Practices and views on fetal heart monitoring: a structured observation and interview study


Dr Mary Dixon-Woods, Social Science Group, Department of Health Sciences, University of Leicester, 22-28 Princess Road West, Leicester LE1 6TP, UK. Email


Objective  To assess and explain deviations from recommended practice in National Institute for Clinical Excellence (NICE) guidelines in relation to fetal heart monitoring.

Design  Qualitative study.

Setting  Large teaching hospital in the UK.

Sample  Sixty-six hours of observation of 25 labours and interviews with 20 midwives of varying grades.

Methods  Structured observations of labour and semistructured interviews with midwives. Interviews were undertaken using a prompt guide, audiotaped, and transcribed verbatim. Analysis was based on the constant comparative method, assisted by QSR N5 software.

Main outcome measures  Deviations from recommended practice in relation to fetal monitoring and insights into why these occur.

Results  All babies involved in the study were safely delivered, but 243 deviations from recommended practice in relation to NICE guidelines on fetal monitoring were identified, with the majority (80%) of these occurring in relation to documentation. Other deviations from recommended practice included indications for use of electronic fetal heart monitoring and conduct of fetal heart monitoring. There is evidence of difficulties with availability and maintenance of equipment, and some deficits in staff knowledge and skill. Differing orientations towards fetal monitoring were reported by midwives, which were likely to have impacts on practice. The initiation, management, and interpretation of fetal heart monitoring is complex and distributed across time, space, and professional boundaries, and practices in relation to fetal heart monitoring need to be understood within an organisational and social context.

Conclusion  Some deviations from best practice guidelines may be rectified through straightforward interventions including improved systems for managing equipment and training. Other deviations from recommended practice need to be understood as the outcomes of complex processes that are likely to defy easy resolution.


Monitoring the fetal heart rate (FHR) is aimed at assessing fetal wellbeing and hypoxia during labour. It can involve listening to fetal heart sounds with an acoustic device (or hand-held Doppler ultrasound), or use of electronic fetal heart monitoring (EFM). Intermittent auscultation, recommended for low-risk pregnancy, is a timed method of listening to fetal heart sounds with an acoustic device, noting rate, rhythm, and variability of the fetal heart beat for at least 1 full minute, usually following a uterine contraction. EFM, first introduced in the 1960s, enables continuous rather than intermittent monitoring of FHR, and is now widely used in obstetrics throughout the developed world.1 EFM uses cardiotocograph (CTG) machines that can monitor the FHR, maternal heart rate, uterine contractions, and (sometimes) the pressure inside the uterus during labour. FHR can be detected either through an ultrasound transducer strapped to the woman's abdomen or an electrocardiogram electrode placed on fetal scalp. It is recommended that EFM be supplemented with fetal scalp blood gas analysis.2

Despite the widespread use of continuous EFM, controversy about its risks and benefits persists.3–5 A Cochrane review6 summarised the evidence from nine randomised controlled trials and found that, apart from a reduction in the rate of neonatal seizures, use of routine EFM has no measurable impact on mortality or morbidity. Some evidence has suggested that intermittent auscultation with a pinard fetal stethoscope is as reliable as continuous EFM as a means of recording fetal wellbeing in normal pregnancies,7 and other work has suggested that continuous EFM contributes to a rise in intervention rates,8 but does not prevent many poor outcomes of labour.9

Amid the debate about the outcomes of EFM, little attention has been given to the evidence that practices in relation to EFM are a cause for concern. The Confidential Enquiry into Stillbirths and Deaths in Infancy have highlighted that errors in interpretations of CTG traces are a major contributor to infant morbidity and mortality.10–12 Recent guidelines from the UK National Institute for Clinical Excellence (NICE) recommend intermittent auscultation for women with low-risk pregnancy2 and make a series of recommendations concerning indications for use of continuous EFM, conduct of EFM, education and training, and risk management. Little is known, however, about the implementation of these guidelines or about why deviations from recommended practice might occur.

