Women who engage in sexual activity weekly or monthly have a lower risk of entering menopause early relative to those who report having some form of sex less than monthly, according to a new UCL study.
The researchers observed that women, who reported engaging in sexual activity weekly, were 28% less likely to have experienced menopause at any given age than women who engaged in sexual activity less than monthly.
Sexual activity includes sexual intercourse, oral sex, sexual touching and caressing or self-stimulation.
The research, published in Royal Society Open Science, is based on data from the USA’s Study of Women’s Health Across the Nation (SWAN). It’s the largest, most diverse and most representative longitudinal cohort study available to research aspects of the menopause transition.
First author on the study, PhD candidate Megan Arnot (UCL Anthropology), said: “The findings of our study suggest that if a woman is not having sex, and there is no chance of pregnancy, then the body ‘chooses’ not to invest in ovulation, as it would be pointless.
There may be a biological energetic trade-off between investing energy into ovulation and investing elsewhere, such as keeping active by looking after grandchildren.
“The idea that women cease fertility in order to invest more time in their family is known as the Grandmother Hypothesis, which predicts that the menopause originally evolved in humans to reduce reproductive conflict between different generations of females, and allow women to increase their inclusive fitness through investing in their grandchildren.”
During ovulation, the woman’s immune function is impaired, making the body more susceptible to disease. Given a pregnancy is unlikely due to a lack of sexual activity, then it would not be beneficial to allocate energy to a costly process, especially if there is the option to invest resources into existing kin.
The research is based on data collected from 2,936 women, recruited as the baseline cohort for the SWAN study in 1996/1997.
The mean age at first interview was 45 years old. Non-Hispanic Caucasian women were most represented in the sample (48%), and the majority of women were educated to above a high school level. On average they had two children, were mostly married or in a relationship (78%), and living with their partner (68%).
The women were asked to respond to several questions, including whether they had engaged in sex with their partner in the past six months, the frequency of sex including whether they engaged in sexual intercourse, oral sex, sexual touching or caressing in the last six months and whether they had engaged in self-stimulation in the past six months. The most frequent pattern of sexual activity was weekly (64%).
None of the women had yet entered menopause, but 46% were in early peri-menopause (starting to experience menopause symptoms, such as changes in period cycle and hot flashes) and 54% were pre-menopausal (having regular cycles and showing no symptoms of peri-menopause or menopause).
Interviews were carried out over a ten-year follow-up period, during which 1,324 (45%) of the 2,936 women experienced a natural menopause at an average age of 52.
This provided a likelihood whereby women of any age who had sex weekly were 28% less likely to experience the menopause compared to those who had sex less than monthly.
By modelling the relationship between sexual frequency and the age of natural menopause, women of any age who had sex weekly had a hazard ratio of 0.72, whereas women of any age who had sex monthly had a hazard ratio of 0.81.
This provided a likelihood whereby women of any age who had sex weekly were 28% less likely to experience the menopause compared to those who had sex less than monthly. Likewise, those who had sex monthly were 19% less likely to experience menopause at any given age compared to those who had sex less than monthly.
The researchers controlled for characteristics including oestrogen level, education, BMI, race, smoking habits, age at first occurrence of menstruation, age at first interview and overall health.
The study also tested whether living with a male partner affected menopause as a proxy to test whether exposure to male pheromones delayed menopause. The researchers found no correlation, regardless of whether the male was present in the household or not.
Last author, Professor Ruth Mace (UCL Anthropology), added: “The menopause is, of course, an inevitability for women, and there is no behavioural intervention that will prevent reproductive cessation. Nonetheless, these results are an initial indication that menopause timing may be adaptive in response to the likelihood of becoming pregnant.”
Funding: The research was funded by the Economic and Social Research Council and Biotechnology and Biological Sciences Research Council.
While women often stop reproducing many years prior to the menopause , the permanent reproductive cessation resulting from the menopause means a woman is no longer physically able to increase her direct fitness.
Many lifestyle factors have been found to associate with ANM, and these are seldom discussed from an evolutionary perspective. Here, we have focused specifically on the relationship between marital status and ANM, testing an original hypothesis that sexual frequency acts as a bio-behavioural mediator between ANM and marital status, in addition to the existing hypothesis that married women enter menopause later owing to male pheromones .
To test the latter hypothesis, three measures of male–female cohabitation were used as a proxy of exposure to male pheromones, and we found no evidence to suggest that menopause timing was responsive to living with a male, and therefore possibly male pheromones.
