The implications of changing hormonal contraceptive use after relationship formation
Introduction
Modern-day humans live in an environment that vastly differs from the one in which they evolved. According to the evolutionary mismatch hypothesis (Li, van Vugt, & Colarelli, 2018), such environmental changes have outpaced the time needed for humans' psychological mechanisms to adapt, and this mismatch has important implications for a wide range of human emotions, cognitions, and behaviors (Buss, 2000; Maner & Kenrick, 2010; Tooby & Cosmides, 1990). For example, despite the greater nutritional value of fruit versus refined sugar and even artificial sweeteners, people frequently prefer candy to fruit.
Given the crucial role of reproduction and pair bonding throughout human evolution, selection pressures have weighed particularly heavily on human mating. Thus, environmental mismatch may have important implications for human mating psychology (see Li et al., 2018). Modern humans struggle to maintain their long-term relationships (Finkel, Hui, Carswell, & Larson, 2014), evidenced by the fact that on average relationship satisfaction declines over time (see Meltzer, McNulty, Jackson, & Karney, 2014b) and marital dissolution rates in industrialized countries hover between 30% and 50% (Amato & James, 2010). Mismatches between our evolved and modern environments may partially account for such relationship difficulties.
Particularly notable in regard to long-term romantic relationship success is the modern advancement of hormonal contraceptives (HCs), which approximately 140 million women worldwide use (Mørch et al., 2017). HCs secrete synthetic progestin (and, for some, ethyl estradiol), suppressing natural production of estrogen and progesterone in women who use (versus do not use) HCs. Consequently, HC-using women do not experience natural cyclical variations in estrogen and progesterone [i.e., spikes in estrogen that facilitate ovulation, spikes in progesterone following ovulation; Fleischman, Navarrete, & Fessler, 2010; though see Grøntvedt, Grebe, Kennair, & Gangestad, 2017 for evidence of continued progestogenic effects]. Likewise, recent evidence suggests HCs alter women's brain structures such that women who use (versus do not use) HCs have more grey matter in their prefrontal cortices, pre- and postcentral gyri, and inferior parietal lobules (Pletzer et al., 2010). Due, at least in part, to these physiological changes, HCs have been implicated in a wide variety of psychological processes and outcomes, including but not limited to depression (Skovlund, Mørch, Kessing, & Lidegaard, 2016; Young, Midgley, Carlson, & Brown, 2000), reward processing (Scheele, Plota, Stoffel-Wagner, Maier, & Hurlemann, 2015), and emotion recognition (Hamstra, De Rover, De Rijk, & Van der Does, 2014; Radke & Derntl, 2016).
With respect to their romantic relationships, many women begin or discontinue using HCs numerous times after meeting their partner, thereby regularly altering their natural hormonal profiles and brain structures from those that evolved to those dictated by modern medicine. And there is reason to believe such changes negatively impact women's long-term relationship outcomes (for a similar argument, see Roberts, Cobey, Klapilová, & Havlíček, 2013; Roberts et al., 2014; Russell, McNulty, Baker, & Meltzer, 2014). Drawing from interdependence theory (Kelley & Thibaut, 1978), which is the predominant relationship-science perspective on how people evaluate their relationships, people's satisfaction depends on the extent to which their relationship experiences and outcomes meet their relationship preferences, or their relationship standards. Some evidence suggests women's sex hormones are associated with their sexual preferences such as heightened desires for sex (Gangestad & Thornhill, 2008; Grøntvedt et al., 2017; Roney, 2018; Roney & Simmons, 2013; also see Jones et al., 2018) as well as their partner preferences such as heightened desires for masculine or symmetrical partners (Meltzer, 2017; Pisanski et al., 2014; Thornhill, Chapman, & Gangestad, 2013; for review, see Alvergne & Lummaa, 2010; for an example of a study that did not detect an association between sex hormones and partner preferences, see Jones, Hahn, & DeBruine, 2019). Thus, beginning or discontinuing HCs after relationship formation may alter women's sexual and relationship preferences. To the extent that women's sexual preferences change when they begin or discontinue using HCs (relative to use at relationship formation) and thus their sexual relationships no longer meet these altered preferences (because they presumably meet their pre-altered preferences), we might expect such women to experience declines in sexual satisfaction. Likewise, to the extent that women's partner and relationship preferences change when they begin or discontinue using HCs (relative to use at relationship formation) and thus their partners and relationships no longer meet these altered preferences, we might expect such women to experience declines in relationship satisfaction.