We aimed to identify deviations from NICE guidelines in relation to fetal monitoring and to explore midwives’ views on fetal heart monitoring in labour.


This study combined a structured observation approach13 with semistructured interviews. Ethical approval was obtained for this study from Leicestershire Local Research Ethics Committee.

The study was conducted in a labour ward (LW) and midwifery-led unit (MU) in a large teaching hospital delivering approximately 5800 babies per annum. The recent NICE guideline on EFM had been adopted as hospital policy. Local policy is that all staff receive training every 6 months on CTG interpretation, in the form of a study day, completing a computer training package (K2 Fetal Monitoring Training System), or a combination of self-directed study and multidisciplinary meetings. Midwives in the hospital have a ‘CTG Passport’ which records CTG training in all its forms.

Structured observations

All observations during labour were undertaken by S.A., a medical student with extensive training and experience in fetal monitoring, over a 1-month period. Labours were purposively selected to include different midwifery staff, day and night shifts, and pregnancies that had been assessed as high- and low risk both antenatally and on admission. Pregnancies likely to be deemed high risk included those of women with previous caesarean section, induced labour, and medical complications. For each observation, S.A. sought consent (following provision of an information sheet and verbal explanation of the study) from both the labouring woman and the midwife caring for her, before the observation took place. S.A. acted in an observational capacity only and was not expected to intervene.

A structured observation proforma based on the NICE guidelines (Box 1. Box 1) was used to record data, and key events during each labour were noted. In addition, each midwife who was being observed was asked to describe indications for EFM, errors commonly made with EFM, and the value of EFM.

Table Box 1. .  Proforma for collection of data from observations of practice based on NICE guidelines on electronic fetal monitoring
Assessment of maternal and fetal heart rates
1. Was the maternal pulse palpated simultaneously with fetal heart rate auscultation in order to differentiate between maternal and fetal heart rates?
Documentation and storage
2. Was there a check to ensure that the date and time clocks on the EFM machine were correctly set?
3. Were the traces labelled with the mother's name, date, and hospital number?
4. Were any intrapartum events that might affect the FHR noted contemporaneously on the EFM trace, signed and the date and time noted?
5. Was any member of staff asked to provide an opinion on the trace? If yes, did s/he note their findings on both the trace and the maternal case notes together with the date, time, and signature?
6. Were the settings on the machine standardised?
7. Was there an appropriate interpretation?
8. Other observations


Midwives were purposively selected to reflect different levels of experience and grade. They were approached directly by S.A., and consent to participate and to tape-record the interviews was obtained. A semistructured prompt guide, based on a literature review and discussions within the project team, was used in the interviews. This covered midwives’ views on the value of EFM, experiences of using EFM, and problems with EFM. Participants who had been observed were not asked to comment on the specific case that had been observed. Interviews were transcribed verbatim.


Classification of the data recorded on each structured observation proforma was undertaken by a consultant obstetrician (C.O.) and S.A. C.O. determined whether CTG traces had been correctly interpreted, using the DR C BRAVADO (determine risk, contractions, baseline rate, variability, accelerations, decelerations, overall assessment) mnemonic structured approach.

For interview transcriptions and observation field notes, a systematic and iterative method of analysis based on the constant comparative method was employed.14 Initially ‘open codes’ were applied to the data representing the significance of sections of text. Open codes were then incrementally grouped into organising categories or themes across the whole corpus (including both field notes and transcripts). These categories were modified and checked constantly in order to develop a coding frame with explicit specifications. An audit trail of the development of categories was maintained. The coding frame was programmed into QSR N5 qualitative software and was used by S.A. to process the data set systematically by assigning each section of text to a category, according to the category specifications. The assignment of data to categories was validated by M.D.-W and R.S.


