Objective: to explore the maternal and neonatal effects of fentanyl administered subcutaneously to women during labour. Design: two methods were used: (1) A retrospective audit of the birth register and maternal and neonatal records for the period from January 2000 to December 2007. (2) A pilot study was also conducted on a convenience sample of women between July 2008 and October 2008. Setting: this study was conducted within a maternity unit at a rural South Australian hospital where approximately 350 babies are birthed each year. Participants: audit participants included women who had uncomplicated pregnancies and birthed at term (37-42 weeks gestation). Women in the experimental group consisted of those who had utilised only subcutaneous fentanyl for pain relief (n=75), or nitrous oxide and oxygen prior to being administered subcutaneous fentanyl (n=196). Stratified random selection based on parity and age was used to determine the control group, which consisted of women who used no pharmacological pain relief (n=196).The pilot study involved a convenience sample of women (n=10) assessed to have an uncomplicated pregnancy and labour occurring at term (≥37 weeks gestation). Measurements: audit variables examined included the women's age, parity, labour duration, mode of birth (spontaneous or assisted), analgesia used, total dosage, time administered prior to birth, time of birth, neonatal Apgar scores, time to establish breathing, naloxone use, days spent in hospital post-birth and breast-feeding outcomes upon discharge.The pilot study explored maternal effects assessed pre- and 30 minutes post-administration of subcutaneously administered fentanyl by observing pain scores, vital signs, sedation levels, nausea/vomiting scores and anti-emetic use. To assess possible adverse effects in the neonate Apgar scores, time to establish respiration, naloxone use, transfer to neonatal nursery and breast-feeding outcomes upon discharge were recorded. Findings: women in the experimental groups were more likely to be induced, experienced a longer duration of labour and had an increased likelihood of an assisted vaginal birth. The average total dose of fentanyl administered was 250. μg. Neonatal outcomes were comparable between groups when examining Apgar scores <7 at 1 and 5 minutes and time to establish breathing. There was, however, a significant difference with naloxone administration between the groups. There was no significant difference between groups in hospital stay or breast-feeding on discharge.The pilot study identified a clinically significant reduction in pain scores for 78% of women following the administration of subcutaneous fentanyl, with the average pain score decreasing from 8.4 (±1.4) to 7.2 (±1.1) (paired t-test, p=0.017). Vital signs were not affected, no anti-emetics were required and all women remained alert with no sedation noted. Key conclusions: the audit identified fentanyl use was associated with a longer length of labour, but this may be explained by more women in the experimental groups requiring induction of labour than those in the control group. However, length of hospital stay, breast-feeding rates and neonatal outcomes were comparable amongst the three groups.Results of the pilot study are consistent with those of the audit in relation to the effects on mother and neonate. In addition, the pilot study begins to provide evidence that fentanyl is efficacious in providing pain relief. Implications for practice: results of this study are the first to explore the effects of fentanyl administered subcutaneously to women during labour. This method of analgesia offers women an additional choice of pain relief during childbirth and may be particularly beneficial in remote and rural settings where resources are often limited and access to specialist services difficult. Further research, however, is required to be able to generalise the outcomes and provide further data to support the clinical effectiveness of this route of administration of fentanyl.