Which factors determine the sexual function 1 year after childbirth?

Authors


Dr HJ van Brummen, Department of Perinatology and Gynaecology, University Medical Centre Utrecht, Room F05.216, Heidelberglaan 100, PO Box 85500, 3508 GA Utrecht, the Netherlands. Email hbrummen@zonnet.nl

Abstract

Objective  To evaluate which factors determine sexual activity and satisfaction with the sexual relationship 1 year after the first delivery.

Design  Prospective longitudinal cohort study.

Setting  Ten midwifery practices.

Population  Three hundred and seventy-seven nulliparous women were included.

Methods  The Maudsley Marital Questionnaire is a standardised and validated questionnaire with 15 items relating to marital and sexual adjustment, with a nine-point (0–8) scale appended to each question. Scores on the sexual scale (MMQ-S) range from 0 to 40. Higher scores are indicative of greater dissatisfaction. Sexual intercourse was dichotomised into having sexual intercourse or not having sexual intercourse. Several obstetric and maternal factors were analysed.

Main outcome measurements  Sexual intercourse at 1 year postpartum and dissatisfaction with the sexual relationship as assessed by the MMQ-S scale.

Results  In multiple logistic regression analysis, the main predictive factor for no sexual intercourse 1 year postpartum was no sexual intercourse at 12 weeks of gestation (β 11.0 [4.01–30.4]). Women were five times less likely to be sexually active after a third/fourth degree anal sphincter tear as compared with women with an intact perineum (β 0.2 [0.04–0.93]). Dissatisfaction with the sexual relationship 1 year after childbirth, assessed with the MMQ-S scale, is associated with not being sexually active at 12 weeks of gestation (β– 0.208, P= 0.004) and with an older maternal age at delivery (β 0.405, P= 0.032).

Conclusion  An important prognostic factor for dissatisfaction with the sexual relationship 1 year postpartum was not being sexually active in early pregnancy. Satisfaction with the sexual relationship seems not to depend on pregnancy- and parturition-associated factors.

Introduction

Although little has been published about the mode of delivery and postpartum sexual functioning, popular media, such as Parents Magazine, report that women are concerned about the potential negative effects of vaginal childbirth on their sexual health. Recent studies suggest that childbirth influences sexual functioning and discuss questions such as ‘does caesarean delivery improves one’s sex life?’.1 These media reports reflect the belief that women who have caesarean delivery (CD) have better subsequent sexual functioning as compared with women who have a vaginal delivery. In addition, between 7 and 24% of obstetricians and 4.4% of midwives preferred CD for themselves or for their partner if male.2–5 Urogynaecologists scored even higher, with 45.5% who would prefer a primary elective CD.5 Fear of childbirth (27%), perineal injury (80–93%), fetal injury (24–39%), anal injury (83%) or urinary incontinence (81%), sexual dysfunction (58–59%), pain (7%), and for convenience (17–39%) are reasons given for this preference.2–5 Women’s concerns about sexual functioning after childbirth are not unfounded. In a recent review, an association between assisted vaginal delivery and degree of perineal pain and dyspareunia was suggested.6 However, the reported associations between CD and sexual dysfunction are inconsistent.7,8

The aim of this prospective study was to evaluate which factors determine sexual activity and satisfaction with the sexual relationship 1 year after a first delivery.

Methods

Study population

A prospective cohort study was undertaken to investigate the impact of first pregnancy and delivery on the pelvic floor. Between January 2002 and July 2003, 524 pregnant women were recruited from ten midwifery practices in the Netherlands. Eligible women had to be nulliparous heterosexual women with a low-risk singleton pregnancy between 12 and 18 weeks of gestation and should have a good knowledge of the Dutch language. Exclusion criteria were previous anorectal or urogynaecological surgery, anorectal and urogenital abnormalities, diabetes mellitus, neurological disorders, and having foster, adopted or stepchildren in the home. The participating midwives gave all consecutive nulliparous pregnant women verbal and written information about the study, and after 1 week, these women were asked to participate by telephone. The Medical Ethics Committee of the University Medical Centre Utrecht approved the study. All participants signed an informed consent.