Structured observations

Over 14 days, 66 hours of observation of 25 labours were undertaken. Two observations were conducted during night shifts and the others during varying times of the day. Thirteen women were assessed as high risk on admission to the hospital and were treated on the LW; 12 women were assessed initially as low risk, with nine of them admitted to the MU and the remaining three to the LW (1Table 1). All except three labours (on the MU) involved some period of EFM. Seven women who were initially assessed as low risk and started on intermittent monitoring were subsequently changed to continuous EFM for a variety of medical reasons. One woman who began on continuous EFM was later changed to intermittent monitoring. All labours resulted in the birth of a live infant, with Apgar scores in the first minute ranging from 6 to 9. Ten labours resulted in a nonelective caesarean section; there were two forceps deliveries, three deliveries by ventouse, and ten normal (unassisted) vaginal deliveries.

Table 1.  Initial classification of risk status of women, initial location, monitoring, form of delivery, and Apgar scores at 1 and 5 minutes
Woman with high-/low-risk pregnancy on admissionInitial locationMonitoringForm of deliveryApgar scores at 1, 5 minutes
  1. CEFM, continuous electronic fetal monitoring; EMCS, emergency caesarean section; NVD, normal vaginal delivery.

 1 LowMUIntermittent followed by CEFMEMCS9, 9
 2 LowMUIntermittent followed by CEFMForceps9, 9
 3 LowLWCEFM followed by intermittentNVD9, 10
 4 HighLWCEFMEMCS9, 10
 5 HighLWCEFMNVD9, 10
 6 HighLWCEFMNVD8, 9
 7 HighLWCEFMNVD8, 9
 8 LowMUIntermittentNVD7, 9
 9 HighLWCEFMEMCS9, 10
11 HighLWCEFMNVD9, 9
12 LowMUIntermittentNVD9, 9
16 HighLWCEFMEMCS9, 10
17 LowMUIntermittent followed by CEFMEMCS9, 9
19 HighLWCEFMNVD9, 9
20 HighLWCEFMVentouse9, 9
21 LowMUIntermittent followed by CEFMVentouse8, 9
22 LowMUIntermittent followed by CEFMForceps9, 9
23 LowLWCEFMVentouse9, 9
24 LowMUIntermittentNVD9, 10
25 LowMUIntermittent followed by CEFMNVD6, 9

Both the MU and the LW were busy environments, with midwives frequently attending more than one woman at a time and sometimes appearing to be under pressure and ‘hassled’. When they were with a woman in labour, midwives performed an executive role, coordinating multiple sources of information and making decisions about when a medical opinion was required. Doctors arrived in the labour room either in response to a call from a midwife or during routine ‘rounds’. They would usually have a consultation with the midwife, look at the CTG trace (sometimes outside the room), and offer an opinion or make a decision.

A total of 243 deviations from NICE guidelines of varying types were identified (2Table 2). Between 2 and 19 deviations from recommended practice, with a median of 11, were observed per labour, indicating that no labour observed was free of deviations from recommended practice in relation to fetal heart monitoring (34Tables 3 and 4). The most common area of deviation from NICE guidelines was in relation to documentation; others included indications for use of EFM and conduct of fetal heart monitoring. None of the deviations from NICE guidelines observed was deemed to have resulted in direct harm to women or their babies.

