Labor induction has been increasing since the early 1990s,1 and the rate is running at about 20% for pregnancies at term.2,3 Induction of labor compared with spontaneous labor is associated with adverse maternal outcomes, including at least a doubling in the caesarean delivery rate,4,5 25–50% increase in instrumental vaginal delivery rate,3,5 higher postpartum hemorrhage rate,5 and prolonged labor.5 Neonates born after induced labor are more likely to have low Apgar score and low umbilical cord blood pH.5
Coital activity at term in healthy women has been reported to be associated with a shortened gestation and less requirement for labor induction for prolonged pregnancy, and there is a direct correlation between the amount of coital activity and expedited onset of labor.6 However, these findings are not consistently reported.7 Although having sex is commonly believed to hasten labor,8 between 20% and 80% of pregnant women have safety concerns.6
Biologic plausibility for sexual activity to promote labor onset is supported by the presence of prostaglandin E in semen,9 observed effect of breast stimulation on labor onset,10 and association of uterine activity with orgasm during sexual intercourse in pregnancy.11 A Cochrane review on sexual intercourse for cervical ripening and induction of labor identified only one study with 28 women, from which no meaningful conclusions can be drawn.12
We sought to investigate a situation in which the woman had an appointment to undergo a nonurgent labor induction at term whether the couple can be persuaded to have vaginal sexual intercourse as a natural method of promoting the onset of labor. We felt that, with labor induction imminent, the motivation would be highest in these couples to respond positively to advice. As a consequence, we anticipated that some couples who would otherwise be abstinent could be persuaded to have sex and those couples who would be sexually active anyway would increase their rate of coital activity. We hypothesized that the increased coital activity would increase the rate of onset of spontaneous labor.
MATERIALS AND METHODS
A randomized trial was performed in which a group of women already given an appointment for nonurgent labor induction at term and advised to have vaginal sex was compared with a control group where sex was neither encouraged nor discouraged. Ethical approval for the study was granted by the University of Malaya Medical Centre Medical Ethics Committee.
We recruited women from our antenatal clinic on the same day that they were given an appointment for labor induction for nonurgent reasons. Appointments were usually made one week in advance. After randomization and allocated counseling, women were allowed to return home so that the opportunity for sexual activity in their home environment existed.
Inclusion criteria were induction of labor scheduled at term (37 weeks or later), a viable singleton fetus, intact membranes, and cephalic presentation. We excluded women with a previous cesarean scar and known gross fetal anomaly.
A previous study from our hospital has shown that 58% of women were sexually active at term. Assuming that advice to have sex was heeded by an additional 20% of women with alpha of 0.05 and power of 80%, 94 women were needed in each arm. Assuming a 10% dropout rate, 209 women in total were needed for the study. Data from the 2003 birth cohort of our hospital indicated that 56% of women who entered the 40th gestational week would be delivered within the next week. Assuming that advising women to have sex increased the spontaneous labor rate by 20–76%, with alpha of 0.05 and power of 0.8 as above, 97 women were needed in each arm.
Women who had been given an appointment for labor induction were identified by antenatal nursing staff and directed to a single investigator (C.M.Y.), who recruited the women. Written consent was obtained from all women.
Randomization was carried out in randomized blocks of 8 or 12 with a computerized random number generator by another investigator (P.C.T.) who also prepared the numbered opaque envelopes containing the individual random allocation to either the advised-coitus group or the no-advice group. The numbered envelopes were used in strict chronological sequence, and an envelope, once opened, was not reused.
We kept the counseling to a single investigator (C.M.Y.) to standardize the process. The counselor-investigator was clearly identifiable to the women as a doctor. Women randomly assigned to the advised-coitus group were told that sexual intercourse in late pregnancy was safe and could promote onset of labor. They were told that induced labor was associated with operative delivery and a more prolonged process compared with spontaneous labor. They were asked to have sex as frequently as possible before their appointment for labor induction. They were also required until delivery to keep a daily diary on vaginal sexual intercourse and orgasms they achieved during sex. Women who wanted more information after the standard counseling were given a response that was supportive of having sex.
For women randomly assigned to no advice, counseling was kept as short as possible. These women were told that sex was safe but the effect was unclear on promotion of labor onset. We also asked them to keep the same diary. To keep our approach to the control group as neutral as possible, women allocated to the no-advice group (control) who wanted additional information were referred to the information leaflet given to all study women.
