A midwifery practice dichotomy on oral intake in labour

ARTICLE IN PRESS Midwifery (2004) 20, 72–81

www.elsevier.com/locate/midw

A midwifery practice dichotomy on oral intake in labour Myra Parsons, BHSc, RN, RM (Postgraduate Student/Independent Midwife) 23, Mansfield Road, Galston, NSW 2159, Australia Received 5 June 2002; received in revised form 29 May 2003; accepted 8 July 2003

Summary Objective: to assess the views and practices of midwives regarding oral intake in labour for women with low-risk pregnancies. Design: an exploratory survey design including open- and closed-ended questions. Setting: four hospitals in Sydney, Australia. Participants: 89 practising midwives who provided care for labouring women. Findings: midwives were divided on the issue of what and when labouring women should, or should not, be allowed to eat and drink. The views and practices of these midwives were influenced by the accepted practice in the hospital in which they were employed and the types of midwifery models in which they have practised. Key conclusions: there is insufficient conclusive research evidence to support any stance on oral intake for labouring women. Most information purported by supporters of oral intake is based on anecdotal evidence and assumptions based on the physiology of the body. ‘Nil by mouth’ policies have never been researched while clear fluid policies are based on research performed with non-obstetric patients. Implications for practice: without reliable research evidence for the management of oral intake for labouring women no hospital practice or policy is valid. This leaves midwives with the responsibility of deciding what they believe is the best management for the oral intake of labouring women in their care. & 2004 Published by Elsevier Ltd.

Introduction Midwifery management of oral intake for labouring women has varied over time and across the world (Broach and Newton, 1988). Today, in the western world, the issue of whether women should be allowed to eat, and to some extent drink, during their labour is very controversial (Pengelley and Gyte, 1998; Sleutel and Golden, 1999). Information regarding this issue was obtained through a search of the CINAHL, MEDLINE and Cochrane Library E-mail address: [email protected] (M. Parsons). 0266-6138/$ - see front matter & 2004 Published by Elsevier Ltd. doi:10.1016/S0266-6138(03)00055-X

databases using the keywords: oral intake, food, labour, obstetric anaesthesia and aspiration. Two very different issues are central to the debate around the oral intake of labouring women. Firstly, there is the fear of complications arising due to aspiration of gastric contents in a labouring woman who may require a general anaesthetic during her labour. Anaesthetists are concerned that to perform a general anaesthetic, especially in an emergency situation when a woman has recently eaten, puts the woman at an increased risk of regurgitation and aspiration (Kallar and Everett, 1993). However, there is no research that supports

ARTICLE IN PRESS A midwifery practice dichotomy on oral intake in labour

the practice of fasting labouring women as a method of preventing gastric aspiration (Pengelley and Gyte, 1998; Sleutel and Golden, 1999). The second issue is the belief that withholding food and some types of fluid during labour may be detrimental to the mother, her baby and the progress of labour (Kristensen et al., 1991). This view is held by many midwives and is believed to out-weigh the concerns of the anaesthetic staff regarding the rare incidence of aspiration (Lewis, 1992; Department of Health, 1993; Pengelley and Gyte, 1998). In light of the controversy and the fact that the management of oral intake for labouring women has been shown to vary from hospital to hospital in the UK, USA and Australia (Michael et al., 1991; Baker, 1996; Parsons, 2001), it was of interest to ascertain whether midwives within hospitals agreed on this issue. Reported here are the findings of a survey conducted in four Sydney teaching hospitals to determine the views and practices of midwives regarding oral intake in labour for women with low-risk pregnancies.

Literature review The literature search regarding the oral intake of labouring women comprised obstetric, anaesthetic, and midwifery texts along with a number of systematic trawls of the computerised databases, CINAHL, MEDLINE and the Cochrane Library. The review was conducted using the keywords: oral intake, food, labour, obstetric anaesthesia and aspiration and limited to English, human, randomised, controlled trials (RCT). Although there were four published studies that met the criteria, only Scrutton et al., (1999) investigated the effect of food consumption during labour on labour outcomes. However, this study had an insufficient sample size, and was more interested in measuring and comparing the gastric contents between groups. Their investigation of the effect of food consumption by labouring women on their labour and birth outcomes appeared to be a side issue. One study explored the effect of intravenous fluid as opposed to food consumption on ketone production during labour (Tourangeau et al., 1999), while the remaining two studies compared the effect of a carbohydrate drink with a placebo drink (Scheepers et al., 2002) and an isotonic ‘sport drink’ with water (Kubli et al., 2002). Neither study found any difference between groups for labour and birth outcomes. Two more unpublished RCTs that had investigated the effect of food consumption on labour outcomes were located. One was a

