that a low-risk result will not guarantee the birth of a ‘normal’ baby (e.g. Ettore, 2000). In a UK setting, the high level of uptake of antenatal screening for Down’s has led some researchers to question the adequacy of the information given to pregnant women and the process of informed consent (Howe et al., 2000). Smith et al. (1994) found that only 36% of their sample of 353 women who had been offered serum screening understood that a negative result would not guarantee that everything was all right with the pregnancy. In addition, only 32% understood that most women with positive results would go on to have a normal baby. Whilst the findings from these studies are compelling, it is important to recognise that for the most part they are based exclusively on interviews or questionnaires with participants after screening has taken place. Those studies that have used recorded data from consultations have tended to employ coding systems designed to identify particular components of the interactions. In general, then, there are few studies examining the ways in which issues are negotiated and the situations described above are achieved in practice. How is it that screening comes to be perceived as routine, that choice is not made explicit, or that women do not understand the nature of a risk figure?
Aims of the study The overall objective of this study was to investigate the relationship between information giving by midwives and decision-making by women offered NT screening. Within this framework, the aim was to examine in more detail the interactional presentation of antenatal screening procedures, and the ways in which they are introduced, discussed and contested by those involved. The study was developed through a recognition of the need to examine NT screening in its own right, to examine whether and to what extent the criticisms of the delivery of serum screening programmes are valid in this scenario. It also aimed to establish to what extent women’s post hoc interview accounts of screening procedures relate to their actual interactions with health-care professionals.
Methods The data presented here are taken from consultations forming part of a newly introduced antenatal
A.M. Pilnick et al.
screening programme. Criticisms of serum screening programmes have not been limited to the psychological and sociological literature, and in the medical field there has been growing concern over the level of accuracy of the results, as well as the short time frame in which women must make decisions about further testing or termination of pregnancy. Influenced by these limitations, NT screening has been introduced in the study hospital. At the time of the study, it was offered to all pregnant women who ‘booked’ to deliver their babies at the hospital and who had their first appointment at the antenatal clinic within the window of time when the screening could be carried out. For the purposes of the programme reported here, a ‘high-risk’ figure was considered to be anything greater than 1 in 200, and those women whose results were in this category would be offered further, diagnostic testing. This figure was chosen in this setting because it equates to the risk of miscarriage from an amniocentesis at 16 weeks, but other centres in the UK use figures ranging from 1 in 100 to 1 in 250.
Sample and recruitment In order to examine as much of the process of the presentation and discussion of NT screening as possible, the data for this project came from three sources: pre-screening discussions of NT screening between community midwives and pregnant women; post-screening consultations between hospital midwives and pregnant women (where the results of the NT scan were given); and face-toface interviews with a subsample of the women who had had their consultations recorded, between two and six weeks after their scan. After obtaining Local Research Ethics Committee approval, and with the co-operation of community and hospital midwives, the aim was to recruit eight pregnant women from each of two different geographical areas in the catchment of the hospital. The first group were recruited by local midwives from an inner-city area where the average age of first pregnancy is below the national average in the UK, and the second by local midwives from an affluent suburban area where the average age of first pregnancy is higher than the national average. These women were asked to take part in a study examining the presentation and discussion of NT screening. A total of 14 women, six from the first group and eight from the second, were eventually recruited, and gave their written consent for their consultations to be tape-recorded. The partners of two women
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
were also present at the hospital consultations, and they also consented verbally to participation and recording. Recruitment of women from the first group was more difficult, as a higher proportion declined to take part. As they were not asked to give a reason for refusal, it is not possible to interpret this finding fully. Whilst recruitment to this group was still ongoing, the NT programme was suspended, making further recruitment impossible. This suspension was due to staff shortages, and the acknowledgement that serum screening at 16 weeks was also available for the women involved. It was initially intended to recruit a third group of women, from a rural area also served by the study hospital, but differences in the way that antenatal care was organised for these women made this problematic, since pregnant women in this area had no contact with a community midwife until after their first hospital antenatal visit. In the two study areas, after a meeting with the community midwife, the woman was subsequently referred to the hospital antenatal clinic. At this first visit to the clinic, NT screening, if it was accepted, took place. If NT screening was declined, women could still undergo a straightforward ultrasound scan to confirm the expected delivery date of the pregnancy, and to exclude the possibility of multiple pregnancy. In order to examine the actual process of discussion about NT screening, it was necessary to tape women’s actual consultations with midwives, rather than relying solely on the post hoc accounts provided by interviews. For the 14 women initially recruited to the study, their first consultations about screening with their community midwife, and, where possible, their subsequent consultations on their first visit to the antenatal clinic, were audiotaped. No researcher was present at these consultations, which all took place in private settings, although in some cases student midwives, partners or children were present. The midwives themselves operated the tape recorders, on the understanding that they could be turned off at any time if any participant so wished. Although women were offered the opportunity of a consultation with a hospital midwife immediately prior to their NT scan if there were any questions they wished to raise at that stage, none of the women recruited to the study took up this offer. All consultations recorded at the hospital clinic during this study took place after NT screening had taken place, and where the results were delivered and discussed by hospital midwives. For the third, interview-based, phase of the study, it was intended to recruit a subsample of
85
eight women from those who had been previously recorded, i.e. four from each area. However, it was only possible to recruit six women (two from area 1 and four from area 2). Two further women from area 1 declined to be interviewed and two did not respond to either a first or follow-up letter requesting an interview. The participants were interviewed by the first author (using semi-structured interview techniques based around broad themes for discussion) between two and six weeks after their visit to the antenatal clinic. The purpose of the interviews was to explore women’s perceptions of NT screening, in particular around whether they had felt their informational needs were met both pre- and post-NT scan.
Analysis The tape-recorded consultations were transcribed and analysed according to conversation analytic (CA) conventions (Jefferson, as published in Atkinson and Heritage, 1984). A simplified key to this transcription system developed by Jefferson, as used in the extracts reproduced here, is included in the appendix. CA is a sociological approach that examines the moment-to-moment organisation of interaction through talk (Heritage, 1984). It relies on a highly detailed transcription system which assumes that no detail of interaction (pauses, hesitations, etc.) can be dismissed a priori as unimportant. By treating talk as a two-way and dynamic process, it avoids the fragmentation and categorisation necessary for some other kinds of analyses. A growing number of CA analyses within health-care settings (e.g. Silverman, 1997; Pilnick, 1999) have shown a clear correlation between the way in which interactional sequences are designed and the responses they generate from their recipients. A CA standpoint has been used here in order to enable examination of the ways in which talk about risk is produced and managed in situ (for a fuller explication of CA, see Heritage, 1984; Sacks, 1992). The interview transcripts were read and coded using a system of open coding (Strauss and Corbin, 1990) which allowed concepts and themes to emerge from the data. The coded data were compared for similarities and differences between the participants. The accounts produced in these interviews were not subject to any kind of ‘verification’ process using the actual taped consultations as a straightforward checklist, in order to establish what had and had not been discussed. Instead, women’s reported perceptions of what was
ARTICLE IN PRESS 86
discussed with their midwives were treated as an analytic resource in their own right.
A.M. Pilnick et al.
scribing conventions used are detailed in the appendix).
Findings Characteristics of sample One factor that was immediately evident before any detailed analysis took place was that all of the pregnant women who were recruited to this study appeared to have made their decisions about NT screening before their visit to the antenatal clinic. None of the sample requested any further discussion with hospital staff prior to the screening. Another factor was that all of the women recruited elected to undergo NT screening, although one woman was found on scanning to be too advanced in her pregnancy to be eligible. Despite the relatively small sample size, the data collected for the project raise many interesting issues. The focus here will be on the way that risk is talked about, in terms of the kind of information that NT screening can provide, the status of the results of screening and, as a consequence, the risks associated with other forms of testing. In order to do this, the analysis will draw on data from all three settings: consultations with community midwives, with hospital midwives, and post-screening interviews.