The allocation of women to the advised-coitus or control groups was not revealed to providers. Recruitment took place between December 2005 and June 2006. Study women received standard obstetric care.
In our center, labor was induced with vaginal dinoprostone if the cervix was unfavorable, and amniotomy was performed if the cervix was favorable. Our standard management for labor induction and intrapartum care has been described previously.13 In the event of premature rupture of membranes at term (PROM), women were given either the option of immediate action, usually an oxytocin infusion, or the option to wait up to 24 hours for spontaneous labor as an inpatient.
The charts of women in our study were obtained after their delivery; admission and delivery details were then transferred to a standard data sheet. Women who did not submit their diaries within a few days after their labor induction appointment were contacted by telephone to obtain diary data. If the women were delivered elsewhere, we obtained as much clinical data from the women as we could by telephone.
Primary outcomes were 1) coital activity from the diaries and 2) the onset of labor. The onset of labor was defined as 1) spontaneous regular contractions leading to cervical changes of at least 3 cm dilatation or 2) PROM, on or before the original appointment date for labor induction. All women who presented after the original appointment date were considered as having failed the outcome even if they were subsequently admitted in spontaneous labor or with PROM. Similarly, we consider all cases of prelabor cesarean delivery as a failed outcome regardless of timing of birth admission.
Secondary outcomes included cesarean delivery, reported orgasms, initial Bishop score at the admission for birth, PROM, use of dinoprostone, use of oxytocin infusion during labor, maternal fever, epidural use in labor, meconium-stained liquor, and various neonatal measures.
Analysis was by intention to treat. Data were entered into SPSS 13 (SPSS Inc, Chicago, IL), and GraphPad Instat (GraphPad Software Inc, San Diego, CA) was also used for data analysis. The t test was used to analyze means and the Kolmogorov-Smirnov test to check distribution. Fisher exact test was used for categorical 2×2 data sets, χ2 for larger categorical data sets, and relative risk and its 95% confidence interval calculated with GraphPad Instat. P<.05 in any test was considered statistically significant, and all tests used two-tailed results.
During our recruitment period of December 2005 to June 2006, 219 women were sent to the investigator for recruitment out of a maximum 569 women who were given an induction of labor appointment. The smaller number was due to investigator availability and study population criteria. Four women declined the invitation to participate. Another five women, all allocated to the control group, decided to withdraw from the study at the counseling phase, a decision we felt might have been due to their counseling having been kept as short and as neutral as possible. This left 210 women for analysis: 108 women assigned to the advised-coitus group and 102 to the no-advice control group (Fig. 1).
Three women who delivered elsewhere were contacted by phone, but data obtained were incomplete. For one woman who delivered at home, data were also incomplete. Another woman did not submit her diary and could not be contacted by telephone, but her chart was available.
The characteristics of study women included in the analysis are shown in Table 1. There was no significant difference in any characteristic between the randomized groups.
More women advised to have sex reported coital activity: 60.2% compared with 39.6% (relative risk 1.5, 95% confidence interval [CI] 1.1–2.0, P=.004) compared with controls, but spontaneous labor onset was no different: 55.6% compared with 52.0% (relative risk 1.1, 95% CI 0.8–1.4, P=.68) (Table 2). Of the 105 women who reported at least one episode of vaginal sex after randomization, 87 (82.9%) reported having at least one orgasm during the study period.
Table 3 shows the clinical presentation of study women at their admission for delivery, stratified according to allocated intervention and timing of delivery admission in relation to appointment date for labor induction. There were no significant differences between the randomized groups with reference to timing of admission for birth or spontaneous labor (PROM included), as well as for presentation at admission for birth (χ2P=.21 and P=.49, respectively).
Table 4 shows the secondary outcome measures; there were no differences in cesarean delivery rate or maternal fever. Neonatal outcome was also not different.
We have demonstrated that it was feasible to motivate women at term who were scheduled for labor induction to have vaginal sex to promote labor. We had taken care to have as controls a group that is not negatively influenced about sex; 39.8% of women assigned to the control group reported coital activity, although this percentage is smaller than the 58% that reported coital activity in a previous study of healthy women at term from our center.6 The previous study covered vaginal sex from 36 weeks of gestation to delivery, whereas the recruitment-to-birth admission interval in our current study population was only 4.7±3.4 days (mean±standard deviation). However, the spontaneous labor rate was not different: 55.6% compared with 52.0% when comparing the randomized groups.
