| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Department of Epidemiology, School of Public Health, the Carolina Population Center and the Department of Obstetrics and Gynecology, School of Medicine, University of North Carolina, Chapel Hill, North Carolina and the Epidemiology Branch, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina.
Address reprint requests to: David A. Savitz, PhD Department of Epidemiology CB #7400 School of Public Health University of North Carolina Chapel Hill, NC 27599-7400 E-mail: david-savitz{at}unc.edu
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: Women at least 16 years old and carrying singleton fetuses were recruited between 24 and 29 weeks’ gestation from prenatal clinics in central North Carolina. They were interviewed by telephone about sexual activity before and during pregnancy. One hundred eighty-seven women delivered between 29 and 36 weeks and had a follow-up interview after delivery. Four hundred nine women who were selected randomly from the cohort served as controls and had a follow-up interview between 29 and 36 weeks (mean gestational age 39.2 weeks).
Results: Intercourse during late pregnancy was associated with a reduced risk of preterm delivery. The conditional odds ratio (OR) was 0.34 and 95% confidence interval (CI) 0.23, 0.51 for preterm delivery within 2 weeks after intercourse. Similar decreased risk for preterm delivery was found with recent female orgasm. Adjusting for race, age, education, and living with a partner had little effect on results. Cases were more likely than controls to report poorer health, medical reasons for reducing sexual activity, less interest in sex, and receipt of advice to restrict sexual activity during pregnancy. Results did not differ substantially according to presence or absence of bacterial vaginosis at 28 weeks.
Conclusion: These data provide evidence against the hypothesis that sexual activity generally increases risk of preterm delivery between 29 and 36 weeks. However, we cannot exclude the possibility that a small subgroup of susceptible women might have adverse consequences of sexual activity.
Preterm birth is an important cause of perinatal death and is a major predictor of neonatal and infant morbidity.1 Sexual activity has long been suspected to be a potential cause of preterm delivery,2,3 and several biologic mechanisms could explain an adverse effect of sexual activity on preterm delivery. Maternal orgasm might release oxytocin and initiate uterine contractions.4 Prostaglandins in seminal fluid also have oxytocic properties.5 Coitus during pregnancy can increase exposure to infectious agents that could result in preterm delivery.5–7
There is inadequate empiric evidence for making recommendations to couples about the safety of sexual activity during pregnancy. Many previous studies had methodologic problems and yielded conflicting results. Some studies included few women with preterm delivery and thus generated imprecise estimates of the association with sexual activity.3,8–10 Only a few studies have controlled for potential confounders.7,11–13 Many studies have assessed sexual activity for preterm cases and term controls at different gestational ages.2–4,13,14 Such comparisons are biased because the frequency of sexual activity declines throughout pregnancy, particularly in the last weeks.8,15,16
We examined the association between sexual intercourse and orgasm in late pregnancy and preterm delivery using information on sexual activity obtained through standardized interviews conducted at similar gestational ages for 187 cases who delivered preterm and 409 controls.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). Study staff members reviewed clinic schedules so they could be present to recruit eligible women. During the recruitment visit, study staff obtained written informed consent and requested permission to collect genital tract swabs in conjunction with routine clinical procedures. A standardized telephone interview was conducted by trained female interviewers within 2 weeks after recruitment (approximately 28 weeks’ gestation) to obtain information on potential risk factors for preterm delivery. This study was approved by the institutional review board at the University of North Carolina School of Medicine and by the institutional review committee at Wake Medical Center.
|
Preterm cases were women who delivered before 37 completed weeks’ gestation; cases received a follow-up telephone interview after delivery. Cases were further categorized by whether preterm delivery was precipitated by premature rupture of membranes (PROM; membrane rupture 4 or more hours before the onset of labor), preterm labor (onset of labor before membrane rupture, or membrane rupture fewer than 4 hours before onset of labor), or medical induction before 37 weeks. Assessment of timing of membrane rupture and onset of labor was made by study obstetricians based on chart review. Information on preterm delivery subtype was available for 98% of cases.
A case-control study was implemented as follows: potential controls were selected randomly from a subset of the cohort and received a follow-up telephone interview at a specific week of pregnancy between 29 and 36 weeks (Figure 1
). Specifically, the RANUNI function in SAS was used to assign a random number to each potential participant when her data were entered into the study tracking system. The range that the number fell into determined whether the woman was assigned to be a control, and if so, what week of gestation her follow-up interview was scheduled to occur. Cases and controls were frequency matched by week of gestation. The distribution of gestational ages at which controls received follow-up interviews was chosen to reflect the expected distribution of preterm deliveries (2:1 control: case ratio for cases delivering in weeks 29–34 and a 1:1 ratio for cases delivering in weeks 35–36). Controls also received a follow-up interview after delivery. Additional information was abstracted from medical charts of cases and controls.