Consistent with this rationale, the HC congruency hypothesis (Roberts et al., 2013) posits that changes in women's HC use relative to relationship formation can negatively impact their subsequent sexual and relationship satisfaction (see Roberts et al., 2014; Russell et al., 2014). The core tenet underlying this hypothesis is that a partnered woman should be less satisfied at times when her HC use is incongruent with her HC use at relationship formation relative to times when her own HC use is congruent with her HC use at relationship formation. That is, HC incongruency versus congruency is inherently a within-person process. Accordingly, one of the strongest tests of the HC congruency hypothesis would utilize a longitudinal design that repeatedly assesses women's HC use and their relationship outcomes. Such a longitudinal design would (a) allow the crucial within-person variance in HC incongruency that is likely associated with changes in women's sexual and relationship satisfaction to be statistically isolated from the between-person variance in HC incongruency in order to demonstrate that any associations are indeed driven by within- versus between-person variance, (b) increase power, and (c) help rule out between-person confounds.
Although some recent empirical work has provided preliminary support for the HC congruency hypothesis (Roberts et al., 2014; Russell et al., 2014; c.f. Jern et al., 2018), none of this work utilized a longitudinal design to statistically isolate the within-person variability in women's HC incongruency. Indeed, three independent studies published in two independent articles (Roberts et al., 2014; Russell et al., 2014) demonstrated that those women whose current HC use was incongruent with their use at relationship formation reported lower sexual satisfaction than those women whose current HC use was congruent with their use at relationship formation [although Russell et al., 2014 used a longitudinal design, their analyses collapsed across within- and between-person variance]. Two of these three studies additionally demonstrated that HC incongruency was associated with lower relationship satisfaction—but only among women whose partners had relatively less attractive faces (Russell et al., 2014). Given the self-selective nature of HC use, however, it is possible that these previously demonstrated effects were due to unmeasured or unknown between-person differences among HC-incongruent versus HC-congruent women rather than within-person changes in HC incongruency. For example, less (versus more) educated women change their HC use more frequently (Frost, Singh, & Finer, 2007) and thus prior HC-incongruency effects could alternatively be attributed to greater financial stress—a correlate of relatively lower education (Dakin & Wampler, 2008). As noted, utilizing a longitudinal design that isolates the within-person variance from the between-person variance would help to rule out such unmeasured or unknown between-person differences. Thus, the primary aim of the current study was to use data drawn from two independent longitudinal studies to examine the association between within-person variance in women's HC incongruency and their sexual and relationship satisfaction, controlling for the between-person variance in HC incongruency.
As previously noted, interdependence theory (Kelley & Thibaut, 1978) posits that women will be most relationally satisfied to the extent that their partners meet their preferences. Given some work suggests women's partner preferences may be linked to their sex hormones and brain structures (e.g., Cooper, Dunne, Furey, & O'Doherty, 2012; Meltzer, 2017; Pisanski et al., 2014; Thornhill et al., 2013; for a review, see Alvergne & Lummaa, 2010; c.f. Jones et al., 2019), we might expect any corresponding changes in relationship evaluations to depend on the extent to which their partners continue to meet their preferences. Although prior research in this area has explored moderation by partner facial attractiveness, theory and supporting evidence regarding the role of partner facial attractiveness provide competing predictions. On the one hand, partner attractiveness serves as a cue of genetic fitness and thus women should benefit (and be satisfied) to the extent that their partners are physically attractive—regardless of changes in their preferences for partner attractiveness. Consistent with this rationale, women value partner attractiveness (Eastwick & Finkel, 2008; Fletcher, Simpson, Thomas, & Giles, 1999). We thus might expect that women with particularly attractive partners are less susceptible to declines in satisfaction following changes in their HC use; data from two independent studies are consistent with this idea (see Russell et al., 2014). On the other hand, however, other research demonstrates partner attractiveness is unassociated or even negatively associated with women's satisfaction—at least in long-term relationships (French, Altgelt, & Meltzer, 2019; Meltzer et al., 2014b), and another line of research suggests partner preferences for attractiveness are not linked to women's sex hormones (Jones et al., 2019). We thus might alternatively expect partner facial attractiveness to play an insignificant role in the association between within-person changes in HC use (relative to use at relationship formation) and women's relationship evaluations. In light of these competing predictions, we did not make clear a priori predictions regarding the role of partner attractiveness. Nevertheless, we had the data available and thus, as a secondary aim, explored whether the associations between women's HC incongruency and their sexual and relationship satisfaction depended on their partners' facial attractiveness.