Table 2.  Observed deviations from recommended practice
DeviationFrequency of error
Indications for EFM
CTG discontinued for epidural, despite labour-inducing drugs having been administered5
Continuous EFM where intermittent was indicated (low-risk pregnancy)2
Assessment of maternal and fetal heart rates
Omitted checking of maternal pulse10
Listened for only 30 seconds, not a full minute4
Maternal pulse checked but not noted when listening intermittently2
Omitted to listen to FHR after contractions1
No palpation of contractions1
Used electronic fetal monitor rather than sonicaid to listen intermittently1
No time recorded on CTG when doctor checked it41
No date recorded on CTG when doctor checked it38
No comment recorded on CTG when doctor checked it28
No maternal pulse recorded on CTG13
Doctor omitted to sign the CTG when checked14
Vaginal examination not recorded on CTG8
Dilation not recorded on CTG6
Oxytocin (used to start or augment labour) commencement or increase not recorded on CTG4
No patient identity sticker on CTG3
Doctor omitted to provide any documentation at all on CTG3
No documentation when CTG stopped3
CTG not secured in notes/loose notes in file3
Epidural procedure not recorded on CTG2
Fetal blood sampling not recorded on CTG2
Change of paper not recorded on CTG2
No full name on section of CTG2
Gap in documentation in notes2
No delivery details recorded on CTG1
Maternal blood glucose level not recorded on CTG1
Antepartum haemorrhage not recorded on CTG1
Commencement of fetal scalp electrode not recorded on CTG1
Cessation of CTG not recorded during instrumental delivery in theatre1
Time of arrival in theatre not recorded on CTG1
Observation of meconium stained liquor following artificial rupture of membranes not recorded on CTG1
Decelerations of FHR not recorded on CTG1
Transfer into theatre not recorded on CTG1
No delivery details in notes1
Missing pages in notes1
Start of second stage of labour not documented1
No documentation of CTG in notes1
Monitor changed—no reason stated1
No check of date and time on monitor1
No time on drug chart for administration of first vaginal prostin (vaginal pessary used for induction)1
CTG signed by any doctor on the round, not the one who had checked it1
Nothing recorded on CTG when recommenced after epidural procedure1
Time on monitor not calibrated1
Failed to check if date and time on CTG were correct1
Nothing recorded in notes for over 2 hours1
Inadequate response to suboptimal CTG trace2
Caesarean section because of poor CTG trace although fetal blood sample normal1
Other deviations from recommended practice
Clinical errors
Combination of continuous and intermittent monitoring—neither done properly1
Staff not monitoring CTG during preparation for theatre1
Midwife failed to explain procedures to labouring women or answer their questions4
Doctor failed to explain procedures to labouring women or answer their questions2
Midwife incorrectly thought transcutaneous electrical nerve stimulation machine would interfere with monitor1
Midwife did not know how to change CTG clock1
Midwife did not know why jelly was used for monitoring1
Problems with tocograph—not set properly or not detecting contractions3
No appropriate leg strap for fetal scalp electrode2
No buttons available for CTG belts; knot had to be used1
Connector needed to operate equipment for fetal scalp electrode not easily available1
Poor quality of trace (loss of contact with fetal heartbeat)1
Table 3.  Number of deviations from recommended practice by type
Deviation typeMinimum per caseMaximum per caseMedian per caseTotal deviations (for 25 cases)
Indications for EFM1117
Assessment of maternal and fetal heart rates265.519
Other deviations13119
Table 4.  Number of deviations by labour
Case numberIndications for EFMAssessment of heart ratesDocumentationInterpretationOtherTotal

Indications for the use of continuous electronic fetal monitoring

Seven occasions were identified on which it was deemed that need for EFM was incorrectly assessed (Table 2, Indications for EFM). For example, continuous EFM was used for two labours where there were no indications of risk; normal deliveries followed. Perhaps the greatest potential for serious risk occurred in labours when EFM was stopped during the administration of an epidural, even though women had been given labour-inducing drugs.

Intermittent auscultation

Observations identified four labours in which 18 deviations from NICE recommendations in the conduct of intermittent auscultation occurred. Many deviations, including omitting to listen for a full minute, would have rendered the procedure useless. Most of these deviations from recommended practice appeared to result from knowledge or skills deficits on the part of those undertaking auscultation.