‘Low risk doesn’t mean no risk’ Previous research on serum screening has suggested confusion for women over the nature of their results (e.g. Smith et al., 1994). In the data collected for this study, however, the midwives appeared to be at great pains to make clear the nature of the result that NT screening would give. This is interesting in the light of Williams et al.’s (2002) finding that practitioners tend to focus more on the actual screening process rather than the condition that is being screened for. Anecdotally, in informal conversations to the researcher, midwives spoke of their concerns over previous clients who had spoken of getting ‘the all clear’ when they received a low-risk result, making it evident that they had not initially understood the limitations of the scan. Extract 1 is a good example of the stress that was placed in all the recorded pre-screening consultations on ensuring that women understood that, to quote one midwife ‘low risk doesn’t mean no risk’ (in all consultation extracts that follow, M ¼ midwife, E ¼ expectant woman, and the tran-
In this case, the midwife uses an example of a hypothetical, very low risk result (one in a thousand) to stress that whilst it is undoubtedly a small risk, it is still a risk. She subsequently presents the same information in several ways: that there is a possibility that the baby would be affected; that the outcome cannot be guaranteed; and that the NT scan is not a diagnostic test that can give a yes or no answer. The expectant mother gives a positive response to both the first and last of these presentations, apparently acknowledging the fact that there is still a risk, and that the test is not diagnostic. Her second acknowledgement, however, overlaps with the midwife’s continued explanation of what the fact that this is not a diagnostic test means (i.e. that there is no ‘yes or no’ answer to be given). Subsequently, the difficulty that may exist in assimilating this information, i.e. in understanding the nature of a risk level, is illustrated by the expectant mother’s question in the next turn, ‘Do they come back and say that (0.2) if there isFCan it be no risk at all?’. The midwife is unequivocal in her response to this, and the woman’s subsequent utterance makes her understanding explicit. Nevertheless, the amount of interactional work that is required here is considerable. The fact that the initial responses to the midwife’s explanations are positive and imply understanding are also noteworthyFit is only the expectant mother’s subsequent question that makes it clear she has not entirely grasped the significance of a risk status at this point. This emphasis on understanding the nature of the results was also echoed in the consultations with hospital midwives. The result from the NT scan was almost always communicated right at the beginning of the consultation, as in Extract 2.
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
Obviously, communicating the result in this way helps to avoid any unnecessary worry or uncertainty on the part of the woman and/or her partner, and as the midwife herself says, in the case of lowrisk results, is designed to reassure. In the case of high-risk results, this communication of results right at the beginning of the consultation provides the maximum opportunity to discuss possible courses of action that may be taken.
87
In this extract the midwife begins the discussion of the results by asking the expectant mother for her own perception of what would take place and what would result from the NT scan. Having seen the potential difficulties that occurred in Extract 1 around ensuring the woman in question understood the nature of a risk figure, this exchange highlights how useful it can be to begin by asking the client to state their own understanding. In Extract 1, the woman’s initial response indicates understanding but her subsequent question makes it clear that there is still some confusion. By asking the client to actually state their understanding rather than asking them if they understand, they do not have the option of simply concurring with what is said by the professional, and any gaps in their understanding become explicit. In this case, however, the woman’s understanding is apparently good, although the midwife follows this up with a question designed to further stress the difference between screening and diagnostic tests. Again, the woman actively demonstrates her understanding of this, and the topic for discussion subsequently changes.
Screening versus diagnostic tests Understanding the status of results Just as the data collected from consultations with community midwives highlighted a stress on understanding the nature and significance of risk figures, so there was a comparable stress in the hospital setting, once the figures have been given, on checking this understanding post-scan. Extract 3 illustrates this stress:
This stress on understanding the status of results from NT scans is highlighted across all the other hospital consultations, although not always in the same verbal form. In Extract 4, a woman has received a low-risk result from her NT scan, but at 38 years of age, had previously expressed concern about her age-related risk of Down’s (1 in 121 at 12 weeks). Having given her a figure of 1 in 498 after scanning, the midwife continues as follows:
ARTICLE IN PRESS 88
In this consultation then, despite the fact that the NT scan has lowered the risk which was calculated on maternal age alone, the expectant mother asks the midwife whether she thinks it is worth having any other tests on the basis of this result. What the midwife does is to reflect this question back to the woman, suggesting that the key point of decision is whether she wants definitive knowledge about the Down’s status of her fetus. The kind of information a diagnostic test would give is explicitly contrasted with the risk level that the woman has at present. At this point in the consultation, the woman herself raises a different but very much related issue, as below:
A.M. Pilnick et al.
figure is may then feel that they want more definitive information. Here, alongside the consideration of whether a risk figure is enough information for her, the woman raises the issue of risks associated with invasive testing. The midwife expands upon these in relation to the specific tests that are available. She goes on to highlight the positive nature of the NT result in that it shows an improvement on age-related risk, but also to stress once again that this is only a screening test without a definitive answer. She ends by emphasising that this is a choice for the couple to make, and there is no further discussion of diagnostic testing in this consultation. This need to weigh up the benefits of any further information against the risks of the tests themselves was also a common theme raised by women in interviews with the researcher.
Talking about dealing with risk In the final section of this paper we focus on women’s accounts of processing risk, and how they described establishing whether or not the risk figure gained from NT screening was satisfactory information for them. Three key themes appeared to recur in the data in relation to this, although as the extracts show, these were not always mutually exclusive. These were external factors affecting risk perception (raised by four women), knowing what action would be taken as a result of any further information (two women), and ‘crossing that bridge when we come to it’ (four women). Theme one: External factors affecting risk perception Extract 5 is taken from an interview with a woman in her first pregnancy, who had a low-risk result from her NT scan and had not considered further testing. She is talking about whether she would undergo NT screening with any subsequent pregnancy. Extract 5 (interview)
Extract 4 illustrates the potentially unsatisfactory nature of a risk figure for some women. Having established that a pregnancy is ‘low risk’ according to official criteria, those women whose understanding is clearest of what a risk
E: Yeah, I’llyIn future pregnancies, yes. I mean, it hasn’t made any diffyI mean obviously the results have been good. There’s, you know, very low risk – I mean I’m a reasonable age group as well, I suppose, in the first instance. In talking about her experiences, then, she draws on the rationale of her age group, as well as her
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
low-risk result, in accounting for her views on the utility of NT alone. Interestingly, however, she starts to refer to the fact that having NT for her made ‘no difference’ to the course of this pregnancy (‘I mean it hasn’t made any diffy). The results were ‘good’ (i.e. low risk) and this, combined with external factors, meant that she felt that this was sufficient information and would be likely to do so in the future. Other kinds of external factors were also described by women as having an effect: Extract 6 is taken from an interview with a woman undergoing her second pregnancy. Her NT scan for her current pregnancy had shown a low-risk result, and so she had not even considered the option of further testing this time. In her first pregnancy, which she is referring to below, she had had serum screening, which had produced a high-risk result. After much discussion with both her partner and health professionals, she decided not to undergo diagnostic testing.
