Abstract 1 2 Objective: This Australian feasibility study aimed to determine; the willingness of women 3 experiencing a post-date pregnancy to participate in a randomised controlled trial (RCT) of 4 acupressure and compliance with the study protocol. The study also aimed to determine the effect 5 size of the primary outcome in order to calculate a sample size for a future appropriately powered 6 RCT. 7 Design: a two-arm randomised controlled trial. Staff providing clinical care were blinded to group 8 allocation unless the participant disclosed study participation. 9 Setting: maternity services at two outer metropolitan public hospitals in New South Wales, Australia 10 Participants: Sixty seven healthy primigravid women experiencing a singleton cephalic pregnancy 11 at 40 weeks +/- 2 days gestation were assessed as eligible to participate and were provided with 12 study information. 13 Intervention: Both groups received standard clinical care, with the intervention group also receiving 14 verbal and written instructions on the self-administration of three acupoints (Spleen 6, Large 15 Intestine 4, and Gall Bladder 21) to be used until spontaneous or induced labour began. 16 Measurements: Assessment of feasibility included determining recruitment rate, acceptability of an 17 RCT for Acupressure, a CAM modality, and treatment compliance via participant surveys. The 18 primary clinical outcome was spontaneous onset of labour. 19 Findings: From the 67 women eligible during the timeframe for the study, 44 women (65.6%) 20 agreed to participate and were randomised. There was no statistically significant difference in rate of 21 spontaneous onset of labour (50% acupressure vs 41% control). Twenty nine participant surveys 22 were returned (65.9%). In the intervention group there was a high compliance with the acupressure 23 protocol (83%) and the use of the three acupoints (94%). 24 Conclusions and implications for practice: This feasibility study revealed that pregnant women 25 are interested in the use of CAM, and acupressure in particular, for the initiation of labour. Most 26 women found it acceptable to be randomised to receive the intervention. While the 9% difference in 1 27 the primary outcome was not statistically significant it is the best estimate of the treatment effect for 28 calculating a sample size of 994 for a future RCT with 80% power, alpha 0.05. 29 30 Trial Registration: Australia and New Zealand Clinical Trials Register (ANZCTR): 31 ANZCTR:12613000145707. 32 Keywords: Acupressure; postdates pregnancy, labour onset, uterine contractions; pilot RCT 2 Introduction 33 34 In an area such as Complementary and Alternative Medicine (CAM) which is endeavouring to 35 establish scientific credibility, randomised controlled trials are an important methodology however 36 few consider the relevance of establishing whether an intervention study can actually be conducted 37 successfully (Hawk et al. 2002; Do et al. 2011). Feasibility studies are conducted before full 38 powered studies in order to answer the question “Can this study be done?” and to estimate 39 important parameters for designing robust randomised controlled trials when there are few 40 previously published studies or limited existing data using a specific intervention technique (Fonteyn 41 & Bauer-Wu 2005; Bowen et al. 2009; Arain et al. 2010). 42 Increasing rates of pharmaceutically induced labour and operative birth have been reported in the 43 UK, US, Canada and Australia since the early 1990s (MacDorman et al. 2008; Mealing et al. 2009; 44 Ehrenthal et al. 2010; Leduc et al. 2013; Stock et al. 2013). Recent Australian government policy 45 recommended that the number of women who embark on a labour and/or go into labour 46 spontaneously needs to increase and the number of labour interventions needs to decrease (NSW 47 Ministry of Health 2010). The discovery of a non-pharmacological, non-invasive technique to 48 stimulate uterine contractions that is simple, safe, effective and without serious side effects may 49 prove beneficial for both mother and baby, especially in areas where pharmacological pain relief 50 may not be available. Acupressure is one technique that has been trialled for this purpose. 51 Acupuncture and acupressure/shiatsu techniques use the same meridian points on the body, 52 however acupuncture uses needle stimulation on the meridian points whereas acupressure/shiatsu 53 uses firm thumb or finger pressure, which is less invasive and does not require an acupuncturist 54 (Kolster & Waskowiak 2007). Shiatsu also includes gentle exercises and massage in the treatment 55 session (Tiran & Mack 2000). These three techniques are based on the traditional Chinese 56 medicine (TCM) philosophy that meridians or pathways flow through the body, enhancing blood 57 flow, nourishing tissue, and facilitating normal functions of the body (Chung et al. 2003; Betts 2006). 