73

newspaper article while the other was found in a book about midwifery research. Both reports were scanty, both failed to consider the effect of extraneous variables such as parity, their data analyses were very basic, and their findings were contradictory (Rodwell, 1992; Yiannouzis and Parnell, 1994). The literature for this paper, therefore, has relied upon the scientific and theoretical knowledge of digestion, gastric emptying, haematological alterations, and aspiration, along with anaesthetic research into drugs and techniques employed during obstetric general anaesthesia and their effect on maternal aspiration. Historical and contemporary opinion tend to provide the less than adequate basis for much of this review. During the first half of the 20th century in England women were strongly encouraged to eat and drink throughout labour to avoid general weakness, delayed labour, and serious postpartum haemorrhage (Ludka and Roberts, 1993). By the mid-1940s in the USA, as various drugs and general anaesthesia (specifically ether and chloroform) for labour, vaginal births, forceps deliveries and caesarean sections became the norm, the practice of feeding labouring women was abandoned for a policy of ‘nil by mouth’ to help reduce the occurrence of aspiration of gastric contents associated with the use of some of these drugs (Ludka and Roberts, 1993). By the 1960s Australia had also instituted this policy (Slater, personal communication 2000). In an endeavour to reduce the number of maternal deaths attributed to anaesthesia-related aspiration of gastric contents, anaesthetists vigorously researched the contributory causes of, and prevention strategies for, aspiration episodes. A number of obstetric and anaesthetic drugs and practices have been tested over the years; some required refinement while others were abolished depending on the effect they had on the incidence of aspiration. Many of the drugs and techniques used were associated, directly or indirectly, with the incidence of aspiration. Fasting labouring women is only one of a number of aspiration prevention strategies employed in an attempt to prevent the incidence of aspiration during general anaesthesia (Oberoi and Phillips, 2000). The purpose of this strategy is to better ensure an empty stomach ‘if’ a general anaesthetic is required during the course of labour (Glosten, 2000), although no research can be located to support this practice. Research has, however, found that despite prolonged fasting the majority of nonobstetric patients have a large gastric volume at the time of surgery (Cote et al., 1982).

ARTICLE IN PRESS 74

Current situation regarding the risk of aspiration The anaesthetic process has changed greatly over the last 50 years, with a subsequent fall in the incidence of aspiration incidents associated with obstetric general anaesthesia, particularly in the last 20 years (NHMRC, 1981, 1988, 2001; Department of Health, 1982, 1991, 2001). There have been vast improvements in anaesthetic and muscle relaxant drug combinations for obstetric anaesthesia (Skinner et al., 1998) along with improvements in anaesthetic techniques (Gibbs and Modell, 1994) and the increased use of epidural anaesthesia (NSW Health Department, 2001). The introduction of, and improvements to, monitoring systems for use during anaesthesia have played a significant part in the detection and prevention of anaesthetic incidents (Runciman et al., 1993). There has not been a report of maternal mortality attributed to aspiration in Australia since 1987 and only one death has been reported in the UK in the last decade (NHMRC, 1988, 2001; Department of Health, 2001) despite a reintroduction of oral intake (fluids and, in some centres, food) over the last 10–15 years (Broach and Newton, 1988; Michael et al., 1991; Baker, 1996; Parsons, 2001). The benefits of oral intake for labouring women The literature supporting oral intake for labouring women revolves around the physiological and psychological disadvantages of restricting food and fluid during labour (e.g. Simkin, 1986; Ludka and Roberts, 1993; Baker, 1996; Chern-Hughes, 1999). It also stresses the lack of evidence demonstrating the association between oral intake and aspiration (e.g. Pengelley and Gyte, 1998). The physical exertion of labour and birth is said to require a large caloric expenditure (ChernHughes, 1999). Most women are able to cope with the exertion of labour and birth without requiring food and sometimes fluids (Odent, 1994). For some however, especially those enduring long labours [longer than 24 hours for nulliparous and 18 hours for multiparous women (Friedman, 1981)], fasting may be detrimental for the mother, her baby and the progress of labour (Kristensen et al., 1991). The presence of ketosis during labour appears to be the key issue. Severe ketosis may progress to ketoacidosis (Ludka and Roberts, 1993) and has been found to be directly related to prolonged lengths of labour, increased need for induction and augmentation in nulliparous women, forceps deliveries and increased blood loss (Foulkes and Dumoulin, 1985). Because of the controversy found within the literature as to whether or not women should be allowed to eat, and sometimes drink, during their

M. Parsons

labour and the paucity of sound, reliable research to support either practice, midwives at four hospitals in Sydney were surveyed to ascertain their beliefs and practices for this issue.