89
risks. However, towards the end of the extract above she also relates that, having thought about her options, she was not sure that she would act if diagnostic testing had indicated fetal abnormality, and so ultimately the information may have been of limited use. This theme of knowing what would be done with the results of any further testing is illustrated more straightforwardly below in Extract 7. Theme two: Knowing what action would be taken The woman quoted in Extract 7 was also in her second pregnancy, and had also had serum screening first time around. Whilst she was clear that she understood the limitations of an NT result in terms of a definitive outcome, she was also unequivocal about what action she and her partner would have taken on the basis of a high-risk result. Extract 7 (interview)
Extract 6 (interview) Expectant mother:y So it was phone calls and midwives and all sorts. I was taken into one of the, is it the early pregnancy centre or something? Interviewer: Yeah, the early pregnancy assessment centre E: And the doctor discussed all, you know severities and things and I think I had another scan. Yeah I had a scan. TheyFshe couldn’t see anything. She measured the neck thing but it wasn’tyAnd she said it was up to me whatyAnd then I went to see Mr X (consultant). And he said, you know, possibly have an amniocentesis. I: Mhmm E: Um, but because I’ve had miscarriages before, and I also had IVy (name) was an IVF baby, to have an amnio and then risk miscarriaging and there’s nothing wrong I: Mmmm E: I didn’t know if something was wrong that I would (0.4) have an abortion. Uh, but we decided not to. But the rest of the pregnancy was quite stressful. After much thought, then, this woman decided not to pursue diagnostic testing, based partly on the risk of miscarriage as a result of the tests as weighed against the difficulty she had had in conceiving her baby and the fact that she had had IVF treatment. Her account of refusing diagnostic testing draws on these external factors in describing how she and her partner viewed comparative
E: If there’d been a very high risk of there being Downs, I’d have had an amniocentesis, and if that showed up a high risk then I probably would have been able to terminate the pregnancy, uh but we both agreed on that. So while I very much hoped that wouldn’t be the case we knew what we’d do next. In both cases above then, in assessing whether or not a risk figure was satisfactory information for them, both women had given some thought to possible future actions, although they arrived at very different conclusions. For the woman in Extract 7, the reality of a high-risk result had not arisen, and so it is impossible to know how her account might have related to her actual choices. Nevertheless, her account reveals a process of discussion and decisionmaking with her partner before she underwent NT screening. Theme three: ‘Crossing that bridge when we come to it’ Not all women, however, had gone through this process of thinking about the possible consequences in advance of having the scan, as the quote in Extract 8 illustrates: Extract 8 (interview) I: And had you thought, before you went into the scan about what you might do on the basis of the information that you would get? Or did you just
ARTICLE IN PRESS 90
think that you would cross that bridge when you came to it? E: Um, to be honest with you, I suppose, uh, I must be a born optimist, but I wouldn’t beFI wouldn’t think about something like that full stop; unless I had reason to believe there was something wrong or that. And it would be very much the case that we’ll cross that bridge when we’ll come to it.
This theme of ‘crossing the bridge when we come to it’ or of ‘worrying about it at the time’ in relation to a high-risk result is one that deserves more consideration, in terms of both the prevalence in this small sample and the personal and practical difficulties that such a strategy may create. Other women in the sample spoke of having the tests for reassurance, and one woman went so far as to turn the question on its head and ask the researcher ‘Why would I not want to have this um really unintrusive test?’. Although she was the only participant to state this quite so baldly, it seems likely that this will be the starting point for many women in their decision-making, an issue that we return to below. For almost all of the women in this sample, the results of NT screening did subsequently provide the reassurance of a low-risk status, although as we have seen, for some women they then wondered whether they wanted more definitive information. However, NT screening will not provide this reassurance for all women. Where unfavourable results are received, the issue of what to do next then becomes particularly salient. In the study sample, two women received NT results that were problematic in some way. One of these women, who had an unambiguously high-risk result, declined to be interviewed. The second woman’s case raises interesting issues, however, in that it highlights the potential problems that can occur not only when women have not previously considered the implications of an undesirable result, but also the conflict that can occur when official and personal notions of high risk do not correspond.
Competing versions of risk In extracts 5 and 6 above, we have seen how women drew on personal external variables e.g. their age or their difficulty in conceiving in order to personalise a risk figure. For the women in these cases, this had not caused them to re-evaluate their risk in a way that came into conflict with official categories, but had instead provided a strategy for decision-making based on their results.
A.M. Pilnick et al.
The woman quoted in Extract 9, however, gave a rather different account. This woman’s NT scan was not high risk in the official sense, in that she was given a risk of 1 in 217 after her scan, but at 25 her age-related risk alone had been 1 in 930 and so the scan result represented a considerable increase in this figure. Understandably, she was alarmed by the result, but was initially told that, in the clinic’s view, she was still low risk. She asked to speak to a consultant and subsequently arranged a CVS. In the extract below, she has related to the interviewer that she only discovered she was pregnant at 10 weeks. Extract 9 (interview) I: So in that case then you didn’t have so much time to decide whether you’d have the NT screening I guess, because if you saw your community midwife around about 10 weeksy E: Yeah, and then, it was sort of in week 12. Yeah so um no I didn’t y umy well as I say I didn’t I didn’t know about I Mmmm E: Um y Secondly, I didn’t it never occurred to me y that there would be a problem. So I just thought, it would kind of be a formality, cause I’m only 25. I: Yeah E: I didn’t (think) there would be any problem at allyUm, so I just thought you know, the more information the better. But I didn’t, you know, y I didn’t really consider it too much. By her own admission then, this woman had not considered the possibility of an unfavourable result, and what action she and her partner might take if they were to receive one. She had assumed that there would not be any problems highlighted by the NT scan, based on her relatively young age, and was therefore particularly shocked by the result. She was also critical of the way in which, as she saw it, the threshold risk of 1 in 200 had initially been applied to her result despite her concerns, saying: Extract 9 contd E: You know, there’s no sort of middle ground. There’s either y high risk or low risk, and um you know I would definitely say I wasn’t low risk, but then I didn’t fall into their high risk. Perhaps unsurprisingly, this woman was the least enthusiastic of all the women interviewed about NT
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
screening, reporting that after the CVS, which had not indicated any fetal abnormality, she had said to her partner that she was never having any screening again. The complexity of this decision, however, was highlighted by a subsequent utterance at the very end of the interview: Extract 9 contd E: Um (2.0) Phew, it’s so difficult to say. IyI think probablyyI think probably I would have to have itybecauseyofythe information being thereyum but I wouldn’t go into a CVS so quickly. Despite the difficulties that this woman felt NT screening had caused for her then, her ultimate view appeared to be that she would undergo the procedure in a subsequent pregnancy. The hesitant way in which this view was expressed makes the difficulty of the thought process clear. However, what she is describing as different for a subsequent pregnancy is the consideration of subsequent action based on NT results. For her first pregnancy, she had not given any consideration to this, but for her second, she proposes a possible course of action in advance.
Discussion The analysis carried out for this project demonstrates the overall success of the implementation of NT screening in this setting. There was a high level of uptake for the service, which was matched by a generally high level of satisfaction reported by those women who underwent NT scanning. This satisfaction was evidenced by the fact that for five of the women interviewed, they were unable to think of any way in which the service could be improved. Extracts from the tape-recorded consultation data demonstrate the pains taken by midwives to explain what NT screening would entail, and what the status of the results that women would receive would be. They also demonstrate the efforts made to ensure that women understand their results post-scan. These results are in stark contrast to previous work on serum screening, which has concluded that inadequate information was given to women beforehand (e.g. Smith et al., 1994; Howe et al., 2000) or that they were given no way of making sense of the results (e.g. Ettore, 2000). This begs the question as to why the findings from this study should be so different. Risk is a difficult subject to
91
communicate, and the nature of a risk figure is hard to grasp for many women. As a result, midwives take great pains interactionally to make clear what the status of NT results are. However, as the analysis above has shown, an attendant problem is that this information, whether high or low risk, may not subsequently be enough for women. For some women, who clearly grasp the nature of a low-risk level, they may think in advance that it will be sufficient, but subsequently feel more of a need for definitive information. Another issue that is highlighted here is in relation to ensuring informed choice. Previous research based on interviews and questionnaires has supported the idea that not all women are aware that there is an active decision to be made, or that they are in fact consenting to an optional test, or that they are clear about the nature of the results that they will receive. Some critics (e.g. Green and Statham, 1996; Press and Browner, 1997) have suggested that this is due to the way in which antenatal screening and testing are presented, but there has been little examination of actual consultation data to verify this hypothesis. The data presented here begin to give a much more clear cut example of the way in which these issues are raised by midwives, and demonstrate the utility of taking a more detailed interactional approach. They also highlight the difficulty of relying solely on post hoc accounts in the absence of information from the actual consultations. Given the stress on ensuring that women understood the purpose of NT screening that has been illustrated here, women’s accounts of presuming that there would be no problem, or that screening would be a formality, might be argued to be linked more to their own external perceptions of risk than to information given by midwives. The limited sample size of this study, combined with the fact that there were no refusers of NT screening and only two women who received problematic results, means that it cannot be generalised without caution. Nevertheless, it highlights the benefits that applying a different analytic approach can bring to the study of this important area. In terms of the practical implications of this study, women’s post hoc accounts of viewing the scan as a formality are also important. For some women undergoing any screening test, the results will not provide the reassurance that they seek. As the woman who was quoted here who found herself in this situation described, not having considered this possibility beforehand can create additional difficulties for decision-making. This apparent viewing of
ARTICLE IN PRESS 92
A.M. Pilnick et al.