According to the HC congruency hypothesis, women are more likely to experience poorer relationship outcomes at times when their HC use is incongruent (versus congruent) with their use at relationship formation—regardless of the direction of such incongruency (i.e., beginning versus discontinuing HCs). Whereas some of the preliminary work testing this hypothesis supports the notion that the direction of HC incongruency is inconsequential (Roberts et al., 2014), other work has demonstrated that women who form their relationships while using (versus not using) HCs are most at risk for declines in satisfaction (Russell et al., 2014). As others have argued (Jern et al., 2018), these studies may have produced inconsistent or even spurious results when testing such directional effects because they were comprised of disproportionately sized comparison groups (i.e., HC-incongruent women who began versus discontinued using HCs). To the extent that the current study has more balanced groups, we additionally sought to explore whether the associations between women's HC incongruency and their sexual and relationship satisfaction depend on whether women began versus discontinued using HCs. Given the prior mixed evidence, we did not make a priori directional predictions.
We used two independent, longitudinal studies of newlywed couples and multilevel modeling to provide what we believe is the strongest test to date of the HC congruency hypothesis. In both studies, we assessed wives' HC incongruency, sexual satisfaction, and marital satisfaction as well as husbands' objective facial attractiveness within the first few months of marriage. We then reassessed wives' HC incongruency, sexual satisfaction, and marital satisfaction every few months across the early years of marriage. We predicted that within-person variance in wives' HC incongruency would be negatively associated with their sexual and marital satisfaction such that a given wife would be less satisfied when her HC use was incongruent (versus congruent) with her use at relationship formation. As previously noted, altered partner preferences may be one mechanism through which HC incongruency might impact relationship outcomes (also see Russell et al., 2014); we thus additionally explored this possibility. Moreover, we additionally explored whether the direction of HC incongruency mattered. Given the parallel designs of both studies, we describe them simultaneously and analyze them together to increase power, though we controlled for idiosyncratic differences across studies.
Section snippets
Participants
The participants in Study 1 were 109 heterosexual women drawn from a broader 4-year longitudinal study of 113 newlywed couples in Dallas, Texas, U.S.A (four women failed to indicate their HC use at relationship formation and thus could not be included in the primary analyses). The participants in Study 2 were 94 heterosexual women drawn from a broader 2-year longitudinal study of 104 newlywed couples in Tallahassee, Florida, U.S.A. (we a priori excluded five same-sex female couples based on the
Descriptive statistics and preliminary analyses
Across both studies, 36.0% (n = 73) of wives became HC incongruent at some point during the study (40 of these wives were not using HCs at relationship formation whereas 33 of these wives were using HCs at relationship formation); of the 64.0% (n = 130) of wives who remained HC congruent, 40 wives were consistent HC users (i.e., they reported using HCs at relationship formation and at each assessment) whereas 90 wives were consistent non-HC users (i.e., they reported not using HCs at
Study rationale and summary of results
Given the evolutionary importance of reproduction and pair bonding throughout human history, the mismatch between evolved and modern environments may have important implications for human mating psychology (Li et al., 2018). Relevant to the current study, the modern advancement of HCs may help explain, at least in part, modern relationship difficulties (see Finkel et al., 2014) such as reduced sexual and relationship satisfaction when women's current HC use is incongruent with their use at
Conclusion
Across the course of their long-term romantic relationships, women begin and discontinue using HCs numerous times, thereby regularly altering their hormonal profiles and brain structures from those that evolved to those dictated by modern medicine. As the current research highlights, such changes may be detrimental for women's relationships—particularly for their sexual relationships. Given the importance of long-term romantic relationships to women's overall psychological and physical
Data availability
The data associated with this research are available at https://osf.io/v2dqc/.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of Competing Interest
None.
References (57)
- et al.
Does the contraceptive pill alter mate choice in humans?
Trends in Ecology & Evolution
(2010) - et al.
Random effects structure for confirmatory hypothesis testing: Keep it maximal
Journal of Memory and Language
(2013) - et al.
Hormonal systems, human social bonding, and affiliation
Hormones and Behavior
(2017) - et al.
Estrogenic and progestogenic effects of hormonal contraceptives in relation to sexual behavior: Insights into extended sexuality
Evolution and Human Behavior
(2017) - et al.
Oral contraceptives may alter the detection of emotions in facial expressions
European Neuropsychopharmacology
(2014) - et al.
A high-powered replication study finds no effect of starting or stopping hormonal contraceptive use on relationship quality
Evolution and Human Behavior
(2018) - et al.
Ovulation, sex hormones, and women’s mating psychology
Trends in Cognitive Sciences
(2019) - et al.
Balancing type I error and power in linear mixed models
Journal of Memory and Language
(2017) - et al.
Changes in salivary estradiol predict changes in women’s preferences for vocal masculinity
Hormones and Behavior
(2014) - et al.
Menstrual cycle and hormonal contraceptive use modulate human brain structure
Brain Research
(2010)