58 A systematic review of randomised controlled trials (RCT) of labour induction and acupuncture (not 3 59 acupressure or shiatsu) found that fewer women receiving acupuncture required cervical ripening 60 prior to labour induction compared to women receiving standard care (Smith et al. 2011). 61 A search of the literature using the search terms, ‘acupressure’, ‘induction of labour’, ‘labour onset’, 62 and ‘post-date pregnancy’, was able to locate six international RCTs that have examined the effect 63 of acupressure on already established labour, uterine contractions, labour duration and pain level 64 (Chung et al. 2003; Lee et al. 2004; Heidari et al. 2008; Hjelmstedt et al. 2010; Kashanian & Shahali 65 2010; El Hamid et al. 2013). All six studies found a shortened first stage of labour and a decrease in 66 labour pain score compared to placebo or standard care (Mollart et al. 2015). One ‘pilot audit’ that 67 examined the effects of shiatsu on initiation of uterine contractions for women experiencing post- 68 date pregnancy was also located (Ingram et al. 2005). This UK study recruited 132 multiparous and 69 primiparous women at 40 weeks gestation, 66 of whom were instructed to use three acupoints 70 (Spleen 6, Large Intestine 4 and Gall Bladder 21) as often as they wanted, together with breathing 71 exercises until the onset of labour (Ingram et al. 2005). The remaining 76 women received standard 72 care. Of those women who used all three acupoints (n= 41/66), 100% laboured spontaneously 73 compared to women who used less than three acupoints and subsequently required 74 surgical/medical induction of labour (Ingram et al. 2005). Overall, women who used acupoints 75 (n=66) compared to women who did not (n=76), were significantly more likely to labour 76 spontaneously (χ24.28, p=0.038). There was no difference in adverse outcomes for mothers or 77 babies (Ingram et al. 2005). Major limitations of this study are the use of the non-randomised 78 design, analysis not stratified by parity and the use of 40 weeks as the point of recruitment, which is 79 not post-term. There is a need for a robust RCT with a standard protocol describing duration and 80 frequency of use of the three acupoints in women who are more than 40 weeks gestation. 81 82 Additionally, there have been few published studies investigating the pregnant woman’s experience 83 of complementary and alternative medicines and/or therapies (CAM) using the RCT design (Mollart 84 2003; Mitchell & Allen 2008; Do et al. 2011). Williams and Mitchell (2007) acknowledged the 85 importance of involving key stakeholders such as women accessing services and health 4 86 professionals, to explore the acceptability and implications of providing CAM in mainstream care 87 (Williams & Mitchell 2007). Our study builds on the current limited evidence by providing a rigorous 88 study protocol focusing on primigravidae, a participant survey and health professional focus groups. 89 This paper reports on the findings of the feasibility RCT, treatment fidelity and on women’s 90 willingness to be randomised to receive acupressure. 91 Aims 92 This Australian feasibility study aimed to determine: the willingness of women experiencing a post- 93 date pregnancy to participate in a randomised controlled trial of acupressure; compliance with the 94 study protocol and to estimate a sample size for a future appropriately powered study. Health 95 professional’s views on the trial and the use of CAM in maternity care were also assessed and will 96 be reported in a related publication. Study Design and Methods 97 98 This study was conducted between 13 February to 30 August 2013 using a two arm blinded RCT 99 design, to compare the experiences and outcomes for primigravid women experiencing a post-date 100 pregnancy using three acupressure points, compared with those who had standard antenatal care. 101 The staff providing clinical care were unaware (blinded) of group allocation unless the participant 102 disclosed study participation. Data were analysed by a statistician blinded to group allocation. 103 The intervention protocol was developed based on the three acupoints used in the Ingram et al 104 study (Ingram et al. 2005). Expert advice from Debra Betts, an internationally recognised expert in 105 perinatal acupuncture and acupressure (Betts 2003; Betts 2006) supported the use of the three 106 acupoints and recommended the pressure duration and frequency of application. The midwife 107 research assistant (MRA) attended a one-day competency based workshop on acupressure and 108 was assessed as competent. All women were asked to complete a self-administered survey 109 following the birth and up to 10 days postpartum. The intervention group also completed an 110 acupressure protocol compliance assessment (daily diary). 