This analysis includes women enrolled between August 21, 1995 and continuing up to those whose estimated due dates were no later than July 10, 1998. Of the 3436 eligible women who attended the clinics during that period, 2038 (59.3%) were recruited successfully, and 1853 were interviewed at approximately 28 weeks’ gestation.
Among the 1853 women, delivery status (term or preterm) was known for 1835 women (99.0%); 218 (11.9%) delivered preterm. After excluding five women who delivered before 29 weeks’ gestation and 26 women who missed the follow-up interview, 187 preterm cases were left for analysis. Follow-up interviews occurred 2–75 days after delivery (median 10 days).
Among the 1853 women interviewed, 469 were selected randomly as potential controls. After excluding 21 women who delivered preterm before the scheduled (control) follow-up interview (and who were thus included as cases) and 39 women who missed the follow-up interview for other reasons, 409 controls remained for analysis (mean gestational age at delivery, 39.2 weeks). Thirty of the 409 controls who were interviewed at 29–36 weeks delivered before 37 weeks and thus became preterm cases. These women have been included as controls in the appropriate period (corresponding to the week of their still-pregnant follow-up interview) and as cases in the appropriate period (corresponding to their week of delivery). Of 469 women originally selected as controls, 356 delivered at term and had after-delivery interviews, thus providing information about sexual activity for the 37–43-week period. When analyses were conducted after excluding from the control group all women who delivered preterm, changes in beta coefficients and odds ratios (OR) were small (generally less than 5%) and changes in standard errors were negligible.
Sexual activity was assessed at several points during pregnancy. The 28-week interview included questions on frequency of sexual intercourse, use of the male superior position during intercourse, orgasm, interest in sex, most recent date of intercourse, and medical advice received about bed rest. Later interviews obtained information from cases and controls about occurrences of sexual intercourse in the past 2 weeks, use of the male superior position, orgasm, most recent dates of intercourse and orgasm, interest in sex, medical advice received about sexual activity in pregnancy, and medical advice received about bed rest. A question about reasons for any decrease in sexual activity was added after the study began, but those data were not available for the complete sample. Follow-up interviews after delivery specifically asked about sexual activity before delivery. At the end of each interview, interviewers rated respondents in terms of cooperation, quality of responses, and recall.
For cases and controls, three measures were constructed to describe sexual activity during the 29–36-week period (ie, relative to the date of delivery for preterm cases, and relative to the date of the 29–36-week interview for controls): intercourse in the past 2 weeks (any or none), time since last intercourse (fewer than 7 days compared with 7 or more days), and orgasm in the past 2 weeks (any or none). Similar variables were constructed for controls who delivered at term to describe sexual activity relative to the 37–43-week period.
Conditional logistic regression models, which took into account the frequency matching of cases and controls by gestational age, were used to calculate ORs and 95% confidence intervals (CI). Among women who reported intercourse in the 2 weeks before the 29–36-week interview (for controls) or before delivery (for cases), the association between preterm delivery and use of the male superior position (any or none) was examined. Odds ratios were calculated among subgroups of women defined by week of delivery (cases) or interview (controls), by preterm delivery subtype (PROM, preterm labor, or medical induction before 37 weeks), and by risk status for preterm delivery. A woman was considered to be at higher risk if she had self-reported poor health during the 29–36-week period, previous miscarriage, stillbirth, or preterm birth, report of medical advice regarding bed rest, or report of medical advice to restrict intercourse or orgasm during pregnancy. Three-quarters of cases and 52% of controls met one of those criteria.
Potential confounders were known or suspected risk factors for preterm delivery, including demographic characteristics (age, race, education, marital status), reproductive variables (parity, gravidity, previous adverse pregnancy outcome, vaginal bleeding), and health behaviors (cigarette smoking, alcohol use, or illicit drug use). These covariates as well as self-reported and Gram stain–diagnosed bacterial vaginosis were evaluated as potential effect modifiers.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). Cases differed from controls in clinic site, race, marital status, parity, and history of preterm delivery.
|
). Cases and controls reported similar frequencies of intercourse before pregnancy and during the first trimester. Cases reported somewhat less frequent intercourse than controls during the month before the main interview (approximately the sixth month of pregnancy) and even less frequent intercourse during the 29–36-week period. During each period between 29 and 36 weeks, a lower proportion of cases than controls reported having any intercourse in the previous 2 weeks (38% compared with 55% for 29–32 weeks, 30% compared with 62% for 33–34 weeks, 24% compared with 52% for 35–36 weeks).