NT screening as a zero cost option, where the question women ask themselves is not why they should have the scan but why they should not have it, is a fundamental issue. Part of the role of the community midwife in this setting is to ensure that these issues are explored. It is suggested, then, that women should be encouraged to consider the possibility of an unfavourable result in advance of screening, to think about what they might want to do as a result, and to discuss this with their partners where appropriate. Not only does this highlight the possibility that the test may not give reassurance, but it may also help women to decide whether the information that may be gained would be useful to them in their preferred course of action.
Acknowledgements We are very grateful to those midwives and pregnant women who participated in the project and allowed their consultations to be tape-recorded. Special thanks are also due to Telsing Andrews, for her assistance with this research, and to the anonymous referees who commented so helpfully on the first version of this paper.
Appendix. Key to transcription symbols [Hi(gap)] [Hello] (0.6) (.)
Hi ¼ ¼ Hello and so:ooo so really? and THEN and then she said JnoJ 4o
Square brackets indicate overlapping talk Figures in brackets indicate length of pauses A dot in brackets indicates a micropause (less than 0.2 seconds)
Equals signs indicate utterances between which there is no pause A colon indicates an extended sound or syllable A question mark indicates a rising intonation (not necessarily a question) Words in upper case are spoken more loudly Underlined words are emphasised comparative to other talk Degree signs indicate talk which is quieter than the surrounding talk When part of an utterance is delivered at a quicker pace than the
surrounding talk, it is enclosed between ‘less than’ signs. Audible aspirations (hhh) and inhalations (.hhh) are inserted in the speech where they occur.
References Atkinson, P., Heritage, J. (Eds.), 1984. Structures of Social Action: Studies in Conversation Analysis. Cambridge University Press, Cambridge. Bewley, S., Roberts, L., Mackinson, A.-M., et al., 1995. First trimester fetal nuchal translucency: problems with screening the general population 2. British Journal of Obstetrics and Gynaecology 102, 386–388. Ettore, E., 2000. Reproductive genetics, gender and the body: ‘please doctor, may I have a normal baby’. Sociology 34 (3), 403–420. Green, J., Statham, H., 1996. Psychosocial aspects of prenatal screening and diagnosis. In: Marteau, T., Richards, M. (Eds.), The Troubled Helix: Social and Psychological Implications of the New Human Genetics. Cambridge University Press, Cambridge. Heritage, J., 1984. Garfinkel and Ethnomethodology. Polity Press, Cambridge. Howe, T., Gornal, R., Wellesley, D., et al., 2000. Six year survey of screening for down’s syndrome by maternal age and midtrimester ultrasound scans. British Medical Journal 320, 606–610. Lane, B., Challen, K., Harris, H.J., et al., 2001. Existence and quality of written antenatal screening policies in the United Kingdom: postal survey. British Medical Journal 322, 22–23. Marteau, T.M., Slack, J., Kidd, J., et al., 1992. Presenting a routine screening test in antenatal care: practice observed. Public Health 106, 131–141. Marteau, T.M., 1995. Towards informed decisions about prenatal testing: a review. Prenatal Diagnosis 15, 1215–1226. Nicolaides, K., Azar, G., Byrne, D., et al., 1992. Fetal nuchal translucency: ultrasound screening for chromosomal defects in first trimester of pregnancy. British Medical Journal 304, 867–869. Pandya, P., Snijders, R., Johnson, S., et al., 1995. Screening for fetal trisomies by maternal age and fetal nuchal translucency thickness at 10 to 14 weeks of gestation. British Journal of Obstetrics and Gynaecology 102, 957–962. Pilnick, A., 1999. ‘Patient counselling’ by pharmacists: advice, information or instruction? The Sociological Quarterly 40, 613–622. Press, N., Browner, C.H., 1997. Why women say yes to prenatal diagnosis. Social Science and Medicine 45 (7), 979–989. Rothman, B.K., 1988. The Tentative Pregnancy: Prenatal Diagnosis and the Future of Motherhood. Pandora, London. Royal College of Obstetricians and Gynaecologists, 1997. Fetal Abnormalities: Guidelines for Screening, Diagnosis and Management. RCOG, London. Sacks, H., 1992. In: Jefferson, G. (Ed.), Lectures on Conversation, Vols. I, II. Blackwell, Oxford. Searle, J., 1997. Routine antenatal screening: not a case of informed choice. Australian and New Zealand Journal of Public Health 21 (3), 268–274.
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
Silverman, D., 1997. Discourses of counselling: HIV counselling as social interaction. Sage, London. Smith, D.K., Shaw, R.W., Marteau, T.M., 1994. Informed consent to undergo serum screening for Down’s syndrome: the gap between policy and practice. British Medical Journal 309, 776.
93
Strauss, A., Corbin, J., 1990. Basics of Qualitative Research. Sage, London. Williams, C., Alderson, P., Farsides, B., 2002. What constitutes ‘balanced information’ in the practitioners’ portrayals of Down’s syndrome? Midwifery 18, 230–237.
www.elsevier.com/locate/midw
Presenting and discussing nuchal translucency screening for fetal abnormality in the UK Alison M. Pilnick, BPharm, MRPharmS, MA, PhD (Senior Lecturer in Sociology)a,*, Diane M. Fraser, BEd, PhD, MTD, RM (Professor of Midwifery/Head)b, David K. James, MA, MD, FRCOG, DCH (Professor of Fetomaternal Medicine)c a
School of Sociology and Social Policy, University of Nottingham, University Park, Nottingham NG7 2RD, UK Academic Division of Midwifery, School of Human Development, University of Nottingham, UK c Division of Obstetrics and Gynaecology, School of Human Development, University of Nottingham, UK b
Received 19 February 2003; received in revised form 17 June 2003; accepted 4 July 2003
Summary Objective: to investigate the relationship between information giving by midwives and decision-making by women offered nuchal translucency (NT) screening. To establish how risk figures are discussed in practice, with the intention of relating this to the existing, and often critical, literature on women’s accounts of antenatal screening. Design: a qualitative study following women through the process of being offered and deciding to undergo NT screening. Tape recording of consultations, analysed in their entirety, was combined with post-screening interviews. Setting: a large teaching hospital in the UK. Participants: fourteen pregnant women eligible for NT screening at the time of recruitment. Data collection: (i) tape recordings of consultations between community midwives and pregnant women where nuchal translucency screening was offered; (ii) tape recordings of consultations between hospital midwives and pregnant women immediately post-screening; (iii) individual face-to-face interviews with pregnant women between two and six weeks after the screening, carried out by the first author. Findings: NT screening was in general well received, particularly by those women who had undergone serum screening with previous pregnancies. However, communicating the nature of a risk figure is an interactionally complex process. A large amount of interactional work is required by midwives both before and after screening to ensure that women comprehend this information. Despite the emphasis placed in these consultations on understanding the purpose of NT screening and the status of the results, women often framed their decision to undergo NT screening in terms of it being a formality, or of presuming that all was well. This sometimes created practical and personal difficulties in terms of decision-making.
*Corresponding author. E-mail address: [email protected] (A.M. Pilnick). 0266-6138/$ - see front matter & 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0266-6138(03)00056-1
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
83
Key conclusions and implications for practice: previous sociological and psychological research has tended to be critical of midwives in terms of ensuring informed choice in screening, but this research is often based on post hoc accounts. Examining actual consultations with these accounts helps to illustrate the other factors that affect women’s perceptions of testing, and the way in which risk, choice and decision-making are introduced and discussed in practice. Encouraging women to consider what action they might take on the basis of a personally unfavourable NT result in advance of undergoing the scan may help them to decide whether the information gained will be useful to them. Recognising the complex interactional work required in making sure that women understand the nature of the results that will be obtained is an important issue for the education and training of midwives. & 2003 Elsevier Ltd. All rights reserved.