It should be noted that this hypothesis could be fundamentally flawed, as there is no conclusive evidence either that humans produce pheromones, or that they are capable of detecting them . Nonetheless, this is, to our knowledge, the first study addressing the pheromonal hypothesis since it was originally proposed, and while male household presence is merely a proxy of pheromones—and it may be that the hypothesis is moot owing to the absence of evidence for human pheromones—it is an indication that the relationship between marital status and ANM is not capturing the effect of male pheromones on the menstrual cycle.
This study did not replicate findings from previous research that married women enter menopause later. In fact, following complete adjustment, the converse was found, with women who were married or in a relationship having an increased risk of entering menopause compared to divorced, separated and single women.
Conflicting results regarding marital status’ effect on ANM have been found elsewhere (e.g. ), and one reason for this may be the way in which the researcher chooses to code the variable. In this analysis, romantic partnerships that may not have been acknowledged in previous studies owing to having not been formalized by a marriage ceremony (e.g. cohabiting but unmarried) were taken into account.
In addition, some prior studies have not included marital status as time-variant and dichotomized the variable as ‘ever married’ or ‘never married’ (e.g. ). Hence, the responsive way in which this study coded marital status may account for the difference in results.
Another reason for this difference may be the cultural setting of previous studies. For example, research originating from Iran found that ever married women experience a later menopause than those who never married . However, in the case of Iran where dowry is still common practice, it means marriage is contingent upon family wealth .
Therefore, the effect of marital status on ANM would be confounded by a woman’s socioeconomic position, which itself would relate to other aspects of her health and life history that have been associated with menopause timing—such as BMI and age of menarche—therefore resulting in a significant difference in ANM between those who have and have not been married.
Within Iran, sex outside of marriage is prohibited both legally and socially, meaning marital status would be highly correlated with sexual behaviour . Hence, it may be that previous studies identifying married women enter menopause later are simply capturing the effect of health and lifestyle patterns that themselves associate with both marital status and menopause timing, rather than demonstrating that marital status itself is a cross-cultural correlate of ANM. Future research should aim to address the cultural setting in which the data were collected when interpreting the results.
Evidence supporting the notion that ANM associated with sexual frequency during the pre- and peri-menopause was found. Even following complete adjustment, results still indicated that women who engage in sexual activity weekly or monthly have a lower risk of entering menopause relative to those who report having some form of sex less than monthly.
If we interpret these results from a fitness-maximizing framework, it may be the physical cues of sex signal to the body that there is a possibility of becoming pregnant, and therefore an adaptive trade-off may occur between continued energetic investment in ovulation and reproductive cessation.
During ovulation, the woman’s immune function is impaired making the body more susceptible to disease [28,29]. Hence, if a pregnancy is unlikely owing to a lack of sexual activity, then it would not be beneficial to allocate energy to a costly process, especially if there is the option to invest resources into existing kin [30,31].
The idea that women cease fertility in order to invest in kin is known as the Grandmother Hypothesis, which predicts that the menopause originally evolved in humans to reduce reproductive conflict between different generations of females, and allow women to increase their inclusive fitness through investing in their grandchildren [30,69,70].
It may be costly for a woman to cease ovulatory function if the chances of her becoming pregnant are still high. In other words, if she is still able to increase her direct fitness, then it may be better to maintain the function of her menstrual cycle for slightly longer.
It should be noted that there may be a bidirectional relationship between the physical condition of the woman when approaching the menopause and sexual engagement. As oestrogen levels decline, women are more likely to experience vaginal dryness and discomfort, making them less inclined to engage in sex .
This study has attempted to control for this factor through adjusting for both oestradiol levels and the woman’s self-perceived overall health, with the association between sexual frequency and ANM still persisting following this adjustment. This suggests that—even when controlling for the complicated relationship between health, hormonal fluctuations and desire for sex—the menopause may be somewhat facultative in response to sexual behaviour, rather than being solely the result of a physiological constraint (e.g. degrading oocyte quality).
This paper demonstrates that sexual frequency is associated with ANM and is also, to our knowledge, the first formal test of the hypothesis that male pheromones have an influence over ANM. While only a proxy of male pheromones was used, no association between male cohabitation and menopause timing was found, indicating that male–female cohabitation is not the driving force behind the relationship between marital status and menopause timing and that pheromones probably do not influence ANM.
We did not replicate the findings from previous research showing that simply being married is associated with a later ANM, most likely owing to the variable cultural and temporal settings of previous studies. However, we did demonstrate that increased sexual frequency during the pre- and peri-menopause decreased the risk of experiencing menopause. While causation cannot be conclusively inferred, we hypothesize that this relationship is the result of an adaptive trade-off relative to the likelihood of pregnancy when approaching the menopause.
Of course, the menopause is an inevitability for women, and there is no behavioural intervention that will prevent reproductive cessation; nonetheless, these results are an initial indication that menopause timing may be adaptive in response to sexual behaviour.