Measurements

Women were asked to complete a questionnaire at 12 and 36 weeks of gestation, and at 3 and 12 months after delivery. The Maudsley Marital Questionnaire (MMQ) was used to evaluate the satisfaction with the sexual relationship during and after pregnancy. The MMQ is a standardised and validated questionnaire with 15 items relating to marital and sexual adjustment, with a nine-point (0–8) scale appended to each question.9–11 Scores on the sexual scale (MMQ-S) range from 0 to 40, while those on the marital scale (MMQ-M) range from 0 to 80. Higher scores are indicative of greater adjustment problems. The sexual function scale consists of the five questions shown in Appendix 1. Sexual intercourse is defined as the situation in which the erect penis enters the vagina at least for a short period, irrespectively of whether ejaculation occurs. The women filled out the MMQ sexual scale at all four time periods, and at each time period, sexual intercourse was dichotomised into having or not having sexual intercourse and the time period when women first had intercourse was noted.

Other potential determinants for sexual functioning in relation to childbirth that were analysed included cohabiting (yes/no), maternal age, partner’s age, duration of the relationship, perineal state (no rupture, first/second degree, third/fourth degree, and episiotomy), mode of delivery (spontaneous vaginal delivery, instrumental vaginal delivery, or CD), infant birthweight, second stage of labour, sexual activity at 12 weeks of gestation, and sexual scale at 12 weeks of gestation.

In the Netherlands, the midwife is responsible for providing obstetric care to healthy low-risk pregnant women. If pregnancy-related problems occur, the obstetrician/gynaecologist is consulted. Deliveries performed by midwives and gynaecologists are registered in the Dutch National Obstetric Database (LVR). Obstetric data of the participants were abstracted from this database.

Statistics

Values are expressed as number (percentage) or mean (standard error [SE]). In univariate analysis, the Fisher’s exact test was used to assess the significance of the associations for categorical variables and the Mann–Whitney test for continuous variables. Where appropriate, odds ratios (OR) with 95% confidence intervals (CI) were calculated. Multivariate analysis was used to calculate adjusted OR (B [95% CI]) for factors that had a P value < 0.10 in univariate analysis. Multiple linear regression analyses were performed to identify associated factors for continuous variables. The McNemar test was used for matched and paired categorical data, and paired continuous data were analysed with the Wilcoxon signed rank test. Data were analysed using SPSS 11.5 (SPSS Inc., Chicogo, IL, USA). A P value < 0.05 was considered to be statistical significant.

Results

Five hundred and twenty-four women agreed to participate in the study. Thirty-seven women were considered lost to follow up because fewer than three questionnaires were completed and returned. Fifty women were excluded from the analysis because they were pregnant again when completing the final questionnaire. Nine single women who did not have a sexual relationship were also excluded. In this analysis, only the 377 women who returned both postpartum questionnaires were included. Women, who only returned one questionnaire postpartum, did not differ in sexual activity (at 3 months postpartum 82.1 versus 81.4%; P= 1.000) as compared with the women included in this analysis. Table 1 shows the characteristics of the study population.

Table 1.  Characteristics of the study population
 MeanSE
  1. IVD, instrumental vaginal delivery; SVD, spontaneous vaginal delivery.

Age at delivery (years)30.30.18
Body mass index at inclusion (kg/m2)23.70.19
 n= 377%
Marital state
Married or cohabiting36596.9
Single or divorced123.1
Educational level
Primary195.1
Secondary or more35894.9
Smoking
Yes4311.5
No33488.5
Mode of delivery
SVD24364.5
IVD6617.5
 Vacuum extraction6116.2
 Forceps delivery51.3
CD6818.0
 Elective297.7
 Emergency3810.3
Perineal state after vaginal delivery
Intact5016.2
First/second degree tear9631.3
Third/fourth degree tear196.1
Episiotomy14446.6
Neonatal dataMeanSE
Gestation (days)278.20.68
Birthweight (grams)3416.233.9

Table 2 shows the data on sexual function at 3 and 12 months postpartum. Univariate and multivariate analyses were performed to identify possible determinants of sexual intercourse 12 months after childbirth (Table 3). None of the women were breastfeeding. In multiple logistic regression analysis, the main predictive factor for no sexual intercourse 1 year postpartum was no sexual intercourse at 12 weeks of gestation (B 11.0 [4.01–30.4]). Women who were not sexually active at 12 weeks of gestation had a 11 times higher chance of not being sexually active 1 year postpartum. The other significant factor was a third/fourth degree anal sphincter rupture. After such a tear, women were five times less likely to be sexually active 1 year postpartum (B 0.2 [0.04–0.93]).