Methods An exploratory survey was conducted between March 2000 and June 2000 with midwives from four public, teaching hospitals in Sydney, Australia regarding their management of labouring women’s oral intake. The four participating hospitals were selected from among 20 public hospitals in Sydney representing 23% of the total Sydney births in 2000. Each hospital served a diverse multi-cultural community (Table 1). Hospitals were selected because of their differing practices for the management of oral intake for labouring women with low-risk pregnancies. Information was collected regarding the oral intake for women with low-risk pregnancies in labour (the interpretation of ‘low-risk’ was left to the discretion of each midwife). Midwifery managers of each Maternity Unit reported that whilst there was an accepted practice among the midwives working in their birthing units, there was no written policy for the management of oral intake for women with low-risk pregnancies. Two hospitals had an agreed practice that restricted oral intake for women with low-risk pregnancies in labour to clear fluids once labour was established (‘Restrictive Hospitals’). The other two hospitals had an agreed practice that permitted the intake of food and free fluid (‘Unrestrictive Hospitals’). All four hospitals kept fresh food (e.g., sandwiches, jelly, ice cream) in their labour ward kitchenette, but in the Restrictive Hospitals this food was meant to be given to the women after they had birthed. The aim of this study was to investigate whether individual midwives followed their hospital’s agreed management regarding this issue and their reason for their practice decision.

Sample Responses were sought from only those midwives who specifically worked with labouring women ðN ¼ 194Þ: Following an information session regarding the study, questionnaires and a ‘questionnaire return’ box were left at every midwives’ station in each midwifery unit. Completion of the questionnaire was voluntary. A questionnaire return rate of 46% ðN ¼ 89Þ was received despite frequent reminders. Forty one per cent ðn ¼ 38Þ of the midwives

ARTICLE IN PRESS A midwifery practice dichotomy on oral intake in labour

Table 1

75

Demographics of the four hospitals. Hospital A

Hospital B

Restricted oral intake Current annual birth rate Models of care available Standard hospital model Team midwifery programme Birth centre

Hospital C

Hospital D

Unrestricted oral intake

2800

2100

2400

1900

Yes No Yes

Yes No No

Yes Yes Yes

Yes No No

Race/ethnic groupa (%) Caucasian Asian Polynesian Middle Eastern

87.4 5.8 3.5 3.5

66.2 19.5 9.1 5.2

82.2 12.2 2.2 3.3

63.6 10.4 5.2 20.8

Agea (%) o20 yrs 20–29 yrs 30–35 yrs 435 yrs

23.0 65.5 9.2 2.3

20.8 63.7 11.7 3.9

3.3 68.9 25.6 2.2

10.4 79.2 7.8 2.6

77.5 6.9

79.0 7.4

63.6 12.5

78.8 6.1

8.2 6.4

6.9 5.1

10.7 12.6

8.0 6.1

10.5 6.9

3.1 11.0

32.2 4.9

11.3 7.4

Birth outcomes 1999a (%) Spontaneous Forceps/ventouse Caesarean section Elective Emergency Epidural ratea (%) General anaesthetic ratea (%) a

Statistics provided by the NSW Department of Health, (2001).

from the ‘Restrictive Hospitals’ (RH) and 53% ðn ¼ 51Þ of the midwives from the ‘Unrestrictive Hospitals’ (UH) returned questionnaires. The questionnaire was developed specifically for the study and contained items generated from the literature (Al-Najjar, 1998) and in response to informal conversations with the midwives employed within the four participating hospitals. Prior to implementation the questionnaire was further critiqued and refined by two separate nurse researchers. The questionnaire contained four closed and six open questions to address the study aim and took no longer than five minutes to complete. The closed questions sought information regarding midwives’ level of experience and their individual practices in the management of labouring women’s oral intake. The open-ended questions sought information as to the midwives’ reasons for restricting or feeding labouring women. This information was analysed using simple content analysis (Beanland et al., 1999). The data were hand coded and themes

derived and these were placed into categories. Inter-rater reliability was provided by having the text coded by a second researcher, providing a 95% agreement. Ethics approval was given by the two Area Health Services involved in the study and the University of Western Sydney.

Findings Midwifery experience Years of midwifery experience The midwives practising within the four hospitals varied in experience level as measured in number of years (not full-time equivalent). Midwifery experience ranged from being a student midwife to having 37 years experience (Restrictive Hospitals: M ¼ 8:6 years, SD ¼ 8.0 years; Unrestrictive Hospitals: M ¼ 10:4 years, SD ¼ 6.42 years).

