Cyna 2011.

Methods	Randomised controlled trial of parallel design conducted in Women's and Children's Hospital, Adelaide, Australia.
Participants	"Inclusion criteria: women > 34 and < 39 weeks' gestation, with a singleton, viable fetus, vertex presentation, who are not in active labour and who are planning a vaginal birth. Exclusion criteria: previous hypnosis preparation for childbirth; poor understanding of English requiring a translator; women who are already enrolled in another pregnancy trial where analgesia requirements are an outcome measure; active psychological or psychiatric problems such as: active depression requiring treatment by a psychiatrist; schizophrenia; prior psychosis; severe intellectual disability. Also women with pain caused by specific pathological entities such as: congenital neuromuscular disorders; spina bifida; metastatic disease; osteoporosis; rheumatoid arthritis; fractures."
Interventions	Intervention Group 1 (n = 154): "antenatal hypnosis training in preparation for childbirth administered by a qualified hypnotherapist with the use of audio compact discs on hypnosis for re‐enforcement". Intervention Group 2 (n = 143): "antenatal hypnosis training in preparation for childbirth using audio compact discs on hypnosis administered by a nurse with no training in hypnotherapy". Controls (n = 151): "participants continue with their usual preparation for childbirth with no additional intervention" (no treatment). The hypnosis interventions were provided in 3 sessions to groups of up to 10 women. The sessions commenced "as closely as possible to 37 weeks' gestation".
Outcomes	"Use of pharmacological analgesia (nitrous oxide; opioids; epidural); use of oxytocics; mode of delivery; Apgar score less than seven at five minutes; admission to HDU/ICU; adverse effects for women (PPH => 600 mL blood transfusion; death; ICU admission); adverse effects for infants (meconium‐stained liquor; admission to neonatal unit); overall experience of pain during labour and childbirth ‐ birth experience was worse/better, same as expected; whether birth rated as positive or negative experience; how well coped with labour/childbirth (postpartum questionnaire); length of labour; length of neonatal nursery stay; length of maternal stay; number women breastfeeding at discharge, 6‐week and 6‐month follow‐up; Edinburgh Postnatal Depression Scale and Spielberger anxiety scales repeated at 6 weeks and 6 months. Hypnotisability was also measured using the Creative Imagination Scale (CIS) with high hypnotisability defined as a score greater than or equal to 23 and low hypnotisability defined as a score < 23."
Notes	Principal investigator contacted on 19/8/2011 and replied "The Cyna trial is complete and it has been written up and accepted in part as a PhD thesis. We are currently preparing the paper for submission to a journal". Update January 2012 ‐ the digital thesis is now available online. The principal investigator has also provided additional data and information about methodology as requested. Update September 2015 ‐ the results of the trial have also been published in BJOG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"computer‐generated random number sequence.” The author provided additional detail that "Study participants were stratified for parity and randomised in (unspecified) blocks of 15 by a computer random number generator."
Allocation concealment (selection bias)	Low risk	“we were provided with group allocation via telephone at the Department of Public Health for the first 6 months of the study and then by a password‐protected computer database program.” “The randomization sequence was inaccessible to research assistants involved in recruiting potential trial participants.” The author provided additional detail that "Allocation concealment was assured by using a computer database assignment to one of three groups, which was only revealed after patient identifiers had been entered."
Blinding of participants (objective outcomes)	Low risk	Participant: "All participants were informed that they may or may not appreciate which group they are in, as we believed that some women might think that the baseline testing for hypnotisability was the intervention. However, we did expect that most women allocated to usual care would probably realise they were not in an intervention group. An assessment of blinding was determined by asking participants if they thought they were in a control or intervention group in the final post‐partum questionnaire."  110 of 134 in the hypnosis believed they were in the hypnosis arm of the trial, 98 of 133 women in the CD believed they were in the hypnosis and 0 of 133 of women in the control of the trial believed they were in the hypnosis. High risk of bias for subjective outcomes (such as satisfaction with pain relief) and low risk of bias for objective outcomes (such as spontaneous vaginal birth).
Blinding of participants (subjective outcomes)	High risk	Participant: "All participants were informed that they may or may not appreciate which group they are in, as we believed that some women might think that the baseline testing for hypnotisability was the intervention. However, we did expect that most women allocated to usual care would probably realise they were not in an intervention group. An assessment of blinding was determined by asking participants if they thought they were in a control or intervention group in the final post‐partum questionnaire."  110 of 134 in the hypnosis group believed they were in the hypnosis arm of the trial, 98 of 133 women in the CD arm believed they were in the hypnosis arm and 0 of 133 of women in the control arm of the trial believed they were in the hypnosis arm. High risk of bias for subjective outcomes (such as satisfaction with pain relief) and low risk of bias for objective outcomes (such as spontaneous vaginal birth).
Blinding of personnel (performance bias)	Low risk	Clinician: confirmed with the author that clinicians caring for the women in labour were blinded to group allocation.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	“All data were collected and analysed by researchers who were unaware of the participants’ group allocation.”
Incomplete outcome data (attrition bias) All outcomes	Low risk	190 women not randomised, (50 declined to participate and 140 did not meet eligibility criteria) 137 were excluded after randomisation due to protocol violations (see below). All other women analysed at birth. 48 women lost to follow‐up at 6 week follow‐up (live hypnosis group = 20, audio CD hypnosis group = 10, control group = 18). 138 women were excluded due to protocol violations 1 in active labour, 137 gestation less than or equal to 34 weeks at randomisation due to human error at the point of randomisation. The author provided additional detail that "After completing nearly two years of recruitment, we became aware that some women, who were ineligible for participation, had been inadvertently randomised, outside our eligibility criteria, prior to 34 weeks' gestation.  We therefore planned to continue to recruit women to the study until our initial planned sample size of eligible women had been reached. Only those women who met all eligibility criteria for inclusion were analysed." "All primary and secondary outcomes of trial participants fulfilling all eligibility criteria were analysed using the 'Intention‐to‐treat' principle."
Selective reporting (reporting bias)	Low risk	All outcomes listed in trial registration reported except maternal rating of control during labour and breastfeeding rates at discharge from hospital (and breastfeeding at 6 months) ‐ data provided by the author. Additional secondary outcomes not listed in trial registration also reported.
Other bias	Low risk	"Our analyses of baseline data shows that the randomisation with stratification for parity produced comparable groups with the exception of the incidence of women with a history of depression, and an EPDS score > 12 being increased in the Hypnosis Group. The distribution of all other participants’ baseline demographic data across the three groups, such as mothers’ use of complementary therapies during their pregnancy, age, weight and country of birth, were also comparable."