|
More cases than controls reported reduced interest in sex during late pregnancy (71% compared with 57%) compared with before pregnancy. Cases were more likely than controls to report having received medical advice directly or indirectly related to sexual activity during pregnancy, for example, being told at any time during pregnancy by a doctor or nurse to stay at home in bed rather than follow their regular schedule (64% compared with 23%), being told anything by a doctor or nurse regarding sex in pregnancy (41% compared with 23%) or being told by a doctor or nurse to stop or limit intercourse or orgasm during pregnancy (32% compared with 12%). Of women asked the reasons for their decreased frequency of intercourse in the 29–36-week period, a higher proportion of cases than controls reported medical reasons (eg, bed rest advised, admitted to hospital, had surgery). A higher proportion of cases than controls reported fair or poor health during late pregnancy (19% compared with 13%).
Frequency of intercourse during the 6 months before pregnancy and during the first trimester was not associated with preterm delivery (Table 2). For the three measures of sexual activity in late pregnancy, the unadjusted OR for preterm delivery anytime between 29 and 36 weeks ranged from 0.25 to 0.38. Odds ratios were similar, subject to some imprecision, for the periods of 29–32 weeks (OR 0.4–0.5), 33–34 weeks (OR 0.1–0.3), and 35–36 weeks (OR 0.3–0.5), indicating that the association was not modified by week of gestation. Among women who were sexually active during late pregnancy, the OR for preterm delivery with use of the male superior position was 1.38 (95% CI 0.74, 2.56). Among women who reported no orgasm during late pregnancy, the OR for intercourse and preterm delivery was 0.34 (95% CI 0.19, 0.59). Among women who reported not having intercourse during late pregnancy, the OR for orgasm and preterm delivery was 0.25 (95% CI 0.07, 0.89).
|
The risk of preterm delivery associated with sexual activity during late pregnancy was even lower among women who were married or living with a partner (Table 3). Results did not differ substantially according to presence or absence of bacterial vaginosis at 28 weeks. When preterm deliveries were divided by subtype, ORs for frequency of intercourse and orgasm were slightly lower for medically induced preterm deliveries than for preterm labor or PROM. Among women who were at lower risk for preterm delivery, ORs for sexual activity during the 29–36-week period and preterm delivery were closer to 1 than for women who were at higher risk.
|
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
When preterm cases and term controls were compared during the time before delivery, the overall decrease in frequency of intercourse during pregnancy confuses the picture. Controls were probably less sexually active than they would have been earlier in gestation (when the preterm cases delivered), thus leading to an upward bias in the estimates of the association between sexual activity and risk of delivering preterm. Most studies making this incorrect comparison reported either no association between sexual activity and preterm delivery or a risk of preterm delivery associated with more frequent sexual activity.2,4,13,14 When we reanalyzed our data to compare sexual activity at the end of pregnancy between preterm cases and term controls we obtained higher ORs (in the range of 0.4–0.6), but sexual activity was still associated with reduced risk of preterm delivery.
We found a small increase in risk associated with use of the male superior position during intercourse. Ekwo et al13 reported an association of the male superior position with preterm PROM (OR 2.4, 95% CI 1.2, 5.0) and with preterm delivery without PROM (OR 1.8, 95% CI 1.0, 3.2). However, their findings could have been biased by use of noncomparable gestational ages for women delivering preterm and those delivering at term. Not only the frequency of intercourse but also the use of the male superior position has been reported to decline throughout pregnancy.8
Most studies that evaluated sexual activity focused on intercourse. Some have considered effects of orgasm but usually only in the context of intercourse. In nonpregnant women, orgasms resulting from masturbation are associated with stronger uterine contractions than orgasms resulting from intercourse.21 We found that orgasm during late pregnancy was associated with reduced risk of preterm delivery regardless of whether women reported having intercourse during the same period. Two previous studies have reported that orgasm with last coitus was associated with a reduced risk of preterm delivery.14,20 Another study found reduced risks of preterm delivery associated with orgasm with or without intercourse.3 In contrast, Goodlin et al4 reported that twice as many cases as controls had orgasm after 32 weeks. Ekwo et al13 reported that orgasm was associated with preterm PROM (OR 1.9, 95% CI 1.0, 3.9), but that study compared orgasm in the 4 weeks before delivery for preterm cases and term controls, which likely overestimated any association because frequency of orgasm would likely decline with advancing gestation.