Introduction At present, there are no specific national guidelines for antenatal screening procedures in the UK, beyond the report of a Royal College of Obstetricians and Gynaecologists (RCOG) (1997) working party on fetal abnormalities, and the RCOG recommendations that all antenatal units should have a written statement of their policy (Lane et al., 2001). National Screening Committee (NSC) recommendations are that there should be a policy of second trimester screening for Down’s syndrome using double analyses as a basic provision, and the National Institute for Clinical Excellence (NICE) is currently formulating evidence-based guidelines with the objective being a national co-ordinated approach to antenatal screening services. What this means in practice is that in the UK current policies vary in specific timing and in tests that are routinely offered on a local and regional basis. Individual hospitals also vary in their policies when women specifically ask for particular tests that are not routinely offered. In most hospitals in the UK, however, serum screening for one or more markers of fetal abnormality is a routinely offered procedure. Serum screening is carried out through a blood test at around 15–18 weeks (ideally at around week 16). Since it is not a diagnostic test, it will not in itself confirm whether or not a fetus is affected, and can give both false-positive and false-negative results. Some centres in the UK now offer an alternative form of screening, nuchal translucency (NT) screening, which can be carried out during an extended ultrasound scan at between 10 and 14 weeks. The nuchal fold is the thickness of skin overlying the cervical spine in the fetus. Nuchal translucency is a measurement of the fluid between the neck skin and the cervical spine. A number of research studies have indicated that a fetal nuchal translucency of X 3 mm is a useful first trimester marker for fetal chromosomal abnormalities, and particularly Trisomy 21 (Down’s syndrome) (e.g. Nicolaides et al., 1992; Pandya et al., 1995), although some
researchers have suggested its feasibility and promise needs to be backed up by further data obtained from low-risk populations (Bewley et al., 1995). The nuchal fold measurement, then, combined with maternal age (which is also a predictor of likelihood of Down’s) is used to produce a risk figure for fetal abnormality, expressed as a probability.
Studies of the implementation of serum screening There is a growing sociological and psychological literature on the practical implementation of serum screening programmes, which is largely critical of the way in which this service is delivered (e.g. Marteau et al., 1992, 1995; Searle, 1997). In particular, the extent to which women recognise that there is a real choice to be made here has been questioned, with research in both UK and US contexts tending to suggest that serum screening is seen by pregnant women as just another part of routine antenatal care (e.g. Rothman, 1988; Green and Statham, 1996; Press and Browner, 1997). Researchers have also explored the information that practitioners give to pregnant women about the conditions that are screened for, e.g. Williams et al.’s (2002) study of practitioners’ portrayals of Down’s syndrome, and have concluded that this can often be focused primarily on potential problems. However, since NT screening is a more recent phenomenon, there is a lack of research looking at its actual implementation in the UK.
Talking about risk The fact that antenatal screening procedures do not produce definitive results, but instead give probabilities, has been another ground on which they have been criticised. Some critical commentators have gone so far as to argue that women have no means of making sense of the information that they have been given, and that they may not
ARTICLE IN PRESS 84
understand that a low-risk result will not guarantee the birth of a ‘normal’ baby (e.g. Ettore, 2000). In a UK setting, the high level of uptake of antenatal screening for Down’s has led some researchers to question the adequacy of the information given to pregnant women and the process of informed consent (Howe et al., 2000). Smith et al. (1994) found that only 36% of their sample of 353 women who had been offered serum screening understood that a negative result would not guarantee that everything was all right with the pregnancy. In addition, only 32% understood that most women with positive results would go on to have a normal baby. Whilst the findings from these studies are compelling, it is important to recognise that for the most part they are based exclusively on interviews or questionnaires with participants after screening has taken place. Those studies that have used recorded data from consultations have tended to employ coding systems designed to identify particular components of the interactions. In general, then, there are few studies examining the ways in which issues are negotiated and the situations described above are achieved in practice. How is it that screening comes to be perceived as routine, that choice is not made explicit, or that women do not understand the nature of a risk figure?
Aims of the study The overall objective of this study was to investigate the relationship between information giving by midwives and decision-making by women offered NT screening. Within this framework, the aim was to examine in more detail the interactional presentation of antenatal screening procedures, and the ways in which they are introduced, discussed and contested by those involved. The study was developed through a recognition of the need to examine NT screening in its own right, to examine whether and to what extent the criticisms of the delivery of serum screening programmes are valid in this scenario. It also aimed to establish to what extent women’s post hoc interview accounts of screening procedures relate to their actual interactions with health-care professionals.
Methods The data presented here are taken from consultations forming part of a newly introduced antenatal
A.M. Pilnick et al.
screening programme. Criticisms of serum screening programmes have not been limited to the psychological and sociological literature, and in the medical field there has been growing concern over the level of accuracy of the results, as well as the short time frame in which women must make decisions about further testing or termination of pregnancy. Influenced by these limitations, NT screening has been introduced in the study hospital. At the time of the study, it was offered to all pregnant women who ‘booked’ to deliver their babies at the hospital and who had their first appointment at the antenatal clinic within the window of time when the screening could be carried out. For the purposes of the programme reported here, a ‘high-risk’ figure was considered to be anything greater than 1 in 200, and those women whose results were in this category would be offered further, diagnostic testing. This figure was chosen in this setting because it equates to the risk of miscarriage from an amniocentesis at 16 weeks, but other centres in the UK use figures ranging from 1 in 100 to 1 in 250.
Sample and recruitment In order to examine as much of the process of the presentation and discussion of NT screening as possible, the data for this project came from three sources: pre-screening discussions of NT screening between community midwives and pregnant women; post-screening consultations between hospital midwives and pregnant women (where the results of the NT scan were given); and face-toface interviews with a subsample of the women who had had their consultations recorded, between two and six weeks after their scan. After obtaining Local Research Ethics Committee approval, and with the co-operation of community and hospital midwives, the aim was to recruit eight pregnant women from each of two different geographical areas in the catchment of the hospital. The first group were recruited by local midwives from an inner-city area where the average age of first pregnancy is below the national average in the UK, and the second by local midwives from an affluent suburban area where the average age of first pregnancy is higher than the national average. These women were asked to take part in a study examining the presentation and discussion of NT screening. A total of 14 women, six from the first group and eight from the second, were eventually recruited, and gave their written consent for their consultations to be tape-recorded. The partners of two women
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
were also present at the hospital consultations, and they also consented verbally to participation and recording. Recruitment of women from the first group was more difficult, as a higher proportion declined to take part. As they were not asked to give a reason for refusal, it is not possible to interpret this finding fully. Whilst recruitment to this group was still ongoing, the NT programme was suspended, making further recruitment impossible. This suspension was due to staff shortages, and the acknowledgement that serum screening at 16 weeks was also available for the women involved. It was initially intended to recruit a third group of women, from a rural area also served by the study hospital, but differences in the way that antenatal care was organised for these women made this problematic, since pregnant women in this area had no contact with a community midwife until after their first hospital antenatal visit. In the two study areas, after a meeting with the community midwife, the woman was subsequently referred to the hospital antenatal clinic. At this first visit to the clinic, NT screening, if it was accepted, took place. If NT screening was declined, women could still undergo a straightforward ultrasound scan to confirm the expected delivery date of the pregnancy, and to exclude the possibility of multiple pregnancy. In order to examine the actual process of discussion about NT screening, it was necessary to tape women’s actual consultations with midwives, rather than relying solely on the post hoc accounts provided by interviews. For the 14 women initially recruited to the study, their first consultations about screening with their community midwife, and, where possible, their subsequent consultations on their first visit to the antenatal clinic, were audiotaped. No researcher was present at these consultations, which all took place in private settings, although in some cases student midwives, partners or children were present. The midwives themselves operated the tape recorders, on the understanding that they could be turned off at any time if any participant so wished. Although women were offered the opportunity of a consultation with a hospital midwife immediately prior to their NT scan if there were any questions they wished to raise at that stage, none of the women recruited to the study took up this offer. All consultations recorded at the hospital clinic during this study took place after NT screening had taken place, and where the results were delivered and discussed by hospital midwives. For the third, interview-based, phase of the study, it was intended to recruit a subsample of
85
eight women from those who had been previously recorded, i.e. four from each area. However, it was only possible to recruit six women (two from area 1 and four from area 2). Two further women from area 1 declined to be interviewed and two did not respond to either a first or follow-up letter requesting an interview. The participants were interviewed by the first author (using semi-structured interview techniques based around broad themes for discussion) between two and six weeks after their visit to the antenatal clinic. The purpose of the interviews was to explore women’s perceptions of NT screening, in particular around whether they had felt their informational needs were met both pre- and post-NT scan.