Table 2.  Sexual function in primiparous women 3 and 12 months after delivery
 3 months postpartum12 months postpartumTestP values
n%n% 
  • The values are number (percentages) or means (SE).

  • *

    Categorical values were compared with the McNemar test.

  • **

    Continuous variables were compared with Wilcoxon’s signed rank test.

Sexual intercourse
Yes30781.435493.9*0.000
No7018.6236.1 
 Mean (SE)RangeMean (SE)Range 
Sexual scale MMQ (range 0–40)11.09 (0.34)0–369.14 (0.33)0–34**0.000
Table 3.  Associated factors for sexual activity 1 year postpartum
 Univariate analysisMultivariate analysis
Sexual intercourseP valueTestOR (95% CI)Adjusted OR (95% CI)
YesNo
MeanSEMeanSE
  • IVD, instrumental vaginal delivery; r, reference group; SVD, spontaneous vaginal delivery.P < 0.05 was considered statistically significant, and if significant, it is highlighted as bold.

  • *

    Categorical values were compared with the Fisher’s exact test.

  • **

    Continuous variables were compared with Mann–Whitney U test.

Maternal age (years)30.290.1931.770.670.041** 0.89 (0.74–1.06)
Body mass index (kg/m2)23.90.2224.80.820.191** 
Age of partner (years)31.90.2433.70.870.032** 0.99 (0.86–1.14)
Duration of relationship (years)6.90.217.40.670.345** 
Infant birthweight (grams)3421.333.063604.944.590.062** 1.00 (0.99–1.001)
Second stage of labour59.32.8469.49.510.239** 
Mode of deliveryn%n%r 
SVD18794.5135.5 
IVD5191.158.90.556*0.71 (0.24–2.08) 
CD5594.835.21.000*1.28 (0.35–4.63) 
Perineal state
No rupture9194.855.2r 
First/second degree tear7392.467.60.310*0.42 (0.10–1.78) 
Third/fourth degree tear1381.3318.70.078*0.27 (0.07–1.04)0.20 (0.04–0.93)
Episiotomy11293.386.70.359*0.49 (0.12–1.89) 
Cohabiting
Yes34693.8236.21.000*1.02 (1.00–1.04) 
No8100.000.0 
Sexually active at 12 weeks of gestation
Yes32397.0103.00.000*12.8 (5.45–30.06)11.02 ( 4.01–30.4)
No3069.81330.2 

Multiple linear regression analyses revealed that dissatisfaction with sexual relationship 1 year after childbirth, assessed with the MMQ-S scale, is associated with not being sexually active at 12 weeks of gestation (β– 0.208, P= 0.004) and with an older maternal age at delivery (β 0.405, P= 0.032). On the average, those who did not have sexual intercourse at 12 weeks of gestation scored 0.208 higher in the MMQ-S scale, and an increase of 1 year in maternal age was associated with an increase of 0.405 on the MMQ-S scale. To obtain better insight in the relationship between maternal age and satisfaction with sexual relationship, we categorised the MMQ-S scale (range 0–34) in three different groups. The first group had a score between 0 and 10, the second group a score between 11 and 20, and the third group had from 21 to 34. The first group had a mean age at delivery of 30.1 (0.21) years, the second group 30.7 (0.40) years of age, and the third group 31.5 (0.77) years of age at delivery.

Discussion

This prospective study in nulliparous women shows that satisfaction with their sexual relationships 1 year after delivery is not related to the mode of delivery. Not being sexually active in early pregnancy was the most important factor predicting dissatisfaction with sexual relationship and for not being sexually active 1 year postpartum.

Several studies have shown that in the first 3 months after childbirth, 22–86% of women report sexual problems.8,12,13 The percentage of women who are sexually active at 3 and 12 months postpartum in our study is in concordance with other studies.13–15

The focus of the MMQ is on satisfaction with sexual relationship.9–11 To our knowledge, we were the first to investigate which factors determine the satisfaction with sexual relationship 1 year after childbirth. We demonstrated that not being sexually active at 12 weeks of gestation and an older maternal age are important determinant factors for being dissatisfied with sexual relationship 12 months postpartum. Most prior research has focused on perineal pain/dyspareunia in relation to mode of delivery. A recent review on postpartum sexual functioning and method of delivery reported a greater risk for perineal pain among women after assisted vaginal delivery.6–8,12,16–18 The reported associations between CD and perineal pain/dyspareunia, and delay in resumption of sexual intercourse are inconsistent.7,8,17,18 Because no questions about perineal pain/dyspareunia are included in the MMQ, we cannot make any statement about such associations with type of delivery.

