A comparative study of the effect of food consumption on labour and birth outcomes in Australia

Abstract

Objective

to explore the effect of volitional food consumption by women during labour on labour and birth outcomes.

Design

a comparative design using concurrent controls.

Setting

four public hospitals in Sydney, Australia.

Participants

217 English-speaking, nulliparous women with low-risk pregnancies. The sample was divided into four sub-groups identified post hoc from reported behaviour: (1) 82 women who chose to eat food during early labour only; (2) 10 who ate during established labour only; (3) 31 who ate during early and established labour and (4) 94 who chose to consume clear fluids only during early and established labour.

Interventions

voluntarily eating food during labour compared with voluntarily consuming clear fluids only.

Measurements

differences between the four eating groups were examined for labour progress using one-way analysis of variance (ANOVA). A hierarchical multiple regression tested the association between eating during labour and labour duration. The relationship between food intake and the incidence of medical interventions was tested using χ² tests.

Findings

eating during the early phase of the first stage of labour was associated with M=2.16 hrs longer labour ($p<0.01$). When women ate food during both their early and established phases of labour, M=3.5 hrs was added to their labour ($p<0.01$). The incidence of vomiting, medical interventions during labour or adverse birth outcomes were unaffected by food intake.

Conclusion

the findings suggest that women should be informed that labour may take longer when they eat food. However, eating does not seem to affect other labour or birth outcomes.

Implication for practice

the findings challenge the belief among many midwives that food intake is beneficial to labour progress. However, women should not be denied food for fear of vomiting or because it may make labour longer. Women with low-risk labours should be informed of the risk, although rare, of aspiration if general anaesthesia is required, and be allowed to respond to their natural desires for oral intake during labour.

Introduction

Oral intake allowance for women during childbirth is governed by individual professional and institutional policies. Unsurprisingly, oral intake policy and practice vary widely (Broach and Newton, 1988; Michael et al., 1991; Baker, 1996; Parsons, 2001). The issue of whether a woman in labour should eat or drink remains controversial. A literature search on the food intake of women in labour comprised obstetric, anaesthetic and midwifery texts, along with a number of systematic trawls of the electronic databases CINAHL 1982–2005, Medline 1966–2005 and the entire Cochrane Library. The search was conducted using the keywords ‘oral intake’, ‘food’, ‘labour’, ‘obstetric anaesthesia’ and ‘aspiration’, and was limited to English-language studies involving humans. Randomised clinical studies have been conducted to compare mothers who consumed food during labour with those who did not, but their methodological differences led to conflicting results (Yiannouzis and Parnell, 1994; Scrutton et al., 1999; Tranmer, 1999). A comparative study conducted by Parsons et al. (2006) found that mothers who ate food only during early labour tended to experience a slightly longer labour and no increase in the incidence of medical intervention. Historical and contemporary opinion provide a less than adequate basis for much of the literature on this topic.

Research has shown that maternal age (Adashek et al., 1993), ethnicity (Albers et al., 1996) and fetal position affect the length of labour (Fitzpatrick et al., 2001). However, empirical information is lacking on the reasons why some women in labour want to eat or the effect food consumption may have on the outcomes of labour.

Anaesthetists are concerned that when a woman in labour has recently eaten, a general anaesthetic will increase her risk of gastric aspiration (Gwinnutt, 1996; American Society of Anesthesiologists, 1999). Although the incidence of gastric aspiration during obstetric general anaesthesia has declined significantly (Department of Health, 1969–2001; NHMRC, 1972–2001), incident reviews show that aspiration is almost always attributable to anaesthetic error or mishap (Hawthorne et al., 1996; Sinclair et al., 1999) not oral intake. The risk of gastric aspiration may offer only a weak justification for restricting oral intake.

Midwives commonly believe that withholding food or fluids may be detrimental to the mother, her fetus and labour progress, and so would permit or encourage food intake during labour (Horner, 1989; Lewis, 1992; Pengelley and Gyte, 1998). Published data over the past 15 years have supported oral intake for women in labour (Broach and Newton, 1988; Chern-Hughes, 1999; Ludka and Roberts, 1993; Champion and McCormick, 2002).

Two arguments have led to opposing hypotheses concerning the effect of food intake on labour progress. The first claims that energy provided by eating during labour will better sustain contractions, leading to a shorter labour than fasting would enable (Pengelley, 2002). An opposing argument claims that digestion may divert resources from the uterus, impairing its performance and increasing labour duration (Enkin et al., 2000). Both propositions can be tested simultaneously by a non-directional hypothesis involving the comparison of labour duration for mothers who have eaten during labour with those who did not eat.

In this paper, we report on a prospective, comparative study into the relationship between food consumption during labour, and labour duration and birth outcomes. Subject to methodological limitations, the aim was to resolve whether volitional eating during the early phase, established phase or both phases of the first stage of labour expedites or extends labour, and whether it increases or decreases the incidence of medical interventions used during labour. A naturalistic, passive approach was used rather than the randomised-controlled design used in previous studies (Yiannouzis and Parnell, 1994; Scrutton et al., 1999; Tranmer, 1999) to investigate voluntary rather than prescribed eating, the latter imposing an ethically questionable consumption regimen.

The outcome variables for this study are (1) the length of natural labour (defined below); (2) the incidence of vomiting, medical interventions (medical augmentation, artificial rupture of membranes, intravenous therapy for hydration, pethidine injection, epidural anaesthesia for labour and forceps or ventouse delivery); and (3) birth outcomes (admission of the newborn baby to a special care nursery facility, Apgar score at five mins and estimated maternal blood loss at birth).

The following criteria have been adapted to define the phases and stages of labour (Cassidy, 1999): (1) early phase of the first stage of labour (0–3 cm cervical dilatation); (2) established phase of the first stage of labour (3–10 cm cervical dilatation); (3) hospital-estimated labour included the established phase of the first stage of labour plus second stage labour (from 3 cm cervical dilatation to birth). The length of this labour period served as the labour duration outcome variable for the study.

We also tested for interactions between fetal position and eating effects on labour. Because there is little a priori or empirical basis for predictions, this aspect of the research is exploratory, with no formal hypotheses proposed.