Infection might modify the effect of sexual activity on risk of preterm delivery. We were able to consider only the effect of bacterial vaginosis because no other infection was present with sufficient frequency. We found no substantial differences in the association between sexual activity and preterm delivery according to bacterial vaginosis status, whether measured by self report or Gram-stain diagnosis at 28 weeks. These results contrast with those of the Vaginal Infections and Prematurity Study, in which frequent intercourse among women with bacterial vaginosis was associated with increased risk of preterm delivery, whereas among women without bacterial vaginosis, frequent intercourse was associated with a reduced risk of preterm delivery.7
Our results could indicate a protective effect of sexual activity, possibly mediated through a mechanism involving social support. However, another explanation could be that women who were at risk of preterm delivery or who otherwise felt ill intentionally reduced their sexual activity in late pregnancy. There is evidence for this in our data. For example, cases were more likely than controls during late pregnancy to report poorer health, to give medically related reasons for reducing sexual activity, to be less interested in sex during late pregnancy, and to report receiving medical advice related to stopping or limiting sexual activity during pregnancy.
To isolate a group of women who were less likely to reduce their sexual activity for reasons related to medical risks and therefore who might be more useful for isolating an effect of sexual activity per se, we conducted separate analyses among women at lower risk of preterm delivery. Among these women, the inverse association between sexual intercourse and preterm delivery persisted but was weaker, and there was no inverse association with orgasm (although confidence limits were wide) suggesting that part of the inverse association among all women was likely driven by women at higher risk who reduced their sexual activity in late pregnancy.
We attempted to minimize the potential for misclassification of sexual activity by asking women about recent activity over a fairly short interval.22,23 In a validation substudy, agreement between prospective daily records and interviews for any intercourse in the past 2 weeks was 82–90% (unpublished data). There is additional evidence from other reliability and validity studies that the reporting of sexual behavior is fairly accurate.22–24
Because cases were interviewed after delivery, whereas controls were still pregnant, recall bias could have distorted the association between sexual activity and preterm birth. Although there were not enough preterm cases in our validation substudy to examine this possibility directly, other evidence in the data suggests that there was not substantial systematic underreporting by cases of behaviors that might be considered undesirable during pregnancy. Odds ratios for the effects of cigarette smoking, alcohol, and illicit drug use during the 29–36-week period were consistent with previously reported results.25 Additionally, results were nearly identical when analyses were restricted to cases and controls who were rated by interviewers in the highest categories of cooperation, quality of responses, and recall (which are possible indicators of data quality). When analyses were restricted to women who had been interviewed within 10 days of delivery ORs for sexual activity changed only slightly, suggesting that the length of time since delivery did not substantially affect accuracy of reporting among cases.
Any uncontrolled variables would have to have had a large confounding effect to fully account for the results. There was little association between preterm delivery and sexual intercourse before or early in pregnancy, suggesting no major differences between groups before pregnancy.
Our data provide evidence against an increased risk of preterm delivery between 29 and 36 weeks associated with sexual activity during late pregnancy. We cannot exclude the possibility that sexual activity might be a risk for a small subgroup of susceptible women. As a whole, however, continued sexual activity during late pregnancy was a strong predictor that a pregnancy will go full term.
|
Footnotes |
|---|
Received March 23, 2000. Received in revised form August 1, 2000. Accepted August 31, 2000.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Pugh WE, Fernandez FL. Coitus in late pregnancy: A follow-up study of the effects of coitus on late pregnancy, delivery, and the puerperium. Obstet Gynecol 1953;2:636–42.
3. Perkins RP. Sexual behavior and response in relation to complications of pregnancy. Am J Obstet Gynecol 1979;134:498–505.[Medline]
4. Goodlin RC, Keller DW, Raffin M. Orgasm during late pregnancy: Possible deleterious effects. Obstet Gynecol 1971;38:916–20.
5. Andersen LF, Fuchs F. Sexual activity and preterm birth. In: Fuchs A, Fuchs F, Stubblefield PG, eds. Preterm birth: Causes, prevention, and management. New York: McGraw-Hill, 1993:161–72.