Analysis The tape-recorded consultations were transcribed and analysed according to conversation analytic (CA) conventions (Jefferson, as published in Atkinson and Heritage, 1984). A simplified key to this transcription system developed by Jefferson, as used in the extracts reproduced here, is included in the appendix. CA is a sociological approach that examines the moment-to-moment organisation of interaction through talk (Heritage, 1984). It relies on a highly detailed transcription system which assumes that no detail of interaction (pauses, hesitations, etc.) can be dismissed a priori as unimportant. By treating talk as a two-way and dynamic process, it avoids the fragmentation and categorisation necessary for some other kinds of analyses. A growing number of CA analyses within health-care settings (e.g. Silverman, 1997; Pilnick, 1999) have shown a clear correlation between the way in which interactional sequences are designed and the responses they generate from their recipients. A CA standpoint has been used here in order to enable examination of the ways in which talk about risk is produced and managed in situ (for a fuller explication of CA, see Heritage, 1984; Sacks, 1992). The interview transcripts were read and coded using a system of open coding (Strauss and Corbin, 1990) which allowed concepts and themes to emerge from the data. The coded data were compared for similarities and differences between the participants. The accounts produced in these interviews were not subject to any kind of ‘verification’ process using the actual taped consultations as a straightforward checklist, in order to establish what had and had not been discussed. Instead, women’s reported perceptions of what was
ARTICLE IN PRESS 86
discussed with their midwives were treated as an analytic resource in their own right.
A.M. Pilnick et al.
scribing conventions used are detailed in the appendix).
Findings Characteristics of sample One factor that was immediately evident before any detailed analysis took place was that all of the pregnant women who were recruited to this study appeared to have made their decisions about NT screening before their visit to the antenatal clinic. None of the sample requested any further discussion with hospital staff prior to the screening. Another factor was that all of the women recruited elected to undergo NT screening, although one woman was found on scanning to be too advanced in her pregnancy to be eligible. Despite the relatively small sample size, the data collected for the project raise many interesting issues. The focus here will be on the way that risk is talked about, in terms of the kind of information that NT screening can provide, the status of the results of screening and, as a consequence, the risks associated with other forms of testing. In order to do this, the analysis will draw on data from all three settings: consultations with community midwives, with hospital midwives, and post-screening interviews.
‘Low risk doesn’t mean no risk’ Previous research on serum screening has suggested confusion for women over the nature of their results (e.g. Smith et al., 1994). In the data collected for this study, however, the midwives appeared to be at great pains to make clear the nature of the result that NT screening would give. This is interesting in the light of Williams et al.’s (2002) finding that practitioners tend to focus more on the actual screening process rather than the condition that is being screened for. Anecdotally, in informal conversations to the researcher, midwives spoke of their concerns over previous clients who had spoken of getting ‘the all clear’ when they received a low-risk result, making it evident that they had not initially understood the limitations of the scan. Extract 1 is a good example of the stress that was placed in all the recorded pre-screening consultations on ensuring that women understood that, to quote one midwife ‘low risk doesn’t mean no risk’ (in all consultation extracts that follow, M ¼ midwife, E ¼ expectant woman, and the tran-
In this case, the midwife uses an example of a hypothetical, very low risk result (one in a thousand) to stress that whilst it is undoubtedly a small risk, it is still a risk. She subsequently presents the same information in several ways: that there is a possibility that the baby would be affected; that the outcome cannot be guaranteed; and that the NT scan is not a diagnostic test that can give a yes or no answer. The expectant mother gives a positive response to both the first and last of these presentations, apparently acknowledging the fact that there is still a risk, and that the test is not diagnostic. Her second acknowledgement, however, overlaps with the midwife’s continued explanation of what the fact that this is not a diagnostic test means (i.e. that there is no ‘yes or no’ answer to be given). Subsequently, the difficulty that may exist in assimilating this information, i.e. in understanding the nature of a risk level, is illustrated by the expectant mother’s question in the next turn, ‘Do they come back and say that (0.2) if there isFCan it be no risk at all?’. The midwife is unequivocal in her response to this, and the woman’s subsequent utterance makes her understanding explicit. Nevertheless, the amount of interactional work that is required here is considerable. The fact that the initial responses to the midwife’s explanations are positive and imply understanding are also noteworthyFit is only the expectant mother’s subsequent question that makes it clear she has not entirely grasped the significance of a risk status at this point. This emphasis on understanding the nature of the results was also echoed in the consultations with hospital midwives. The result from the NT scan was almost always communicated right at the beginning of the consultation, as in Extract 2.
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
Obviously, communicating the result in this way helps to avoid any unnecessary worry or uncertainty on the part of the woman and/or her partner, and as the midwife herself says, in the case of lowrisk results, is designed to reassure. In the case of high-risk results, this communication of results right at the beginning of the consultation provides the maximum opportunity to discuss possible courses of action that may be taken.
87
In this extract the midwife begins the discussion of the results by asking the expectant mother for her own perception of what would take place and what would result from the NT scan. Having seen the potential difficulties that occurred in Extract 1 around ensuring the woman in question understood the nature of a risk figure, this exchange highlights how useful it can be to begin by asking the client to state their own understanding. In Extract 1, the woman’s initial response indicates understanding but her subsequent question makes it clear that there is still some confusion. By asking the client to actually state their understanding rather than asking them if they understand, they do not have the option of simply concurring with what is said by the professional, and any gaps in their understanding become explicit. In this case, however, the woman’s understanding is apparently good, although the midwife follows this up with a question designed to further stress the difference between screening and diagnostic tests. Again, the woman actively demonstrates her understanding of this, and the topic for discussion subsequently changes.
Screening versus diagnostic tests Understanding the status of results Just as the data collected from consultations with community midwives highlighted a stress on understanding the nature and significance of risk figures, so there was a comparable stress in the hospital setting, once the figures have been given, on checking this understanding post-scan. Extract 3 illustrates this stress:
This stress on understanding the status of results from NT scans is highlighted across all the other hospital consultations, although not always in the same verbal form. In Extract 4, a woman has received a low-risk result from her NT scan, but at 38 years of age, had previously expressed concern about her age-related risk of Down’s (1 in 121 at 12 weeks). Having given her a figure of 1 in 498 after scanning, the midwife continues as follows:
ARTICLE IN PRESS 88
In this consultation then, despite the fact that the NT scan has lowered the risk which was calculated on maternal age alone, the expectant mother asks the midwife whether she thinks it is worth having any other tests on the basis of this result. What the midwife does is to reflect this question back to the woman, suggesting that the key point of decision is whether she wants definitive knowledge about the Down’s status of her fetus. The kind of information a diagnostic test would give is explicitly contrasted with the risk level that the woman has at present. At this point in the consultation, the woman herself raises a different but very much related issue, as below:
A.M. Pilnick et al.
figure is may then feel that they want more definitive information. Here, alongside the consideration of whether a risk figure is enough information for her, the woman raises the issue of risks associated with invasive testing. The midwife expands upon these in relation to the specific tests that are available. She goes on to highlight the positive nature of the NT result in that it shows an improvement on age-related risk, but also to stress once again that this is only a screening test without a definitive answer. She ends by emphasising that this is a choice for the couple to make, and there is no further discussion of diagnostic testing in this consultation. This need to weigh up the benefits of any further information against the risks of the tests themselves was also a common theme raised by women in interviews with the researcher.