5 111 Study Population 112 All eligible women at two outer metropolitan public hospitals in New South Wales, Australia were 113 invited to participate. 114 Sample size 115 It was estimated that a sample size of 60 women (30 in each group) would be sufficient to provide 116 data to address the feasibility study aims (Hertzog 2008; Arain et al. 2010). It is recommended that 117 a feasibility study have an adequate sample size to estimate the critical parameters such as 118 acceptability, recruitment, practicality, and adaptability (Bowen et al. 2009; Arain et al. 2010). 119 Control and Intervention groups 120 Control: Standard care 121 Women randomly allocated to the control group received standard clinical antenatal care at either of 122 the two study sites. Each woman was advised of her next clinic/ Day Assessment Unit (DAU) 123 appointment at approximately 40wks+10 days (+/-2 days) for a maternal and fetal assessment that 124 included abdominal palpation, vaginal examination to determine cervical favorability for induction of 125 labour (Bishop Score), and fetal heart pattern (electronic monitoring). This assessment determined 126 the method of medical induction and set an induction date before 40 weeks and 14 days. 127 Intervention: standard care plus acupressure 128 In addition to standard care, women randomly allocated to the intervention group received verbal 129 and written information on the self-administration of acupressure to three acupoints Spleen 6 [SP6 - 130 lower leg], Large Intestine 4 [LI4 -hand], Gall Bladder 21 [GB21 -shoulder] (see Figure 1). The 131 participants were asked to apply sustained bilateral pressure using thumb or finger for: 132 • 2 minutes on point SP6 (right/ left) followed by 2 minutes on LI4 (right/left) every 2 hours 133 during the day; and 134 • 2 minutes on point GB21 (right/left) twice a day (morning and evening). 135 6 136 The MRA demonstrated, on the participant, the acupoints and level of pressure that results in a 137 feeling of numbness, warmth, tingling, or a buzzing sensation (known as Deju) (Betts 2006). The 138 participant was then asked to locate the acupoints and self-apply pressure whilst being observed by 139 the MRA. An acupressure diary sheet was provided for the women to record acupoints used on a 140 daily basis from recruitment (40 weeks+5 days) to commencement of labour i.e. spontaneous or 141 medically induced at 40 weeks+11-14 days. A CD with short videos demonstrating the acupressure and 142 location of the three acupoints was also supplied. The women were contacted one to two days after 143 recruitment to discuss compliance and if further education was required. 144 Inclusion Criteria 145 Women who fulfilled the following conditions were included: healthy primigravid woman of gestation 146 ≥40 weeks+5days with a singleton pregnancy; cephalic fetal presentation; English speaking; ≥18 147 years; and receiving midwifery-led antenatal care. 148 Exclusion criteria 149 Participants who met any of the following criteria were excluded: wanting to use or currently using 150 acupressure (due to concern of crossover after randomisation); experiencing regular uterine 151 contractions; any contra-indications for vaginal birth; and highly dependent on medical 152 care/requiring specialist medical/obstetric consultation and likely to have medical intervention prior 153 to 41 completed weeks of gestation 154 Trial Recruitment 155 The MRA accessed the clinic appointment books and hospital obstetric database (ObstetriX) to 156 determine the eligibility of women to participate, identified their next clinic appointment and placed 157 the study information package (information sheet, consent form and revocation of consent form) in 158 the woman’s records ready for her next antenatal visit. 159 Between 39-40 weeks gestation each eligible woman received verbal and written information on the 160 study. At 40 weeks+5-7days visit, the MRA obtained written consent if the woman agreed to participate. 161 Once written consent was obtained, the woman’s details were entered into the trial register and the 162 remote, computer generated randomisation allocation service was contacted for allocation to group. 7 163 Group Allocation 164 Allocation concealment was assured by using a remote internet-based allocation service. The MRA 165 provided the woman’s hospital medical record number and eligibility criteria, the allocation service 166 randomised the woman based on a block size of four and immediately provided the group 167 allocation. After the MRA received confirmation of the woman’s group allocation and study number, 168 the woman was informed of her group allocation. If allocated to the intervention group, the MRA 169 provided the woman with the acupressure information as outlined previously. Participant 170 demographic details, date of randomisation, group allocation, and date of demonstrating 171 acupressure techniques were kept in the Trial Register. 