The relationship between maternal age and dissatisfaction with sexual relationship cannot be explained by the possibility that other children may disproportionately burden older women, given that all women in this study were nulliparous. Dissatisfaction may reflect lifestyle and career differences between older and younger women. The association may have a more biologic basis that is related to the healing process because previous research found an association between pain on sexual intercourse and older maternal age.19

Most studies on the effect of method of delivery on sexual functioning have been of a retrospective design, did not use validated measures, or measured only short-term outcomes. One study that used a validated sexual health measurement had a retrospective study design.12 All studies, except one, had a follow up of less than 9 months. Unfortunately, the one study with a follow-up period of 18 months used a retrospective design.17 Recall bias is a major drawback in all retrospective studies, and therefore, results from these types of studies have to be interpreted with caution. One study did have a prospective design but a relatively short follow-up period of 24 weeks.8 In these various studies, factors such as assisted vaginal delivery, perineal damage, parity, and marital state were suggested to be associated with sexual dysfunction after childbirth.7–12,16–18 In our prospective study, whether woman engaged in sexual intercourse 1 year postpartum was independently associated only with a third/fourth degree anal sphincter tear and being sexually active in early pregnancy. It is known that perineal damage is associated with dyspareunia and that pain can influence whether women engage in sexual intercourse.6–8,12,16–18 Whether women were sexually active at 12 weeks of gestation may be a reflection of sexual function before pregnancy. We did not collect information about the woman’s sexual relationship before pregnancy and we cannot make any statement about the effect of sexual relationship before pregnancy on its postpartum status. Further research is necessary not only to confirm our data but also to evaluate the effect of subsequent pregnancies, sociodemographic variables, cultural influences, the partner’s prospective and the quality of sexual relationship before pregnancy.

There are several limitations of our study. First, selection bias may be a problem. However, the mode of delivery in our study population was not different from the mode of delivery of low-risk Dutch nulliparous women. This makes significant participation bias unlikely.

Second, the validity of results on human sexuality can be impaired by terminological problems, which are difficult to manage adequately in studies using self-administered questionnaires. Interview studies have concluded that many women do not adequately understand the terms ‘orgasm’ or ‘intercourse’.20,21 We attempted to avoid this by using a validated questionnaire for the Dutch population, by giving clear instructions and by emphasising that the participants could also contact the main investigator if they had questions.

Third, we did not ask about the partner’s perceptions of their sexual relationship. As women can be reluctant to speak about their own sexuality, much less their partner’s, we focused on their own experiences in the hope that this would increase participation.

Fourth, only 23 of the 377 women were not sexually active 1 year postpartum. This is a relatively small number of women. Further research with greater power is necessary to confirm our findings.

In conclusion, the most important prognostic factor for dissatisfaction with sexual relationship 1 year postpartum was not being sexually active in early pregnancy. Satisfaction with sexual relationship 1 year after first childbirth seems not to depend on factors associated with pregnancy and parturition.

Acknowledgements

We would like to specially thank all midwifery practises for their participation: De Lekbrug, Utrecht Noord, Maarssen, Maarssenbroek, Houten, ‘Luna’ Leusden, ‘De Hazelaar’ Wijk bij Duurstede, University Medical Center Utrecht, ‘Corver and Joosten’ Woerden, ‘Gram’ Geldermalsen. We would also like to thank Dr SE Schraffordt Koops for his input.

Appendix

Table Appendix1..  The MMQ
 Questions of sexual scale MMQ
  1. A nine-point (0–8) scale is appended to each question.

Sexual intercourseHow often have you had sexual intercourse with your partner in the last month?
Satisfaction with sexAre you satisfied with the present frequency of sexual intercourse?
Sexual enjoymentDo you find sex enjoyable?
Likelihood to achieve an orgasmWhat proportion of the time, in relationship sex, do you experience a climax?
Physical contactHow did you feel during physical contacts between you and your partner?

Ancillary