ARTICLE IN PRESS 76

M. Parsons

Midwives employed in the Unrestrictive Hospitals (UH) tended to have more years of midwifery experience than midwives employed in the Restrictive Hospitals (RH) although the difference was not statistically significant. Experience in various models of midwifery care It was believed that midwives who had experience in models of midwifery care outside the standard hospital model where midwifery care is segmented, were more likely to encourage labouring women to eat and drink throughout their labour. This assumption was based on a perceived increase in the level of autonomy experienced in other midwifery models (Homer et al., 2001) and the associated increase in confidence that comes with autonomy (Christmas, 1991; McLeod and Sherwin, 2000). There were four midwifery models identified by the surveyed midwives: (a) standard hospital model, (b) team midwifery programme, (c) birth

centre and (d) homebirth. Each model of care provides midwives with various levels of autonomy in the care of women with low-risk pregnancies (Box 1). All surveyed midwives had experience in the standard model of maternity care. However, opportunities to practice in other models of midwifery are not as easily available in the Australian state of New South Wales (NSW). There are only eight birth centres and seven team midwifery programmes operating within the 129 NSW hospitals according to the latest report (NSW Health Department, 2000), and only 34 midwives are listed as providing a homebirth service (Robinson, 2003). Despite one of the four hospitals in this study having a birth centre and team midwifery programme and all hospitals being based in Sydney, where the majority of birth centres ðn ¼ 5Þ and team midwifery programmes ðn ¼ 6Þ are located, only 39% ðn ¼ 35Þ of these midwives had experience

Box 1 Midwifery practice within various midwifery models in Australia Midwifery Model

Practice

Standard hospital model

The standard hospital model is operated through a hospital with an obstetrician supervising care. All women are seen by a doctor throughout their pregnancy care or, if attending the midwives’ clinic, they are seen by a midwife for the majority of their pregnancy care while a doctor sees them for a minimum of two visits. Midwives provide care during labour and the early postnatal period. Midwives have ‘little’ to ‘no’ autonomy in this type of midwifery model.

Team midwifery programme

The team midwifery programmes are operated through a hospital. A team of midwives, usually between four to eight per team, provide total care for women throughout their pregnancy, labour and postnatal period. The team may care for women with low-risk pregnancies or may have a shared-care arrangement with an obstetrician for women with high-risk pregnancies. An obstetrician is available for consultation and has input into the care provided through this midwifery model. Due to the continuity of care provided in this midwifery model and the lack of medical presence, midwives tend to acquire a level of autonomy.

Birth centre

Birth centres are set within a hospital and have a similar arrangement to the team midwifery programmes with an obstetrician having input into the policies and procedures of the centre and are available for medical management where required. However, these midwives only provide care for women with low-risk pregnancies during the antepartum and intrapartum period. Some may also provide postpartum care for the first 24 hours after birth. Due to the environment created and the separateness of these settings from the routine labour/delivery unit, these midwives also acquire a level of autonomy.

Homebirth

Midwives who participate in homebirths do so independently and outside the confines of the hospital system. These midwives provide total care for women with low-risk pregnancies throughout their pregnancy, labour and an extended postnatal period. This group of midwives attain the highest level of autonomy of all four midwifery models. (Halliday et al. 1999).

ARTICLE IN PRESS A midwifery practice dichotomy on oral intake in labour

in other midwifery models (Table 2). Midwives from the ‘UH’ tended to have more midwifery experience by way of other models of midwifery care.

specific about the type and amount of food they would permit: for example: a bite of chocolate or hot chips if requested. (R.13, RH) A small amount of bread if the woman insisted. (R.22, RH) may suck barley sugar [lollies]. (R.25, RH)

Midwifery practice for the oral intake of labouring women Each respondent gave information about their own practice for oral intake of food and fluid for labouring women regardless of the accepted practices in their hospital. Practices varied from ‘nil by mouth’ to ‘whatever the woman desired’. Restrictive Hospitals As labour progressed midwives from the ‘RH’ became more restrictive about food intake and encouraged fluid intake only (Table 3). Although some midwives did allow food, some were quite

Table 2 models.

Midwives’

experience

in

midwifery

Midwifery Models

Restrictive Hospitals (N ¼ 38)

Unrestrictive Hospitals (N ¼ 51)

Standard hospital model Birth centre Team midwifery Homebirth

38

51

2 0 2

13 14 4

77

Unrestrictive Hospitals All midwives from the ‘UH’ allowed food during the early phase of the first stage of labour. As with the midwives from the ‘RH’, the number of midwives allowing food as the first stage of labour progressed declined (Table 3).

The relationship between midwifery experience and practice Information regarding midwives’ experience level was obtained in order to establish if their level of experience was associated with changes in practice. Some comments made by respondents indicated that midwives with less experience did not have the confidence, or were given no choice other than, to restrict oral intake for labouring women. For example, a midwife with one year of postgraduate experience stated: [I fast labouring women] only once in established labour, [then] oral fluids only as it is hospital policy; not what I’ve been taught but I do not have seniority to challenge it. (R.29, RH)

Table 3 Midwifery practice for oral intake of women with low-risk pregnancies during the first stage of labour.