172 Data Collection 173 Clinical Data 174 The majority of clinical data required for the study were routinely collected and available in hospital 175 records. The hospital’s maternity data manager, who was blinded to the participants’ allocation, 176 extracted the following information from the hospital obstetric database (ObstetriX): 1) 177 Demographic information including maternal age, model of midwifery care, and past obstetric, 178 medical and surgical history (these variables are the same as used in previous studies of 179 acupressure to ensure comparability); and 2) Clinical outcome measures for labour and birth 180 including onset of labour (spontaneous or induced), analgesia and any other pain relief options 181 used, and mode of birth. 182 Analysis 183 Categorical data were analysed using Chi Square and Fisher’s exact test (due to the small sample 184 size) and results are reported as odds ratios and 95% confidence intervals. T-tests were used to 185 analyse normally distributed continuous data. All levels of significance are reported as p< 0.05. Data 186 were analysed using the Statistical Package for Social Science V21.0 (SPSS). 8 187 Ethical aspects 188 Human Research Ethics Committee (HREC) approval for two sites was provided by the Northern 189 Sydney Local Health District HREC (1209-293M). Research governance approval for two hospital 190 study sites was also obtained (SSA 1209-307M; 1210-354M). 191 Data safety and monitoring 192 A Data and Adverse Event Monitoring Committee (DAVMC) assessed any serious adverse events. Findings 193 194 Eighty (80) nulliparous women at 40 weeks+5 days gestation were identified as eligible for the study. 195 Thirteen women were not recruited, as seven women birthed prior to being contacted by the MRA 196 (40weeks+5days to 40weeks+6days), and six women were unable to be contacted. From the remaining 197 67 eligible women, 23 declined to participate because 11 preferred to use, or had already started to 198 use acupressure and 12 women were not interested (n=6/12) or reason unknown (n=6/12). As a 199 result of the decline to participate rate, recruitment was stopped once 44 women had been 200 randomised, 22 to the intervention group and 22 to the control group (Figure 2). It was determined 201 by the authors there were sufficient participants to address the feasibility parameters such as 202 acceptability, practicality, adaptability and recruitment. 203 Demographics 204 Baseline characteristics of study participants were similar between groups (Table 1). Participants’ 205 ages ranged between 19 and 39 years, gestation at time of birth between 40weeks and 6 days to 40 206 weeks and 14 days and Body Mass Index (BMI) ranged between 20 and 36. 207 Table 1 - Participant baseline characteristics Acupressure Control p value N=22 N=22 Caucasian N (%) 22 (100) 21 (95.5) - Mean Age (SD) 27 (5.23) 27 (6.02) 0.939 Mean BMI (SD) 25.46 (4.14) 28.72 (5.75) 0.037† 9 Mean Gestation at birth 290.6 (1.78) 290.4 (2.17) 0.707 - days (SD) 208 † statistically significant 209 210 Recruitment and Acceptability 211 Twenty nine participant postnatal surveys were returned (65.9%) with 18 from the acupressure 212 group (82%) and 11 from the control group (50%). The majority of women in both groups would 213 agree to participate in the study again if presented with the opportunity (94% acupressure; 82% 214 control). 215 Four women (36%) randomised to the control group disliked being randomised with the following 216 comments: 217 “I was excited to participate in this study but very disappointed when I wasn’t selected & felt 218 let down” 219 “Didn't experience much as didn't get selected, was just info I could go through and discuss. 220 Thought it should be something available to all without being random if consent given” 221 One woman in the acupressure group disliked being randomised and she clarified her answer by 222 commenting “I was lucky to be in the acupressure group. I would have been disappointed if I was in 223 the control group”. 224 Involvement in a RCT 225 The women commented on what they liked about being in the study as shown in Figure 3. Many 226 women in both groups liked assisting with research and liked having other options available to bring 227 on labour. Some of the women in the acupressure group explained: 228 “feeling like I was doing something empowering”, 229 “being able to try non-drug/non-invasive options”, 230 “I'm glad some natural options are available to try help bring on labour”. 231 One woman in the control group commented 10