A post hoc analysis on reported coitus compared with spontaneous labor onset was performed to explore the possibility that our study might be underpowered. Although statistical significance was not achieved, the spontaneous labor rate was higher in women who reported no coitus (60.6% compared with 46.7%, relative risk 1.3, 95% CI 1.0–1.7, P=.052). This finding differed from that of an observational study from our center that excluded from analysis women who had labor induction for any medical indications apart from those indicated by prolonged pregnancy.6 A possible explanation for our post hoc finding might be that women close to spontaneous labor onset had reduced libido and consequently less coitus; the mean recruitment-to-admission interval was only 4.7 days in our study.
A recent study has indicated that women who are sexually active at term have slightly longer gestations and slightly lower Bishop scores.7 Orgasm has also been reported to be associated with a lower occurrence of preterm delivery.14 Our post hoc analysis was more consistent with these findings.
Of women in our study who reported having sex during the study period, 82.9% also reported at least one episode of orgasm. Therefore, regardless of its effect on labor onset, sex at term was orgasmic for a large majority of sexually active women in the study.
Adverse neonatal outcome was uncommon in our study and was no different between the randomized groups. Post hoc analysis, stratified according to coital activity to maximize power, similarly showed no difference in neonatal outcome.
Our study has some shortcomings. We relied on reported coital activity, and we could not independently verify the reports. That having been said, we did not see any indication that women were systemically biased in their reporting. Women were typically counseled without their partners being present, and we also did not record partner involvement during the study. Ideally, the couple should be seen together, because the male partner is often the main initiator of sex during pregnancy.15 Our study also involved a single investigator doing all the counseling, which was helpful for standardization of the delivery of our intervention but which might reduce generalizability. Based on the findings of our study, women scheduled for induction of labor at term should not be given advice to have sex for the purpose of promoting labor onset.
1. Rayburn WF, Zhang J. Rising rates of labor induction: present concerns and future strategies. Obstet Gynecol 2002;100:164–7.
2. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F, Munson ML. Births: final data for 2003. Natl Vital Stat Rep 2005;54:1–116.
3. The Information Centre, Community Health Statistics. NHS Maternity Statistics, England: 2004–2005.Available at: http://www.ic.nhs.uk/webfiles/publications/maternityeng2005/NHSMaternityStatistics260506_PDF.pdf
. Retrieved September 5, 2006.
4. Ben-Haroush A, Yogev Y, Bar J, Glickman H, Kaplan B, Hod M. Indicated labor induction with vaginal prostaglandin E2 increases the risk of cesarean section even in multiparous women with no previous cesarean section. J Perinat Med 2004;32:31–6.
5. Heimstad R, Romundstad PR, Eik-Nes SH, Salvesen KA. Outcomes of pregnancy beyond 37 weeks of gestation. Obstet Gynecol 2006;108:500–8.
6. Tan PC, Andi A, Azmi N, Noraihan MN. Effect of coitus at term on length of gestation, induction of labor, and mode of delivery. Obstet Gynecol 2006;108:134–40.
7. Schaffir J. Sexual intercourse at term and onset of labor. Obstet Gynecol 2006;107:1310–4.
8. Schaffir J. Survey of folk beliefs about induction of labor. Birth 2002;29:47–51.
9. Taylor PL, Kelly RW. 19-Hydroxylated E prostaglandins as the major prostaglandins of human semen. Nature 1974;250:665–7.
10. Kavanagh J, Kelly AJ, Thomas J. Breast stimulation for cervical ripening and induction of labour (Cochrane Review). In: The Cochrane Library, Issue 3, 2005. Oxford: Update Software.
11. Chayen B, Tejani N, Verma UL, Gordon G. Fetal heart rate changes and uterine activity during coitus. Acta Obstet Gynecol Scand 1986;65:853–5.
12. Kavanagh J, Kelly AJ, Thomas J. Sexual intercourse for cervical ripening and induction of labour (Cochrane Review). In: The Cochrane Library, Issue 2, 2001. Oxford: Update Software.
13. Tan PC, Jacob R, Omar SZ. Membrane sweeping at initiation of formal labor induction: a randomized controlled trial. Obstet Gynecol 2006;107:569–77.
14. Perkins RP. Sexual behavior and response in relation to complications of pregnancy. Am J Obstet Gynecol 1979;134:498–505.
15. Naim M, Bhutto E. Sexuality during pregnancy in Pakistani women. J Pak Med Assoc 2000;50:38–44.