Restrictive Hospitals Food (e.g., eat and drink as desired; light snacks; sandwiches) Free fluids (no food) (e.g., milk drinks, clear fluids, tea & coffee with milk) Clear fluids (e.g., water, cordial, black tea or coffee, ice) Nil by mouth Unanswered Unrestrictive Hospitals Food (e.g., eat and drink as desired; light snacks; sandwiches) Free fluids (no food) (e.g., milk drinks, clear fluids, tea & coffee with milk) Clear fluids only (e.g., water, cordial, black tea or coffee, ice) Nil by mouth Unanswered

Early phase n ¼ 38 (%)

Active phase n ¼ 38 (%)

Transition n ¼ 38 (%)

76

37

16

3

16

16

8 0 13

29 3 16

47 8 13

96

88

61

0

4

29

0

0

0

0 4

0 8

2 8

ARTICLE IN PRESS 78

M. Parsons

Table 4 Comparison of experience (in years), midwifery model exposure and oral intake practice for labouring women among the midwives in the Restrictive Hospitals (RH) and Unrestrictive Hospitals (UH). Years of experience

Less than 5 years Five or more years

Standard hospital model only

Exposure to other midwifery models

RH ðn ¼ 31Þ

UH ðn ¼ 23Þ

RH ðn ¼ 2Þ

Feed

Fast

Feed

Fast

Feed

Fast

Feed

Fast

3 6

8 14

0 13

2 8

0 2

0 0

12 13

0 3

Another comment made by a midwife with two years post-graduate experience was that: [If I fed these women] I would be chastised for doing it by the anaesthetic staff. (R.62, UH)

Although this type of comment was rare, this information suggests that midwives’ management of the labouring woman’s oral intake may also be influenced by their experience and confidence level; less experienced midwives may be more likely to follow the instructions of the more senior staff regardless of their own beliefs. By comparing the midwives’ years of experience, exposure to various midwifery models and their individual practice for the management of oral intake for labouring women and consequently their capacity to breach accepted practice, it can be seen that midwives’ decision-making in practice has more to do with confidence. The comparisons will be demonstrated separately for the ‘Restrictive Hospital’ and the ‘Unrestrictive Hospital’ midwives. Experience versus practice among the ‘Restrictive Hospital’ midwives Thirty-four of the 38 midwives from the ‘RH’ provided information regarding their midwifery practices and experiences (all were qualified midwives) – see Table 4. The first two factors (midwives’ years of experience and exposure of various midwifery models) may assist in explaining these midwives’ confidence and willingness to challenge or disregard hospital practice. The majority of midwives from these hospitals had only practised in the standard hospital model of maternity care, a model that provided the least amount of autonomous practice for midwives. Experience versus practice among the ‘Unrestrictive Hospital’ midwives All 51 midwives from the ‘UH’ provided information regarding their midwifery practices and experiences (Table 4).

UH ðn ¼ 28Þ

Information from these midwives suggests that greater exposure to models of midwifery that enable greater autonomy may have the most influence on midwives’ decision to allow oral intake for labouring women; however, a larger sample of midwives would be required to substantiate this inference.

Reasons for restricting or feeding Four categories evolved from the open-ended responses regarding reasons for either restricting or allowing oral intake of food and fluids during labour. These categories were labelled: ‘Fast, just in case’, ‘Need fluids not food’, ‘Need food and fluids’ and ‘There is no research’. Not all midwives responded to the open-ended question, while some of the respondents provided more than one answer. Fast, just in case Three midwives from the ‘RH’ (8%) stated that they fasted women with low-risk pregnancies at some phase of the first stage of labour. Midwives fasted labouring women because they might vomit, might require a caesarean section or because it was a hospital policy stipulation. Examples were: yin case of an emergency caesarean section and also [because] women tend to vomit if they eat during labour. (R.12, RH)

Need fluids not food Twenty-three (61%) respondents from the ‘RH’ compared with two (4%) from the ‘UH’ stated that labouring women need fluids during labour, not food. Another 17 (19%) midwives (RH ¼ 12; UH ¼ 5) commented that labouring women do not want to eat, for example: Usually no need to fast [labouring women] because they lean towards fluids anyway. They don’t want to eat. (R.26, RH)

Need food and fluids Eight (21%) midwives from the ‘RH’ compared with 37 (73%) from the ‘UH’ responded that they

ARTICLE IN PRESS A midwifery practice dichotomy on oral intake in labour

believed labouring women needed food and fluids during labour. The most commonly cited reason given for providing food and fluids for labouring women was their physiological need for nourishment and hydration, for example: they need fluid for hydration and glucose for uterine activity and effective contractions. (R.52, UH)

Avoidance of ketosis during labour was another reason given. For example, one respondent commented: sometimes it is early labour and they are already showing ketones [on urinalysis testing] because they don’t realize they can eat. (R.81, UH)

The psychological benefit of eating and drinking was another belief expressed by this group of midwives. One commented: They need to feel they have some control and may feel nauseated or dehydrated if not allowed to do what their body is telling them to do. (R.66, UH)

There is no research The lack of research to support the restriction of oral intake for labouring women was identified by seven (14%) midwives, all from the ‘UH’. One midwife stated that: There is no research claiming women should not eat in labour, high or low risk. (R.42, UH)

Another midwife wrote about the large amount of literature that supports eating and drinking for labouring women and the improvement to labour progress when these women eat.