NICE guidelines note that concise, accurate, and contemporaneous documentation of intrapartum events is an important factor in obstetric litigation, and that CTG traces are important medical and legal documents. However, deviations from recommended practice in relation to documentation accounted for 80% of the total deviations observed. Examples include a case where the date and time was incorrect on the machine and was not corrected for over 3 hours, meaning that all documentation during those hours of labour was inaccurate. On occasions, midwives did not record all the necessary patient details, including name and hospital number. Midwives also omitted to document intrapartum events, including vaginal examinations, the initiation of fetal scalp blood sampling, and insertion of epidural catheters. Consultants, registrars, and senior house officers were also all observed deviating from guidance on documentation. Examples included a consultant offering an opinion on a trace, but not recording it and not documenting the initiation of fetal scalp blood sampling.

NICE guidelines state that the EFM trace should be stored securely with the maternal notes at the end of the monitoring process, as some claims associated with failure to respond to abnormal EFM cannot be defended because of missing EFM documentation. Our observations identified occasions when the trace was not stored in a wallet, meaning the print quality would deteriorate over time and result in illegibility; sometimes it was stored upside down (making it easy to fall out) and sometimes it was not inserted in the notes at all.

Observations suggested that many deviations in relation to documentation occurred during midwives’ attempts to refocus attention on the labouring woman, and where there were multiple claims on midwives’ and doctors’ attention, including the need to attend to more than one woman. Midwives sometimes appeared irritated at having to put stickers on traces, while doctors were observed to offer verbal information and move to the next patient without making records either in the patients’ notes or on the CTG trace.


Three errors of interpretation were noted. In two cases, suboptimal CTGs which should have activated a clinical response were not correctly assessed; in another case a failure to interpret a CTG in the light of evidence from a fetal blood sample indicating no cause for alarm resulted in a possibly unwarranted caesarean section.

Other issues

Other issues observed included problems with communication, including failure to explain issues relating to EFM to labouring women. On one occasion, for example, the partner of a woman in labour asked why jelly was used for the abdominal monitor but the midwife did not know.

The availability and maintenance of equipment was mentioned as a problem by all midwives who were being observed. Midwives described frustration at not being able to easily find equipment that was in good order and could be relied on to function correctly. Observations identified five labours in which these problems were evident, for example in the lack of leg straps, which resulted in sticky tape or abdominal belts with knots being used.


Interviews were undertaken with 20 midwives of varying grades. Nine of them usually worked with low-risk, midwifery-led labours, while the remainder usually worked with high-risk labours. Years of experience of working as a midwife ranged from 6 months to 30 years.

Role of the technology

Three different orientations towards the use of EFM could be identified in participants’ accounts: faith, caution, and scepticism. Many of the differences in these orientations could be traced to variations in beliefs about the extent to which CTG traces could be treated as authentic and reliable in providing an indication of the wellbeing of the fetus.

Faith in EFM.  One orientation involved feeling reliant on EFM and ‘trusting’ it. For some, especially newly qualified midwifery staff, EFM was important because they did not yet feel sufficiently confident to trust their own professional judgement; they tended to see the technology as offering a more authoritative resource:

In general midwives tend to value the EFM as they can then be reassured that nothing is going wrong and as I [am] a newly qualified midwife even more so. (Participant 7, high-risk labours)

Others in this group felt that EFM had become incorporated into their standard practice, and pointed to the ability of EFM to provide a documented account of the monitoring.

I find CTG monitoring an important part of my role, whether that's because I am a high risk midwife and I am used to them and you get used to things, don’t you. It gives you a nice hard copy and I think that suits everybody, it just settles your mind and you’ve got proof and with intermittent you have no copy. (Participant 20, high-risk labours)

Cautious about EFM.  A second group was more circumspect in the role they saw for EFM, seeing it as only one element of the care of women, and a limited one at that. Midwives in this group emphasised the need to see the CTG trace as just one element in a range of inputs to judgements about fetal wellbeing.