Talking about dealing with risk In the final section of this paper we focus on women’s accounts of processing risk, and how they described establishing whether or not the risk figure gained from NT screening was satisfactory information for them. Three key themes appeared to recur in the data in relation to this, although as the extracts show, these were not always mutually exclusive. These were external factors affecting risk perception (raised by four women), knowing what action would be taken as a result of any further information (two women), and ‘crossing that bridge when we come to it’ (four women). Theme one: External factors affecting risk perception Extract 5 is taken from an interview with a woman in her first pregnancy, who had a low-risk result from her NT scan and had not considered further testing. She is talking about whether she would undergo NT screening with any subsequent pregnancy. Extract 5 (interview)
Extract 4 illustrates the potentially unsatisfactory nature of a risk figure for some women. Having established that a pregnancy is ‘low risk’ according to official criteria, those women whose understanding is clearest of what a risk
E: Yeah, I’llyIn future pregnancies, yes. I mean, it hasn’t made any diffyI mean obviously the results have been good. There’s, you know, very low risk – I mean I’m a reasonable age group as well, I suppose, in the first instance. In talking about her experiences, then, she draws on the rationale of her age group, as well as her
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
low-risk result, in accounting for her views on the utility of NT alone. Interestingly, however, she starts to refer to the fact that having NT for her made ‘no difference’ to the course of this pregnancy (‘I mean it hasn’t made any diffy). The results were ‘good’ (i.e. low risk) and this, combined with external factors, meant that she felt that this was sufficient information and would be likely to do so in the future. Other kinds of external factors were also described by women as having an effect: Extract 6 is taken from an interview with a woman undergoing her second pregnancy. Her NT scan for her current pregnancy had shown a low-risk result, and so she had not even considered the option of further testing this time. In her first pregnancy, which she is referring to below, she had had serum screening, which had produced a high-risk result. After much discussion with both her partner and health professionals, she decided not to undergo diagnostic testing.
89
risks. However, towards the end of the extract above she also relates that, having thought about her options, she was not sure that she would act if diagnostic testing had indicated fetal abnormality, and so ultimately the information may have been of limited use. This theme of knowing what would be done with the results of any further testing is illustrated more straightforwardly below in Extract 7. Theme two: Knowing what action would be taken The woman quoted in Extract 7 was also in her second pregnancy, and had also had serum screening first time around. Whilst she was clear that she understood the limitations of an NT result in terms of a definitive outcome, she was also unequivocal about what action she and her partner would have taken on the basis of a high-risk result. Extract 7 (interview)
Extract 6 (interview) Expectant mother:y So it was phone calls and midwives and all sorts. I was taken into one of the, is it the early pregnancy centre or something? Interviewer: Yeah, the early pregnancy assessment centre E: And the doctor discussed all, you know severities and things and I think I had another scan. Yeah I had a scan. TheyFshe couldn’t see anything. She measured the neck thing but it wasn’tyAnd she said it was up to me whatyAnd then I went to see Mr X (consultant). And he said, you know, possibly have an amniocentesis. I: Mhmm E: Um, but because I’ve had miscarriages before, and I also had IVy (name) was an IVF baby, to have an amnio and then risk miscarriaging and there’s nothing wrong I: Mmmm E: I didn’t know if something was wrong that I would (0.4) have an abortion. Uh, but we decided not to. But the rest of the pregnancy was quite stressful. After much thought, then, this woman decided not to pursue diagnostic testing, based partly on the risk of miscarriage as a result of the tests as weighed against the difficulty she had had in conceiving her baby and the fact that she had had IVF treatment. Her account of refusing diagnostic testing draws on these external factors in describing how she and her partner viewed comparative
E: If there’d been a very high risk of there being Downs, I’d have had an amniocentesis, and if that showed up a high risk then I probably would have been able to terminate the pregnancy, uh but we both agreed on that. So while I very much hoped that wouldn’t be the case we knew what we’d do next. In both cases above then, in assessing whether or not a risk figure was satisfactory information for them, both women had given some thought to possible future actions, although they arrived at very different conclusions. For the woman in Extract 7, the reality of a high-risk result had not arisen, and so it is impossible to know how her account might have related to her actual choices. Nevertheless, her account reveals a process of discussion and decisionmaking with her partner before she underwent NT screening. Theme three: ‘Crossing that bridge when we come to it’ Not all women, however, had gone through this process of thinking about the possible consequences in advance of having the scan, as the quote in Extract 8 illustrates: Extract 8 (interview) I: And had you thought, before you went into the scan about what you might do on the basis of the information that you would get? Or did you just
ARTICLE IN PRESS 90
think that you would cross that bridge when you came to it? E: Um, to be honest with you, I suppose, uh, I must be a born optimist, but I wouldn’t beFI wouldn’t think about something like that full stop; unless I had reason to believe there was something wrong or that. And it would be very much the case that we’ll cross that bridge when we’ll come to it.
This theme of ‘crossing the bridge when we come to it’ or of ‘worrying about it at the time’ in relation to a high-risk result is one that deserves more consideration, in terms of both the prevalence in this small sample and the personal and practical difficulties that such a strategy may create. Other women in the sample spoke of having the tests for reassurance, and one woman went so far as to turn the question on its head and ask the researcher ‘Why would I not want to have this um really unintrusive test?’. Although she was the only participant to state this quite so baldly, it seems likely that this will be the starting point for many women in their decision-making, an issue that we return to below. For almost all of the women in this sample, the results of NT screening did subsequently provide the reassurance of a low-risk status, although as we have seen, for some women they then wondered whether they wanted more definitive information. However, NT screening will not provide this reassurance for all women. Where unfavourable results are received, the issue of what to do next then becomes particularly salient. In the study sample, two women received NT results that were problematic in some way. One of these women, who had an unambiguously high-risk result, declined to be interviewed. The second woman’s case raises interesting issues, however, in that it highlights the potential problems that can occur not only when women have not previously considered the implications of an undesirable result, but also the conflict that can occur when official and personal notions of high risk do not correspond.
Competing versions of risk In extracts 5 and 6 above, we have seen how women drew on personal external variables e.g. their age or their difficulty in conceiving in order to personalise a risk figure. For the women in these cases, this had not caused them to re-evaluate their risk in a way that came into conflict with official categories, but had instead provided a strategy for decision-making based on their results.
A.M. Pilnick et al.
The woman quoted in Extract 9, however, gave a rather different account. This woman’s NT scan was not high risk in the official sense, in that she was given a risk of 1 in 217 after her scan, but at 25 her age-related risk alone had been 1 in 930 and so the scan result represented a considerable increase in this figure. Understandably, she was alarmed by the result, but was initially told that, in the clinic’s view, she was still low risk. She asked to speak to a consultant and subsequently arranged a CVS. In the extract below, she has related to the interviewer that she only discovered she was pregnant at 10 weeks. Extract 9 (interview) I: So in that case then you didn’t have so much time to decide whether you’d have the NT screening I guess, because if you saw your community midwife around about 10 weeksy E: Yeah, and then, it was sort of in week 12. Yeah so um no I didn’t y umy well as I say I didn’t I didn’t know about I Mmmm E: Um y Secondly, I didn’t it never occurred to me y that there would be a problem. So I just thought, it would kind of be a formality, cause I’m only 25. I: Yeah E: I didn’t (think) there would be any problem at allyUm, so I just thought you know, the more information the better. But I didn’t, you know, y I didn’t really consider it too much. By her own admission then, this woman had not considered the possibility of an unfavourable result, and what action she and her partner might take if they were to receive one. She had assumed that there would not be any problems highlighted by the NT scan, based on her relatively young age, and was therefore particularly shocked by the result. She was also critical of the way in which, as she saw it, the threshold risk of 1 in 200 had initially been applied to her result despite her concerns, saying: Extract 9 contd E: You know, there’s no sort of middle ground. There’s either y high risk or low risk, and um you know I would definitely say I wasn’t low risk, but then I didn’t fall into their high risk. Perhaps unsurprisingly, this woman was the least enthusiastic of all the women interviewed about NT
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
screening, reporting that after the CVS, which had not indicated any fetal abnormality, she had said to her partner that she was never having any screening again. The complexity of this decision, however, was highlighted by a subsequent utterance at the very end of the interview: Extract 9 contd E: Um (2.0) Phew, it’s so difficult to say. IyI think probablyyI think probably I would have to have itybecauseyofythe information being thereyum but I wouldn’t go into a CVS so quickly. Despite the difficulties that this woman felt NT screening had caused for her then, her ultimate view appeared to be that she would undergo the procedure in a subsequent pregnancy. The hesitant way in which this view was expressed makes the difficulty of the thought process clear. However, what she is describing as different for a subsequent pregnancy is the consideration of subsequent action based on NT results. For her first pregnancy, she had not given any consideration to this, but for her second, she proposes a possible course of action in advance.