Discussion These findings were limited by the poor response rate. It cannot be assumed that the respondents represented the ‘typical’ views held by all the midwives in the group. The findings of this research are, therefore, not generalisable (Polit and Hungler, 1991). Postal surveys are known to have a very low response rate possibly due to the impersonal nature of the approach (Polit and Hungler, 1991). It is possible that this survey’s response rate may have been improved if each midwife had been approached personally. Despite this setback, this study demonstrated that beliefs and practices regarding the management of oral intake in labour for women with low-risk pregnancies vary from hospital to hospital (as demonstrated by the differing accepted practices among the surveyed

79

hospitals) and midwife to midwife regardless of whether policy guidelines exist. The findings of this survey demonstrated that women may be allowed to eat food during labour depending on their stage of labour, the hospital in which they were giving birth and the midwife to which they were assigned for care. There was no scientific basis to these decisions. Whilst many midwives followed their hospital’s accepted practice for oral intake not all did so, particularly when the practice restricted the women’s oral intake. That is, when employed in a hospital where the policy restricted oral intake to clear fluids, once labour was established, twothirds of these midwives restricted oral intake to fluids, although not necessary clear fluids. When employed in a hospital which allowed food and fluids throughout labour for women with low-risk pregnancies, approximately two-thirds of these midwives allowed food throughout the labour process. From the individual woman’s perspective it could be considered ‘the luck of the draw’ as to whether or not food or fluids were provided during labour. The wishes of a woman with definite ideas about oral intake or with a physical desire to eat would depend on the hospital in which she birthed her baby, and the experience, views and practice of the midwife to whom she was allocated for care. A woman’s choice regarding oral intake does not seem to exist in hospitals that restrict this form of labour management. It could be presumed that in situations where the woman (and her wishes) was previously known to the midwife, practices may also differ. In this study, it appears that a midwife’s decision to restrict or allow oral intake during labour may have been influenced by three interacting factors: (1) midwifery experience, (2) the hospital’s accepted practice and (3) the arguments used by individual midwives in their decision. The differing views on the management of oral intake for labouring women seemed to depend more on the midwives’ experiences in a variety of midwifery models that were perceived to enable a degree of autonomy, than the number of years practising as a midwife. In spite of similar years of midwifery experience midwives who had experience in differing models of midwifery care were more likely to feed labouring women, compared to those who had no experience outside the standard hospital model. The accepted practice or ‘culture’ among the midwives in each hospital also appeared to play a crucial role in midwives’ ability to critically assess the needs of the labouring woman. There is midwifery literature that supports the practice of

ARTICLE IN PRESS 80

offering food and fluids for labouring women (e.g. Kristensen et al., 1991; Pengelley and Gyte, 1998; Sleutel and Golden, 1999). However, the anaesthetic literature for the last 50 years has opposed, and continues to oppose, this view (Kallar and Everett, 1993; Gibbs and Modell 1994; Scrutton et al., 1999). Neither side of the debate has substantial, current, reliable research that reports the effect on the woman, her baby, the labour progress or the incidence of aspiration. This debate and lack of substantial evidence on which to base practice is the reason for midwives’ practice dichotomy regarding oral intake for labouring women. It has been suggested that some nurses are taskorientated and have a need to adhere to the authority of hospital policies while others are inclined to be ‘thinkers’ who combine their experiences with information from current literature to individualise and improve care (Greenwood, 2000); this could also be said of some midwives. Whether this is the case in this situation is unknown, however, only midwives employed in the ‘RH’ stated that the hospital policy (although no written policy existed) was their reason for restricting oral intake for labouring women. Other midwives may have restricted oral intake because of insufficient contact with the literature and a lack of awareness of any debate regarding the issue. On the contrary, it is possible that these midwives were aware of the literature and the inconclusive evidence supporting the oral intake debate. The most current and valid evidence available for the management of oral intake is research showing that clear fluids are safe up to two hours before a general anaesthetic (Splinter and Schaefer, 1990) and that food consumption as opposed to water consumption by labouring women has no effect on labour and birth outcomes (Scrutton et al. 1999). There is no research, however, that reports the risks or benefits of fasting labouring women. There are a number of premises on which midwives in this survey may have based their reasons for allowing labouring women to eat. Maternal mortality associated with aspiration during general anaesthesia is a rare event and has not occurred in Australian obstetrics for over a decade. This can be attributed, in part, to the vast improvements in anaesthetic drugs and techniques (Kallar and Everett, 1993; Oberoi and Phillips, 2000) and the increased use of epidural anaesthesia (Glosten, 2000). There is a lack of evidence to support the view that fasting during labour reduces the incidence of aspiration (Michael et al., 1991; O’Sullivan, 1994; Jacob, 1998). There has been a large volume of literature published over the last