6. Naeye RL, Ross S. Coitus and chorioamnionitis: A prospective study. Early Hum Dev 1982;6:91–7.[Medline]
7. Read JS, Klebanoff MA. Sexual intercourse during pregnancy and preterm delivery: Effects of vaginal microorganisms. Am J Obstet Gynecol 1993;168:514–9.[Medline]
8. Solberg DA, Butler J, Wagner NN. Sexual behavior in pregnancy. N Engl J Med 1973;288:1098–103.
9. Wagner NN, Butler JC, Sanders JP. Prematurity and orgasmic coitus during pregnancy: Data on a small sample. Fertil Steril 1976;27:911–5.[Medline]
10. Neilson JP, Mutambira M. Coitus, twin pregnancy, and preterm labor. Am J Obstet Gynecol 1989;160:416–8.[Medline]
11. Mills JL, Harlap S, Harley EE. Should coitus late in pregnancy be discouraged? Lancet 1981;2:136–8.[Medline]
12. Klebanoff MA, Nugent RP, Rhoads GG. Coitus during pregnancy: Is it safe? Lancet 1984;2:914–7.[Medline]
13. Ekwo EE, Gosselink CA, Woolson R, Moawad A, Long CR. Coitus late in pregnancy: Risk of preterm rupture of amniotic sac membranes. Am J Obstet Gynecol 1993;168:22–31.[Medline]
14. Georgakopoulos PA, Dodos D, Mechleris D. Sexuality in pregnancy and premature labour. Br J Obstet Gynaecol 1984;91:891–3.[Medline]
15. Morris NM. The frequency of sexual intercourse during pregnancy. Arch Sex Behav 1975;4:501–7.[Medline]
16. Reamy K, White SE, Daniell WC, Le Vine ES. Sexuality and pregnancy: A prospective study. J Reprod Med 1982;27:321–7.[Medline]
17. Savitz DA, Dole N, Williams J, Thorp JM, McDonald T, Carter AC, et al. Determinants of participation in an epidemiological study of preterm delivery. Paediatr Perinat Epidemiol 1999;13:114–25.[Medline]
18. Harger JH, Hsing AW, Tuomala RE, Gibbs RS, Mead PB, Eschenbach DA, et al. Risk factors for preterm premature rupture of fetal membranes: A multicenter case-control study. Am J Obstet Gynecol 1990;163:130–7.[Medline]
19. Kurki T, Ylikorkala O. Coitus during pregnancy is not related to bacterial vaginosis or preterm birth. Am J Obstet Gynecol 1993; 169:1130–4.[Medline]
20. Rayburn WF, Wilson EA. Coital activity and premature delivery.Am J Obstet Gynecol 1980;137:972–4.[Medline]
21. Masters WH, Johnson VE. Human sexual response. Boston: Little, Brown, and Company, 1966.
22. Catania JA, Gibson DR, Marin B, Coates TJ, Greenblatt RM.Response bias in assessing sexual behaviors relevant to HIV transmission. Eval Program Plann 1990;13:19–29.
23. Kauth MR, St. Lawrence JS, Kelly JA. Reliability of retrospective assessments of sexual HIV risk behavior: A comparison of biweekly, three-month, and twelve-month self-reports. AIDS Educ Prev 1991;3:207–14.[Medline]
24. Hornsby PP, Wilcox AJ. Validity of questionnaire information on frequency of coitus. Am J Epidemiol 1989;130:94–9.
25. Savitz DA, Pastore LM. Causes of prematurity. In: McCormick M, ed. Effectiveness of prenatal care. New York: Cambridge University Press, 1999:63–104.
This article has been cited by other articles:
|
|
 |
|
  P. C. Tan, A. Andi, N. Azmi, and M. N. Noraihan Effect of coitus at term on length of gestation, induction of labor, and mode of delivery. Obstet. Gynecol., July 1, 2006; 108(1): 134 - 140. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Schaffir Sexual intercourse at term and onset of labor. Obstet. Gynecol., June 1, 2006; 107(6): 1310 - 1314. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A Savitz, N. Dole, and A. H Herring Methodologic issues in the design and analysis of epidemiologic studies of pregnancy outcome Statistical Methods in Medical Research, April 1, 2006; 15(2): 93 - 102. [Abstract] [PDF]
|
|
|
|
 |
|
 C. E. Adair, L. Kowalsky, H. Quon, D. Ma, J. Stoffman, A. McGeer, S. Robertson, M. Mucenski, and H. D. Davies Risk factors for early-onset group B streptococcal disease in neonates: a population-based case-control study Can. Med. Assoc. J., August 5, 2003; 169(3): 198 - 203. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. C. Goodlin, A. E. Sayle, D. A. Savitz, and J. M. Thorp Jr SEXUAL ACTIVITY DURING LATE PREGNANCY AND RISK OF PRETERM DELIVERY Obstet. Gynecol., June 1, 2001; 97(6): 1025 - 1025. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