I think it has got a place, but I think it has also got to have a backup. There is no point in having electronic fetal monitoring if you have not got the […] system to back it up […] I don’t think it is reliable by itself. (Participant 3, low-risk labours)

Sceptical about EFM.  A third group was extremely sceptical about the value of EFM, believing that it was not necessary or that it was used inappropriately. Six midwives, for example, felt that EFM involved ‘looking for trouble’.

I don’t believe in doing them really. I think the evidence is that it is not necessary to do them and I certainly … it is just like you’re looking for trouble and you end up with the cascade of intervention, so as soon as you pin the lady down to the bed in a position you know that's it. You always find something on a CTG, we always seem to find some reason for it to stay on a bit longer and then it never comes off. (Participant 18, high-risk labours)

Rather than seeing EFM as a reliable, authoritative, and well-documented way of monitoring FHR, midwives in this third group saw use of EFM as eroding and undermining professional skills that were more likely to produce an authentic assessment of fetal wellbeing.

On [high risk] labour ward the midwives sort of lose their skills. If they are on [the midwifery-led unit] listening in, [they are] relying on their own skills picking up abnormalities. But on labour ward they are trusting a machine and a piece of paper. (Participant 1, low-risk labours)

Notwithstanding these differing orientations, all midwives, including those who were generally in favour of EFM, expressed concerns about its use. All accounts referred to the potential for discomfort and anxiety in labouring women. There was a strong perception that EFM was overused, particularly with low-risk women, and for reasons that were clinically unjustified, including defensive practice. Use of EFM was seen to give rise to important tensions for midwives. Contrasts were drawn between the need to give attention and care to the woman in labour and the need to attend to the CTG machine: when the machine was in use, it was seen as deflecting attention from personalised midwifery care.

Everybody in the room focuses on it, if the conversation dries up everybody looks at the monitor. You become fixated by it. (Participant 20, high-risk labours)

Midwives’ accounts identified a key tension between giving what was seen to be ‘personalised care’ for women and the need for thorough documentation of fetal monitoring. Midwives described how they felt that focusing on the machine and its documentation meant a reduction in focus on the woman, resulting in significant role strain.

It takes your attention away from the woman because you’re anxious that you need to keep looking at it and you’ve got to keep your notes up to date, it just becomes an extra thing to keep writing in the notes, and you feel as though you are spending all your time note keeping and not actually looking after the woman. (Participant 17, high-risk labours)

Professional roles

Midwives’ accounts expressed considerable concern about issues of professional roles in relation to EFM. Some midwives, who typically looked after ‘low-risk’ women and who rarely used EFM, saw EFM as encouraging a ‘deskilling’ of midwives, in which the expertise of midwives comes to be seen as less authoritative than the machine. ‘Looking at the monitor rather than the woman’ was seen as a particular problem.

[…] lot of midwives tend to look at the monitor rather than at the progress in labour and about the woman's body language and you know taking cues from all of that, so I think they can be, they can distort things a little I think. […] you can get twitched over them when there's no need to be. (Participant 5, high-risk labours)

Midwives working routinely with high-risk labours, particularly if they were relatively experienced, generally expressed confidence in their ability to use a CTG machine and interpret traces. However, they showed concern about the involvement of doctors as an occupational group in responding to CTG traces. There was a perception among some midwives that doctors tended to rely too heavily on CTG traces rather than making a full assessment of situations based on all the information. Midwives argued that as an occupational group, midwives were in a position to make a holistic assessment of fetal and maternal wellbeing based on multiple sources of evidence and their presence with the woman, sometimes over many hours. Doctors as an occupational group were, by contrast, perceived to focus their attention too narrowly on CTG traces. There was concern that a doctor's opinion based solely on a CTG, particularly if junior or inexperienced, could trump that of an experienced midwife who believes that she has more insight into a particular labour:

[..] they’ve not been with the woman for the amount of time that the midwife's got, they’re breezing in and out of rooms, so really it's their it's their only tool. (Participant 15, high-risk labours)