M. Parsons

15 years suggesting that labouring women have better outcomes if allowed to eat and drink as they wish (Foulkes and Dumoulin, 1985; Jacob, 1998; Pengelly and Gyte, 1998; Chern-Hughes, 1999; Sleutal and Golden, 1999). There is, however, a paucity of reliable evidence, such as that obtained from a large, randomised controlled trial, to guide midwives and women in their decision-making.

Conclusion The accepted practice for the management of oral intake in labour for women with low-risk pregnancies varies from hospital to hospital (Parsons, 2001). In the main, midwives in this study tended to follow the accepted practice of the hospital in which they were employed. However, the findings from this survey demonstrated that many midwives were questioning the 50-year-old tradition of fasting or restricting oral intake for women in their care, and some were practising contrary to hospital accepted practice which restricts oral intake. Whether or not allowing food and fluid throughout labour is beneficial or harmful can only be determined by further research. A large, quasiexperimental trial is in progress in Sydney, Australia to compare the birth outcomes of labouring women who choose to eat or not eat during their labour. The results of this trial will increase the body of knowledge pertaining to this issue and in the future enable a more evidence-based approach to midwifery practice decisions and offer women informed choice in labour.

References Al-Najjar, J., 1998. Policy? what policy? hospital guidelines on food and drink in labour. The Practising Midwife 1 (4), 10–11. Baker, C., 1996. Nutrition and hydration in labour. British Journal of Midwifery 4 (11), 568–572. Beanland, C., Schneider, Z., LoBiondo-Wood, G., et al., 1999. Nursing Research: Methods, Critical Appraisal and Utilisation, 1st Edition. Mosby, Sydney. Broach, J., Newton, N., 1988. Food and beverages in labour. Part I: cross-cultural and historical practices. Birth 15, 81–85. Chern-Hughes, B., 1999. Oral intake in labor: trends in midwifery practice. Journal of Nurse-Midwifery 44 (2), 135–138. Christmas, J., 1991. Autonomy and personal history. Canadian Journal of Philosophy 21 (1), 1–14. Cote, C.J., Goudsouzian, N.G., Liu, L.M.P., et al., 1982. Assessment of risk factors related to the acid aspiration syndrome in pediatric patients F gastric ph and residual volume. Anesthesiology 56, 70–72. Department of Health, 1982. Report on confidential enquiries into maternal deaths in England and Wales 1976–1978. HMSO, London.

ARTICLE IN PRESS A midwifery practice dichotomy on oral intake in labour

Department of Health, 1991. Report on confidential enquiries into maternal deaths in the United Kingdom 1985–1987. HMSO, London. Department of Health, 1993. Changing childbirth: report of the expert maternity group. HMSO, London. Department of Health, 2001. Report on confidential enquiries into maternal deaths in the United Kingdom 1997–1999. TSO, London. Foulkes, J., Dumoulin, J.G., 1985. The effects of ketonuria in labour. British Journal of Clinical Practice 39 (2), 59–62. Friedman, E.A., 1981. The labour curve. Clinics in Perinatology 8 (1), 15–25. Gibbs, C.P., Modell, J.H., 1994. Pulmonary aspiration of gastric contents: pathophysiology, prevention, and management. In: Miller, R.D. (Ed.), Anesthesia, Vol. 2, 4th Edition. Churchill Livingstone, New York. Glosten, B., 2000. Anesthesia for obstetrics. In: Miller, R.D. (Ed.), Anesthesia, Vol. 2, 5th Edition. Churchill Livingstone, Philadelphia. Greenwood, J., 2000. On donkeys, ingratiates, thinkers andyclinical development units. Collegian 7 (2), 34–39. Halliday, J., Ellis, I., Stone, C., 1999. Who usually delivers whom and where: report on models of antenatal care. Perinatal Data Collection Unit, Victoria. Homer, C., Brodie, P., Leap, N., 2001. Establishing models of continuity of midwifery care in Australia: a resource for midwives and managers. Faculty of Nursing, Midwifery & Health, University of Technology Sydney, Sydney. Jacob, S., 1998. Food deprivation in labour. RCM Midwifery Journal (April mid-month Suppl), 6. Kallar, S.K., Everett, L.L., 1993. Potential risks and preventative measures for pulmonary aspiration: new concepts in preoperative fasting guidelines. Anesthesia and Analgesia 77, 171–182. Kristensen, M.S., Gellett, S., Bach, A.B., et al., 1991. Hemodynamics and arterial oxygen saturation during preoperative emptying of the stomach. Acta Anaesthesiologica Scandinavica 35, 342–344. Kubli, M., Scrutton, M.J., Seed, P.T., et al., 2002. An evaluation of isotonic ‘sport drinks’ during labor. Anesthesia and Analgesia 94 (2), 404–408. Lewis, P., 1992. Feeding or fasting in labour? New Generation 11 (2), 14–15. Ludka, L., Roberts, C.C., 1993. Eating and drinking in labour: a literature review. Journal of Nurse-Midwifery 38 (4), 199–207. McLeod, C., Sherwin, S., 2000. Relational autonomy, self-trust, and health care for patients who are oppressed. In: Mackenzie, C., Stoliar, N. (Eds.), Relational Autonomy: Feminist Perspectives on Autonomy, Agency and the Social Self. Oxford University Press, New York. Michael, S., Reilly, C.S., Caunt, J.A., 1991. Policies for oral intake during labour: a survey of maternity units in england and wales. Anesthesiology 46, 1071–1073. NHMRC, 1981. Report on Maternal Deaths in Australia 1976–1978. Australian Government Publishing Service, Canberra. NHMRC, 1988. Report on maternal deaths in Australia 1982–1984. Australian Government Publishing Service, Canberra.