Discussion The analysis carried out for this project demonstrates the overall success of the implementation of NT screening in this setting. There was a high level of uptake for the service, which was matched by a generally high level of satisfaction reported by those women who underwent NT scanning. This satisfaction was evidenced by the fact that for five of the women interviewed, they were unable to think of any way in which the service could be improved. Extracts from the tape-recorded consultation data demonstrate the pains taken by midwives to explain what NT screening would entail, and what the status of the results that women would receive would be. They also demonstrate the efforts made to ensure that women understand their results post-scan. These results are in stark contrast to previous work on serum screening, which has concluded that inadequate information was given to women beforehand (e.g. Smith et al., 1994; Howe et al., 2000) or that they were given no way of making sense of the results (e.g. Ettore, 2000). This begs the question as to why the findings from this study should be so different. Risk is a difficult subject to
91
communicate, and the nature of a risk figure is hard to grasp for many women. As a result, midwives take great pains interactionally to make clear what the status of NT results are. However, as the analysis above has shown, an attendant problem is that this information, whether high or low risk, may not subsequently be enough for women. For some women, who clearly grasp the nature of a low-risk level, they may think in advance that it will be sufficient, but subsequently feel more of a need for definitive information. Another issue that is highlighted here is in relation to ensuring informed choice. Previous research based on interviews and questionnaires has supported the idea that not all women are aware that there is an active decision to be made, or that they are in fact consenting to an optional test, or that they are clear about the nature of the results that they will receive. Some critics (e.g. Green and Statham, 1996; Press and Browner, 1997) have suggested that this is due to the way in which antenatal screening and testing are presented, but there has been little examination of actual consultation data to verify this hypothesis. The data presented here begin to give a much more clear cut example of the way in which these issues are raised by midwives, and demonstrate the utility of taking a more detailed interactional approach. They also highlight the difficulty of relying solely on post hoc accounts in the absence of information from the actual consultations. Given the stress on ensuring that women understood the purpose of NT screening that has been illustrated here, women’s accounts of presuming that there would be no problem, or that screening would be a formality, might be argued to be linked more to their own external perceptions of risk than to information given by midwives. The limited sample size of this study, combined with the fact that there were no refusers of NT screening and only two women who received problematic results, means that it cannot be generalised without caution. Nevertheless, it highlights the benefits that applying a different analytic approach can bring to the study of this important area. In terms of the practical implications of this study, women’s post hoc accounts of viewing the scan as a formality are also important. For some women undergoing any screening test, the results will not provide the reassurance that they seek. As the woman who was quoted here who found herself in this situation described, not having considered this possibility beforehand can create additional difficulties for decision-making. This apparent viewing of
ARTICLE IN PRESS 92
A.M. Pilnick et al.
NT screening as a zero cost option, where the question women ask themselves is not why they should have the scan but why they should not have it, is a fundamental issue. Part of the role of the community midwife in this setting is to ensure that these issues are explored. It is suggested, then, that women should be encouraged to consider the possibility of an unfavourable result in advance of screening, to think about what they might want to do as a result, and to discuss this with their partners where appropriate. Not only does this highlight the possibility that the test may not give reassurance, but it may also help women to decide whether the information that may be gained would be useful to them in their preferred course of action.
Acknowledgements We are very grateful to those midwives and pregnant women who participated in the project and allowed their consultations to be tape-recorded. Special thanks are also due to Telsing Andrews, for her assistance with this research, and to the anonymous referees who commented so helpfully on the first version of this paper.
Appendix. Key to transcription symbols [Hi(gap)] [Hello] (0.6) (.)
Hi ¼ ¼ Hello and so:ooo so really? and THEN and then she said JnoJ 4o
Square brackets indicate overlapping talk Figures in brackets indicate length of pauses A dot in brackets indicates a micropause (less than 0.2 seconds)
Equals signs indicate utterances between which there is no pause A colon indicates an extended sound or syllable A question mark indicates a rising intonation (not necessarily a question) Words in upper case are spoken more loudly Underlined words are emphasised comparative to other talk Degree signs indicate talk which is quieter than the surrounding talk When part of an utterance is delivered at a quicker pace than the
surrounding talk, it is enclosed between ‘less than’ signs. Audible aspirations (hhh) and inhalations (.hhh) are inserted in the speech where they occur.
References Atkinson, P., Heritage, J. (Eds.), 1984. Structures of Social Action: Studies in Conversation Analysis. Cambridge University Press, Cambridge. Bewley, S., Roberts, L., Mackinson, A.-M., et al., 1995. First trimester fetal nuchal translucency: problems with screening the general population 2. British Journal of Obstetrics and Gynaecology 102, 386–388. Ettore, E., 2000. Reproductive genetics, gender and the body: ‘please doctor, may I have a normal baby’. Sociology 34 (3), 403–420. Green, J., Statham, H., 1996. Psychosocial aspects of prenatal screening and diagnosis. In: Marteau, T., Richards, M. (Eds.), The Troubled Helix: Social and Psychological Implications of the New Human Genetics. Cambridge University Press, Cambridge. Heritage, J., 1984. Garfinkel and Ethnomethodology. Polity Press, Cambridge. Howe, T., Gornal, R., Wellesley, D., et al., 2000. Six year survey of screening for down’s syndrome by maternal age and midtrimester ultrasound scans. British Medical Journal 320, 606–610. Lane, B., Challen, K., Harris, H.J., et al., 2001. Existence and quality of written antenatal screening policies in the United Kingdom: postal survey. British Medical Journal 322, 22–23. Marteau, T.M., Slack, J., Kidd, J., et al., 1992. Presenting a routine screening test in antenatal care: practice observed. Public Health 106, 131–141. Marteau, T.M., 1995. Towards informed decisions about prenatal testing: a review. Prenatal Diagnosis 15, 1215–1226. Nicolaides, K., Azar, G., Byrne, D., et al., 1992. Fetal nuchal translucency: ultrasound screening for chromosomal defects in first trimester of pregnancy. British Medical Journal 304, 867–869. Pandya, P., Snijders, R., Johnson, S., et al., 1995. Screening for fetal trisomies by maternal age and fetal nuchal translucency thickness at 10 to 14 weeks of gestation. British Journal of Obstetrics and Gynaecology 102, 957–962. Pilnick, A., 1999. ‘Patient counselling’ by pharmacists: advice, information or instruction? The Sociological Quarterly 40, 613–622. Press, N., Browner, C.H., 1997. Why women say yes to prenatal diagnosis. Social Science and Medicine 45 (7), 979–989. Rothman, B.K., 1988. The Tentative Pregnancy: Prenatal Diagnosis and the Future of Motherhood. Pandora, London. Royal College of Obstetricians and Gynaecologists, 1997. Fetal Abnormalities: Guidelines for Screening, Diagnosis and Management. RCOG, London. Sacks, H., 1992. In: Jefferson, G. (Ed.), Lectures on Conversation, Vols. I, II. Blackwell, Oxford. Searle, J., 1997. Routine antenatal screening: not a case of informed choice. Australian and New Zealand Journal of Public Health 21 (3), 268–274.
ARTICLE IN PRESS Nuchal translucency screening for fetal abnormality
Silverman, D., 1997. Discourses of counselling: HIV counselling as social interaction. Sage, London. Smith, D.K., Shaw, R.W., Marteau, T.M., 1994. Informed consent to undergo serum screening for Down’s syndrome: the gap between policy and practice. British Medical Journal 309, 776.
93
Strauss, A., Corbin, J., 1990. Basics of Qualitative Research. Sage, London. Williams, C., Alderson, P., Farsides, B., 2002. What constitutes ‘balanced information’ in the practitioners’ portrayals of Down’s syndrome? Midwifery 18, 230–237.