If [doctors] haven’t got something to look at they don’t want to know. If I said to a doctor the fetal heart is perfect … I mean I have had it happen where you have had a perfect fetal heart and a baby born who is not a problem and they will not actually believe you. (Participant 4, low-risk labours)

These concerns did not simply relate to territorial disputes over professional boundaries, but to a reluctance to accept an individual doctor's competence at face value over something as equivocal as the interpretation of a CTG trace. Midwives repeatedly stressed the importance of doctors having sufficient experience and tacit knowledge to make an appropriate judgement, describing how they often had to negotiate to find a doctor whom they could trust to produce a credible and reliable interpretation:

You have to get a doctor in who has even less experience than yourself and that is sometimes a bit worrying. You just have to jump a few heads to get the person you need. (Participant 9, high-risk labours)

Training and equipment

Midwives were generally satisfied with the training available for fetal heart monitoring, but many also emphasised the role of experience in using the equipment and making appropriate judgements. Considerable frustration was expressed in relation to the quality, availability, and maintenance of equipment for fetal heart monitoring. Problems with paper getting stuck, having no system for reporting repairs, dirty or damaged belts, and missing parts were reported. The problem of ‘ownership’ of problems relating to equipment, its availability, cleanliness, and maintenance was prominent in midwives’ accounts.

Well we have not actually got the connecting things on the monitors anymore, we have to go to the core staff, find one of them, find the keys, get the electrode out of the cupboard and then hook it up and it all depends on which monitor you have to which connector you require. It's a bit frustrating and obviously if you need one you need it for a reason, so you have not got the time to go looking for all the bits and bobs to go with it. (Participant 1, low-risk labours).


Our mixed method of study involving structured observations of labours and interviews with midwives suggests that when fetal heart monitoring practices are compared against national guidelines, a high rate of deviation from recommended practice, particularly in relation to documentation, is evident. Midwives and doctors routinely do not fully document their opinions and actions in relation to fetal heart monitoring, or record intrapartum events. There is also evidence of difficulties with availability and maintenance of equipment, and some deficits in staff knowledge and skill. Encouragingly, none of these deviations was observed to have resulted in direct harm to any of the women or their babies involved in this study, as all babies were safely delivered, but clearly many practices observed posed the potential for adverse outcomes. Deviations from recommended practice need to be understood within an organisational and social context: this study shows that the initiation, management, and interpretation of fetal heart monitoring is a highly complex process distributed across time, space, and professional boundaries.

This study has a number of limitations. It was based in only one hospital, though we have little reason to believe that practices in this hospital are likely to be different from those nationally. This was a relatively small-sample study, which was not necessarily representative of the population from which it was drawn; for example the spontaneous vaginal delivery rate for our study was 40%, compared with 65–70% for the whole population. A wider range of events, including more examples of serious outcomes, might have been identified if more observations had been undertaken. Assessment of CTG traces was undertaken by one consultant only, and other professionals might have formed a different view of some traces. Similar research in other hospitals around the UK would be welcomed, as would interviews with medical staff of varying grades to complement our focus on midwives. Nonetheless, this study provides important insights into the nature of professional practices and organisational influences in relation to fetal heart monitoring.

Our findings confirm previous work that suggests that routine practice very often does not incorporate guidelines on best practice in a straightforward way.15 There has been variable success in use of guidelines and other interventions to influence professional behaviour, but little progress in understanding why any particular intervention does or does not work.16 Recent empirical work has begun to show that the rational–technical thrust of evidence-based practice has failed to take sufficient account of vital aspects of professional practice that rely on intuitive, embodied, tacit knowledge,17 and the local and experiential nature of the world of clinical practitioners.18 Our study demonstrates the relevance of such insights while also illustrating the need for opportunities for simple deficits in equipment and training to be identified and corrected.