81

NHMRC, 2001. Report on maternal deaths in Australia 1994– 1996. Australian Government Publishing Service, Canberra. NSW Health Department, 2000. NSW Framework of Maternity Services. Better Health Centre, Gladesville. NSW Health Department, 2001. New South Wales Mothers and Babies 1997: NSW Public Health Bulletin (2). NSW: NSW Health Department, Sydney. Oberoi, G., Phillips, G., 2000. Anaesthesia and Emergency Situations: A Management Guide. McGraw-Hill, Sydney. Odent, M., 1994. Labouring women are not marathon runners. Midwifery Today 31, 23–24, 43, 51. O’Sullivan, G., 1994. The stomach F fact or fantasy: eating and drinking during labour. International Anesthesiology Clinics 32, 31–44. Parsons, M., 2001. Policy or tradition: oral intake in labour. Australian Journal of Midwifery 14 (3), 6–12. Pengelley, L., Gyte, G., 1998. Eating and drinking in labour (1). A summary of medical research to facilitate informed choice about the care of mother and baby. Practising Midwife 1 (7/8), 34–37. Polit, D.F., Hungler, B.P., 1991. Nursing Research: Principles and Methods, 4th Edition. Lippincott, Philadelphia. Robinson, J., 2003. Australian Society of Independent Midwives Members Directory. Australian Society of Independent Midwives, Sydney. Rodwell, R.R., 1992. Giving birth on an egg. London Times, December 17. Runciman, W.B., Sellen, A., Webb, R.K., et al., 1993. Errors, incidents and accidents in anaesthesia practice. Anaesthesia Intensive Care 21 (5), 506–519. Scheepers, H.C.J., Thans, M.C.J., de Jong, P.A., et al., 2002. A double-blind, randomised placebo controlled study on the influence of carbohydrate solution intake during labour. British Journal of Obstetrics and Gynaecology 109 (2), 178–181. Scrutton, M.J.L., Metcalfe, G.A., Lowy, C., et al., 1999. Eating in labour: a randomised controlled trial assessing the risks and benefits. Anaesthesia 54, 329–334. Simkin, P., 1986. Stress, pain, and catecholamines in labour: Part 1. A Review. Birth 13 (4), 227–239. Skinner, H.J., Biswas, A., Mahajan, R.P., 1998. Evaluation of intubating conditions with rocuronium and either propofol or etomidate for rapid sequence induction. Anaesthesiology 53, 702–710. Sleutel, M., Golden, S.S., 1999. Fasting in labour: relic or requirement. Journal of Obstetrics, Gynaecology and Neonatal Nursing 28 (5), 507–512. Splinter, W.M., Schaefer, J.D., 1990. Unlimited clear fluid ingestion two hours before surgery in children does not affect volume or pH of stomach contents. Anaesthesia and Intensive Care 18 (4), 522–526. Tourangeau, A., Carter, N., Tansil, N., et al., 1999. Intravenous therapy for women in labour: implementation of a practice chance. Birth 26, 31–36. Yiannouzis, K., Parnell, C., 1994. Randomized controlled trial measuring the effects on labour of offering a light, low fat diet. In: Simms, C., McHaffie, H., Renfrew, M.J., et al. (Eds.), The Midwifery Research Database MIRIAD: A Sourcebook of Information About Research in Midwifery. Haigh & Hochland, Cheshire, England.