Our findings point to a number of mundane, but nonetheless important, issues that affect practices in relation to fetal monitoring. Problems with equipment and its maintenance were mentioned by all staff and were observed as a source of difficulty. LW need to have systems in place for ‘owning’ problems concerning equipment and ensuring its maintenance and availability. Some deviations from recommended practice reflected lack of knowledge or skill, and here appropriate training is likely to be of benefit.

More generally, however, our study suggests that many deviations in relation to fetal heart monitoring are not a simple function of deficits in training or staff vigilance, but are, rather, the outcomes of forms of social organisation and understandings of professional role. Our findings show that EFM has a very ambivalent status in midwifery practice and confirm earlier suggestions of disquiet about EFM among (at least some) health professionals,19,20 including concerns that EFM can cause the ‘art of midwifery’ to be lost to technology,21 and that EFM is lacking in credibility as an intervention. Midwives’ views and practices in relation to fetal heart monitoring are not homogeneous, and their orientations towards EFM in particular varies from hostility to welcome. These orientations appear to be strongly related to what midwives see as the proper definition of their ‘work’ and their domain.

Observations and midwives’ accounts suggested that midwives see their primary responsibilities as two-fold: first, providing ‘personalised care’ to women in labour, and second, acting in an executive capacity in coordinating multiple sources of information about the wellbeing of the woman and the fetus. Continuous EFM by some midwives is seen as a source of anxiety and discomfort for labouring women, and tending to the CTG machine rather than the woman was often seen as a potential violation of midwives’ first responsibility. This finding may help to explain the high level of omissions in documentation. Documentation omissions seem to occur in large part because of the strains midwives experience between trying to maintain the standard of documentation required and trying to provide ‘personalised care’ in the context of multiple demands on their time. A redefinition of midwifery responsibilities that places documentation at the core of ‘what midwives do’ might require profound shifts in understandings of professional role.

The second responsibility—that of the ‘executive’—was a source of further ambivalence in relation to fetal monitoring. Midwives varied in the extent to which they saw CTG traces as offering an authentic and reliable means of making fetal wellbeing visible, or, in Atkinson's terms, legible.22 Relying on a ‘machine and a piece of paper’ was seen by some as offering a poor substitute for using traditional midwifery skills, but most concern focused on the fallibility of interpretations of CTG traces. Many midwives’ accounts saw CTG traces not as providing reliable guides to what should happen next, but as documents that were vulnerable to being ‘read’ incorrectly and used to support inappropriate decisions. Some midwives drew contrasts between the types of practices engaged in by midwives as an occupational group and doctors as an occupational group. The CTG trace appeared to signal in some accounts an opportunity for symbolic power23—the power to define a situation—to slip from midwifery control to medical control, and was therefore a focus of dilemmas about professional boundaries and judgements. Midwives were frequently critical of doctors’ training and competence, although the basis for these concerns was unclear. These findings demonstrate the tensions between the kinds of experiential, tacit knowledge that experienced practitioners value (particularly in the context of contested technologies) and the rational model of decision making and practice implied by guidelines. They also point to the need for team-based, multidisciplinary approaches to training and reflective practice.24

Deviations from recommended practice in relation to EFM need to be understood as the outcome of complex processes. Some deviations from best practice guidelines occur due to the unsatisfactory storage of necessary equipment or due to multiple demands on time and attention, which may be rectified through relatively straightforward interventions. Other deviations from recommended practice are likely to defy easy resolution. The multiple inputs to decision making in relation to fetal heart monitoring; the negotiations and values attached to individuals as well as to occupational groups; and the exigencies of LW, where the human and physical resources are often very stretched, will need to be recognised in any efforts to improve practices in relation to fetal heart monitoring.


Thanks to the maternity unit staff for welcoming the research and thanks to the participants for their cooperation. Thanks to Sheila Bonas, Department of Health Sciences, University of Leicester, for help with preparation of tables and advice on interpretation. Thanks to David Taylor, Doug Charlton, and Jennifer Kurinczuk for providing comments on an earlier draft.